problem gambling · problem gambling in people seeking treatment for mental illness lubman et al....

responsiblegambling.vic.gov.au

CONNECT WITH US ON:vicrgf

responsiblegambling

Victorian Responsible Gambling Foundation

ResponsibleGambling


July 2017

Problem gambling in people seeking treatment for mental illness

Problem gam

bling in people seeking treatment for m

ental illness


RESEARCH REPORT

July 2017

RESEARCH REPORT

vicrgf

responsiblegambling


ResponsibleGambling

Our vision: A Victoria free from gambling-related harm

© Victorian Responsible Gambling Foundation, July 2017

This publication is licensed under a Creative Commons Attribution 3.0 Australia licence. The licence does not apply to any images, photographs, branding or logos.

This report has been peer reviewed by two independent researchers. For further information on the foundation’s review process of research reports, please see responsiblegambling.vic.gov.au.

This project was funded by the Victorian Responsible Gambling Foundation through the Grants for Gambling Research Program.

For information on the Victorian Responsible Gambling Foundation Research Program visit responsiblegambling.vic.gov.au.

DisclaimerThe opinions, findings and proposals contained in this report represent the views of the authors and do not necessarily represent the attitudes or opinions of the Victorian Responsible Gambling Foundation or the State of Victoria. No warranty is given as to the accuracy of the information. The Victorian Responsible Gambling Foundation specifically excludes any liability for any error or inaccuracy in, or omissions from, this document and any loss or damage that you or any other person may suffer.

Conflict of interest declarationThe authors declare no conflict of interest in relation to this report or project.

To cite this reportLubman, D, Manning, V, Dowling, N, Rodda, S, Lee, S, Garde, E, Merkouris, S & Volberg, R 2017, Problem gambling in people seeking treatment for mental illness, Victorian Responsible Gambling Foundation, Melbourne.

EnquiriesRosa Billi +61 3 9452 2625 [email protected]

Victorian Responsible Gambling FoundationLevel 6, 14–20 Blackwood StreetNorth MelbourneVictoria 3051

PO Box 2156Royal Melbourne HospitalVictoria 3050

Telephone: +61 3 9452 2600Facsimile: +61 3 9452 2660ABN: 72 253 301 291

See Inside gambling for the latest information, evidence and expert opinion on gambling issues

in Australia and overseas.

From betting on the races with Dad to professional punter: my journey

Teenage gambling: a business that’s only getting riskier

Helping people in their own language

Five minutes with...

Community educator, David Clark

Online magazine

insidegambling.com.au

Hear from:• public health experts • gambling sector professionals • people with firsthand experience

of gambling problems

Let us know what you’d like to read

If you have ideas for themes or stories for Inside gambling, we would love to hear from you.

Go to: insidegambling.com.au/contact-us

Subscribe to Inside gambling

Problem gambling in

people seeking treatment

for mental illness

Dan Lubman

Victoria Manning

Nicki Dowling

Simone Rodda

Stuart Lee

Erin Garde

Stephanie Merkouris

Rachel Volberg

July 2017

Problem gambling in people seeking treatment for mental illness Lubman et al. 2017

Victorian Responsible Gambling Foundation Page ii

Acknowledgements

The authors wish to thank Professor Jayashri Kulkarni for her support and expert advice, as well as the

many researchers who assisted with project administration, data collection and analysis, including Fiona

Barker, Ramez Batish, Tomas Cartmill, Nyssa Fergusson, Gabriella Flaks, Mollie Flood, Erin Garde,

Joshua Garfield, Andrew Larner, Mathan Maglan, Janette Mugavin, Annabeth Simpson, Laura Gorrie,

Pinar Thorn, Christopher Greenwood and Erin Oldenhoph.

The authors are extremely grateful to the clinicians, team leaders, practice managers, support workers,

consumer representatives and most of all the patients attending mental health services for their support

and participation in the study.

Finally, the authors would like to express gratitude to the invaluable contribution of the clinical reference

panel, which included consumers, clinicians and service managers across the mental health and

gambling sectors, Victorian Responsible Gambling Foundation senior staff and Department of Health

representatives, as well as Jan Rice and Jane Evans from Eastern Health for their assistance in

facilitating the workshop.

Funding

This project was funded by the Victorian Responsible Gambling Foundation.


Victorian Responsible Gambling Foundation Page iii

Table of contents

Acknowledgements ..................................................................................................................................... ii

Table of contents ........................................................................................................................................ iii

List of tables ................................................................................................................................................ v

List of figures ............................................................................................................................................ viii

Key terms .................................................................................................................................................... xi

Executive summary ..................................................................................................................................... 1

Purpose of the report .................................................................................................................................. 1

Conclusion ................................................................................................................................................ 12

Background ................................................................................................................................................ 17

Overall aims and activities ........................................................................................................................ 19

Study 1: Literature review ........................................................................................................................ 21

Introduction ............................................................................................................................................... 21

Section 1: Background .............................................................................................................................. 21

Section 2: Problem gambling and mental health disorders ...................................................................... 26

Section 3: Screening and assessment of problem gambling within primary care, AOD and mental

health services .......................................................................................................................................... 33

Section 4: Treatment of problem gambling in primary care, AOD and mental health settings ................. 53

Section 5: Effective partnerships between gambling and primary care, AOD and mental health

services ..................................................................................................................................................... 59

Section 6: Overall conclusion of literature review ..................................................................................... 69

Study 2a: Clinician survey: Current practice and responses to gambling ......................................... 71

Aims .......................................................................................................................................................... 71

Method ...................................................................................................................................................... 71

Results ...................................................................................................................................................... 74

Discussion ................................................................................................................................................. 90

Study 2b: Barriers and facilitators to responding to problem gambling ............................................. 93

Aims .......................................................................................................................................................... 93

Method ...................................................................................................................................................... 93

Results ...................................................................................................................................................... 94

Discussion ............................................................................................................................................... 116

Study 3a: Patient survey: Gambling behaviours and prevalence of problem gambling ................. 120

Aims ........................................................................................................................................................ 120

Method .................................................................................................................................................... 120

Results .................................................................................................................................................... 125


Victorian Responsible Gambling Foundation Page iv

Discussion ...............................................................................................................................................161

Study 3b: Psychometric testing of gambling screens .........................................................................166

Aims ........................................................................................................................................................166

Method ....................................................................................................................................................166

Results ....................................................................................................................................................171

Discussion ...............................................................................................................................................176

Discussion and recommendations ........................................................................................................180

Clinical reference panel...........................................................................................................................182

Conclusion ..............................................................................................................................................183

References ...............................................................................................................................................189

Appendix 1: Study 2a: Clinician survey ................................................................................................201

Appendix 2: Study 2b: Detailed data tables of clinician survey results ............................................205

Appendix 3: Study 3a: Patient survey ...................................................................................................210

Appendix 4: Study 3a: Detailed data tables of patient survey results ...............................................227


Victorian Responsible Gambling Foundation Page v

List of tables

Table 1. Recent Australian national surveys: Prevalence rates by gambling risk category .................. 22

Table 2. Prevalence of comorbid mental health disorders in community-representative samples of

people with gambling problems ............................................................................................................. 26

Table 3. Prevalence of comorbid current psychiatric disorders in people with gambling problems

seeking treatment .................................................................................................................................. 27

Table 4. Prevalence of comorbid current personality disorders in people with gambling problems

seeking treatment .................................................................................................................................. 28

Table 5. Prevalence of problem gambling in mental health populations ............................................... 29

Table 6: Summary of the brief screening instruments for problem gambling ........................................ 37

Table 7. Randomised trials evaluating interventions for comorbid drug use/mental health disorders

and problem gambling ........................................................................................................................... 58

Table 8. Structure of the clinicians' survey ............................................................................................ 73

Table 9. Demographic characteristics of the clinicians who participated in the survey ........................ 74

Table 10. Clinicians' responses on the current knowledge about gambling and mental illness

survey .................................................................................................................................................... 78

Table 11. Key themes arising from the reasons clinicians gave for not screening for problem

gambling ................................................................................................................................................ 80

Table 12. Clinicians' current attitudes towards problem gambling ........................................................ 84

Table 13. Summary of barriers and facilitators to screening for problem gambling .............................. 94

Table 14. Proportion of patients in each of the gambling categories .................................................. 143

Table 15. Odds of gambling harm (any) gambling among gamblers with current mental health

diagnoses ............................................................................................................................................ 148

Table 16. Number of mental health diagnoses and proportion experiencing gambling harm ............. 149

Table 17. Participation in specific gambling activity by gambling category ......................................... 154

Table 18. Types of services gamblers in each risk category report they would be likely to use if

they develop a gambling problem ....................................................................................................... 160

Table 19 Description of how indices of classification accuracy are calculated. .................................. 167

Table 20. Single items employed as one-item screening instruments ................................................ 169

Table 21. Classification accuracy of the single items and brief screening instruments using PGSI

score of 8+ as reference standard ....................................................................................................... 172





Table 24. Mental health illnesses clinicians commonly observe comorbid problem gambling ........... 205

Table 25. Frequency clinicians ask patients about gambling, and screen for problem gambling ....... 205


Victorian Responsible Gambling Foundation Page vi

Table 26. Methods or instruments clinicians use to screen for problem gambling ............................. 205

Table 27. Clinicians' level of comfort in asking patients about gambling behaviour .......................... 206

Table 28. Clinicians' level of confidence in detecting, screening and assessing patients for problem

gambling ............................................................................................................................................. 206

Table 29. Clinicians' current responses when patients experiencing problem gambling are

identified.............................................................................................................................................. 206

Table 30. Frequency clinicians report referring patients experiencing problem gambling to outside

services ............................................................................................................................................... 206

Table 31. Services to which clinicians report referring patients experiencing problem gambling ...... 207

Table 32. Clinicians' level of confidence in referring patients to outside services for problem

gambling ............................................................................................................................................. 207

Table 33. Clinicians' opinions about referring patients to outside agencies when problem gambling

is identified .......................................................................................................................................... 207

Table 34. Clinicians' opinion about gambling help services being equipped to deal with patients

with mental health illnesses ................................................................................................................ 208

Table 35. Frequency clinicians report treating patients for problem gambling ................................... 208

Table 36. Type of treatment provided by clinicians and/or their service or practice .......................... 208

Table 37. Clinicians' level of confidence in treating patients for problem gambling ........................... 209

Table 38. Clinicians' understanding of the external Gambler’s Help services and programs ............ 209

Table 39. Clinicians' understanding of the types of treatments effective for problem gambling ........ 209

Table 40. Clinicians' level of agreement to the statement: Mental health and problem gambling

clinicians can work effectively together to support patients ................................................................ 209

Table 41. Demographic characteristics of patient sample .................................................................. 227

Table 42. Lifetime mental health diagnoses self-reported by patients ............................................... 230

Table 43. Specific current anxiety mental health diagnoses self-reported by patients ...................... 231

Table 44. Current medications patients report being prescribed for mental health conditions .......... 231

Table 45. Patients' self-reported compliance with currently prescribed medication for mental health

conditions ............................................................................................................................................ 231

Table 46. Patients' self-reported participation in gambling activity in the past year ........................... 232

Table 47. Proportion of past year gamblers falling in to each risk category on the PGSI for the

whole sample ...................................................................................................................................... 233

Table 48. Gender breakdown of each of the gambling categories ..................................................... 233

Table 49. Age breakdown of gambling categories ............................................................................. 233

Table 50. Country of birth and gambling categories ........................................................................... 234

Table 51. Language spoken at home and gambling categories ......................................................... 234

Table 52. Ethnic identity and gambling category ................................................................................ 234

Table 53. Aboriginal and Torres Strait Islander status and gambling category .................................. 235

Table 54. Marital status and gambling category ................................................................................. 235


Victorian Responsible Gambling Foundation Page vii

Table 55. Highest level of education achieved and gambling category .............................................. 235

Table 56. Current employment status and gambling category ............................................................ 235

Table 57. Lifetime mental health diagnoses and gambling category .................................................. 237

Table 58. Current mental health diagnoses and gambling category ................................................... 238

Table 59. Average monthly spend on gambling activity (any) for patients in each gambling harm

category ............................................................................................................................................... 239

Table 60. Proportion of gamblers in each risk category who reported being asked about or

screened for gambling problems by a clinician ................................................................................... 240


Victorian Responsible Gambling Foundation Page viii

List of figures

Figure 1. Proportion of clinicians with previous training in problem gambling by profession ...................75

Figure 2. Clinicians' estimates of the proportion of their caseloads involving patients with

gambling problems ....................................................................................................................................76

Figure 3. Mental health illnesses clinicians associate with comorbid problem gambling .........................77

Figure 4. Frequency that clinicians ask patients about gambling .............................................................79

Figure 5. Frequency that clinicians screen patients for problem gambling ...............................................79

Figure 6. Methods clinicians used to identify problem gambling ..............................................................80

Figure 7. Clinicians' level of confidence in detecting or screening patients for problem gambling ...........81

Figure 8. Clinicians' current responses when patients experiencing problem gambling are

identified ....................................................................................................................................................82

Figure 9. Frequency clinicians report referring patients experiencing problem gambling to external

services .....................................................................................................................................................85

Figure 10. Services to which clinicians report referring patients experiencing problem gambling ...........86

Figure 11. Clinicians' level of confidence in referring patients to outside services for problem

gambling ....................................................................................................................................................86

Figure 12. Clinicians' opinions about the importance of referring patients to outside agencies

when problem gambling is identified .........................................................................................................87

Figure 13. Clinicians' response to the statement ‘Gambling Help Services are not equipped to

deal with clients with mental illness’ ..........................................................................................................87

Figure 14. Frequency clinicians report treating patients for problem gambling ........................................88

Figure 15. Type of treatment provided by clinicians and/or their service or practice ...............................88

Figure 16. Clinicians' level of confidence in treating patients for problem gambling ................................89

Figure 17. Clinicians' level of agreement to the statement: ‘I have a good understanding of the

Gamblers' Help service system and the programs available’ ...................................................................89

Figure 18. Clinicians' understanding of the types of treatments effective for problem gambling .............90

Figure 19. Clinicians' level of agreement to the statement: ‘Mental health and problem gambling

clinicians can work effectively together to support clients’ ........................................................................90

Figure 20. Description of the patient survey sample ...............................................................................123

Figure 21. Age categories of the total patient sample ............................................................................126

Figure 22. Type of mental health service patients attended at the time of participation ........................128

Figure 23. The four most commonly reported lifetime and current mental health diagnoses by

patients ....................................................................................................................................................129

Figure 24. Proportion of patients with one or more current mental health diagnosis .............................129

Figure 25. Current medications patients report being prescribed for mental health conditions .............130

Figure 26. Distribution of scores across ATOP items assessing psychological health, physical

health, and overall quality of life ..............................................................................................................131


Victorian Responsible Gambling Foundation Page ix

Figure 27. Frequency of patients’ self-reported alcohol consumption in the past year among

patients who reported any past-year alcohol use ................................................................................... 131

Figure 28. Number of self-reported cigarettes smoked on a typical day in the past year among

smokers................................................................................................................................................... 132

Figure 29. Specific drugs used among the total sample (N = 841) and among patients reporting

drug use in the past year (N = 203) ........................................................................................................ 133

Figure 30. Frequency of drug use in the past year ................................................................................. 134

Figure 31. Proportion of patients who reported engaging in specific gambling activities in the past

year ......................................................................................................................................................... 135

Figure 32. Proportion of gamblers who reported engaging in specific gambling activities in the

past year ................................................................................................................................................. 135

Figure 33. Frequency of gamblers' participation in specific gambling activities at physical venues

and online in the past year (n = 348) ...................................................................................................... 137

Figure 34. Frequency of gamblers' participation in specific gambling activities (at venue or online)

in the past year (n = 348) ........................................................................................................................ 137

Figure 35. Gamblers' spend on specific gambling activities at venues and online in the past year ....... 139

Figure 36. Gambling risk across the whole sample ................................................................................ 140

Figure 37. Gender breakdown of gambling categories ........................................................................... 141

Figure 38. Rates of gambling participation and harm in current patient sample and general

Victorian population ................................................................................................................................ 142

Figure 39. Level of harm past year gamblers report currently experiencing .......................................... 142

Figure 40. Age breakdown of gambling category ................................................................................... 143

Figure 41. Marital status and gambling category .................................................................................... 144

Figure 42. Current employment status and gambling harm category .................................................... 145

Figure 43. Proportion of gamblers presenting to mental health services with moderate-risk or

problem gambling ................................................................................................................................... 146

Figure 44. Current mental health diagnoses and gambling harm category ............................................ 147

Figure 45. Mean scores on ATOP scales for each gambling category .................................................. 150

Figure 46. Proportion of smokers in each of the gambling categories ................................................... 151

Figure 47. Past year drug use status by gambling category .................................................................. 151

Figure 48. Average number of times in the past year patients reported playing Pokies or

electronic gambling ................................................................................................................................. 155

Figure 49. Average number of times in the past year patients reported betting on horse of

greyhound racing .................................................................................................................................... 155

Figure 50. Average number of times in the past year patients reported playing Lotto, Powerball or

the Pools ................................................................................................................................................. 156

Figure 51. Average monthly spend on gambling activity (any) for patients in each gambling harm

category .................................................................................................................................................. 157


Victorian Responsible Gambling Foundation Page x

Figure 52. Mean PGSI score as a function of frequency of participation in the top four gambling

activities ..................................................................................................................................................158

Figure 53. Proportion of patients in each gambling category who reported being asked about or

screened for their gambling by a clinician ...............................................................................................159

Figure 54. Types of help patients from the total sample were likely to seek if they develop a

gambling problem ....................................................................................................................................159

Figure 55. Mean scores on readiness to change items of gamblers in each risk category ....................161


Victorian Responsible Gambling Foundation Page xi

Key terms

Acronym Description

AOD Alcohol and other drugs

ASI-PG Addiction Severity Index for pathological gambling

ATOP Australian Treatment Outcome Profile

AUDIT Alcohol Use Disorders Identification Test-Consumption

BBGS Brief Biosocial Gambling Screen

BPGS Brief Problem Gambling Screen

Clinicians Mental health workers, support workers

CPGI Canadian Problem Gambling Index

DSM The Diagnostic and Statistical Manual of Mental Disorders

EIGHT screen Early Intervention Gambling Health Test

EGMs Electronic gaming machines/poker machines (pokies)

GA20 Gamblers Anonymous Twenty Questions

Gambling harm category

Gambling status of whole sample (i.e., non-gambler, non-problem gambler, low-risk gambler, moderate-risk gambler or problem gambler)

Gambling harm category

PGSI determined level of harm or risk for problem gambling among gamblers (i.e., non-problem gamblers and low-risk, moderate-risk and problem gamblers)

GP General Practitioner

HSI Heaviness of Smoking Index

Lie-Bet Lie/Bet Questionnaire

MAGS Massachusetts Gambling Screen

NODS-CLIP National Opinion Research Center DSM-IV Screen for Gambling Problems – Control, Lying, and Preoccupation Measure (three items)

NODS-CLIP2 National Opinion Research Center DSM-IV Screen for Gambling Problems – Control, Lying, and Preoccupation Measure (five items)

NODS-PERC National Opinion Research Center DSM-IV Screen for Gambling Problems – Preoccupation, Escape, Chasing, and Risked Relationships Measure

Patients Mental health patients, clients, service users or consumers

PGSI Problem Gambling Severity Index

PGSI – three item

Problem Gambling Severity Index Short-Form

PPGM Problem and Pathological Gambling Measure

SOGS South Oaks Gambling Screen

SLUGS Sydney Laval University Gambling Screen

VGS Victorian Gambling Screen


Victorian Responsible Gambling Foundation Page xii


Victorian Responsible Gambling Foundation Page 1

Executive summary

Purpose of the report

This report presents the findings of a multi-phased research project examining problem gambling

rates and responses across mental health services in Victoria, Australia. Focusing on mental

health settings, the research provides important insights into:

current clinician and service responses to problem gambling

the prevalence of problem gambling in these settings and its relationship to comorbid

psychiatric disorders

valid problem gambling screening measures.

Background

While gambling is a popular pastime for most Australians, a small but significant minority

experience harm from gambling. The international literature suggests that rates of problem

gambling are elevated in individuals with mental health disorders and its compounding impact

includes increased psychiatric morbidity, poorer health and wellbeing and significant psychosocial

disadvantage. Unlike comorbid drug or alcohol problems, which can be difficult to mask, problem

gambling is often hidden. This means the problem often remains undetected and untreated until

associated problems (e.g. financial and relationship difficulties etc.) become overt. It is therefore

essential that clinicians and services working with those who are at increased risk of problem

gambling take steps to proactively identify gambling harms so that appropriate assessment and

treatment responses can be offered. Yet international research indicates that screening for

problem gambling rarely takes place in treatment settings, and the factors preventing and

facilitating problem gambling screening remain poorly understood.

To date, few studies have examined responses to problem gambling within mental health settings

and there is limited understanding of the skills, knowledge and attitudes of mental health clinicians

required to identify and respond to problem gambling. Research is needed to identify the

prevalence of problem gambling within Australian mental health settings, as well as existing

approaches, gaps and service needs. This information is fundamental to planning effective inter-

agency working relationships and providing treatment approaches that ensure the optimal and

timely delivery of care to those with concurrent gambling and mental health issues.

Aims

The overarching aims of the project were:

to examine the ways in which clinicians currently respond to problem gambling in Victorian

mental health settings

to examine the gambling behaviours of patients attending Victorian mental health services

to determine the prevalence of problem gambling and gambling-related harm and explore

its relationship with comorbid psychiatric disorders



to explore suitable problem gambling screening instruments and service responses within

Victorian mental health settings.

Overview of methodology

To achieve the aims of the project, a mixed-methods approach was used across five distinct study

components.

Study 1 was a comprehensive literature review undertaken to explore problem gambling

rates and the relationship with comorbid psychiatric disorders; screening, assessment and

treatment approaches for problem gambling in a mental health context; and best-practice

service models/partnerships for managing problem gambling in patients with mental health

disorders.

Study 2 was an organisational workforce survey to identify current screening practices,

referral systems and treatment approaches to problem gambling across a range of mental

health service settings, as well as clinician attitudes towards responding to problem

gambling.

Study 3 was an in-depth, qualitative investigation of current screening, assessment,

referral and treatment practices, and barriers and facilitators to responding to problem

gambling within mental health settings.

Study 4 was a cross-sectional study of patients attending multiple mental health settings to

examine their gambling behaviours and estimate the prevalence of problem gambling and

gambling-related harm.

Study 5 was a psychometric study of problem gambling screening tools to identify the

optimal brief screening tools to identify problem gambling and gambling-related harm

within mental health settings.

The emerging results from the five studies were presented at a clinical reference panel comprising

clinicians, consumers, senior managers and funders across the Victorian mental health and

gambling sectors. The panel assisted with translating the research findings into a series of

implications and actions for improved responses to problem gambling at the service system and

mental health service level.

Study 1: Literature review

The first project component was an extensive narrative review of the national and international

literature on gambling and mental health, divided into five sections.

Section 1 presents the literature on the prevalence of problem gambling and profiles of Victorians

who are at risk of developing gambling problems. The key findings revealed a gambling

participation rate of 70.1 per cent in Victoria1, a problem gambling rate of 0.8 per cent, a moderate-

risk gambling rate of 2.8 per cent and a low-risk gambling rate of 8.9 per cent. Despite the

existence of considerable gambling harms, rates of help-seeking for gambling problems in the

Victorian population are low (around 22 per cent). Since people experiencing gambling problems

1 Note that the participation rate excluding raffles, which were not assessed in the present study, was 61.6% [95% CI 59.1% - 64.0%]



have an increased likelihood of being diagnosed with depression, anxiety disorders and alcohol

use disorder and frequently attend their GP, the literature highlights the important role health

professionals play in the identification and management of gambling problems.

Section 2 presents literature examining rates of co-occurring gambling and mental health problems

in the community and in different treatment settings, as well as the temporal relationship between

the two. The emergent findings indicated that problem gambling is consistently associated with a

range of comorbid mental health disorders including alcohol and other drug use disorders, mood

disorders, anxiety disorders, impulse control disorders and personality disorders. People with

gambling problems are grossly over-represented in primary care, AOD settings and mental health

treatment populations, with rates of problem gambling identified in up to 30 per cent of treatment

seekers. Evidence suggests that problem gambling typically precedes and predicts the onset of

other mental health conditions, suggesting that it has the potential to complicate treatment plans

and hamper treatment outcomes, particularly if it goes unidentified and untreated. The findings

highlight the importance of identifying problem gambling through routine screening in mental health

settings with a view to appropriate management and/or referral to specialist gambling services.

Section 3 presents literature on screening for problem gambling, examines and evaluates brief

screening instruments for problem gambling and presents evidence on the most appropriate

screening instruments for use in different treatment settings. The key findings from the review were

that rates of routine screening in mental health and other services are typically low. Barriers to

screening include lack of time, a lack of knowledge and skills, the presence of gambling-related

stigma and a perception that problem gambling has a low burden of disease. However, the slow

development of appropriate and valid brief screening instruments may be a further obstacle to

screening. While there are several brief instruments available, there is limited information to guide

the selection of screens that are effective in mental health settings and a need to determine their

classification accuracy in Australian mental health settings.

The literature suggests that comprehensive clinical assessment is needed to determine diagnostic

status and problem severity when screening positive for problem gambling, and that this

assessment is best conducted within mental health services if time and resources permit. Once a

gambling problem has been identified and assessed, mental health services must have the time,

skills and resources to treat the gambling problem or have appropriate referrals in place.

Section 4 reviews psychological and pharmacological treatments for problem gambling, as well as

interventions for comorbid problem gambling and mental health disorders. Taken together, the

extant research indicates that individuals with gambling problems must be offered appropriate

management, provided by mental health services or managed via referral to specialist gambling

treatment services. Although a diverse range of psychological and pharmacological options for the

treatment of problem gambling are available, the evidence base is limited and is confounded by

generally low quality standards.

The Australian NHMRC-endorsed clinical practice guideline provides a low-grade recommendation

involving the cautious use of naltrexone in the treatment of problem gambling and several higher-

grade recommendations for the use of Cognitive Behaviour Therapy (CBT) and Motivational

Interviewing (MI). Despite our understanding of the comorbidity between problem gambling,

substance use problems and mental health issues, there is very little evidence on which to base

treatment recommendations for different subpopulations of problem gamblers based on their

psychiatric comorbidity. Brief interventions and online self-help programs based on MI and CBT

may be useful resources for clinicians working with clients with comorbid gambling problems within

primary care, AOD and mental health services.



The final section of the review examines treatment models and effective partnerships in gambling

and mental health services. The findings revealed limited empirical knowledge about models of

care for gambling and mental health, prompting the need to examine models addressing comorbid

mental health and alcohol and drug issues more broadly, where integrated working is more

established. Indeed, there are few specialist treatment services available for individuals with

comorbid mental health disorders and problem gambling. While many evaluations of these models

reveal promising results, with reported reductions in substance use and psychiatric

symptomatology, the quality of evaluations are often poor (e.g. small samples and an absence of

comparison groups). Nonetheless, common components of integrated models include the use of

multidisciplinary teams, continuous treatment teams, an integrated treatment philosophy and

stage-wise or tailored treatment approaches. A core minimum set of model features include

universal screening across all mental health practitioners and services, risk assessment and

diagnosis of symptoms, supportive therapies, prevention and psychoeducation, collaboration with

AOD services and/or GPs, supportive policies and cross-sectoral collaboration.

Study 2a: Clinician survey: Current practice and responses to

gambling

The second study component aimed to assess problem gambling-related knowledge, attitudes and

practices of mental health clinicians, with a focus on role legitimacy, screening, assessment,

referral and treatment of clients with problem gambling within mental health settings. Using a

cross-sectional design, clinicians from a broad range of mental health services (youth, adult,

public, private etc.) and sites (metropolitan, regional etc.) were invited to participate in an

anonymous survey (online or hard-copy) about how they respond to problem gambling.

Participating services included three public area mental health services (AMHS) which offer

catchment-based inpatient and outpatient clinical care and case management; a statewide mental

health community support service (MHCSS) which offers outreach psychosocial rehabilitation and

support; two private psychiatry outpatient clinics; and one community health service that provides

general health and psychological support to clients with mental health issues.

In total, 311 clinicians (representing 71 per cent of the workforce) from 14 different sites across

nine different mental health services completed the survey. The sample comprised doctors,

nurses, psychologists, case managers, social workers, support workers, occupational therapists

and other mental health clinicians working at the frontline and directly responsible for patient care.

Clinicians were predominantly female (73 per cent), averaged 40 years of age and had been

working in mental health for an average of 12 years. Clinicians estimated that one in ten of their

patients were affected by gambling problems. Encouragingly, most clinicians were knowledgeable

about problem gambling and an overwhelming majority considered screening, assessment and

referral to be a core part of their role. However, it is concerning that as many as 40 per cent of

clinicians never, or only rarely, ask their patients about their gambling behaviour and that 55 per

cent never, or only rarely, screen for problem gambling. Less than 5 per cent of clinicians reported

routinely (‘often or always’) screening for problem gambling, and only 2 per cent reported using a

standardised/formal screening tool, with most clinicians relying on informal discussion. The

majority (60 per cent) of clinicians had low confidence in their capacity to detect problem gambling.

Reported barriers to screening related to systemic issues (e.g. screening not being a part of

standard intake assessment, or not an organisational requirement), knowledge base and training

issues (e.g. a lack of knowledge of how to screen for gambling problems) and conflicting treatment

priorities (e.g. other issues that patients present with having a higher priority). However, low rates



of screening could be explained by the fact that only 10 per cent reported being aware of screening

and assessment tools.

Clinicians’ confidence in managing gambling problems was particularly low, with less than one-

quarter reporting that they were knowledgeable about effective treatments and only one-third

reporting that they could treat the gambling problem. While concerning, these results are perhaps

unsurprising given that only a minority (12 per cent) had ever received training in problem

gambling. Although two-thirds of clinicians reported that they could refer patients with gambling

problems to external treatment providers, few reported actually doing this, and the data suggest

that this may reflect concerns that specialist services are not equipped to deal with mental health

issues or a lack of understanding of what Gambler’s Help services offer (which was expressed by

80 per cent). Greater levels of knowledge, confidence and responding to problem gambling was

observed among clinicians who had received prior training in problem gambling, which provides a

compelling justification for greater investment in workforce training.

Study 2b: Barriers and facilitators to responding to problem

gambling

The third activity was an exploration of the barriers and facilitators to responding to problem

gambling, through 30 in-depth qualitative interviews with clinicians drawn from 11 of the mental

health services that participated in Study 2a. The sample comprised 19 female and 11 male

clinicians and included registered nurses, social workers, occupational therapists, case workers

and a clinical psychologist. Findings revealed a marked disparity in how clinicians currently ask

about and respond to gambling and suggest there is a need to address barriers to screening.

Overall, the majority of clinicians indicated that gambling was not included in the assessment tool

used at their service, either as a direct question or as a prompt. For the most part, screening was

ad-hoc or at the discretion of individual clinicians, often only occurring if the client disclosed or in

some way indicated that they may have a gambling problem. Commonly reported red flags were

financial difficulties such as recurrent quests for food parcels or applications for financial

assistance.

Overall, clinicians identified a wide range of interrelated factors that acted as barriers to routine

screening for gambling problems. Specifically, these included:

being overburdened with the scope of their current role, the breadth of issues they were

required to assess, and having limited time

limited knowledge of screening tools

reluctance to include more tools/screens and add a burden to patients

resistance to screening in general

greater focus on immediate risk or disorders associated with greater harm

perception that gambling is not a common problem, or that it is not a mental health issue

low level of training about problem gambling and its impact on patients with mental health

disorders, and inadequate access to training and education to upskill

low level of confidence in ability to respond when a gambling problem is identified



concerns about patient openness to gambling screening, willingness to disclose a

gambling problem, and readiness to change, related to the shame and stigma associated

with problem gambling.

In contrast, facilitators to screening (i.e. factors associated with screening practices) emerging from

the interviews included:

awareness of the high prevalence of gambling problems among mental health populations

recognition of problem gambling as an underlying issue

a readiness to use screening tools as part of assessment processes

the availability of advice/training in how to respond when a gambling problem is identified.

The majority of clinicians acknowledged the benefit of having a formal screening tool to identify

problem gambling, yet the inclusion of a new tool should take a number of factors into

consideration:

existing screening and assessment requirements

provision of specific training to facilitate use of the tool and identification process

provision of specific training to support appropriate responses to the identification of

problem gambling.

Training needs were identified in five main areas:

1. awareness of problem gambling, including prevalence in mental health services

2. knowing the signs and how to introduce the issue especially in those that have not been

screened

3. identification and application of appropriate screens (and therefore training in what the

screens are, how to administer, what the results mean)

4. knowledge of how to assess

5. treatment/referral options, especially minimal or brief interventions.

Clinicians also indicated a preference for group over online training with practical examples that

they could relate to and with an opportunity to practise skills. Multiple clinicians noted the

importance of follow-up, booster sessions or other post-training contact and the need for training to

be evaluated.



Study 3a: Patient survey: Gambling behaviours and prevalence of

problem gambling

Using a cross-sectional design, the fourth research activity entailed a survey of 841 patients

attending community mental health services in Victoria, to examine gambling participation and

determine the prevalence of problem gambling and gambling-related harm among patients seeking

treatment for mental health disorders. The sample was drawn from eight separate mental health

services and 12 individual sites/teams participated, representing a broad mental health population

in terms of the types of services accessed (i.e. private, public, adult and youth), geographical

locations (metropolitan, regional) and population demographic. The 841 patients represented

around 55 per cent of all patients attending those services during the data collection period. Just

over half of the sample were male, the mean age was 38 years (range = 16–95), 77 per cent were

Australian-born, 65 per cent were single or never married and only one-third (32 per cent) were

employed, with one in five having a personal income of less than $500 per fortnight. More than half

of the sample (52 per cent) had been attending the mental health service for more than one year.

Most patients (87 per cent) reported having a current mental health diagnosis, most commonly

depression, anxiety, a psychotic disorder and bipolar disorder. Few patients reported having a

diagnosis of substance use disorders (only 10 per cent), with 2.2 per cent and 0.8 per cent of

patients reporting a lifetime or current gambling disorder respectively. However, the majority of

patients (67.7 per cent) reported consuming alcohol in the past year and 55 per cent were

identified as being problem drinkers on the AUDIT-C, just under half of the sample (49.1 per cent)

reported smoking in the past year and almost a quarter (24.1 per cent) reported using an illicit drug

or a prescription medication for non-medical use (drug use used henceforth for both terms) in the

past year.

Gambling participation rates in the past year among patients were considerably lower than in the

Victoria adult general population: 41.4 per cent [95 per cent CI = 38.1 – 44.7], versus 61.6 per cent

[95 per cent CI 59.1 per cent – 64.0 per cent]2. The most common gambling activities among

patients were pokies or electronic gaming, closely followed by Lotto, Powerball or the Pools,

followed by betting on horses or greyhounds, and then scratch tickets. Most gambling took place in

venues, although the most common activities online were betting on sports, horse or greyhound

racing and pokies/EGMs. Gamblers reported spending a mean of $119 per month, although the

median spend was $20.00 and the mode among those who had gambled in the past month was

$50.00.

Gambling harm

The Problem Gambling Severity Index (PGSI) was used to determine gambling harm and the

mean total score among the gamblers (n = 348 patients) was 3.2 (SD = 5.1, range = 0–27), just

inside of the moderate-risk range. 19.6 per cent (n = 165) had a PGSI score in the non-problem

gambling range, 7.1 per cent (n = 60) had a PGSI score in the low-risk range, 8.3 per cent (n = 70)

had a PGSI score in the moderate-risk range risk range, and 6.3 per cent (n = 53) were identified

as problem gamblers.

The following table shows that patients attending mental health services report elevated rates of

gambling harm. Compared to the general population in Victoria, patients were eight times as likely

2 Note differences in rates of gambling participation and gambling harm based on PGSI category may reflect differences in the demographic characteristics of the two samples



to be a problem gambler and three times as likely to be a moderate-risk gambler2. However, given

the shift to conceptualising gambling harm on a spectrum, with recent evidence that the greatest

harm is experienced by low-risk gamblers in Victoria (Browne et al., 2016), it can be seen that one

in five patients (21.7 per cent) experienced at least some gambling harm (i.e. were identified as

low-risk, moderate-risk or problem gamblers), while 52.5 per cent of all gamblers experienced at

least some gambling harm. This suggests that if a patient in a mental health treatment setting has

engaged in gambling in the past year, there is a strong possibility they are experiencing gambling-

related harm. The total sample reported low levels of physical wellbeing, psychological wellbeing

and overall quality of life. Moderate-risk gamblers reported significantly poorer levels of physical

wellbeing.

Mental health sample

(N = 841)

Victorian general population

(N = 13,554)

% [95% CI] % [95% CI]

Non-gamblers 58.6 [55.3 – 61.9] 29.9 [27.5 – 32.4]

Non-problem gamblers 19.6 [16.9 – 22.3] 57.6 [55.0 – 60.2]

Low-risk gamblers 7.1 [5.4 – 8.8] 8.9 [7.2 – 11.0]

Moderate-risk gamblers 8.3 [6.4 – 10.2] 2.8 [1.8 – 4.2]

Problem gamblers 6.3 [4.6 – 7.9] 0.8 [0.5 – 1.4]

Differences in gambling participation and harm across subgroups

Male patients were significantly more likely to be gamblers and to be experiencing gambling-

related harm, and gamblers in the 35–44 year age category were most likely to be experiencing

gambling harm. While gamblers in the 65+ age group had the highest rates of participation, they

experienced the lowest rates of gambling harm. Gambling harm category did not differ if the patient

identified with an ethnic minority group or as Aboriginal or Torres Strait Islander, was Australian-

born or spoke a different main language at home. Similarly, there were no differences in gambling

harm category related to marital status or education level; however, a significantly higher

proportion of moderate-risk and problem gamblers reported that they were currently neither

employed nor studying. There were a few differences by mental health service type, with higher

rates of moderate-risk and problem gambling observed in mental health community support

services (outreach) versus specialist public mental health services (31.5 per cent vs. 13.4 per

cent), in public vs. private services (17.3 per cent vs. 10.3 per cent) and in adult vs. youth services

(15.8 per cent vs. 5.4 per cent). In terms of substance use, moderate-risk and problem gamblers

were significantly more likely to be smokers and a larger proportion of low-risk gamblers were

identified as problem drinkers.

Gambling harm by mental health diagnosis

Higher rates of gambling harm were reported among patients with certain mental health disorders.

Patients with a drug use disorder were 3.6 times as likely and patients with psychotic disorder 2.4

times as likely to be experiencing gambling-related harm. Patients with drug use disorder were 3.4

times as likely and patients with borderline personality disorder were 2.6 times as likely to be a

problem gambler. The proportion of patients falling into the moderate-risk and problem gambling

categories increased as the number of current mental health diagnoses increased, suggestive of

increased gambling harm among those experiencing more comorbid psychiatric disorders.



Gambling harm by frequency and spend

Compared to non-problem gamblers, problem gamblers spent approximately 11 times more money

per month on gambling activities (a mean of $440), moderate-risk gamblers spent approximately

three times as much per month (a mean $124) than non-problem gamblers, and low-risk gamblers

spent approximately 30 per cent more than non-problem gamblers (a mean of $50). For the top

four most common gambling participation categories, daily gamblers had significantly higher PGSI

scores than weekly, monthly or less frequent gamblers on EGMs, horse and greyhound racing, but

not on Lotto, Powerball or the Pools and scratch tickets.

Screening for problem gambling, readiness to change and help-seeking preferences

Further evidence of inadequate screening practices emerged in that only 43 per cent of the sample

reported having been asked about their gambling since attending the mental health service. The

majority of patients (two-thirds of participants) indicated that speaking to their mental health worker

or using self-help-strategies were the preferred methods for seeking help if they were to

experience a gambling problem, although there were differences in the preferred method

according to gambling-risk category. Finally, a significant difference was found in the mean score

of the item ‘How important is it for you to reduce or stop gambling?’, with increasing importance

reported by patients displaying greater severity of gambling harm along with a lower mean score

on the item assessing confidence in their ability to reduce or stop gambling.

Study 3b: Psychometric testing of gambling screens

The fifth study component aimed to determine the optimal brief screening tools for use in mental

health settings. Using data from gambling participants identified in the patient survey (n = 348), the

psychometric properties of 10 brief, two- to five-item problem gambling screening instruments and

the 23 single items that comprised those screening instruments were examined (the PGSI, the

Lie/Bet, the two-, three-, four- and five-item BPGS, the PGSI-short form, NODS-CLIP, NODS-

CLIP2, BBGS, NODS-PERC). The sensitivity, specificity and overall classification accuracy were

compared across the screens.

The findings were that several single items and all of the brief screening instruments displayed

adequate sensitivity, specificity and overall diagnostic accuracy in a mental health population. The

psychometric properties of item 1 of the PSGI (‘Thinking about the past 12 months, have you bet

more than you could really afford to lose?’) indicated that this single question accurately detected

96 per cent of clients with mental health problems classified as problem gamblers on the full PGSI

and 84 per cent of patients classified as moderate-risk or problem gamblers on the full PGSI.

However, the three-item screen BPGS most effectively detected 92.7 per cent of patients classified

as moderate-risk or problem gambling and the five-item screen BPGS most effectively detected

80.3 per cent of patients classified as low-risk, moderate-risk or problem gambling within a mental

health sample.



Overall study strengths and limitations

When considering the study findings, it is important to acknowledge a number of limitations. Firstly,

the clinician survey and qualitative interviews may have been affected by a sampling (self-

selection) bias, with those with the greatest interest or with the most knowledge and experience of

problem gambling being more likely to agree to participate. However, given that we were able to

sample over 70 per cent of eligible clinicians, this limits the likelihood of this bias occurring.

Secondly, the survey and interviews were reliant upon self-report and not observations of actual

practice, and thus social desirability effects may have encouraged clinicians to respond to

questions in line with best practice. While this is possible, the anonymous nature of the survey is

likely to have countered many of these biases. Nonetheless, both these issues could portray

current clinician responding to problem gambling in a more positive light than we have identified in

this report.

In terms of limitations with the patient survey, a convenience sample was used, representing

approximately 55 per cent of all available patients attending mental health services during the data

collection period. It was not possible to recruit the most severely unwell patients (e.g. acutely

psychotic) or those with significant risks. Since the literature and current data are indicative of

greater gambling-related harm among these patients, the 21 per cent estimate of gambling harm is

a likely underestimate. Similarly, shame and stigma and social desirability effects could contribute

to an under-reporting of gambling behaviour and harm by patients. This limitation was addressed

to some extent by the anonymous nature of the survey and by reassuring participants about the

confidentiality of their responses, and that their responses would not affect their treatment in the

mental health service. The PGSI was only administered to individuals who reported participation in

gambling activities in the previous 12 months; however, it is possible that patients could have

experienced gambling issues prior to the past 12 months, which would have again led to an

underestimation of gambling harm in the sample. The patient survey was also entirely reliant on

patient estimates of their gambling behaviours, with no objective or corroborated data. Finally,

although a wide range of services were recruited, the results may not generalise to other mental

health settings (e.g. acute inpatient settings, outreach services that serve patients who are

generally too unwell to attend community settings), where the prevalence of gambling harm could

be greater still.

The main limitation of the psychometric study of problem gambling screening instruments was that

the classification accuracy was assessed against the nine-item PGSI, which is the longer measure

from which several of the brief screening instruments are derived, and the overlap between the

reference standard and the brief screening instruments may have inflated the classification

accuracy coefficients. Further limitations include the PGSI and brief screening instruments being

administered only to individuals who reported participation in gambling activities in the previous 12

months, and the timeframes for all brief screening instruments developed as lifetime measures

(e.g. the Lie/Bet, NODS-CLiP, and NODS-PERC) were converted to 12 month timeframes. Finally,

the data are based on participant self-report, whereby the validity of the data may not be accurate

if participants wished to conceal the severity of their gambling.

Despite these limitations, there are a number of important strengths to the project that increase our

confidence in the research findings. These include the adoption of multiple robust methodologies

to address the key research questions, the large number of clinicians (N = 311) and patients (N =

841) recruited, the breadth of mental health services involved, the number of different service sites

across Victoria from which participants were drawn, and the expertise and insights of the research

team and clinical reference panel.



Clinical reference panel

The project findings highlight the gaps and opportunities for improved responding to problem

gambling within mental health settings. To help develop practical approaches to the way in which

mental health services can improve their responding to problem gambling, a clinical reference

panel was convened. A half-day event was attended by a diverse range of clinicians (N = 32)

working at the frontline, as well as team leaders and service managers in the mental health and

gambling sectors, consumers, VRGF senior staff and Department of Health and Human Services

representatives. Findings from the five studies were presented to the panel, and then using a

RAPID Process Improvement Methodology, panel members examined strengths and weakness of

current practices, and priorities for action.

There was a clear consensus that there are many areas for improvement. Priorities for action

identified were:

reducing stigma

raising consumer and clinician awareness of problem gambling

addressing the lack of routine screening and assessment for problem gambling

improving a lack of lived experience/knowledge in training or service provision

addressing fragmentation of mental health and gambling services and funding models.

In terms of solutions, panel members identified several low-cost/effort initiatives that could

immediately improve mental health service response to problem gambling:

1. implementing a validated brief screen for problem gambling (the single-item PGSI as a

minimum) at intake

2. raising consumer awareness that gambling problems and mental health issues commonly

occur, and help is available through the service

3. identifying champions/role models for good clinical practice around problem gambling

(who keep gambling on the agenda and build capacity within a team)

4. sharing existing education packages around problem gambling and clinical approaches

5. educating staff about the Gambler’s Help system and developing local referral pathways.

In terms of medium-cost/effort initiatives, the panel identified three key initiatives:

1. investment in clinical supervision and support, and clearly defined models of care for

responding to gambling within mental health settings

2. specialist training for clinicians in screening, assessment, brief intervention and treatment

(where relevant) of problem gambling

3. development of partnerships and integrated working between mental health and gambling

services with established memoranda of understanding (MOU) to promote accountability.



Conclusion

Consistent with previous research, the report’s findings highlight major gaps in current responding

to problem gambling in Victorian mental health services. Given the elevated rates of problem

gambling observed in patients attending mental health services, and the finding that one in two

gamblers are experiencing gambling-related harm, there are significant opportunities for

prevention, early identification and intervention for a population at elevated risk.

Encouragingly, clinicians are broadly positive towards working with patients with problem gambling

and consider it part of their role. However, only a minority have received any form of training, and

they tend to see gambling as a secondary or non-critical issue. In addition, the lack of a

standardised screening process or tool within mental health services, and concerns among

clinicians that asking about gambling too early would increase stigma, means that identifying and

responding is both discretional and ad-hoc. Low rates of detection (and subsequent referral) is

further exacerbated by clinicians’ low confidence in treating problem gambling, as well as their

poor understanding of the Gambler’s Help system and concerns about the capacity of Gambler’s

Help services to manage patients with mental illness.

In terms of supporting earlier identification of problem gambling within mental health services, the

findings indicate that a one- to three-item problem gambling screen is an effective method, and this

could be easily added to existing intake processes. The adoption of a brief screening tool into

routine clinical practice was widely supported by clinicians across all stages of the project;

however, the need for this to be part of a broader system response was also identified. This should

include comprehensive training in the assessment and management of problem gambling (given

the breadth of evidence-based treatment and support options now available), as well as improved

intersectorial partnerships and referral pathways. Together, these approaches will ensure that

patients with mental health and gambling issues receive timely and appropriate intervention to

optimise their recovery and wellbeing.

Implications of research findings

This multi-component study provided consistent findings relating to the identification and

management of patients with gambling problems within mental health services. The findings

highlighted gaps in current service provision and opportunities for improved responding to the needs

of this population. The implications of this work, informed by discussion with the clinical reference

panel and targeted at services and the service system, are listed below together with

recommendations for future research.

1. Implications for the service system

(a) Raising consumer and carer awareness that the risk of gambling-related harm is

greater among those with mental health issues. The research revealed that among

patients with mental health problems, gamblers were more likely to be experiencing

gambling-harm than not. Furthermore, rates of problem gambling in this population

were eight times higher and rates of moderate-risk gambling three times higher than

they were in the general adult population. Clinicians estimated that one in every 10

patients on their caseload was affected by gambling. These findings support earlier

research studies included in the literature review which point towards elevated rates of

gambling-related harm for individuals with mental health disorders. One mechanism

for increasing consumer or carer awareness could be through the development of



targeted health promotion materials or resources (and related campaigns), ensuring

these are readily available in primary health, AOD and mental health services.

(b) Reducing the stigma associated with problem gambling and instilling hope that

recovery is possible for patients, families and the broader community. Both the

literature review and the qualitative clinician interviews highlighted stigma as a

common barrier to seeking help for a gambling problem. One potential mechanism for

reducing stigma among patients with a mental health problem could be to establish a

platform for the dissemination of gambling recovery success stories, for example

through peer-led initiatives that normalise the experience of gambling problems and

help-seeking behaviours and promote the reality of recovery success. One related

recommendation from the clinical reference panel was that the contribution of ‘lived-

experience’ is not tokenistic, but an integral part of any training or awareness-raising

program, to help communicate from a patient perspective why addressing gambling is

important and what has worked to help individual patients.

(c) Improving intersectorial partnerships and establishing joint policy and funding

initiatives to enhance the integration of care for patients with gambling and

mental illness. One of the key outputs of the clinical reference panel was identifying

the need to harness opportunities for joint-working between commissioners of mental

health and gambling services, moving away from fragmented, siloed operations and

towards shared joint strategic planning. As proposed in a recent paper on policy and

service delivery approaches for patients with comorbid problem gambling and mental

health issues, cross-sector integration is likely to require incentives and long-term

government commitment and support (Martyres & Townshend, 2016). The Victorian

Responsible Gambling Foundation (VRGF) and Victorian Department of Health and

Human Services (DHHS) should consider forming a joint working group to examine

opportunities for joint policy and funding initiatives to enhance the integration of

service delivery.

(d) Increasing clinical skills in screening, assessment and referral pathways. In

support of the findings from the literature review, the clinician survey and qualitative

interviews evidenced low rates of screening, assessment and referral. Over half the

clinicians reported rarely or never screening for problem gambling. Similarly, less than

half of the patients surveyed had been asked about their gambling problem. In the

qualitative interviews, clinicians indicated a preference for group over online training,

and a preference for training with practical examples (e.g. vignette-based training) they

could relate to and with opportunities to practise skills. Multiple clinicians noted the

importance of follow-up, booster sessions or other post-training support and the need

for training to be evaluated. Specialist training must aim to increase knowledge,

confidence and capacity to recognise the signs of problem gambling, and include

guidance on how to introduce the issue of gambling, how to identify and apply

appropriate screens, how to administer the screens and interpret scores,

recommended responses for different gambling risk severity and referral options. An

existing resource that could be adapted for mental health clinicians is the ‘Slots and

Shots’ guideline produced by Rowe, White, Long, Roche, and Orr (2015).

(e) Increasing confidence and capacity to deliver treatments targeting problem

gambling. The clinician survey and qualitative interviews demonstrated that few

clinicians were confident in managing or treating a gambling problem, but that rates of

confidence and general willingness to respond were higher among clinicians who had



previously received training in problem gambling. With evidence from the literature that

integrated CBT (which simultaneously addresses cognitions and behaviours that result

in psychological distress and problematic gambling behaviour) is an effective

treatment option, there are opportunities to upskill key clinicians within mental health

services to provide this treatment option where indicated, especially where they are

already providing intensive psychological support to patients with multiple morbidities.

Training could be offered to relevant services across the breadth of potential

intervention options, including harm reduction and self-help strategies, brief

intervention or CBT (for services with capacity), as well as joint treatment planning with

Gambler’s Help agencies for services without treatment capacity.

(f) Ensuring Gambler’s Help services are equipped to meet the needs of patients

with mental illness. The clinician survey and qualitative interviews identified concerns

regarding the capacity of Gambler’s Help services to meet the needs of patients with

mental illness. It is recommended that capacity is assessed and training provided if

necessary. This is pertinent since the patient survey indicated that patients with the

most complex needs (those with multiple disorders, drug use disorder, psychosis and

borderline personality disorder) are more likely to be experiencing gambling-related

harm. Gambler’s Help services should identify staff who are skilled in managing

patients with comorbid gambling problems and mental illness, and be offered relevant

training and supervision to support this work.

2. Implications for services

(a) Improving processes for early identification of patients with gambling problems.

The finding in the literature review that mental health disorders typically predate the

onset of problem gambling highlights the importance of targeting mental health

populations. The literature review, clinician survey and patient survey data all point to

the need to implement and standardise processes to identify problem gambling. The

qualitative interviews indicated that on the rare occasions when routine screening does

happen, clear protocols have been established. As highlighted through the qualitative

research with clinicians, mental health services should, as a minimum, consider

embedding a screening tool within the service’s intake process (as most have done

with alcohol and drugs). The findings from the study examining the psychometric

properties of problem gambling screening tools indicated that a range of screen

instruments performed adequately in mental health service populations. A single item

screen (PGSI-item 1) or a three-item screen (BPGS) most effectively detected

moderate-risk or problem gambling, and a five-item screen (BPGS) most effectively

detected low-risk, moderate-risk or problem gambling within a mental health

population.

(b) Raising the profile of problem gambling within mental health services. While

patients reported that mental health workers were the preferred method for seeking

support if they experienced a gambling problem, clinicians had low confidence in

managing problem gambling and it was typically seen as a ‘low priority’ within a

patient’s treatment plan. It is therefore important to communicate to mental health

clinicians that problem gambling can serve as a stressor that impacts on a person’s

mental state, increasing the risk of relapse (and potential acute/crisis service use) as

well as their ability to adhere to treatment (e.g. significant financial loss impacts

housing stability, ability to pay for medication/supportive therapy). One potential

method (that has been effective in addressing substance use within mental health



services) is to identify a ‘champion’, portfolio holder or working group, so that gambling

is given a higher priority and treatment and referral options are actively promoted

within the service.

(c) Supporting staff training opportunities and clinical supervision in managing

problem gambling. The clinician survey revealed that only 12 per cent of the

workforce had received at least some specialist training in problem gambling,

highlighting opportunities for further professional development. The clinical reference

panel argued that one of the strengths of existing shared-care initiatives was the

opportunity for joint participation in professional development (e.g. forums supporting

reciprocal arrangements, where mental health clinicians provide consultation and

training to gambling counsellors and gambling counsellors provide consultation and

training to mental health clinicians in their respective areas of expertise), and this is a

useful model to consider. Services should support staff opportunities for ongoing

professional development on topics relating to identifying and responding to problem

gambling and provide a mechanism for facilitating clinical supervision related to

problem gambling.

(d) Developing working partnerships between mental health and local Gambler’s

Help services. The findings from the literature review on models of care and outputs

of the clinical reference panel highlight the need to invest in gambling–mental health

shared-care partnerships. This is particularly important as the clinician survey

indicated a poor understanding of what Gambler’s Help services offer and few

clinicians were regularly referring patients to these services. This could be achieved by

establishing service agreements or memoranda of understanding, with shared care

models defining clear referral pathways between Gambler’s Help and mental health

services. This could include the establishment of local processes to facilitate

collaboration (e.g. streamlined referral processes, joint assessment, shared-care plans

and in-reach work), as well as in-service training about the Gambler’s Help system and

local processes.

3. Recommendations for future research

(a) Determining rates and patterns of gambling harm across all mental health

populations. While this study surveyed a diverse range of community mental health

services, the findings cannot be generalised to other mental health settings (e.g. acute

inpatient services and public mental health outreach services) where gambling harm

may differ. In addition, the high rates of gambling harm among patients with substance

use problems identified in the literature and patient survey highlights the importance of

a more robust examination of gambling harm within AOD treatment settings. Finally,

given the high rate of gambling harm among patients with mental health problems,

further research is needed to understand why this population is at such risk and the

most effective strategies to minimise harm.

(b) Determining effective models of care for patients with comorbid problem

gambling and mental illness. The literature review highlighted the dearth of empirical

knowledge about models of care for gambling in mental health settings. However,

drawing on the more extensive AOD literature, current evidence suggests that models

should comprise multidisciplinary teams, continuous treatment teams, an integrated

treatment philosophy, and stage-wise or tailored treatment approaches. A core

minimum set of model features includes universal screening across all mental health



practitioners and services, risk assessment and diagnosis of symptoms, supportive

therapies, prevention and psychoeducation, collaboration with other services and/or

GPs, and supportive policies and procedures. The literature review also draws

attention to the limitations of integrated treatment model evaluations to date. Further

research is needed to determine the impact and cost-effectiveness of service models,

at both a patient/use and service level through rigorous evaluation of existing models

as well as pilot/feasibility studies of integrated care.

(c) Developing effective treatments for individuals with comorbid gambling and

mental health disorders. The literature review of treatment approaches concluded

that there is a paucity of evidence on which to base treatment recommendations for

different subpopulations of problem gamblers based on their psychiatric comorbidity

(as comorbid patients have typically been excluded from treatment trials for gambling

problems). This gap in evidence on treatment effectiveness was also identified as a

key priority area by members of the clinical reference panel. Research demonstrating

a clear relationship between comorbidity and poorer treatment outcomes remains

limited. Further evidence of this relationship may encourage clinicians to screen and

respond to gambling problems. This could potentially be achieved through investment

in systems that support the monitoring of outcomes of patients with comorbid issues

and greater focus on trials of integrated psychosocial interventions for these

populations.



Background

Gambling is ubiquitous, and major technological advances in the past two decades (e.g. growth in

online gambling and smart phone applications) have led to unprecedented accessibility and

availability to gambling activities. In Australia, most states and territories offer a wide range of

gambling activities, with an estimated 70 per cent of the Victorian adult population having gambled

in the past year. Total expenditure on gambling in 2014–2015 was $22.7 billion, over half of which

(51 per cent) was lost on poker machines in pubs and clubs, $2.6 billion in Victoria alone. Indeed,

the Australian Gambling Statistics for 2014–15 suggest adult Australians lose, on average, $1,242

a year on gambling.

While gambling is a popular pastime for most individuals, a small but significant minority

experience harm from gambling, which can have a detrimental impact on psychological, social,

familial and occupational functioning. Problem gambling is estimated to affect 0.5 per cent to 7.6

per cent of the adult population worldwide, with the average rate across all countries being 2.3 per

cent (Williams, Volberg, & Stevens, 2012). The most recent estimates from the Victorian

prevalence survey (Hare et al., 2015) are that 0.8 per cent of the adult Victorian population are

problem gamblers (approximately 35,000 adults). In the Australian context, the term ‘problem

gambling’ is used to describe harms associated with difficulties in limiting time and/or money spent

on gambling (Neal, Delfabbro, & O'Neil, 2005), and is intended to encompass a continuum of

severity that includes the diagnostic classification of pathological or disordered gambling. Harm

extends far beyond the gambler, and can have a profound impact on family members, friends,

colleagues, employers and the surrounding community. It has been estimated that for every

problem gambler, seven others are adversely affected (Productivity Commission, 1999).

Despite rapid expansion in problem gambling services in recent years, research suggests few

individuals seek treatment at specialist centres. Data from the Victorian Prevalence Study 2014

indicates that only 22 per cent of respondents identified as problem gamblers had sought help for

their gambling in the past 12 months (Hare, 2015). Common barriers to seeking treatment for

gambling problems include cost, stigma, availability and the perceived effectiveness of treatment

(Gainsbury, Hing, & Suhonen, 2014; Rockloff & Schofield, 2004). Poor uptake of treatment may be

a consequence of poor awareness, screening for problem gambling and underutilised/inefficient

referral systems.

Problem gambling often presents alongside mental health conditions (Lorains, Cowlishaw, &

Thomas, 2011). There is also evidence that among gamblers with comorbid psychiatric disorder

are elevated rates of problem gambling, impulsivity, psychosocial difficulties and suicide (Brown,

Oldenhof, Allen, & Dowling, 2016; Pietrzak & Petry, 2005; Stinchfield, Kushner, & Winters, 2005;

Waluk, Youssef, & Dowling, 2016), further highlighting the importance of early identification and

treatment and the need for tailored interventions and treatment models.

The most thorough examination of comorbid psychiatric disorders among treatment-seeking

problem gamblers to date is a recently completed systematic review and meta-analysis (Dowling,

Cowlishaw, et al., 2015). The findings based on 36 studies (including three Australian studies)

conducted between 1990 and 2011 were that approximately three-quarters of treatment seekers

display comorbid DSM-IV Axis I (clinical) disorders, with the most common being a current mood

disorder (23.1 per cent) and/or any substance use disorder (22.2 per cent). Specifically, the

prevalence estimates for current disorders were highest for nicotine dependence (56.4 per cent)

and major depressive disorder (29.9 per cent), with smaller estimates for alcohol abuse (18.2 per

cent), alcohol dependence (15.2 per cent), social phobia (14.9 per cent), generalised anxiety



disorder (14.4 per cent), panic disorder (13.7 per cent), and post-traumatic stress disorder (PTSD;

12.3 per cent). A meta-analytic study on DSM-IV Axis II (personality) disorders from the same

systematic search based on 15 studies meeting inclusion criteria (including two Australian studies),

found that almost half of those seeking treatment for problem and pathological gambling had

personality disorders, with narcissistic (16.6 per cent), antisocial (14.0 per cent), avoidant (13.4 per

cent), obsessive-compulsive (13.4 per cent) and borderline (13.1 per cent) personality disorders

being the most commonly identified (Dowling, Cowlishaw, et al., 2015). There is also growing

national and international evidence to suggest that people with gambling problems are over-

represented in primary care, alcohol and other drug (AOD) and mental health populations

(Cowlishaw, Merkouris, Chapman, & Radermacher, 2014; Dowling, Jackson, et al., 2014;

Goodyear-Smith et al., 2006).

Since only a minority of problem gamblers seek treatment, the early identification of those

experiencing gambling-related harm is paramount, and health care professionals have a critical

role to play. According to Berkson's bias (1946), individuals accessing treatment services are more

likely to have comorbid disorders and more severe problems than in the community, as the

compounding effect of both mental illness and addiction can precipitate treatment seeking. This

highlights the importance of routinely screening for gambling disorders among those seeking help

for mental health problems. Indeed, as noted by Miller (2014) people with mental health problems

are at increased risk of gambling harm, however, there is limited evidence on the prevalence of

problem gambling among patients attending mental services in Victoria.

An Australian study of veterans entering PTSD treatment programs reported that 28 per cent met

the criteria for probable pathological gambling (Biddle, Hawthorne, Forbes, & Coman, 2005). A

Victorian study, conducted at the Alfred Hospital, examined gambling problems among mental

health service users (de Castella et al., (2011). The study assessed 893 consecutive acute

psychiatric admissions over a six-month period, with staff asked to screen their caseloads using a

four-item problem gambling screening tool developed specifically for the study. Staff screened 290

patients (32.5 per cent), with 17.2 per cent of those screened reporting gambling problems

(although this included reports of gambling problems in someone close to the person being

assessed). Co-occurring mental health conditions identified were mainly anxiety, depression

(especially in women) and substance use problems (especially in men). Gambling problems were

quite severe, with 81 per cent of those completing a subsequent in-depth interview meeting criteria

for pathological gambling on the Massachusetts Gambling Screen (Shaffer, LaBrie, Scanlan, &

Cummings, 1994), yet only 47 per cent had spoken to their doctor about their gambling.

Nonetheless, 76 per cent reported that it was helpful to talk about their gambling, 38 per cent found

it helpful speaking with Gambler’s Help, while 19 per cent found Gamblers Anonymous to be

helpful. However, it is important to note that the exceptionally high prevalence of gambling

problems in this study may have been related to the proximity of the Victorian Casino to the

hospital catchment, as well as the screening tool used (given it also included asking about

gambling issues in key supports). Nevertheless, a recent study by Haydock, Cowlishaw, Harvey,

and Castle (2015) identified 5.8 per cent of patients with psychotic disorders attending two

community mental health services in Melbourne as problem gamblers. Despite these studies, it is

clear more research is needed across multiple mental health sites and settings to achieve a more

robust prevalence estimate for Victoria.

Unlike comorbid drug or alcohol problems which can be difficult to mask, problem gambling is often

hidden. This means the problem often remains undetected and untreated, until associated

problems (e.g. financial and relationship difficulties etc.) become overt. Indeed, studies suggest

that patients with mental health problems are more vulnerable to develop problem gambling as

they are frequently marginalised, stigmatised and isolated, with low levels of social support



(Johnstone, 2001; Overton & Medina, 2008). They are also more likely to be unemployed,

receiving government assistance and to have low incomes, and can struggle with impulsivity and

difficulties regulating their emotions (Gross & Jazaieri, 2014; Jazaieri, Urry, & Gross, 2013;

Kessler, Heeringa, et al., 2008; Kring & Sloan, 2010; Kring & Werner, 2004; Sareen, Afifi,

McMillan, & Asmundson, 2011; Stuart, 2006). As such, the compounding impact of a gambling

problem and its sequelae on individuals with existing mental health problems includes increased

psychiatric morbidity, poorer wellbeing and significant psychosocial disadvantage.

It is therefore essential that clinicians and services working with those who are at increased risk of

problem gambling, take steps to proactively identify gambling harms so that appropriate

assessment and treatment responses can be offered. Yet international research indicates that

screening for gambling problem rarely takes place in treatment settings (Sanju & Gerada, 2011; S.

A. Thomas, Piterman, & Jackson, 2008; Tolchard, Thomas, & Battersby, 2007), and the barriers to

and facilitators of problem gambling screening remain poorly understood. Few studies have

examined rates of screening among individuals accessing mental health care, with limited data

available on the skills, knowledge and attitudes of mental health clinicians towards problem

gambling, and the nature of any existing barriers or how they may be overcome. As such, more

research is needed to identify the scale of the problem within a local context, existing approaches,

the feasibility or barriers to routine screening and the training and support needs of services

working with this population. This information is fundamental to the planning of effective inter-

agency working relationships and treatment approaches that ensure the optimal and timely

delivery of care to this vulnerable population.

Overall aims and activities

The overarching aims of the project were:

to identify current clinician responses to problem gambling in Victorian mental health

services

to identify the prevalence of problem gambling and its relationship with comorbid

psychiatric disorders

to explore suitable problem gambling screening instruments and service responses.

To achieve these aims, five distinct activities were conducted:

An extensive literature review of national and international research on gambling and

comorbid psychiatric disorders, screening, assessment, treatment approaches and an

examination of existing service models/partnerships between gambling/addiction and

mental health services;

An organisation and workforce survey to examine attitudes towards problem gambling,

existing training opportunities, current screening practices, referral systems and treatment

approaches to problem gambling across a range of mental health treatment settings;

A prevalence estimate of problem gambling among people seeking treatment for mental

health issues across multiple settings;

Psychometric testing of gambling screens currently available for use in Australia, to identify

which are most effective; and,



Presentation of findings to a clinical reference panel, comprising senior clinicians,

managers and consumers across the Victorian mental health and gambling sectors, to

inform the development of recommendations for best practice, as well as potential

implementation challenges and approaches.



Study 1: Literature review

Introduction

Structure of this review

In this narrative review of the existing literature (which includes both peer-reviewed and grey

literature), we describe research findings relating to the screening, assessment, and management

of problem gambling within Victorian mental health services. In Section 1 of this review, we provide

background for this project by describing the prevalence of problem gambling in Victoria, the

profiles of Victorians who are at risk of developing gambling problems, and the help-seeking

behaviour of Victorians with gambling problems. In Section 2, we explore the prevalence of mental

health disorders in problem gambling populations and investigate the prevalence of problem

gambling in primary care, alcohol and other drug (AOD) services, and mental health populations.

In this section, we also explore the temporal relationship between problem gambling and mental

health disorders. In Section 3, we identify and evaluate brief screening instruments for problem

gambling and discuss the evidence base for the selection of the most appropriate screening

instruments within primary care, AOD, and mental health services. In Section 4, we describe

psychological and pharmacological treatments for problem gambling, with a particular emphasis on

interventions for comorbid problem gambling and mental health disorders. We conclude the

literature review in Section 5 with a description of treatment models for gambling and mental health

services, which draws from the more extensive dual diagnosis (AOD and mental health) literature.

Finally, we provide a brief summary of the literature review in Section 6.

Section 1: Background

The fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) has

reclassified pathological gambling as an addiction and related disorder, along with alcohol and

other drug use disorders, and renamed it gambling disorder (American Psychiatric Association,

2013). In Australia, however, the term problem gambling refers to all forms of gambling that lead to

adverse consequences for the gambler, others, or the community (Neal et al., 2005). Australian

estimates of gambling participation have dropped in the last decade, with the most recent national

gambling surveys identifying participation rates of 64 per cent. In contrast, rates of gambling

problems have plateaued, with national surveys using the Problem Gambling Severity Index

(PGSI) revealing estimates between 0.4 and 0.6 per cent for problem gambling, 1.9 and 3.7 per

cent for moderate-risk gambling, and 3.0 and 7.7 per cent for low-risk gambling (see Table 1).



Table 1. Recent Australian national surveys: Prevalence rates by gambling risk category

Study Sample size

Gambling participation

Problem gambling

Moderate-risk gambling

Low-risk gambling

To whom PGSI was administered PGSI response options

Sampling frame

(95% CI) (95% CI) (95% CI) (95% CI)

(Dowling, Youssef, et al., 2016)

2,000 63.9% (61.4–66.3)

0.4% (0.2–0.8)

1.9% (1.3–2.6)

3.0% (2.2–4.0)

Everyone who was over the age of 18 and had participated in at least one gambling activity in the past 12 months.

Standard options:

Never = 0, sometimes = 1, most of the time = 2, almost always = 3

Dual frame (50% landline, 50% mobile telephone)

(Gainsbury, Russell, Hing, Wood, & Blaszczynski, 2013; Gainsbury, Russell, et al., 2014)

15,006 64.3%

(no CI reported)

0.6%

(no CI reported)

3.7%

(no CI reported)

7.7%

(no CI reported)

All interactive gamblers and a random sample (13%) of non-interactive gamblers, except non-weekly lottery only and bingo only gamblers (N = 1768)

Standard options:

Never = 0, sometimes = 1, most of the time = 2, almost always = 3

Single frame



Gambling participation and problems in Victoria

The most recent gambling prevalence survey in the Australian state of Victoria, the 2014 Victorian

Prevalence Study (Hare, 2015), employed a dual-frame design. This methodology, which involves

sampling both landline and mobile telephone respondents, is primarily designed to capture

individuals who exclusively use mobile telephone (Dowling, Youssef, et al., 2016; Jackson,

Pennay, Dowling, Coles-Janess, & Christensen, 2014). The findings of the 2014 Victorian

Prevalence Study revealed a gambling participation rate of 70.1 per cent (69.9 per cent excluding

mobile telephone respondents), which is a significant decrease from the estimate of 73.1 per cent

in 2008. In this study, the most common forms of gambling were lotteries (46.9 per cent),

raffles/sweeps (41.6 per cent), horse, greyhound or harness racing (20.6 per cent), electronic

gaming machines (EGMs) (16.7 per cent), and scratch tickets (10.7 per cent). Smaller proportions

of the population participated in competitions where you pay money to enter (5.8 per cent), casino

table games such as blackjack, roulette, and poker (5.1 per cent), sports betting (4.8 per cent),

keno (3.7 per cent), informal private games for money (2.8 per cent), bingo (2.6 per cent), and

event betting (such as on elections and television shows) (0.5 per cent). Excluding mobile

telephone only respondents, these findings suggest that from 2008 to 2014, there has been a

significant decline in participation on EGMs, instant scratch tickets, and phone/SMS competitions,

but a significant increase in race, sports, and event betting.

The Victorian Prevalence Study 2014 revealed a problem gambling rate of 0.8 per cent, a

moderate-risk gambling rate of 2.8 per cent, and a low-risk gambling rate of 8.9 per cent. The

findings suggest that from 2008 to 2014, there has been a significant increase in low-risk gambling,

but no significant change in the proportion of moderate-risk or problem gambling.

Profiles of Victorian at-risk gamblers

In the Victorian Prevalence Study (Hare, 2015), people classified as problem gamblers were

significantly more likely to be from an Indigenous background and aged between 35 and 44 years

than people classified as non-problem gamblers. They were also significantly less likely to be

employed part-time and to be aged 65 years and older. The most common gambling activities in

which people classified as problem gamblers participated during the past year were lotteries (67.4

per cent), EGMs (66.6 per cent), and race betting (52.5 per cent). Compared to people classified

as non-problem gamblers, people classified as problem gamblers were significantly more likely to

bet on almost all gambling activities, including informal private games, EGMs, casino table games,

race betting, sports betting, event betting, keno, scratch tickets, and bingo. The top three highest-

spend gambling activities for these gamblers were EGMs (50.6 per cent), race betting (31.0 per

cent), and lotteries (9.2 per cent). The prevalence of regular gambling (weekly or more often) for

people classified as problem gamblers was 47.9 per cent for EGMs, 44.6 per cent for race betting,

22.5 per cent for lotteries, 13.6 per cent for casino table games, 12.6 per cent for sports betting,

and 4.4 per cent for keno.

People classified as moderate-risk gamblers in the 2014 Victorian Prevalence Study were

significantly less likely than people classified as non-problem gamblers to have an income of

$1,500 or more per week, to be aged between 25 and 34 years, and to be aged 55 to 65 years or

older. The most common gambling activities in which people classified as moderate-risk gamblers

participated during the previous year were lotteries (80.6 per cent), EGMs (58.5 per cent) and race

betting (53.0 per cent). Compared to people classified as non-problem gamblers, people classified

as moderate-risk gamblers were significantly more likely to bet on most gambling activities,

including informal private games, EGMs, casino table games, race betting, sports betting, event

betting, keno, lotteries, scratch tickets and bingo. The top three highest-spend gambling activities



of these gamblers were EGMs (38.7 per cent), lotteries (30.4 per cent), and race betting (12.1 per

cent). The prevalence of regular gambling (weekly or more often) for people classified as

moderate-risk gamblers was 38.6 per cent for lotteries, 23.9 per cent for EGMs, 15.6 per cent for

keno, 10.3 per cent for race betting, 5.3 per cent for sports betting, and 0.8 per cent for casino

table games.

In the 2014 Victorian Prevalence Study, people classified as low-risk gamblers were significantly

less likely than people classified as non-problem gamblers to have an income of $1500 or more

per week, have an income of $1000 to $1499 per week, and to own a home with a mortgage. The

most common gambling activities in which people classified as low-risk gamblers participated

during the past year were lotteries (68.9 per cent), raffles/sweeps (57.8 per cent), and racing (43.5

per cent). Compared to people classified as non-problem gamblers, people classified as low-risk

gamblers were significantly more likely to bet on most gambling activities, including informal private

games, EGMs, casino table games, race betting, sports betting, event betting, keno, scratch

tickets, bingo, and phone/SMS competitions. The top three highest-spend gambling activities of

these gamblers were lotteries (41.3 per cent), EGMs (17.7 per cent), and race betting (14.3 per

cent). The prevalence of regular gambling (weekly or more often) for people classified as low-risk

gamblers was 26.2 per cent for lotteries, 4.6 per cent for EGMs, 1.3 per cent for keno, 6.2 per cent

for race betting, 1.8 per cent for sports betting, and 0.1 per cent for casino table games.

Help-seeking by people with gambling problems in Victoria

The 2014 Victorian Prevalence Study revealed that 41.6 per cent of people classified as problem

gamblers and 3.1 per cent of people classified as moderate-risk gamblers positively endorsed a

question relating to whether they had experienced problems because of their gambling in the

previous year. A significant proportion of people classified as problem gamblers (71.5 per cent)

and people classified as moderate-risk gamblers (18.3 per cent) described these problems as

serious or moderate. Despite experiencing these gambling-related harms, only 44.2 per cent of

people classified as problem gamblers and 6.6 per cent of people classified as moderate-risk

gamblers reported they had ever sought help (formal or informal) for their gambling problem.

Moreover, only 22.1 per cent of people classified as problem gamblers and 1.9 per cent of people

classified as moderate-risk gamblers had sought help in the previous 12 months. The main triggers

for seeking help were financial problems, feeling depressed or worried, and relationship problems.

The main type of help sought included face-to-face counselling, telephone counselling, and peer or

friendship support. Approximately two-thirds who had sought help rated the help as useful. People

classified as moderate-risk or problem gamblers who did not seek help did not do so because they

believed they could solve the gambling problem independently, thought seeking help was

inconvenient, did not think it was a serious enough problem, or were too embarrassed or shy.

While not yet examined in the Victorian context, international literature suggests there is also a

delay in accessing treatment (Petry, 2002; Séguin et al., 2013).

Potential for screening for problem gambling in Victorian mental

health services

Although the above findings confirm that Victorians with gambling problems display relatively low

rates of help-seeking for problem gambling, the findings of the 2014 Victorian Prevalence Study

suggests that there may be the potential for general practitioners (GPs) and other health

professionals to play a role in the identification and management of gambling problems. On

average, people classified as problem or moderate-risk gamblers attended their GP seven to eight

times per year, suggesting that people within both risk categories have higher levels of contact with

GPs relative to the Australian population (statistics from Medicare indicate that Australians visit a



GP approximately 5.6 times per year). People classified within both risk categories also displayed

an elevated likelihood of being diagnosed with affective disorders, with 41.9 per cent of people

classified as problem gamblers and 24.1 per cent of people classified as moderate-risk gamblers

positively endorsing a question that they had ever been diagnosed by a medical professional with

depression; and 39.5 per cent of people classified as problem gamblers and 20.0 per cent of

people classified as moderate risk positively endorsing a question that they had ever been

diagnosed by a medical professional with an anxiety disorder. Of those gamblers who consumed

alcohol while gambling, 31.3 per cent of people classified as problem gamblers, 40.7 per cent of

people classified as moderate-risk gamblers, and 19.1 per cent of people classified as low-risk

gamblers showed signs of either being at risk for an alcohol use disorder or showed signs of more

significant problem gambling (measured using the four-item CAGE Screen). There was also

evidence that people classified as both problem and moderate-risk gamblers experience low

quality of life, with 23.3 per cent of people classified as problem gamblers and 10.6 per cent of

people classified as moderate-risk gamblers describing their overall quality of life as ‘poor’ or ‘very

poor’ on an item in which they rated their quality of life in the past four weeks.

Conclusion

The findings of the 2014 Victorian Prevalence Study revealed a gambling participation rate of 61.6

per cent, a problem gambling rate of 0.8 per cent, a moderate-risk gambling rate of 2.8 per cent,

and a low-risk gambling rate of 8.9 per cent. Despite experiencing considerable gambling harms,

there are relatively low rates of help-seeking for gambling problems in the Victorian population.

People experiencing gambling problems, however, frequently attended their GP, suggesting that

there may be the potential for health professionals to play a role in the identification and

management of gambling problems. Victorians classified as problem or moderate-risk gamblers

also displayed an elevated likelihood of being diagnosed by a medical professional with depression

and anxiety disorders, and also reported high rates of alcohol abuse. In Section 2, we expand on

the prevalence of mental health disorders in problem gambling populations. In this section, we also

investigate the prevalence of problem gambling in primary care, alcohol and other drug (AOD)

services, and mental health populations, and explore the temporal relationship between problem

gambling and mental health disorders.



Section 2: Problem gambling and mental health

disorders

Prevalence of mental health disorders in problem gambling

populations

Internationally, the co-occurrence between problem gambling and comorbid mental health

conditions has been empirically examined in both epidemiological and clinical samples. By

randomly sampling the general population, epidemiological studies provide results that are

representative of people with gambling problems in the community. They are, however, generally

only able to determine the co-occurrence of problem gambling with highly prevalent comorbid

disorders because they yield very small numbers of people with gambling problems (Lorains et al.,

2011; Westphal & Johnson, 2007). Lorains et al. (2011) conducted a systematic review and meta-

analysis of comorbid disorders in population-representative samples of people with gambling

problems. These analyses revealed high rates of mental health disorders, including nicotine

dependence (60.1 per cent), alcohol and other drug use disorders (57.5 per cent), mood disorders

(37.9 per cent) and anxiety disorders (37.4 per cent, see Table 2).

Although population-representative studies provide important information about the prevalence of

psychiatric comorbidity in people with gambling problems living in the community, caution is

required in generalising results from these samples to treatment-seeking populations. There is

emerging evidence that people with gambling problems seeking treatment display more severe

gambling problems and a greater variety and severity of comorbid mental health disorders

compared with their non-treatment seeking counterparts (Crockford & el-Guebaly, 1998; Slutske,

2006; Specker, Carlson, Edmonson, Johnson, & Marcotte, 1996). Although they generally employ

non-random samples, treatment-seeking studies generally comprise larger samples of people with

gambling problems, which allows for the investigation of lower prevalence comorbid disorders and

an investigation of how the level of gambling severity co-varies with comorbid disorders (Westphal

& Johnson, 2007).

Table 2. Prevalence of comorbid mental health disorders in community-representative

samples of people with gambling problems

Disorder No. of estimates

Lowest estimate (%)

Highest estimate (%)

Summary effect

(%)

Any alcohol/other drug use disorder 3 26.0 76.3 57.5

Alcohol use disorder 8 9.9 73.2 28.1

Illicit drug abuse/dependence 3 1.6 39.9 17.2

Nicotine dependence 4 34.9 76.3 60.1

Any mood disorder 3 11.6 55.6 37.9

Major depression 6 8.8 38.6 23.1

Bipolar disorder/manic episodes 6 0 32.5 9.8

Any anxiety disorder 3 14.0 60.3 37.4

Generalised anxiety disorder 3 7.7 16.6 11.1

Antisocial personality disorder 2 23.3 35.0 28.8



A systematic review and meta-analysis (Dowling, Cowlishaw, et al., 2015) indicates that almost

three-quarters (74.8 per cent) of treatment-seeking gamblers report comorbid psychiatric disorders

(formerly known in the DSM-IV as Axis I disorders, see Table 3). Although there are a broad range

of estimates for many disorders, the summary estimates suggest that there are high rates of mood

disorders (23.1 per cent), alcohol use disorders (21.2 per cent), anxiety disorders (17.6 per cent),

and other drug use disorders (7.0 per cent) in these samples. Specifically, the most common

psychiatric disorders were nicotine dependence (56.4 per cent) and major depressive disorder

(29.9 per cent), with smaller proportions of treatment-seeking gamblers reporting alcohol abuse

(18.2 per cent), alcohol dependence (15.2 per cent), social phobia (14.9 per cent), generalised

anxiety disorder (14.4 per cent), panic disorder (13.7 per cent), posttraumatic stress disorder

(PTSD: 12.3 per cent), cannabis use disorder (11.5 per cent), attention deficit/hyperactivity

disorder (ADHD: 9.3 per cent), adjustment disorder (0.2 per cent), bipolar disorder (8.8 per cent),

and obsessive-compulsive disorder (8.2 per cent).

Table 3. Prevalence of comorbid current psychiatric disorders in people with gambling

problems seeking treatment


Lowest estimate (%)


Summary effect

(%)

Any Axis I disorder 5 21.0 100 74.8

Any alcohol or other drug use disorder

10 7.5 47.8 22.2

Any alcohol use disorder 12 5.2 38.1 21.2

Alcohol abuse 9 10.9 32.7 18.2

Alcohol dependence 7 4.3 32.5 15.2

Any other drug use disorder

7 0 54.8 7.0

Other drug abuse 8 0 24.8 6.6

Other drug dependence 6 0 12.1 4.2

Nicotine dependence 3 37.3 68.6 56.4

Cannabis use disorder 3 0 19.0 11.5

Any mood disorder 10 4.8 83.0 23.1

Major depressive disorder 17 10.4 82.5 29.9

Dysthymic disorder 3 5.6 7.5 6.7

Bipolar disorder 10 0 66.6 8.8

Any anxiety disorder 10 4.3 94.4 17.6

Obsessive-compulsive disorder

7 2.1 50.0 8.2

Panic disorder 6 3.8 38.9 13.7

Generalised anxiety disorder

4 3.8 50.0 14.4

PTSD 4 5.0 34.2 12.3

Social phobia 3 5.0 50.0 14.9

Intermittent explosive disorder

3 2.1 7.5 4.6

Kleptomania 3 1.9 5.0 2.7

Psychotic disorder 5 0 6.0 4.7




Lowest estimate (%)


Summary effect

(%)

Somatoform disorder 5 0 10.0 3.6

Adjustment disorder 5 0 17.4 9.2

ADHD 4 1.3 20.0 9.3

Moreover, almost one half (47.9 per cent) of treatment-seeking gamblers displayed comorbid

personality disorders (see Table 4) (Dowling, Cowlishaw, et al., 2014). They were most likely to

display Cluster B disorders (characterised as dramatic, erratic, or emotional personality disorders:

17.6 per cent), with smaller proportions reporting Cluster C disorders (characterised as anxious or

fearful personality disorders: 12.6 per cent) and Cluster A disorders (characterised as odd or

eccentric personality disorders: 6.1 per cent). The most prevalent personality disorders were

narcissistic (16.6 per cent), antisocial (14.0 per cent), avoidant (13.4 per cent), obsessive-

compulsive (13.4 per cent), and borderline (13.1 per cent) personality disorders. The prevalence

estimates presented in both of these meta-analyses were consistently robust to the inclusion of

clinical trials and self-selected samples. Although there was significant variability in reported rates,

there were no consistent patterns according to gambling problem severity, measure of comorbidity

employed, type of treatment facility and study jurisdiction.

Table 4. Prevalence of comorbid current personality disorders in people with gambling

problems seeking treatment


Lowest estimate (%)


Summary effect

(%)

Any personality disorder 9 12.4 93.0 47.9

Any Cluster A disorder 4 2.2 24.0 6.1

Paranoid personality disorder

8 2.5 40.2 10.1

Schizoid personality disorder 8 0 20.7 6.0

Schizotypal personality disorder

7 0 37.8 4.1

Any Cluster B disorder 4 6.5 42.0 17.6

Antisocial personality disorder

14 0 29.3 14.0

Borderline personality disorder

8 0 69.5 13.1

Histrionic personality disorder

7 0 65.9 6.3

Narcissistic personality disorder

8 5.0 57.3 16.6

Any Cluster C disorder 4 3.7 27.0 12.6

Avoidant personality disorder

6 0 36.6 13.4

Dependent personality disorder

8 0 48.8 6.0

Obsessive-compulsive personality disorder

6 0 37.5 13.4



Prevalence of problem gambling in mental health populations

In addition to high rates of mental health disorders in problem gambling populations, there is also

growing international evidence to suggest that people with gambling problems are over-

represented in primary care, AOD, and mental health populations. Rates of past year problem

gambling (including pathological gambling) range from 3.0 per cent to 15.7 per cent in primary care

settings (Goodyear-Smith et al., 2006; Levens, Dyer, Zubritsky, Knott, & Oslin, 2005; Morasco,

Vom Eigen, & Petry, 2006; Pasternak & Fleming, 1999) and up to 26.0 per cent in university

medical and dental clinics (Ladd & Petry, 2002). A systematic review and meta-analysis of the

prevalence of gambling problems in AOD treatment services (Cowlishaw et al., 2014) found that

the prevalence of pathological gambling ranged from 2.8 per cent to 52.7 per cent, with a mean

prevalence of 13.7 per cent. Moreover, the prevalence of problem gambling (including pathological

gambling) in these services ranged from 10.0 to 43.4 per cent, with a mean prevalence of 22.8 per

cent. Rates of past year and lifetime problem and/or pathological gambling are also elevated in a

range of other mental health populations (see Table 5).

Table 5. Prevalence of problem gambling in mental health populations

Mental health population

Lifetime problem gambling

prevalence3

Past year problem gambling prevalence3

Lifetime pathological gambling prevalence

Past year pathological gambling prevalence

Study

Psychiatric outpatient

4.0% 2.0–4.4% 2.0–2.3% 0.7% Dowling, Jackson, et al., 2014; Henderson, 2004; Nehlin, Gronbladh, Fredriksson, & Jansson, 2013; Zimmerman, Chelminski, & Young, 2006a, 2006b

Psychiatric inpatient

9.0% 6.7–8.9% 6.9% Aragay et al., 2012; Grant, Levine, Kim & Potenza, 2005; Lesieur & Blume, 1990

Any affective disorder

3.1% 1.4% Cowlishaw, Hakes, & Dowling, 2016

Any mood disorder

2.2% 3.2–5.4% 2.2–10.0% Aragay et al., 2012; Kennedy et al., 2010; McIntyre et al., 2007; Quilty, Watson, Robinson, Toneatto, & Bagby, 2011

Major depressive disorder

3.1% 0.9%–6.0% 2.8–12.0% Cowlishaw et al., 2016; Jones et al., 2015; Kennedy et al., 2010; Lejoyeux, Arbaretaz, McLoughlin, & Ades, 2002; McIntyre et al., 2007; Quilty et al., 2011

Dysthymia 3.6% 0.9% Cowlishaw et al., 2016

Bipolar disorders

4.6% 2.3–6.3% 3.2–9.0% Cowlishaw et al., 2016; Di Nicola et al., 2010; Jones et al., 2015; Karakus & Tamam, 2011; Kennedy et

3 including pathological gambling



Mental health population

Lifetime problem gambling

prevalence3

Past year problem gambling prevalence3

Lifetime pathological gambling prevalence

Past year pathological gambling prevalence

Study

al., 2010; McIntyre et al., 2007; Quilty et al., 2011

PTSD 29.1% 17.0% Biddle et al., 2005

Panic disorder

3.1% 1.2% Cowlishaw et al., 2016

Generalised anxiety disorder


Specific phobia


Social phobia


Psychotic disorders

17.1% 5.8% 14.3% Aragay et al., 2012; Haydock et al., 2015

The prevalence estimates of problem and/or pathological gambling are highly variable across

included studies, which is likely due to a number of methodological considerations, such as

differences in problem gambling measurement instruments employed (e.g. structured clinical

interview versus self-report), characteristics of the samples evaluated (treatment-seeking versus

non-treatment seeking), the severity of gambling problems measured (pathological or problem

gambling), the settings in which the studies were conducted (e.g. inpatient versus outpatient), the

timeframe assessed (e.g. lifetime versus past-year prevalence), the range and severity of mental

health disorders included in the sample, recruitment methodology (consecutive versus non-

consecutive recruitment), method of data collection (e.g. clinician versus researcher), and

jurisdictions in which the studies were conducted.

The temporal relationship between problem gambling and mental

health conditions

Although the findings of the studies presented in this section suggest that problem gambling is

comorbid with many mental health disorders, the cross-sectional nature of these studies precludes

an explication of the temporal order between these disorders. It may be that some mental health

disorders are risk factors for problem gambling, while others might be consequences of problem

gambling. Inferences regarding the temporal nature of problem gambling and comorbid mental

health conditions are provided by studies exploring the onset and pattern of these disorders.

Kessler, Hwang, et al. (2008) explored the possible temporal relationship between problem

gambling and several psychiatric disorders using retrospective age of onset data in a large

nationally representative sample of the US population. Although there were some exceptions,

these data suggest that the clear majority of psychiatric disorders typically predate the onset of

problem gambling. In 74.3 per cent of cases where the respondent with problem gambling met

criteria for another lifetime disorder, at least one of these disorders began at an earlier age than

the gambling problem. Overall, most respondents reported that anxiety disorders (82 per cent),

mood disorders (65 per cent), and alcohol or other drug use disorders (57 per cent) began at an

earlier age than problem gambling. Specifically, panic disorder (82 per cent), generalised anxiety

disorder (80 per cent), major depressive disorder or dysthymia (74 per cent), alcohol or other drug

abuse (71 per cent), alcohol or other drug dependence (56 per cent), and bipolar disorder (46 per



cent) predated the onset of problem gambling. Moreover, all phobias and comorbid impulse control

disorders (ADHD, oppositional defiant disorder, conduct disorder, and intermittent explosive

disorder) began at an earlier age than the problem gambling. The only exceptions were PTSD and

nicotine dependence. An equal number of respondents indicated that the onset of PTSD predated

the onset of problem gambling and the onset of problem gambling predated the onset of PTSD. In

contrast, nearly two-thirds of respondents (61 per cent) reported that they were dependent on

nicotine after developing gambling problems. Age of onset data was also reported by Hodgins,

Peden, and Cassidy (2005) in a naturalistic sample of pathological gamblers who had recently quit

gambling. The findings suggest that major depressive disorder was equally likely to precede (40

per cent) or follow (44 per cent) the age of onset of problem gambling. Both alcohol (74 per cent)

and other drug (84 per cent) disorders were more likely to precede the age of onset of problem

gambling.

Kessler, Hwang, et al. (2008) conducted two parallel survival analyses using the retrospective age

of onset data to explore the possibility that some mental health disorders might be risk factors for

problem gambling, while others might be consequences of problem gambling. The first analysis

employed information about temporally primary mental health disorders to predict the subsequent

onset of problem gambling, while the second analysis employed information about temporally

primary problem gambling to predict the subsequent onset of mental health disorders. Although

there were significant time-lagged predictive associations for problem gambling predicting the

subsequent onset of other mental health disorders, there were many more associations for other

disorders predicting the subsequent onset of problem gambling. Specifically, the onset of problem

gambling was predicted by alcohol or other drug abuse, alcohol or other drug dependence, major

depressive disorder or dysthymia, bipolar disorder, panic disorder, generalised anxiety disorder,

phobia, oppositional defiant disorder, conduct disorder, and intermittent explosive disorder. In

contrast, the onset of problem gambling only significantly predicted the onset of bipolar disorder,

phobia, PTSD, alcohol or other drug dependence, and nicotine dependence. The authors note

several asymmetries in these associations, whereby problem gambling did not predict panic

disorder, generalised anxiety disorder, or intermittent explosive disorder, while all of these

disorders predicted problem gambling. In contrast, problem gambling predicted PTSD and nicotine

dependence, but these disorders did not predict problem gambling.

Similar cross-lagged analyses were conducted by Quilty, Watson, Robinson, Toneatto, and Bagby

(2011) in psychiatric samples with a history of bipolar disorder and depressive disorder to examine

the longitudinal association of problem gambling and mood disorder symptoms according to

retrospective report. The analyses exploring the associations between problem gambling and

mood disorder symptom severity for each month across the most recent four months revealed that

monthly mood symptoms did not significantly predict problem gambling symptoms the subsequent

month. Moreover, monthly problem gambling symptoms did not significantly predict mood

symptoms the subsequent month. This pattern of non-significant findings were identified for each

year across the most recent four years. These findings suggest that when the stability of symptom

severity is accounted for, any direct association between problem gambling and mood symptoms

is eliminated.

Although retrospective age of onset data is helpful in exploring the possible temporal or causal

relationships between problem gambling and comorbid mental health conditions, they are limited

by a reliance on retrospective study designs that may introduce recall and reporting biases.

Moreover, the issue of causality is confounded in these studies by the fact that some disorders

naturally have an earlier age of onset. It has been long established that the ‘gold standard’ for

determining the course of a clinical phenomenon, such as problem gambling, is a longitudinal-

based approach. Longitudinal surveys have become highly valued by researchers and policy



makers for their ability to provide insights that cannot be obtained by any other means. There is an

emerging literature of prospective and longitudinal research on the determinants of problem

gambling. A systematic review and meta-analysis exploring the early risk and protective factors

that are longitudinally associated with the development of gambling problems (Dowling, Merkouris,

et al., 2015) reveals that alcohol use frequency (two studies), cannabis use (two studies), illicit

drug use (five studies), tobacco use (two studies), and depressive symptoms (six studies) were

positively associated with subsequent problem gambling, with small but significant moderate effect

sizes. In contrast, anxiety symptoms (two studies) were not predictive of subsequent problem

gambling.

There is also limited evidence that problem gambling is a risk factor for the subsequent occurrence

of mental health disorders. In a large nationally representative US sample, Chou and Afifi (2011)

found that problem gambling only predicted the incidence of some Axis I disorders three years

later, after adjusting for socio-demographic characteristics. These included any psychiatric disorder

(formerly Axis I disorders), any mood disorder, any drug use disorder, and specifically, bipolar

disorder, generalised anxiety disorder, PTSD, alcohol use disorder, and alcohol dependence

disorder. After controlling for medical conditions, health-related quality of life and recent stressful

life events, however, problem gambling only significantly predicted the subsequent onset of any

mood disorder, generalised anxiety disorder, PTSD, alcohol use disorders, and alcohol

dependence.

Expanding these findings, Parhami, Mojtabai, Rosenthal, Afifi, and Fong (2014) found a graded or

dose–response relationship between different levels of problem gambling and the subsequent

onset of several disorders (any mood disorder, any anxiety disorder, any drug use disorder,

specific phobia, PTSD, any alcohol use disorder, and any other drug [non-nicotine] disorder) after

controlling for socio-demographic characteristics. In addition, problem gambling was associated

with the onset of major depressive episodes, dysthymia, hypomanic episodes, generalised anxiety

disorder, and social phobia. Using the same dataset to explore the predictive ability of problem

gambling on Axis I psychopathology in older adults (aged 55 to 90 years), Pilver, Libby, Hoff, and

Potenza (2013) found that relative to low-frequency gambling/non-gambling, at-

risk/problem/pathological gambling was positively associated with the incidence of any Axis I

disorder, mania, panic disorder, any drug use disorder, alcohol abuse/dependence, and nicotine

dependence. Only the associations with generalised anxiety disorder and any drug use disorder

remained significant, however, after controlling for demographic characteristics, psychiatric

comorbidity, health behaviours, physical health, and stressful life events. There is also some

evidence from Dussault, Brendgen, Vitaro, Wanner, and Tremblay (2011) that gambling problems

at age 17 predicted an increase in depressive symptoms from age 17 to age 23.

Taken together, the findings of these age of onset and longitudinal studies suggest that the clear

majority of alcohol and drug use, mood, anxiety, and impulse control disorders typically predate

and predict the onset of problem gambling. There is evidence that alcohol and other drug use

disorders (alcohol or other drug abuse, alcohol or other drug dependence, as well as use of

cannabis, illicit drugs, and tobacco), mood disorders (major depressive disorder, dysthymia, bipolar

disorder), and anxiety disorders (panic disorder, generalised anxiety disorder, phobia), and

impulse control disorders (ADHD, oppositional defiant disorder, conduct disorder, and intermittent

explosive disorder) are risk factors for the development of problem gambling, although there is

some equivocal data for mood and anxiety disorders. There is, however, also some evidence that

some disorders, such as PTSD and nicotine dependence, typically occur after the development of

problem gambling (although the results for PTSD were equivocal). Moreover, at least one other

psychiatric disorder began at a later age than the gambling problem in approximately one-quarter

of cases. There is also evidence that problem gambling is a risk factor for the development of



alcohol and other drug use disorders (alcohol use disorder, alcohol dependence, other drug

dependence, nicotine dependence), mood disorders (major depression, bipolar disorder,

hypomania), anxiety disorders (panic disorder, specific phobia, social phobia, PTSD, generalised

anxiety disorder). Some, but not all, of these associations were attenuated after controlling for

psychiatric comorbidity, health behaviours, physical health, medical conditions, health-related

quality of life and stressful life events.

Conclusion

Problem gambling has been consistently associated with a range of comorbid mental health

disorders, including alcohol and other drug use disorders, mood disorders, anxiety disorders,

impulse control disorders, and personality disorders. There is also growing evidence to suggest

that people with gambling problems are over-represented in primary care, AOD settings, and

mental health populations, with up to 30 per cent of people presenting to these services

experiencing problems with their gambling. Moreover, evidence that problem gambling precedes

and predicts the onset of other mental health conditions suggests that it has the potential to

complicate treatment plans and hamper treatment outcomes for mental health treatment,

particularly if it goes unidentified and untreated (Brett et al., 2014; Chou & Afifi, 2011). These

findings highlight the importance of identifying problem gambling through routine screening within

primary care, AOD, and mental health settings, with a view to appropriate management and/or

referral to specialist gambling services. In Section 3 of this review, we will discuss research

findings relating to screening for problem gambling, with particular emphasis on screening within

primary care, AOD, and mental health services.

Section 3: Screening and assessment of problem

gambling within primary care, AOD and mental health

services

Routine screening has the potential to improve care and reduce healthcare costs (Tiet, Finney, &

Moos, 2008). Primary care, AOD, and mental health services are logistically well placed to identify

problem gambling as they are frequently attended by individuals experiencing gambling problems

(Billi, Stone, Marden, & Yeung, 2014). These services may therefore enhance the identification of

people with gambling problems and provide appropriate referral or generalist first level

interventions for problem gambling (Brett et al., 2014; Goodyear-Smith, Arroll, & Coupe, 2009;

Rockloff, Ehrich, Themessl-Huber, & Evans, 2011; Sullivan, McCormick, Lamont, & Penfold,

2007). In Section 2, we highlighted the high rates of comorbid problem gambling in these services

and provided evidence that problem gambling can lead to incident and persistent psychiatric

disorders and that these significant associations cannot be explained by a range of other factors,

such as psychiatric comorbidity, health behaviours, physical health, medical conditions, health-

related quality of life and stressful life events. Taken together, these findings suggest that comorbid

problem gambling has the potential to compromise engagement in, and the effectiveness of, AOD

and mental health treatment, particularly if it goes unidentified and unmanaged (Brett et al., 2014;

Chou & Afifi, 2011; Himelhoch et al., 2015). It is therefore important for health service providers to

accurately screen for problem gambling so that people with gambling problems in these services

can be identified and offered appropriate management or referral. This conclusion is supported by

a consensus-based recommendation that ‘screening could be used in primary care settings where

at risk patients may be presenting for services. These may include people who present for other

mental health problems’ in the Australian National Health and Medical Research Council

(NHMRC)-endorsed clinical guideline (p. 55). In this section, we explore the prevalence of



screening and barriers to screening within primary care, AOD, and mental health services. We

then identify brief screening instruments for possible use within these services.

Prevalence of screening for problem gambling within primary care,

AOD and mental health services

Although health providers acknowledge that they have a role to play in helping patients with

gambling problems (Corney, 2011; Sanju & Gerada, 2011; Sullivan, Arroll, Coster, Abbott, &

Adams, 2000; Sullivan et al., 2007; Temcheff, Derevensky, St-Pierre, Gupta, & Martin, 2014;

Tolchard et al., 2007), their opinions about what this role involves vary substantially. For example,

a study of GPs in New Zealand found that while they believe that intervening in patients with

gambling problems is within their role, they were less supportive of screening, with the majority of

GPs indicating that they have no opinion on the matter (Sullivan et al., 2000). Few studies have

explored the frequency with which primary care, AOD, and mental health clinicians screen for

problem gambling. In a sample of 71 Swiss GPs, Achab et al. (2014) found that none

systematically screened for excessive gambling, 7 per cent often screened for excessive gambling,

and the remainder of the sample rarely or never screened for excessive gambling.

Barriers to screening for problem gambling within primary care,

AOD and mental health services

Limited research has examined barriers to screening and intervening in gambling problems within

primary care, AOD and mental health services. With the exception of one study that explored the

beliefs and attitudes of mental health professionals with respect to gambling and other high risk

behaviours in schools (Temcheff et al., 2014), this research has focused on GPs in primary care

settings (Achab et al., 2014; Corney, 2011; Rowan & Galasso, 2000; Sullivan, 2011; Sullivan et al.,

2000; Sullivan et al., 2007), including opinion pieces highlighting the need for GPs to have a

greater involvement in screening and intervening in patients with gambling problems

(McCambridge & Cunningham, 2007; Sanju & Gerada, 2011; S. A. Thomas et al., 2008). These

studies have identified several barriers to screening and intervening in patients with gambling

problems within these services. These include:

Lack of time: Lack of time is identified as the greatest impediment to intervening in

gambling problems (McCambridge & Cunningham, 2007; Sullivan et al., 2007).

Effectiveness of screening: Gambling problems are unlikely to be prioritised for

screening by health professionals as there is a lack of evidence that screening will reduce

their associated morbidity and mortality, increase service engagement, or improve

treatment outcomes (Problem Gambling Research and Treatment Centre, 2011; Rowan &

Galasso, 2000; Sullivan, 2011).

Training, knowledge and skills: Although there are some practice guidelines for the

screening and treatment of problem gambling available (Problem Gambling Research and

Treatment Centre, 2011), many health professionals lack confidence that they have the

necessary training, knowledge or skills to address gambling problems (Achab et al., 2014;

Corney, 2011; Problem Gambling Research and Treatment Centre, 2011; Rowan &

Galasso, 2000; Sullivan, 2011; Sullivan et al., 2000; Temcheff et al., 2014; Tolchard et al.,

2007).



Stigma: Health professionals may hold a stigma about problem gambling, whereby they

perceive that gambling problems result from a lack of self-control, as compared to alcohol

and other drug use disorders, which involve the influencing presence of a psychoactive

substance (Sullivan, 2011).

Prevalence and burden of disease: It has been argued that health professionals will be

unmotivated to screen for problem gambling if they perceive problem gambling is a

relatively rare disorder or that it has a relatively low burden of disease (Sullivan, 2011).

Appropriate interventions: A barrier to screening and intervening with problem gambling

may be the absence of appropriate interventions that have recognised effectiveness

(Sullivan, 2011).

Funding and resource constraints: There is generally no funding available to

compensate for the additional time required to screen, assess and manage gambling

problems (McCambridge & Cunningham, 2007; Sullivan, 2011).

Specialist referral services: Access to and confidence in specialist referral services may

also affect whether those employed in primary care, AOD and mental health services

screen for problem gambling (Sullivan, 2011).

Another major impediment to the identification and management of gambling problems in primary

care, AOD and mental health services may have been the relatively slow development of brief

screening instruments. It is essential that screening instruments employed in these clinical settings

are easy to use, have short administration times, are easily scored, and require minimal training

due to limited time and money for screening (Brett et al., 2014; Goodyear-Smith et al., 2008;

Rockloff et al., 2011; Stinchfield & McCready, 2014; Toce-Gerstein, Gerstein, & Volberg, 2009; R.

A. Volberg, Munck, & Petry, 2011). Consistent with the literature in the broader addiction field, one-

to five-item instruments are considered brief in this review because they can be administered in a

short period of time (such as one to two minutes) (Stinchfield & McCready, 2014).

Brief screening instruments for problem gambling

Although slow to develop, several brief screening instruments are now available to screen for

lifetime or current problem gambling among adults or adolescents in population or clinical samples.

These include the one-item screen (S. A. Thomas, Jackson, Browning, & Piterman, 2010), Lie/Bet

Questionnaire (Johnson, Hamer, & Nora, 1997), the Case Finding and Help Assessment Tool

(CHAT) (Goodyear-Smith et al., 2008), the Brief Problem Gambling Screen (BPGS) (R. A. Volberg

& Williams, 2011), the Problem Gambling Severity Index (PGSI) Short Form (R. A. Volberg &

Williams, 2012), the National Opinion Research Center Diagnostic Screen for Gambling Disorders

– Loss of Control, Lying and Preoccupation (NODS-CLiP) (Toce-Gerstein et al., 2009), the Brief

Biosocial Gambling Screen (BBGS) (Gebauer, LaBrie, & Shaffer, 2010), the Consumption Screen

for Problem Gambling (CSPG) (Rockloff, 2012), the National Lottery screen – Loss of Control,

Lying and Preoccupation (NLCLiP) (Lepper & Haden, 2013), the National Opinion Research

Centre Diagnostic Screen for Gambling Disorders – Preoccupation, Escape, Chasing and Risked

Relationships (NODS-PERC) (R. A. Volberg et al., 2011), the National Opinion Research Center

Diagnostic Screen for Gambling Disorders – Loss of Control, Lying and Preoccupation 2 (NODS-

CLiP2) (Volberg et al., 2011) and the Short South Oaks Gambling (SOGS) Screen (Room, Turner,

& Ialomiteanu, 1999). Most of these instruments were derived from DSM-IV criteria or from existing

assessment instruments based on DSM-IV diagnostic criteria and most have been developed and



validated in North American samples. In Table 6, we provide a summary description of each of the

available brief screening instruments.

The selection of a brief screening instrument is generally determined by the classification accuracy

of the instruments within the relevant setting. Classification accuracy, which refers to how well the

instrument identifies those with and without gambling problems, can be assessed using a number

of coefficients, including sensitivity (the true positive rate, i.e., the proportion of positive test results

among those with the disorder) and specificity (the true negative rate, i.e., the proportion of

negative test results among those without the disorder). The available evidence (see Table 6)

suggests that several brief screening instruments, including the Lie/Bet, NODS-CLiP, NODS-

CLiP2, BBGS, and CSPG display good classification accuracy in community samples. Several

brief screening instruments, such as the Lie/Bet, NODS-CLiP, BBGS, and one-item screen, also

display good psychometric properties in discriminating between problem gambling treatment and

community samples. Although the performance of many of these instruments was assessed in

relation to the longer screens from which they are derived, the available evidence appears to

support their continued use in these contexts.



Table 6: Summary of the brief screening instruments for problem gambling

Brief screening instrument

Development information

No. of items

Timeframe Items Scoring information Sample type Sensitivity Specificity References

One item screen Developed in Australia for use in medical practice but was validated in an adult age- and sex-representative community sample.

1 Lifetime but has been converted to past year (Rockloff et al., 2011; Stinchfield & McCready, 2014

Have you ever had an issue with your gambling?

The response options for the item are dichotomous (yes/no) and the original screening instrument employs a lifetime timeframe.

Adult community samples

0.21–0.79 0.96–0.98 Thomas et al., 2010; Rockloff et al., 2011

Discrimination between adult problem gambling treatment and community samples

0.75–0.99 0.87–1.00 Stinchfield & McCready, 2014

Lie/Bet Questionnaire

Items from a 12-item questionnaire measuring the DSM-IV diagnostic criteria that best discriminated between male Gamblers Anonymous members and non-problem gambling controls (Veterans Administration Medical Centre employees).

2 Lifetime (1) Have you ever had to lie to people important to you about how much you gambled?

(2) Have you ever felt the need to bet more and more money?

Response options: (1) Yes, (0) No. Positive endorsement of one or both items is indicative of problem gambling.


0.92–0.96 0.89–0.97 (Colasante et al., 2013; Gotestam, Johansson, Wenzel, & Simonsen, 2004)


0.86–1.00 0.78–1.00 Stinchfield & McCready, 2014; Johson et al., 1997, 1998





No. of items


Adolescent community/school samples

0.65–0.93 0.85–0.94 Gotestam et al., 2004; Rossow & Molde, 2006

Brief Problem Gambling Screen (BPGS) (2 item)

Developed for the purpose of screening for problem gambling in both clinical settings and population research. Best-performing 2 items selected from a pool of 30 items from the four most widely used problem gambling assessment measures that best captured the largest proportions of pathological, problem, and at-risk gamblers identified via clinical assessment in community and online samples.

2 Past year In the past 12 months:

(1) Would you say you have been preoccupied with gambling?

(2) Have you often gambled longer, with more money or more frequently than you intended to?



0.73 0.99 Volberg & Williams, 2012

Case Finding and Help Assessment Tool (CHAT)

The 24-item CHAT is a composite health screen that screens for nine current lifestyle and mental health conditions (tobacco use, alcohol and other drug misuse, problem gambling, depression, anxiety and stress, abuse, anger problems, inactivity, and eating disorders).

2–3 Not reported; Presumably current

(1) Do you sometimes feel unhappy or worried after a session of gambling?

(2) Does gambling sometimes cause you problems?

(3) If yes to either or both of


Not been validated in community or problem gambling treatment samples





No. of items


these 2 questions, do you want help with this? (No; Yes, but not today; Yes)

Problem Gambling Severity Index (PGSI) Short Form

Developed for the purpose of tracking the prevalence of problem gambling in the general population. Comprised of three items from the PGSI.

3 Thinking about the last 12 months:

(1) Have you bet more than you could really afford to lose?

(2) Have people criticised your betting or told you that you had a gambling problem, regardless of whether or not you thought it was true?

(3) Have you felt guilty about the way you gamble or what happens when you gamble?

Response options: (0) Never; (1) Sometimes; (2) Most of the time; and (3) Almost always. Scores range from 0 to 9. Cut-offs: 3+ = problem gambling, 1–2 = at-risk gambling, 0 = non-problem gambling. Classification accuracy provided is based on the cut-off of ≥ 1 because it is most desirable in monitoring changes in the overall gambling harms in the population and maximises sensitivity over the alternative cut-off of ≥ 3 [54].


0.87–1.00 0.63–0.97 Volberg & Williams, 2012

National Opinion Research Center Diagnostic Screen for Gambling

Derived from the NODS, which is a 17-item measure based on the DSM-IV criteria for pathological

3 Lifetime (although has been converted to lifetime

(1) Have there ever been periods lasting 2 weeks or longer when you spent

Response options: (1) Yes, (0) No. Positive endorsement of one or more items is


0.80–0.99 0.82–0.94 Toce-Gerstein et al., 2009; Volberg et al., 2011





No. of items


Disorders – Loss of Control, Lying and Preoccupation (NODS-CLiP)

gambling. The NODS was designed to improve the efficiency of identifying people with gambling problems in population research. NODS-CLiP comprises the 3 NODS items that best identified problem gambling across eight separate community surveys.

(Volberg et al., 2011)

a lot of time thinking about your gambling experiences, or planning out future gambling ventures or bets?

(2) Have you ever tried to stop, cut down, or control your gambling?

(3) Have you ever lied to family members, friends, or others about how much you gamble or how much money you lost on gambling?

indicative of problem gambling.



Adult problem gambling treatment samples

0.98–1.00 0.17–0.30 Volberg et al., 2011

Brief Biosocial Gambling Screen (BBGS)

Derived from the DSM-IV diagnostic criteria for pathological gambling as measured by the Alcohol Use Disorder and Associated Disability Interview Schedule (AUDADIS) in the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC).

3 Past year During the past 12 months:

(1) Have you become restless, irritable, or anxious when trying to stop and/or cut down on gambling?

(2) Have you tried to keep your family or friends from knowing how

Response options: (1) Yes, (0) No. Positive endorsement of one or more items is indicative of problem gambling.


0.96 0.99 Gebauer et al., 2010




1.00 0.14–0.26 Brett et al., 2014





No. of items


much you gambled?

(3) Did you have such financial trouble as a result of gambling that you had to get help with living expenses from family, friends, or welfare?

Brief Problem Gambling Screen (3 item)



(1) Have you needed to gamble with larger amounts of money to get the same feeling of excitement?


(3) Have you made attempts to either cut down, control or stop gambling?

Response options: (1) Yes, (0) No. Positive endorsement of one or more items is indicative of problem gambling [45].



Consumption Screen for Problem Gambling (CSPG)

A conceptual analogue of the Alcohol Use Disorders Identification Test – Consumption

3 Past year The questions and response options are:

Like the AUDIT-Consumption, the CSPG has different response options for


1.00 0.93 Rockloff, 2012





No. of items


(Bush, Kivlahan, McDonell, Fihn, & Bradley, 1998). It was developed in an adult community (email panel) sample in Australia and measures consumption rather than harm. Although it was developed and validated in a community sample, the authors suggest that the CSPG may most useful in clinical settings, such as general practice, as items about consumption may be less intrusive than those about gambling harms.

Item 1: How often did you gamble in the past 12 months?

(0) I have NEVER gambled OR I have not gambled at all in the past 12 months (1) Monthly or less (2) 2 to 4 times a month (3) 2 to 3 times a week (4) 4 to 5 times a week (5) 6 or more times a week

Item 2: How much time did you spend gambling on a typical day in which you gambled in the past 12 months?

(0) Less than 30 min (1) More than 30 min but less than 1 hour (2) More than 1 hour but less than 2 hours (3) More than 2 hours but less than 3 hours (4) More than 3 hours.

each item. A score of 4 or more is indicative of problem gambling.





No. of items


Item 3: How often did you spend more than 2 h gambling (on a single occasion) in the past 12 months?

(0) Never (1) Less than monthly (2) Monthly (3) Weekly (4) Daily or almost daily.

National Lottery screen – Loss of Control, Lying and Preoccupation (NLCLiP)

Designed to assess changes in problem gambling prevalence and harms in children and adolescents over time. An adaptation of the NODS-CLiP, in which the wording of CLiP items was simplified for use among children and multiple response options were provided for each item.


(1) How often have you found yourself thinking about gambling or planning to gamble?

(2) How often have you tried to cut down how much you gamble?

(3) How often have you lied to your family, friends, or anyone else about how much you gamble?

Multiple response options are provided for each item: (3) Quite a lot; (2) Only sometimes; (1) Don’t know; and (0) Not at all. A score of 3 or more is indicative of problem gambling.

Adolescent community/school samples

0.55 0.85 Lepper & Haden, 2013

National Opinion Research Centre Diagnostic Screen for

Designed to be used in patients attending alcohol and other drug treatment and other

4 Lifetime (1) Have there ever been periods lasting 2 weeks or longer

Response options: (1) Yes, (0) No. Positive endorsement of one or more items is

Discrimination between adult problem gambling treatment and






No. of items


Gambling Disorders – Preoccupation, Escape, Chasing and Risked Relationships (NODS-PERC)

clinical settings. Comprised of the four NODS items that best identified problem gambling in a problem gambling treatment sample recruited from alcohol and other drug and medical treatment settings.

when you spent a lot of time thinking about your gambling experiences or planning out future gambling ventures or bets?

(2) Have you ever gambled as a way to escape from personal problems?

(3) Has there ever been a period when, if you lost money gambling one day, you would return another day to get even?

(4) Has your gambling ever caused serious or repeated problems in your relationships with any of your family members or friends?

indicative of problem gambling.

community samples


1.00 0.18–0.39 Volberg et al., 2011


Developed for the purpose of screening for problem gambling in both clinical settings and population research. Best-performing 4 items


(1) Have you needed to gamble with larger amounts of money to get








No. of items


selected from a pool of 30 items from the four most widely used problem gambling assessment measures that best captured the largest proportions of pathological, problem, and at-risk gamblers identified via clinical assessment in community and online samples.

the same feeling of excitement?


(3) Have you made attempts to either cut down, control or stop gambling?

(4) Have people criticised your betting or told you that you had a gambling problem, regardless of whether or not you thought it was true?

National Opinion Research Center Diagnostic Screen for Gambling Disorders – Loss of Control, Lying and Preoccupation 2 (NODS-CLiP2)

The NODS-CLiP2 adds Chasing and Escape to the 3 NODS-CLiP items. Validated in a problem gambling treatment sample recruited from alcohol and other drug and medical treatment settings.

5 Lifetime The two additional items are:

(1) Has there ever been a period when, if you lost money gambling one day, you would return another day to get even?

(2) Have you ever gambled as

Response options: (1) Yes, (0) No. Positive endorsement of one or more items is indicative of problem gambling.


0.90–0.93 0.62–0.90 Volberg et al., 2011





No. of items


a way to escape from personal problems?




(1) Would you say you have been preoccupied with gambling?

(2) Have you needed to gamble with larger amounts of money to get the same feeling of excitement?

(3) Have you often gambled longer, with more money and more frequently than you intended to?

(4) Have you made attempts to either cut down, control, or stop gambling?

(5) Have you borrowed money or sold anything to get money to gamble?



0.91–0.99 0.61–0.99 Volberg & Williams, 2012







No. of items


Short South Oaks (SOGS) Gambling Screen

Derived from the 20 SOGS items in a Canadian population survey

5 Previous year

(1) Was there ever a time when you gambled more than you intended to?

(2) Have people criticised your gambling?;

(3) Have money arguments centred on your gambling?;

(4) Have you felt guilty about the way you gamble or what happens when you gamble; and

(5) Have you claimed to be winning money gambling when you were not?

Response options: (1) Yes, (0) No. Positive endorsement of two or more items is indicative of problem gambling.





Screening for problem gambling in primary care, AOD and mental

health services

The selection of a brief screening instrument within a clinical setting, such as a primary care, AOD,

and mental health service, is generally determined by the classification accuracy of the instruments

within the relevant setting. The classification accuracy of a screening instrument is, however, a

function of the base prevalence rate of the disorder within the population to which it is used

(Gambino, 2006; Stinchfield, 2010; Stinchfield, Govoni, & Frisch, 2007; Stinchfield & McCready,

2014). For example, an instrument that was developed for the measurement of problem gambling

in clinical samples, in which the base rate is high, will have weaker classification accuracy when

applied to the general population, in which the base rate is very low (Gambino, 2006; Stinchfield,

2010; Stinchfield et al., 2007). It is therefore important that brief screening instruments are

validated for use within primary care, AOD, or mental health services.

The selection of a screening instrument may also be determined by the needs of the clinical

setting, such as the number of items or duration of administration, the age of the clinical

population, and the timeframe of the instrument (Himelhoch et al., 2015). First, given time and

resource demands, many services may only have the capacity to administer very short screening

instruments that require little time to administer, use direct questions, employ simple response

options and scoring algorithms, and require minimal training. Second, most of the brief screening

instruments have been developed for use in adult samples. This practice, which assumes that

existing adult instruments are appropriate for the measurement of adolescent problem gambling, is

questionable given that problem gambling behaviour or harms among adolescents of different

ages and developmental stages may be qualitatively different to those of adults (Derevensky,

Gupta, & Winters, 2003; Stinchfield, 2010; Stinchfield et al., 2007; Stinchfield & McCready, 2014).

Moreover, children and adolescents are very sensitive to context and responses may be influenced

by the wording of questions (Lepper & Haden, 2013). Finally, the timeframe used by the instrument

should be determined by the purpose of the screening (Stinchfield et al., 2007). For example,

screening instruments measuring lifetime problem gambling do not adequately discriminate

between individuals experiencing current problem gambling and those in remission or recovery as

they will include respondents who have experienced problem gambling in the past, but not the

present (Problem Gambling Research and Treatment Centre, 2011; Stinchfield et al., 2007;

Stinchfield & McCready, 2014).

Despite these considerations, there is currently limited information available to guide the selection

of a brief screening instrument for problem gambling in primary care, AOD, or mental health

services. To date, items from the CHAT have been used to screen for gambling problems in

primary care settings (Goodyear-Smith et al., 2009; Goodyear-Smith et al., 2008), the NODS-

PERC was developed to identify problem gambling in clients attending AOD treatment and other

clinical settings (R. A. Volberg et al., 2011), and the classification accuracy of four brief screening

instruments (Lie/Bet, NODS-CLiP, NODS-PERC, and BBGS) has been evaluated in AOD settings

(Himelhoch et al., 2015).

Goodyear-Smith et al. (2008) developed the CHAT for the screening of nine lifestyle and mental

health conditions, including problem gambling, of adult clients of primary care services. The CHAT

displays adequate sensitivity (0.80 and 0.88) and good specificity (0.97 and 0.98) compared to the

SOGS in adult clients of primary care practices. A subsequent study (Goodyear-Smith et al., 2009)

was designed to assess the value of a help question that is added after a client positively endorsed

a CHAT issue. In these studies, the CHAT displayed adequate sensitivity (0.80 and 0.88) and good

specificity (0.97 and 0.98) in relation to the SOGS. The addition of the help question increased the



specificity of the CHAT gambling items (0.98 to 0.99) without compromising sensitivity. It was

concluded that the CHAT provides an important instrument for routine use within primary care

settings.

R. A. Volberg et al. (2011) developed the NODS-PERC for identifying problem gambling in clients

attending AOD treatment and other clinical settings. The sample on which the screening

instrument was validated, however, was a problem gambling treatment sample. Participants were

involved in a study of brief interventions for problem gambling who were recruited using

advertisements and screening efforts in AOD and medical treatment settings that served inner city

populations with high rates of drug use problems. R. A. Volberg et al. (2011) found that the NODS-

PERC identified 99.7 per cent of all people classified as problem gamblers and 100 per cent of all

people classified as pathological gamblers as assessed by the full NODS. The NODS-CLiP also

displayed high sensitivity as it identified 98.4 per cent of all people classified as problem gamblers

and 100 per cent of all people classified as pathological gamblers. This high sensitivity for both

screening instruments, however, appeared to be at the expense of including relatively large

numbers of at-risk or non-problem gamblers (NODS-PERC specificity = 0.39 for problem gambling

and 0.18 for pathological gambling; NODS-CLiP specificity = 0.30 for problem gambling and 0.17

for pathological gambling). The NODS-CLiP therefore did not perform as well in this clinical sample

as it does in the general population, while the NODS-PERC seemed to demonstrate slightly

superior sensitivity and specificity in this clinical sample compared with the NODS-CLiP. Based on

these findings, the authors concluded that the use of the NODS-PERC is preferable to the NODS-

CLiP in services in which the base prevalence rate of problem gambling is high, such as AOD

treatment services and inner city medical clinics.

Finally, Himelhoch et al. (2015) evaluated the diagnostic accuracy of four brief screening

instruments (Lie/Bet, NODS-CLiP, NODS-PERC, and BBGS) in a sample of 300 people recruited

from intensive outpatient AOD treatment or methadone maintenance programs. Compared to the

cut-off of four or more DSM-5 criteria, sensitivity was high for all of the brief screens (0.94 for the

Lie/Bet, 1.00 for the NODS-PERC, 1.00 for the NODS-CLiP, and 0.91 for the BBGS). Specificity

was more variable (0.66 for the Lie/Bet, 0.57 for the NODS-PERC, 0.54 for the NODS-CLiP, and

0.87 for the BBGS).

Taken together, the available evidence suggests that the CHAT and BBGS display satisfactory

classification accuracy in primary care settings and AOD settings, respectively. While the Lie/Bet,

NODS-CLiP, and NODS-PERC appear to display high sensitivity in these services, they are limited

by low specificity. Moreover, the BBGS, NODS-CLiP, and NODS-PERC also display good

sensitivity, but poor specificity, in problem gambling treatment samples (see Section 3.3). Although

it has been recommended that satisfactory classification accuracy is indicated by sensitivity and

specificity greater or equal to 0.80 (DiStefano & Morgan, 2011; Glascoe, 2005), R. A. Volberg et al.

(2011) argue that the key performance characteristic of a brief screening instrument in clinical

settings is sensitivity as the aim in these settings is to capture people with gambling problems,

even at the expense of including relatively large numbers of at-risk or non-problem gamblers. They

suggest that clinical settings provide opportunities for further assessment to determine the

accuracy of the screening instrument.

It therefore appears that the BBGS, CHAT, Lie/Bet, NODS-CLiP, and NODS-PERC have the

strongest evidence base for their use in primary care, AOD, or mental health settings. They also

appear to identify people who otherwise would not seek help for their gambling problems.

Himelhoch et al. (2015) found that few clients (6 per cent) who screened positive had previously

spoken with a health professional about their gambling. Moreover, the completion of these

measures appears acceptable to clients within these settings. Himelhoch et al. (2015) found that



almost all of the clients (91 per cent) indicated that they felt comfortable answering the brief

screening instrument items, although people classified as problem gamblers were significantly less

comfortable than people classified as non-problem gamblers.

As previously indicated, the selection of one of these instruments may best be determined by the

needs of the clinical setting, such as the number of items or duration of administration, the age of

the clinical population, and the timeframe of the instrument (Himelhoch et al., 2015). There is little

difference in the number of items between these five screening instruments, with the Lie/Bet and

CHAT comprising two items, the BBGS and NODS-CLiP comprising three items, and the NODS-

PERC comprising four items. Moreover, they all require only one minute to administer, use direct

questions, employ dichotomous response options, have simple scoring algorithms and

interpretations, and require minimal training. With regard to the age of the clinical population, all of

the five brief screening instruments have been developed for use in adult samples, with only the

Lie/Bet having been applied to adolescent samples. The only brief screening instrument that has

been adapted for the specific application to child or adolescent samples is the three-item NLCLiP

(Lepper & Haden, 2013). This instrument, however, was not developed for use in clinical samples

and displays poor sensitivity in community child and adolescent samples. Finally, in terms of the

timeframe measured, the CHAT screens for current problem gambling and the BBGS screens for

problem gambling in the previous 12 months. In contrast, the Lie/Bet, NODS-CLiP, and NODS-

PERC screen for lifetime problem gambling. The Lie/Bet, NODS-CLiP, and NODS-PERC may

therefore not adequately discriminate between individuals experiencing current problem gambling

and those in remission or recovery (Problem Gambling Research and Treatment Centre, 2011;

Stinchfield et al., 2007; Stinchfield & McCready, 2014).

Assessment of problem gambling within primary care, AOD and

mental health settings

A positive screen on these brief screening instruments should trigger a more comprehensive

clinical assessment to determine problem severity or diagnostic status and provide information that

can assist in referral decisions and/or treatment planning (Stinchfield et al., 2007; R. A. Volberg et

al., 2011). Given that services may not have the resources and capacity to conduct this

assessment, referral for further clinical assessment by specialist gambling practitioners may be

required. In the next section, we briefly describe the available self-report measures and clinical

interviews available should clinicians within Victorian primary care, AOD, and mental health

services wish to conduct further assessment of gambling problems.

Self-report measures

Many self-report assessment measures have been developed to measure problem gambling,

which have varying levels of classification accuracy for the particular population or objective of the

assessment. Self-report measures for adults include the Problem Gambling Severity Index

(henceforth PGSI; Ferris & Wynne, 2001), South Oaks Gambling Screen (henceforth SOGS;

Lesieur & Blume, 1987), National Opinion Research Centre DSM Screen for Gambling Problems

(henceforth NODS; Gerstein et al., 1999), Problem and Pathological Gambling Measure

(henceforth PPGM; Williams & Volberg, 2010), Early Intervention Gambling Health Test

(henceforth EIGHT screen; Sullivan, 1999), Victorian Gambling Screen (henceforth VGS; Ben-

Tovim, Esterman, Tolchard, & Battersby, 2001), Gamblers Anonymous Twenty Questions

(henceforth GA20; Gamblers Anonymous), Sydney Laval University Gambling Screen (henceforth

SLUGS; Blaszczynski, Ladouceur, & Moodie, 2009), Massachusetts Gambling Screen (henceforth

MAGS; Shaffer et al., 1994), and Addiction Severity Index for pathological gambling (henceforth

ASI-PG; Lesieur & Blume, 1992).



The PGSI (Ferris & Wynne, 2001) has been adopted as the preferred measurement tool for

population-level research in Australia (Neal et al., 2005). It has however, also been adopted for

use in evaluating problem gambling severity in Victorian problem gambling specialist counselling

services, including the face-to-face Gambler’s Help services and the national online counselling

service, Gambling Help Online. Given its use in Victorian specialist gambling services, generally

good psychometric properties, and simple scoring and interpretation (Ferris & Wynne, 2001;

Holtgraves, 2009; McMillen & Wenzel, 2006; Neal et al., 2005), the PGSI is an obvious choice for

the assessment of problem severity within Victorian primary care, AOD, and mental health

services. It should be noted however, that the majority of the studies validating the PGSI have

been conducted on community samples, with fewer studies exploring the validity of the PGSI in

samples from gambling, primary care, AOD, or mental health settings.

The nine items of the PGSI are:

Thinking about the last 12 months …

1. Have you bet more than you could really afford to lose?

2. Have you needed to gamble with larger amounts of money to get the same feeling of

excitement?

3. When you gambled, did you go back another day to try to win back the money you lost?

4. Have you borrowed money or sold anything to get money to gamble?

5. Have you felt that you might have a problem with gambling?

6. Has gambling caused you any health problems, including stress or anxiety?

7. Have people criticized your betting or told you that you had a gambling problem,

regardless of whether or not you thought it was true?

8. Has your gambling caused any financial problems for you or your household?

9. Have you felt guilty about the way you gamble or what happens when you gamble?

The PGSI requires respondents to indicate how often each item applies to them in the previous 12

months on a four-point scale: (0) Never; (1) Sometimes; (2) Most of the time; and (3) Almost

always. Scores range from 0 to 27 and higher scores indicate higher problem severity. Scores on

the PGSI can be used to classify individuals as non-problem gamblers (score of 0), low-risk

gamblers (scores of 1 or 2), moderate-risk gamblers (scores between 3 and 7), or problem

gamblers (scores of 8 or higher). Given evidence that the low-risk and moderate-risk categories

displayed poor discriminant validity using the existing scoring protocol, Currie, Hodgins, and Casey

(2013) recommend a simple modification to the scoring system in which low-risk gambling is

defined by scores between one and four and moderate-risk gambling is defined by scores between

5 and 7. This change to the moderate-risk gambling cut-off score improves the distinctiveness of

the PGSI categories on many dimensions. The original cut-offs, however, are still predominantly

employed in most research and clinical settings in Australia.

The PGSI has the added benefit of being embedded within the Canadian Problem Gambling Index

(CPGI), which contains an additional 22 items measuring gambling involvement (types of gambling

activity; frequency; duration; expenditure) and correlates of problem gambling (e.g. distorted



cognitions; first gambling experiences; family background of problem gambling and alcohol or drug

problems; comorbidity; gambling, drinking or drug use to self-medicate; treatment for physical or

emotional problems brought on by stress; feelings of depression; and gambling-related suicide

ideation or attempts). These follow-up questions can assist service providers in their referral and/or

management decisions.

The Australian NHMRC-endorsed guideline for the screening and assessment of problem

gambling includes a practice point that ‘a structured clinical interview may be required for a full

assessment’ (p. 56). Several diagnostic interviews are available to confirm diagnostic status after

initial screening, but most have limited data on their psychometric properties, particularly in relation

to their classification accuracy. These include the Diagnostic Interview for Gambling Schedule –

DIGS (Winters, Specker, & Stinchfield, 2002), the Structured Clinical Interview for Pathological

Gambling –SCI-PG (Grant, Steinberg, Kim, Rounsaville, & Potenza, 2004), the World Mental

Health Composite International Diagnostic Interview – WMH-CIDI (Kessler & Üstün, 2004), the

Gambling Behaviour Interview – GBI (Stinchfield, Govoni, & Frisch, 2005), the Gambling

Assessment Module – GAM (Cunningham-Williams et al., 2005) and the Structured Clinical

Interview for Pathological Gambling – SCIP (Walker, Anjoul, Milton, & Shannon, 2006).

Most of these clinical interviews have the added benefit of including additional questions across

multiple domains, including gambling behaviour (such as current and past gambling frequency,

duration, expenditure, and locations), treatment history, onset of gambling, sources of borrowed

money, gambling-related consequences (such as financial, social, psychological, and legal

problems), mental health, other impulse control disorders, medical status, and social and family

functioning. These follow-up questions can assist service providers in their referral and/or

treatment planning.

Conclusion

It is important for health professionals to accurately screen for problem gambling as comorbid

problem gambling has the potential to compromise the effectiveness of treatment for drug use and

mental health problems. There seem to be, however, low rates of routine screening for problem

gambling within primary care, AOD, and mental health services. Perceived barriers to such

screening include lack of time, an absence of information about the effectiveness of screening, a

lack of knowledge and skills, the presence of gambling-related stigma, a perception that problem

gambling has a low burden of disease, an absence of effective interventions and limited access to

specialist referral services. The low rate of screening may also be, in part, due to the relatively

slow development of brief screening instruments that are easy to use, have short administration

times, are easily scored, and require minimal training.

Although there are now several brief instruments to screen for problem gambling, there is currently

limited information available to guide their selection for use within primary care, AOD, or mental

health services. The BBGS, CHAT, Lie/Bet, NODS-CLiP and NODS-PERC appear to have the

strongest evidence base for their use in these settings. Although some of these screening

instruments are characterised by low specificity in clinical samples, it has been argued that the aim

in clinical settings is to capture people with gambling problems, even at the expense of including

relatively large numbers of non-problem gamblers, as there are likely opportunities for further

assessment of problem gambling. The selection of one of these instruments, however, may best

be determined by the needs of the clinical setting, such as the number of items or duration of

administration, the age of the clinical population, and the timeframe of the instrument. It is

important to note that clients in clinical settings appear comfortable with answering questions

relating to their gambling behaviour and that these instruments identify clients who have not



previously discussed their gambling with a health professional. Given that they have been

developed and validated in other countries, however, the classification of these screening

instruments may not generalise to Victorian primary care, AOD, and mental health settings.

A positive screen on these brief screening instruments should trigger a more comprehensive

clinical assessment to determine problem severity or diagnostic status and provide information that

can assist in referral decisions and/or treatment planning. If time and resources permit, this

assessment can be conducted by the primary care, AOD or mental health service. Alternatively,

this assessment can be conducted by a specialist gambling service. This assessment can involve

the use of a simple self-report measure, such as the PGSI, or a clinical interview. Following this

assessment, primary care, AOD, and mental health services need to have the time, skills and

resources to treat the gambling problem or have appropriate referrals in place. This conclusion is

supported by a consensus-based recommendation from the Australian NHMRC-endorsed clinical

guideline that ‘those who screen positive for problem gambling using an initial brief screening tool

could be referred for further assessment and treatment by appropriately trained specialist

practitioners in problem gambling’ (p. 55). In Section 4 of this review, we briefly describe best-

practice psychological and pharmacological treatments for problem gambling, with a particular

emphasis on interventions for comorbid problem gambling and mental health disorders.

Section 4: Treatment of problem gambling in primary

care, AOD and mental health settings

Psychiatric comorbidity in problem gambling is associated with more complex clinical

presentations, including more severe gambling problems, psychiatric symptoms, impulsivity,

suicidality, personality disorder pathology, and other psychosocial difficulties (Brown et al., 2016;

Pietrzak & Petry, 2005; Stinchfield, Kushner, et al., 2005; Waluk et al., 2016). In Section 2, we

presented findings suggesting that comorbid problem gambling has the potential to compromise

engagement in, and the effectiveness of, AOD and mental health treatment, particularly if it goes

unidentified and unmanaged. Although it may be important to consider the functional relationship

between the drug use or mental health disorder and the problem gambling, gambling problems

may impact on the effectiveness of treatment, even when multiple disorders within the one

individual are aetiologically independent (Hollander, Sood, Pallanti, Baldini-Rossi, & Baker, 2005;

Winters & Kushner, 2003). Managing comorbid problem gambling may therefore lead to

individually tailored treatment approaches that could maximise treatment response, enhance client

satisfaction, reduce attrition, and lower treatment costs.

It is therefore important for health service providers to accurately screen for problem gambling so

that people with gambling problems in these services can be identified and offered further

assessment and appropriate management or referral. In Section 3, we identified five brief

screening instruments that have the best evidence base for use within primary care, AOD and

mental health settings, as well as several self-report measures and/or clinical interviews that can

be used by health providers or specialist gambling practitioners to provide a more comprehensive

clinical assessment. Following confirmation of diagnostic status and problem severity, people with

gambling problems may require appropriate management. Should time and resources permit, this

management can occur through the primary care, AOD, or mental health service. Alternatively,

individuals with gambling problems can be referred for treatment of the gambling problem by

specialist gambling practitioners. A diverse range of psychological and pharmacological options for

the treatment of problem gambling are available, with varying levels of empirical support. In

Section 4 of this review, we will briefly describe recent available meta-analytic evidence for the



efficacy of psychological and pharmacological interventions for problem gambling and explore the

limited evidence base for interventions for comorbid problem gambling and mental health

disorders.

Psychological treatments for problem gambling

There is recent meta-analytic evidence for the efficacy of cognitive behavioural therapies (CBT)

(Cowlishaw et al., 2012;(Gooding & Tarrier, 2009) and motivational interviewing (MI) or

motivational enhancement therapies (MET) (Cowlishaw et al., 2012; Yakovenko, Quigley,

Hemmelgarn, Hodgins, and Ronksley (2015). Although there is some preliminary evidence

suggestive of some possible benefits from integrative and other psychological therapies, there are

few too studies on which to evaluate these therapies (Cowlishaw et al., 2012). Recent meta-

analyses suggest that CBT is effective in reducing gambling behaviours and depressive and

anxiety symptoms following treatment, but that there are few studies employing long follow-up

evaluation periods (Cowlishaw et al., 2012; Gooding & Tarrier, 2009). There is also preliminary

meta-analytic evidence from a smaller number of studies that MI/MET interventions are associated

with significant short-term reductions in gambling expenditure and reduced gambling frequency at

longer term follow-up evaluations (Cowlishaw et al., 2012; Yakovenko et al. (2015). There is,

however, no improvement in depressive or anxiety symptoms resulting from MI/MET interventions

(Cowlishaw et al., 2012). CBT and MI are both therefore likely to have robust short-term effects on

gambling behaviour that likely endure, although the durability of these effects requires additional

investigation.

CBT and MI/MET interventions have been successfully delivered as brief interventions (Abbott et

al., 2012), as well as online and self-help interventions (Merkouris et al., In Press).These less

intensive interventions may be helpful resources for clinicians in primary care, AOD, and mental

health services who are working with clients with comorbid gambling problems.

Pharmacological treatments for problem gambling

Pharmacological treatment options for problem gambling are generally classified into three

categories: opioid antagonists (such as naltrexone and nalmefene), antidepressants (most

commonly selective serotonin reuptake inhibitors [SSRIs]), and mood stabilisers or

anticonvulsants. Less commonly evaluated pharmacological agents include antipsychotics, such

as olanzapine (Fong, Kalechstein, Bernhard, Rosenthal, & Rugle, 2008; McElroy et al., 2008) and

amino acids, such as N-acetyl cysteine (Grant, Kim, & Odlaug, 2007). There is, however, little

empirical data to guide the selection of one pharmacological intervention over another, with few

differences in outcome between the main classes of pharmacological interventions. While meta-

analytic evidence suggests opiate antagonists demonstrate a significant benefit compared to

placebo (Bartley & Bloch, 2013; S. A. Thomas et al., 2011), Bartley and Bloch (2013) argue that

this treatment effect may be driven by several early trials of opiate antagonists that have employed

small sample sizes and non-intention-to-treat analyses. Bartley and Bloch (2013) note other

pharmacological agents, such as antidepressant medications and antipsychotic agents, have effect

sizes similar in magnitude to opiate antagonists but fail to cross the threshold for statistical

significance because of the smaller samples employed in these trials.



Interventions for comorbid drug use/mental health disorders and

problem gambling

The treatment of drug use and mental health disorders is complicated by the substantial

comorbidity with problem gambling. There is, however, little empirical evidence regarding the

identification of specific treatment strategies best suited to individuals with concurrent drug use and

mental health disorders and problem gambling. Dowling, Merkouris, and Lorains (2016) conducted

a mini-review with a view to highlighting the gaps in the literature that preclude the identification of

treatment recommendations for subpopulations of problem gamblers with comorbid psychiatric

disorders. They drew upon a systematic search of the literature to identify randomised trials

evaluating the efficacy of intervention approaches for problem gambling with specific psychiatric

comorbidities. They identified six randomised trials that explored the efficacy of interventions for

problem gambling with psychiatric comorbidities (see Table 7). There are also several non-

randomised trials, case studies, or small sample designs that provide promising results for the

treatment of concurrent problem gambling and drug use/mental health disorders.

Alcohol and other drug use disorders

Three of the randomised trials identified in the mini-review (Dowling, Merkouris, et al., 2016)

related to the efficacy of interventions for concurrent alcohol and other drug use problems and

problem gambling. Korman et al. (2008) found that a 14-week integrated treatment (modified

dialectical behaviour therapy addressing anger and addiction problems) was more effective in

gambling behaviour, anger and drug use at a 12 week follow-up evaluation than a specialised

treatment-as-usual for 42 individuals with comorbid problem gambling, anger and drug use.

In an 11-week randomised, double-blind, placebo-controlled trial, Toneatto, Brands, and Selby

(2009) found that naltrexone (in combination with CBT) was no more effective than placebo for

alcohol use and gambling behaviours following treatment or at one-year follow-up for 52 individuals

with concurrent problem gambling and alcohol abuse/dependence. The authors suggest that the

significant improvements in alcohol and gambling identified for both groups may be attributed to

the adjunctive CBT. Finally, Grant et al. (2014) found that the addition of N-acetylcysteine to

tobacco support/imaginal desensitisation/MI was associated with a significant reduction in problem

gambling severity, but not nicotine dependence severity, at a three-month follow-up evaluation

compared to augmentation with a placebo pill for 28 individuals with comorbid problem gambling

and nicotine dependence. The authors suggest that N-acetylcysteine may facilitate maintenance of

behavioural interventions for problem gambling following the completion of treatment.

Since the publication of the mini-review, Petry, Rash, and Alessi (2016) examined the

effectiveness of three brief interventions to reduce problem gambling over a two-year period

among 217 clients attending AOD settings:

a 10–15-minute psychoeducation gambling intervention

a 10–15-minute brief advice intervention addressing gambling norms, risk factors, and

methods to prevent additional problems

four 50-minute sessions of MET plus CBT for reducing gambling.

The findings revealed that, for the whole sample, gambling days, money wagered and gambling

problems decreased significantly after five months of treatment, with only modest but significant

further reductions (though not in gambling days) thereafter. The brief advice condition significantly

reduced days gambled at five months relative to the brief psychoeducation session. Moreover, the



MET plus CBT condition was found to be more effective in decreasing money wagered and

gambling problems than the brief advice condition and was also associated with a reduction in

alcohol use and problems, but not illicit drug use.

There are also several non-randomised trials and case studies evaluating the efficacy of

interventions applied to individuals with concurrent alcohol and other drug dependence and

problem gambling. Lesieur and Blume (1991) found that a psychological treatment program

involving individual psychotherapy, group counselling, client education and a 12-step self-help

group targeting gambling, alcohol and other drug use concurrently was effective in reducing

alcohol use, drug use, and gambling for 72 individuals with comorbid alcohol and drug abuse and

problem gambling. Crockford and el-Guebaly (1998) reported that four weeks of combined

naltrexone/SSRI treatment achieved abstinence from gambling, ceased gambling cravings, and

reduced alcohol use in a 49 year old male with comorbid alcohol dependence and problem

gambling.

Bipolar disorder

Only one of the randomised trials identified in the mini-review (Dowling, Merkouris, et al., 2016)

evaluated an intervention for individuals with concurrent bipolar disorder and problem gambling. In

a 10-week double-blind placebo-controlled trial, Hollander, Pallanti, Allen, Sood, and Baldini-Rossi

(2005) found that sustained-release lithium simultaneously improved gambling and bipolar disorder

symptoms compared to placebo in 29 individuals with concurrent bipolar spectrum disorders and

problem gambling. Several case studies have also successfully applied targeted interventions with

co-occurring bipolar spectrum disorders. Moskowitz (1980) found that an open-label lithium

treatment resulted in abstinence of gambling behaviours and cessation of mania symptoms for

three problem gamblers with bipolar features. Dell'Osso and Hollander (2005) reported that a 10-

week lithium trial reduced both gambling behaviour and mania symptoms in a 30 year old female

problem gambler with comorbid bipolar disorder. Nicolato, Romano-Silva, Correa, Salgado, and

Teixeira (2007) found that an eight-week combined lithium and topiramate treatment reduced

gambling behaviour and improved affective symptoms in a 57 year old female with concurrent

bipolar disorder II and problem gambling. Although lithium alone improved affective symptoms, the

gambling behaviour improved only after topiramate was added.

Anxiety disorders

Dowling, Merkouris, et al. (2016) identified one randomised trial that explored the efficacy of

interventions for concurrent anxiety disorders and problem gambling. Grant and Potenza (2006)

conducted an 11-week open-label study of escitalopram followed by an eight-week double-blind

discontinuation study for individuals with comorbid anxiety disorders and problem gambling. The

open-label phase revealed that 8 out of 13 (61.5 per cent) individuals responded to escitalopram

and anxiety and gambling symptoms improved concurrently. Although not statistically or clinically

significant, the three responders to escitalopram who were randomised to escitalopram reported

slightly worse gambling symptoms in the discontinuation phase.

In addition, Najavits (2011) found that individuals with comorbid PTSD and problem gambling

preferred treatments for PTSD over those for problem gambling and expressed preferences for

individual therapy, seeking safety therapy (a present-focused therapy to help people attain safety

from trauma/PTSD and drug abuse), exposure therapy, and CBT. In a pilot study of seeking safety

with seven individuals with comorbid PTSD and problem gambling, Najavits et al. (2013) found

significant improvements in both PTSD and problem gambling.



Schizophrenia

The mini-review conducted by Dowling, Merkouris, et al. (2016) identified one randomised trial that

investigated the efficacy of an intervention for individuals with concurrent schizophrenia and

problem gambling. In this study, Echeburua, Gomez, and Freixa (2011) compared CBT for

problem gambling/standard drug therapy for schizophrenia to a waiting list control group (standard

drug therapy for schizophrenia). In 44 clients diagnosed with chronic schizophrenia and problem

gambling, therapeutic success for gambling behaviour was significantly higher for CBT/standard

drug therapy (73.9 per cent) than the control condition (19.0 per cent) at the three month follow-up

evaluation. This study did not measure psychotic symptoms as an outcome of treatment.

The efficacy of targeted interventions for individuals with comorbid schizophrenia and problem

gambling has also been reported in several case studies. Potenza and Chambers (2001) reported

that olanzapine, an atypical antipsychotic drug with mood stabilising properties, resulted in

improvement in psychotic symptoms and gambling behaviour in a 31 year old woman with

comorbid problem gambling and schizophrenia. This improvement, however, also correlated

temporally with the introduction of psychosocial interventions targeting her gambling. In contrast,

N. Smith, Kitchenham, and Bowden-Jones (2011) reported three case studies in which individuals

with concurrent psychosis and problem gambling were treated with aripiprazole, an atypical

antipsychotic. Although all three individuals reported a positive response in their psychotic

symptoms with aripiprazole, they all demonstrated an adverse gambling response, including

increased gambling urges and escalation of gambling behaviour. Finally, Shonin, Van Gordon, and

Griffiths (2014) found that a 20 week treatment course of CBT followed by meditation awareness

training (a mindfulness based meditation intervention) was effective at reducing both schizophrenia

and problem gambling symptoms in a 32 year old female problem gambler with comorbid

schizophrenia.



Table 7. Randomised trials evaluating interventions for comorbid drug use/mental health disorders and problem gamblinga

Study Sample size Country Treatment type Control group Psychiatric comorbidity

Post-treatment assessments

Gambling outcomes

Korman et al. (2008)

42 Canada Modified DBT Specialised treatment-as-usual (TAU) for gambling and drug use (eclectic including CBT relapse prevention)

Anger and alcohol or other drug use disorder

Post-treatment and 3 month follow-up

PGSI; percentage of monthly income spent gambling

Toneatto et al. (2009)

52 Canada Naltrexone plus CBT Placebo plus CBT Alcohol or other drug use disorder


Days gambling and expenditure/gambling day

Grant et al. (2014)

28 USA N-acetylcysteine plus Ask-

Advise-Refer therapy plus imaginal desensitisation/MI

Placebo plus Ask-

Advise-Refer therapy plus imaginal desensitisation/MI

Nicotine dependence


PG-YBOCS

Petry et al. (2016) 217 USA Brief psychoeducation; or brief advice; or MET plus CBT

N/A Alcohol, cocaine, opioid or marijuana use disorder

2, 5, 8, 12, 16, 20, & 24 months post-treatment

SOGS; days gambled; dollar amounts risked; ASI, breath/urine samples

Grant and Potenza (2006)

13 (open-label); 4

(RCT)

USA Escitalopram Placebo Anxiety disorders (SCID)

Post-treatment (11 weeks open-label; 8 weeks RCT)

Treatment response (30% or greater reduction in PG-YBOCS total score); G-SAS; CGI-Severity

Hollander, Pallanti, et al. (2005)

29 USA Sustained-release lithium

Placebo Bipolar spectrum disorders

Post-treatment Treatment response (a 35% or greater reduction in PG-YBOCS total score at end point compared with baseline and a score of 1 or 2 on the CGI Improvement Scale [‘very much improved’ or ‘much improved’])



Conclusion

Given that problem gambling may reduce the outcomes of treatment of mental health disorders, it

is important that individuals with gambling problems identified in primary care, AOD, and mental

health services are offered appropriate management. If time and resources allow, this

management can occur within the primary care, AOD, and mental health service. Alternatively, the

gambling problem could be managed via referral to specialist gambling treatment services.

Although a diverse range of psychological and pharmacological options for the treatment of

problem gambling are available, the evidence base is limited and is confounded by generally low

quality standards. Interventions with the strongest evidence base include CBT, MI/MET, and opioid

antagonists. Despite our awareness of the comorbidity between problem gambling, drug use

problems, and mental health issues, there is very little evidence on which to base treatment

recommendations for different subpopulations of problem gamblers based on their psychiatric

comorbidity. Brief interventions and online self-help programs based on MI and CBT therapies may

be useful resources for clinicians working with clients with comorbid gambling problems within

primary care, AOD, and mental health services.

Section 5: Effective partnerships between gambling

and primary care, AOD and mental health services

The presence of problem gambling in individuals presenting to primary care, AOD and mental

health services may compromise the effectiveness of treatments for their presenting issue. It is

therefore important that people with gambling problems in these services are appropriately

identified and that the gambling problem is appropriate managed through the primary care, AOD,

or mental health service, or via referral to specialist gambling practitioners. In this review, we have

identified several screening and assessment instruments that have the best evidence base for use

in these services (Section 3) and a limited number of interventions that have been successfully

applied to individuals with concurrent mental health disorders and problem gambling (Section 4). In

Section 3, however, we identified several perceived barriers to the screening and intervention for

gambling problems in these services, which include a lack of time; a lack of confidence from health

providers that they have the necessary training, knowledge, or skills to address gambling

problems; a lack of funding to compensate for the additional time required to screen, assess, and

manage gambling problems; and access to, and confidence in, specialist referral services. These

barriers highlight the need to examine effective inter-agency working relationships, in which

screening, referral and treatment can be integrated. In Section 5 of this review, we provide a

description of treatment models for gambling and mental health services, which draws from the

more extensive dual diagnosis (AOD and mental health) literature.

Treatment models for gambling and mental health services

In contrast to those for AOD, there are few specialist treatment services available for individuals

with comorbid mental health disorders and problem gambling. We are aware of only two services

that have been designed to treat people with comorbid problem gambling and mental health

issues. In Victoria, a state-wide multidisciplinary service run by Alfred Health exists to support

gambling treatment services with individuals with co-occurring mental health disorders and

gambling problems. This service, The Problem Gambling and Mental Health Program, consists of

a specialist multidisciplinary team that aims to increase the knowledge and capacity of specialist

gambling counselling services and provide optimal outcomes for individuals with co-occurring

mental health issues and problem gambling. The services provided by this program include

assessment, short-term treatment that addresses the pharmacological, social and psychological



needs of the client, secondary support through referrals from primary service providers such as

specialist gambling practitioners and GPs, professional support and consultation services to

specialist gambling practitioners, and outreach consultations to specialist gambling services. It

remains unclear if the outcomes for clients are optimised, or if the assessment and referral

procedures are efficient and effective as evaluation data for this service has not yet been

published. A large trial is currently underway in Canada exploring the effectiveness of an online

mental health and gambling intervention versus a gambling-only online intervention for problem

gamblers with and without co-occurring mental health symptoms on gambling outcomes at

6 months (Cunningham et al., 2016). Given the limited evidence base examining models of care

for gambling and mental health services, however, the remainder of this section will describe

models of care in the dual diagnosis (AOD and mental health) literature.

Effective partnerships between AOD and mental health services

Historically, the dominant model for treating dual diagnosis clients has been the single service

model, which provides either sequential or parallel treatment for drug use and mental health

problems. In sequential treatment models, individuals are treated for one condition first and are

then referred to another service for the treatment of the second condition, while parallel treatment

models involve the simultaneous treatment by different services, independent of one another

(Donald, Dower, & Kavanagh, 2005). These treatment models are both flawed as they fail to

acknowledge the possible relationship between the co-occurring disorders by treating each

disorder separately and/or through a different service. Moreover, in these models, the individual

with the co-occurring disorders is responsible for bringing together the treatment messages from

each service, which can be difficult when treatment philosophies differ across services (NSW

Health, 2015).

In contrast, integrated models that treat the co-occurring conditions simultaneously using the same

provider or service are becoming increasingly prevalent. It has been argued that integrated

treatment models are advantageous as they allow for the relationship between the two disorders to

be explored rather than viewing them as mutually exclusive, thereby overcoming the issue of

fragmented or contradictory treatment (Australian Institute for Primary Care, 2009; Donald et al.,

2005; Drake, O'Neal, & Wallach, 2008; Mueser, Noordsy, Drake, & Fox, 2003). Moreover, these

models may improve access to treatment, limit the financial burden, and enable clients to receive

individualised and clinically relevant treatment (Australian Institute for Primary Care, 2009; Donald

et al., 2005; Drake et al., 2008; Mueser et al., 2003). There is, however, considerable variability in

the literature regarding the definition of an integrated treatment model, primarily in relation to the

level at which integration occurs. For example, models can be integrated at the client/program

level, service/system level, or at the sector level (Australian Institute for Primary Care, 2009). At

the client/program level, integration refers to the treatment of both disorders by a single treatment

service or clinician, either by a multidisciplinary team or by clinicians trained in both mental health

and drug use problems. At the service/system level, integration refers to the coordination or

collaboration between the mental health and AOD service providers to enable integrated treatment

for the individual. Lastly, the single sector integrated model refers to the use of the mental health or

AOD sector as the primary provider of integrated treatment; these models are usually limited to a

particular type or level of psychiatric comorbidity (Australian Institute for Primary Care, 2009;

Donald et al., 2005; Drake, Mercer-McFadden, Mueser, McHugo, & Bond, 1998). It has been

argued that models of care that attempt to integrate treatment by simply adding a treatment

component to an already existing service may not truly be integrated and that treatment must

directly acknowledge and manage the presence of the co-occurring disorders (Donald et al., 2005).



The remainder of this section will provide an overview of existing services and models of care,

broken down based on the service structure:

services that are structured around a specialised comorbidity service

services that are structured so that they can be incorporated into an existing AOD or

mental health service.

Specialised comorbidity service structure

In the dual diagnosis literature, there are few descriptions and/or evaluations of specialised

services created specifically to care for individuals with a dual diagnosis. Illustrative examples,

however, include the Burnaby Treatment Centre for Mental Health and Addiction – BCHMA

(Schütz et al., 2013) and the Triple Care Farm (Mission Australia, 2011). These models and their

associated evaluations are described in this section.

Burnaby Treatment Centre for Mental Health and Addiction

The BCHMA is a Canadian specialised inpatient treatment facility for co-occurring mental health

and drug use problems (Schütz et al., 2013). A panel of experts with a range of specialities,

including providers of drug abuse treatment, rehabilitation and psychological therapy, developed

the key principles of the BCHMA. These principles include values of strength-based care, and

treatment concepts based on assertiveness, motivation, time-unlimited comprehensive

programming, treatment tailoring to stage of change, harm reduction that leads to abstinence,

stepped care, and cultural sensitivity and competence. The BCHMA emphasises that client

recovery is a long process based on managing relapses and crises. The treatment program itself is

a nine-month inpatient program, with flexibility for each client to have shorter stays. Using a

multidisciplinary team (e.g. psychiatrists, nurses, counsellors, and art and music therapists), this

service provides evidence-based treatment that can include individual or group interventions,

addressing relapse prevention, anger management, contingency management, and motivational

interviewing. When required, the service also includes pharmacological treatment. The intensity of

treatment is tailored depending on the client’s needs, with treatment goals discussed in

collaboration with the team and the client. To be eligible for inclusion in the BCHMA, one must

have failed other programs on a regional level and must have issues in all four of the following

domains: mental health, drug use, physical health and behaviour. Ultimately, the BCHMA program

is designed to help the most vulnerable populations (i.e. individuals who are difficult to engage in

treatment and present with multi-morbidity). Schütz et al. (2013) evaluated the BCHMA program

using a pre-post design. Of the 47 participants that completed the follow-up evaluation, promising

results were identified, with a significant decrease in symptoms of psychopathology and drug use.

Triple Care Farm

Another specialised service identified is the Triple Care Farm, a residential facility in NSW,

Australia, which aims to treat young people (16–24 years old) with a comorbid mental health

(including psychosis) and drug use problem (Mission Australia, 2011). Intake and assessment in

Triple Care Farm occurs while clients are still living in the community. Clients within the Triple Care

Farm are called students and the philosophy underpinning the Triple Care Farm model is that the

students are responsible for their own treatment and progress. Within the Triple Care Farm model,

students work and progress through three stages:



1. the Gateway stage, in which goals are set and relationships are built

2. the Explorer stage, in which students work towards the goals they set in the Gateway

stage and engage in skills training

3. the Outbound stage, in which students are assisted with their transition into the

community.

Within each of these three stages are three program elements that each relate to aspects of

treatment. In the Residential program element, students concentrate on acquiring life skills. The

Therapeutic program elements relates to the medical, psychological and behavioural problems that

are associated with a dual diagnosis. Lastly, the Engagement and Wellbeing program element

focuses on healthy living and physical fitness, and helps students re-engage with education and

work. Triple Care Farm also offers an aftercare program that aims to decrease the risk of relapse

when graduates return to the community. The three treatment stages occur over approximately

three months, and the aftercare program is available for up to six months after graduation. The

team at Triple Care Farm is multidisciplinary, including psychologists, medically trained clinicians

AOD counsellors, social welfare and case workers, social scientists, educators and care workers.

The treatment, psychological and medical, is provided in a consolidated way through collaboration,

sharing of information in management and administration of treatment.

The Triple Care Farm has conducted an in-house evaluation of the program (Mission Australia,

2011). This evaluation examines key indicators of the therapeutic program outcomes, including

alcohol use, drug use, psychological wellbeing and engagement, and wellbeing program outcomes

(such as employment, and housing) at three and six months following entry into the aftercare

program. While no statistical comparisons were conducted, the results revealed that at the six-

month follow-up, 45 per cent of the 160 graduates were abstaining from drug use and 52 per cent

were abstaining from alcohol use. In addition, graduates showed improvements in psychological

health, demonstrated an increased participation in work, education and training, and displayed

stability in housing/accommodation. Although this evaluation shows promising findings, an

independent evaluation of this specialised service is required, in which statistical comparisons are

conducted to determine its effectiveness in treating dual diagnosis individuals.

Integration within existing service structures

There are several models described in the dual diagnosis literature that are structured so that they

can be incorporated into an existing AOD or mental health service. Illustrative examples of models

that have undergone evaluation include the Assertive Community Treatment model and case

management models (Stein & Test, 1980; Wood & Anderson, 1995), the Integrated Dual Disorders

Treatment model (Drake, Teague, & Warren, 1990), the COMPASS program (Graham, Copello,

Birchwood, Maslin, et al., 2003), the Comprehensive, Continuous, Integrated System of Care

model (Minkoff & Cline, 2004), the Collaborative Early Identification model (Staiger et al., 2008),

and a model of care for youth within the alcohol or other drug sector (Lubman, Hides, & Elkins,

2008). There are also several models of care which have yet to be evaluated. These include the

Substance Use and Mental Illness Treatment Team (SUMITT) (Melbourne Health, 2016), the

Hunter New England Mental Health and Substance Use Service (MHSUS: (NSW Health, 2015),

Co-Exist (Co-Exist NSW, 2008), the Collaborative Practice Development Model (Monisse-Redman,

2015), the New South Wales (NSW) pathway of care (NSW Health, 2015), and the model for

responding to dual diagnosis (Victorian Government: Department of Human Services, 2007).

These models and their associated evaluations are described in the remainder of this section.



Assertive Community Treatment and case management models

Conceptualised by Stein and Test (1980) in the US in the 1980s, the Assertive Community

Treatment (ACT) model was one of the first models of care developed for dual diagnosis

populations. The ACT model is a community-based service that consists of a multidisciplinary team

which is responsible for the constant and continuous care (i.e., 24 hours a day) of their clients

(Drake, McHugo, et al., 1998). The ACT model uses a low staff-to-client ratio of approximately

1:10, and most of their services are directly provided within the community by the ACT team

(Essock et al., 2006). The case management model (also known as the care coordination model)

is similar to the ACT model, in that it uses a team of professional staff, such as social workers, to

coordinate the care of dually diagnosed individuals (Wood & Anderson, 1995). The case

management model, however, differs from the ACT model in that one staff member is designated

the case manager and the services provided are brokered to other clinicians or services, meaning

clients have to travel to receive the care they need (Drake, McHugo, et al., 1998; Essock et al.,

2006; Wood & Anderson, 1995; Young, Clark, Moore, & Barrett, 2009).

The ACT and case management models of care have been extensively examined, with several

studies comparing the effectiveness of the ACT and case management models of care in the

treatment of dually diagnosed individuals with severe mental health disorders such as

schizophrenia (Drake, McHugo, et al., 1998; Essock et al., 2006; Mueser, Drake, & Miles, 1997).

Drake, McHugo, et al. (1998) demonstrated that an ACT group showed greater improvements on

measures of alcohol use, life satisfaction and financial support at a three year follow-up than a

case management group. There were, however, no differences between the groups on measures

of drug use, housing stability, psychiatric symptom severity and other areas of quality of life, such

as social and family contacts. Similarly, Essock et al. (2006) found that both ACT and standard

case management groups showed similar improvements in drug use, alcohol use, psychiatric

symptoms, general life satisfaction, and global functioning. The ACT group, however, showed

significantly greater improvements on self-reported days of drinking and days spent in residential

services. In a randomised trial, Mueser et al. (1997) found that both programs were associated

with decreased alcohol and other drug use, a reduced number of hospitalisations, and

improvements in symptom severity and other areas of life functioning over the three year follow-up

period. Taken together, these studies suggest that the ACT and case management models both

result in improvement on various alcohol use, other drug use, psychiatric symptomatology, quality

of life and housing stability measures, with minimal differences between the groups.

Integrated Dual Disorders Treatment model

In the 1990s, the Integrated Dual Disorders Treatment (IDDT) model was designed to assist with

the treatment of severely mentally ill adult clients with a co-occurring drug use disorder within an

outpatient mental health service system in the US (Drake et al., 1990). Assumptions underlying the

IDDT model are that treating drug use disorders in a chronically mentally ill population should take

place within the mental health service system, that the mental health system should provide truly

integrated treatment for mental health and drug use problems, and that all staff must have an

increased awareness and the relevant skills to treat dually-diagnosed clients for treatment to be

truly integrated (Drake et al., 1990). Importantly, the IDDT model integrates the mental health and

AOD services at the administrative, hospital-based, community mental health centre-based

treatment, residential service, system linkage, clinical training program and research based levels

(Drake et al., 1990). Moreover, important factors in this model include the reliance on continuous

treatment teams and case managers to link and coordinate the care of dually diagnosed clients

and the need for continuous staff training (Drake, Antosca, Noordsy, Bartels, & Osher, 1991; Drake

et al., 1990). Key treatment components of the IDDT model include the use of continuous



treatment teams, the promotion of stable housing, and the use of residential treatment facilities

such as psychiatric hospitals to assist with detoxification and treatment for chemical dependency

(Drake et al., 1990). The IDDT model also uses a stage-wise treatment approach, whereby the

goal of treatment and the intervention provided differs depending on the client’s stage (Mueser et

al., 1997). These stages include:

1. engagement, which refers to building relationships with clients, usually through intensive

outreach by case managers

2. persuasion, in which clients are motivated to deal with their drug use disorders and accept

long-term abstinence-oriented treatment, usually through clear and consistent education

3. active treatment, which refers to the provision of treatment (e.g. behavioural, group or

educational interventions) directly to clients to develop the skills needed to address their

drug use behaviours

4. relapse prevention, in which clients are taught that relapses should be anticipated and are

assisted to develop strategies to decrease vulnerability to relapse (Drake et al., 1990;

Mueser et al., 1997).

The IDDT model has been adapted for use in several jurisdictions, including the USA (Drake et al.,

1991; Drake et al., 1990), Sweden (Blix & Eek, 2005) and the Netherlands (Boyle & Kroon, 2006).

It has also been employed in AOD services (Kola & Kruszynski, 2010) and inpatient settings

(Wieder, Lutz, & Boyle, 2006). Despite wide adoption, the effectiveness of this model has been

examined in a limited number of studies (Blix & Eek, 2005). Blix and Eek (2005) examined the

effectiveness of the IDDT model in Sweden using an uncontrolled study design. This study used

the Substance Abuse Treatment Scale (a measure of assessing client’s stage of AOD treatment,

ranging from pre-engagement to in remission or recovery) as an indicator of successful treatment

at discharge. Using this scale, clients in stage 7 (relapse prevention) are considered to be

successful discharges. Of the 82 clients admitted, 11 were successfully discharged and a further

11 were considered to have shown improvement. In addition, the results suggest that up to 30 per

cent of clients who left treatment were abstinent.

The COMPASS program

The COMPASS program was developed in the UK as an integrated shared-care approach to

treating dual diagnosis individuals (Graham, Copello, Birchwood, Maslin, et al., 2003). The

COMPASS program uses a specialist multidisciplinary team (e.g. clinical staff, clinical psychologist,

psychiatric nurses and occupational therapist) to train and support existing mental health and

addiction services to provide integrated treatment, as opposed to creating a new and separate dual

diagnosis team (Graham, Copello, Birchwood, Maslin, et al., 2003). In the COMPASS program, a

single clinician treats both the mental health and drug use disorder concurrently. If additional

expertise is required, a shared care approach between the mental health and AOD services is then

utilised. This model aims to integrate at the clinician and service level and to build working

relationship between the mental health and AOD services. In practice, the COMPASS program

works towards engaging and case-managing clients with a mental health disorder within the

mental health service, and then involves AOD services when necessary.

The services provided by the COMPASS program include an intensive intervention, delivered by

an Assertive Outreach Team, and a consultation-liaison service. The intensive intervention, called

cognitive-behavioural integrated treatment (C-BIT), was specifically designed to treat dually



diagnosed individuals, especially when clients do not believe their drug use is a problem. C-BIT

uses a harm reduction approach and aims to promote self-management of recovery. Similar to the

IDDT model, C-BIT consists of four treatment phases (Graham, Copello, Birchwood, Orford, et al.,

2003):

1. engagement phase, which involves establishing and building a therapeutic relationship

with clients without discussing the drug use behaviour

2. negotiating behaviour change phase, which attempts to increase client awareness of their

problematic alcohol or other drug use and increase motivation to change the behaviour

3. early relapse prevention phase, in which therapists help clients reduce their alcohol or

other drug use and achieve abstinence

4. relapse prevention/management phase, which attempts to increase client awareness that

relapses are common.

In addition, the C-BIT has a screening and assessment phase that aims to:

1. assess the types of drugs used and frequency of use

2. assist in treatment planning

3. assist clients in acknowledging the problems associated with their alcohol or other drug

use.

The consultation-liaison service, provided by the COMPASS program, is offered to inpatient

facilities and non-assertive outreach community mental health or AOD services. As part of this

service, clinicians can assess and, where required, provide brief treatment to individuals with more

severe forms of mental health and drug use problems. An evaluation of the COMPASS program

was conducted by training and supervising five existing assertive outreach teams to deliver the C-

BIT intervention in the UK (Graham et al., 2006). The aim of this study was to determine if

integration within existing services could be achieved and if it would have positive outcomes at the

client level. The results indicate that integration went well and changes in teams occurred and that

the staff members reported an increased confidence in their ability to deliver the C-BIT. Results

relating to client outcomes were limited, however, due to the small study sample. Overall, the

results indicated that all clients, regardless of whether the team had received the C-BIT training

immediately or delayed (wait-list control group equivalent), showed improvements in client

engagement and reductions in alcohol use and positive alcohol-related beliefs.

Comprehensive, Continuous, Integrated System of Care model

Another more recent model of care is the Comprehensive, Continuous, Integrated System of Care

(CCISC) model, which has been recognised as one of the best co-occurring mental health and

drug use disorder treatment protocols (Minkoff & Cline, 2004). The CCISC model follows four

characteristics, including change at a system level, use of existing resources, incorporation of best

practices, and a philosophy of integrated treatment (Harrison, Moore, Young, Flink, & Ochshorn,

2008; Young et al., 2009). It was built on eight evidence-based organising principles that ultimately

provide a framework that can be used to design a ‘welcoming, accessible, integrated, continuous,

and comprehensive system of care’ (Minkoff & Cline, 2004, p. 4). These principles are that:



dual diagnosis should be treated as an expectation, not an exception

the four-quadrant model for categorising co-occurring disorders should be utilised as a

service planning guide (categorised as high or low severity for each disorder)

integrated treatment should be continuous

clients can be empowered by balancing caretaking and confrontation

both co-occurring disorders should be viewed as primary

both disease and recovery models should be utilised

treatment should be individualised based on each client’s motivation and diagnoses

each disorder should receive an individualised care assessment (Minkoff & Cline, 2004).

In an evaluation of the CCISC model, Harrison et al. (2008) conducted a pre-post design study in a

sample of 76 dually diagnosed homeless participants. The results demonstrated significant

improvements from baseline to the six-month follow-up evaluation in measures of housing and

employment status, mental health symptomatology, and drug use. The effectiveness of the CCISC

model has also been compared to an ACT model in a sample of homeless individuals with

complex behavioural needs (Young et al., 2009). In this study, the CCISC model was implemented

in an inpatient treatment facility, whereas the ACT model was implemented in a community mental

health centre and had an additional supportive housing component. The findings revealed that

both groups had significant reductions in drug use, decreases in mental health symptoms and

improvements in residential stability. The CCISC group, however, was associated with marginally

greater reductions in the mental health symptoms, while the ACT group were more likely to have

their own place of residence.

Collaborative Early Identification model

Following a consultation with consumers and service providers, the Collaborative Early

Identification Model was developed as a practical model for dually diagnosed individuals in

Australia (Staiger et al., 2008). Key elements of the Collaborative Early Identification Model

include:

a no wrong door approach which allows for a centralised screening process

systematic screening across all relevant entry points (i.e., mental health and AOD

services) using brief, easy to administer and psychometrically sound screening tools

a two-step approach to screening including an initial brief screening measure followed by a

more comprehensive assessment measure for those who screened positive

services tailoring the strategy to suit the service’s individual requirements

integration into existing intake/assessment procedures

employing approaches that enable client engagement



clinical guidelines that direct client care after early identification to the appropriate care

pathway/referral.

The Collaborative Early Identification Model was implemented into a mental health emergency

crisis assessment team and an outpatient alcohol and drug services intake system in Australia.

Evaluation of this implementation indicated that clients responded either positively or neutrally to

the screening process and that clinicians reported that the screening process provided a

supportive and structured chance for clients to open up about any drug use or mental health

problems (Staiger et al., 2008). Although some clinicians reported experiencing difficulty in asking

the screening questions due to their sensitive nature, they also reported that screening process

enabled them to have an appropriate and respectful discussion with clients about treatment and

referrals and risky drug use.

Youth treatment model

A model of care for youth within the alcohol or other drug sector (Lubman et al., 2008) was

developed in Australia, originally as a way to improve the capacity of AOD service staff to manage

young people with co-occurring disorders. This model included training alcohol or drug workers in

the use of an evidence-based brief intervention, which used MI, CBT, and mindfulness skills within

a harm minimisation framework. An evaluation of this model indicated that while training AOD

workers in the use of this brief intervention had a positive impact on their knowledge and skills,

there were a number of systemic issues that influenced the use of this training in their everyday

practice. These included high caseloads and limited time to deliver the intervention, a lack of

confidence in delivering the intervention, inconsistencies in theoretical orientation, and difficulties

attending supervision sessions. Based on this evaluation, changes were made to the model to

include universal screening of mental health problems in the AOD services and brief intervention

conducted by a clinical psychologist embedded within the service, in addition to the standard

treatment for alcohol and other drug use problems, when treatment for mental health problems

was required. The team, which includes the embedded clinical psychologist, psychiatric trainee

and consultant, also provide secondary consultation to AOD workers and the embedded GPs who

provide medical care to clients of this service.

Other treatment models

Finally, several models of care have been described in the dual diagnosis literature for which there

is no available evaluation data. This includes a range of existing services within the mental health

sector in Australia, such as the Substance Use and Mental Illness Treatment Team (SUMITT:

(Melbourne Health, 2016), the Hunter New England Mental Health and Substance Use Service

(MHSUS: (NSW Health, 2015) and Co-Exist (Co-Exist NSW, 2008). As part of the SUMITT model,

a clinician co-located at various mental health services provides clinical services to dually

diagnosed individuals to improve their health outcomes, upgrade the skills of clinicians across the

mental health and AOD services, and improve the effectiveness of the service system in providing

care and treatment to dually diagnosed individuals. The MHSUS consists of an inpatient service

and a community team that provides individual and group treatment programs to assist clients with

engaging with treatment and increasing motivation. This service emphasises the need for

management plans that take into consideration the needs of the client, family, carers and others

involved in the care of the individual. Both of these models also involve a comprehensive

biopsychosocial assessment of each of the disorders. Co-Exist NSW uses the clinical brokerage

model to provide assistance with issues of comorbidity in a linguistically and culturally informed

manner.



Additional models of care developed based on clinical expertise or reviews of the literature that

have been adopted in Australia include the Collaborative Practice Development Model (Monisse-

Redman, 2015), the New South Wales (NSW) pathway of care (NSW Health, 2015) and a model

for responding to dual diagnosis employed by the Victorian government (Victorian Government:

Department of Human Services, 2007). The Collaborative Practice Development Model for

adolescents with co-occurring mental health, drug use and impulse control disorders involves a

partnership between AOD services, local GPs and a clinical psychologist. The aim of the program

is to provide a simultaneous clinical psychology service to young people with comorbid mental

health and drug use problems that enables accessible and affordable clinical interventions, ease of

referral from the AOD service to the clinical psychology service, the provision of flexible,

collaborative and comprehensive assessment and treatment of clients in the AOD service, and

ensures that the focus on youth is promoted and maintained. The Victorian government has also

developed a three level schema for responding to dual diagnosis across the mental health and

AOD service sector, whereby recommendations for the service sector responsible for treating the

individual is provided based on the severity level of each co-occurring disorder. Most recently, the

NSW government proposed a pathway of care to be used at a practice level in mental health

services, based on the Lubman et al. (2008) model of care for youth in the alcohol or other drug

sector. Further information on these models of care can be found in a recent evidence review

check conducted by the NSW government (NSW Health, 2015).

Conclusion

NSW Health (2015) has outlined a core minimum set of model features. Despite these features

focusing on models of care situated within the mental health service sector, these features are also

applicable to the AOD service sector (NSW Health, 2015). These features include universal

screening across all mental health practitioners and services, conducting risk assessment and

diagnosis of symptoms, providing supportive therapies (e.g. MI, CBT and medication),

incorporating prevention and psychoeducation in relation to alcohol and other drug use, and

collaborating with AOD services and/or GPs, when required. Finally, policies and procedures

should be in place that enable assertive follow-up and allow the service to act as the primary care

coordinator until the client is accepted within an alternative service.

Taken together, the findings presented in this section suggest that there is considerable overlap

between models of care. Common components include the use of multidisciplinary teams,

continuous treatment teams, an integrated treatment philosophy, and stage-wise or tailored

treatment approaches. Overall, the earlier models of care, such as the ACT and case management

models, focused on integration at a treatment component level and the development of

community-based treatment approaches. Subsequent models of care, such as the IDDT model,

however, are more comprehensive or holistic and consider how integration should occur at other

levels, such as the administrative and service levels. Additionally, some of the more recent models

of care, such as the IDDT, COMPASS, CCISC, and the model implemented by the Victorian

government, have acknowledged the potential variability in the severity levels of the co-occurring

disorders and have included approaches that facilitate the tailoring of treatment based on severity

level and motivation for change. Moreover, some of the more recently developed models of care,

such as the Collaborative Early Identification, youth treatment, and NSW government pathway of

care models, have clearly outlined the pathways into the relevant treatment and service streams,

highlighted the need for universal screening, and streamlined referral and follow-up processes,

dependent on the severity of the co-occurring conditions.



Evaluations of these models reveal promising results, with most appearing to result in reductions in

alcohol use, drug use, and psychiatric symptomatology. Caution should, however, be taken when

interpreting the results of these evaluations due to the design of the evaluations (such as small

samples and an absence of comparison groups). Moreover, most of the available models of care

have limited data evaluating their effectiveness. There is a clear need for future research to

determine the effectiveness of each model, not only at a client/user level (e.g. outcomes relating to

alcohol or other drug use), but also at a service level to determine the effectiveness of the system,

including the screening and referral processes, as well as the cost-effectiveness of the model.

Section 6: Overall conclusion of literature review

The literature review described research findings relating to the screening, assessment, and

management of problem gambling within Victorian mental health services. Despite experiencing

considerable gambling harms, there are relatively low rates of help-seeking for gambling problems

in the Victorian population. Problem gambling, however, has been consistently associated with a

range of comorbid mental health disorders and people with gambling problems appear to be over-

represented in primary care, AOD, and mental health populations. This suggests that there may be

the potential for health professionals to play a role in the identification and management of

gambling problems. Moreover, there is some evidence that problem gambling precedes and

predicts the onset of several mental health conditions and that psychiatric comorbidity in problem

gambling is associated with more complex clinical presentations. Taken together, these findings

suggest that comorbid problem gambling has the potential to compromise engagement in

treatment, complicate treatment plans and hamper treatment outcomes for mental health

treatment, particularly if it goes unidentified and untreated. These findings highlight the importance

of identifying problem gambling through screening within primary care, AOD, and mental health

settings, with a view to generalist first level gambling interventions conducted within the service or

appropriate referral to specialist gambling services for management of the gambling problem.

Routine and accurate screening has the potential to improve care and reduce healthcare costs.

Although health providers acknowledge that they have a role to play in helping clients with

gambling problems, there is some evidence to suggest that there are low rates of screening for

problem gambling within primary care, AOD, and mental health services. Perceived barriers to

such screening include lack of time, an absence of information about the effectiveness of

screening, a lack of knowledge and skills, the presence of gambling-related stigma, a perception

that problem gambling has a low burden of disease, an absence of effective interventions and

limited access to specialist referral services.

The low rate of screening in primary care, AOD, and mental health services may also have been,

in part, due to the slow development of brief screening instruments that are easy to use, have short

administration times, are easily scored, and require minimal training. While there are now several

brief screening instruments for problem gambling available, there is currently limited information

available to guide the selection of an instrument in these services. To date, the BBGS, CHAT,

Lie/Bet, NODS-CLiP and NODS-PERC have the strongest evidence base for their use in these

settings. Although some of these screening instruments are characterised by low specificity in

clinical samples, it has been argued that the aim in clinical settings is to capture people with

gambling problems, even at the expense of including relatively large numbers of non-problem

gamblers, as there are likely opportunities for further assessment of problem gambling. The

selection of a brief screening instrument, however, may best be determined by the needs of the

clinical setting, such as the number of items or duration of administration, the age of the clinical

population, and the timeframe of the instrument.



A positive screen on a brief instrument should trigger a more comprehensive clinical assessment

via a simple self-report measure, such as the PGSI, or a clinical interview to determine problem

severity or diagnostic status and provide information that can assist in referral decisions and/or

treatment planning. Following this assessment, it is important that individuals with gambling

problems are offered appropriate management. The assessment and intervention of the gambling

problem can be conducted within the primary care, AOD or mental health service if time, skills, and

resources permit or can be conducted by a specialist gambling service. Although a diverse range

of psychological and pharmacological options for the treatment of problem gambling are available,

there is very little evidence on which to base treatment recommendations for different

subpopulations of problem gamblers based on their psychiatric comorbidity. Brief interventions and

online self-help programs based on MI and CBT therapies may, however, be useful resources for

clinicians working with clients with comorbid gambling problems within primary care, AOD, and

mental health services. Although there is limited empirical knowledge about models of care for

gambling and these services, common components of treatment models from the more extensive

dual diagnosis (drug use and mental health) literature include the use of multidisciplinary teams,

continuous treatment teams, an integrated treatment philosophy, and stage-wise or tailored

treatment approaches. A core minimum set of model features include universal screening across

all mental health practitioners and services, risk assessment and diagnosis of symptoms,

supportive therapies, prevention and psychoeducation, collaboration with AOD services and/or

GPs, and supportive policies and procedures.



Study 2a: Clinician survey: Current practice and

responses to gambling

Aims

The aim of this study was to examine clinicians’ attitudes towards problem gambling, existing

training opportunities, current screening practices, referral systems and treatment approaches to

problem gambling across a range of mental health treatment settings.

Method

Design and participants

Using a cross sectional design, mental health workers (N = 311) with a diverse range of

backgrounds were recruited for this survey. This included professionals working at the forefront in

mental health services, seeing a large number of patients (e.g. nurses, psychologists, case

managers, doctors), as well as service and program managers. Workers were recruited from a

range of mental health services in Victoria between September 2014 and March 2015.

Sites

Mental health services selected as study sites in this project were representative of the wide range

of services from which individuals seek help for mental health conditions, the various target

populations (e.g. adult or youth), and geographical spread. Participating services included three

public area mental health services (including inner metropolitan, outer metropolitan and regional

sites) which offer catchment-based inpatient and outpatient clinical care and case management; a

state-wide mental health community support services (MHCSS) which offers outreach

psychosocial rehabilitation and support; two private psychiatry outpatient clinics; and a primary

healthcare community health service that provides general health and psychological support to

clients with mental health issues. The metropolitan services themselves also represented areas

with differing socio-economic status. There were a total of 436 clinicians working at the frontline in

the services at the time the study was conducted, meaning the sample represented 71.3 per cent

of all eligible clinicians.

Measures

To explore clinician practices, knowledge and attitudes across the targeted services, a self-

completion questionnaire was developed specifically for the study (see Table 8 for a summary of

survey sections and Appendix 1 for the actual questionnaire). The questionnaire included items

adapted from Hides et al. (2007), a previous study conducted by the research team to determine

the feasibility and acceptability of a mental health screening tool among healthcare workers in the

alcohol and drug sector.

Items on the clinician questionnaire included a demographics section (including time worked with

the service and training history) and an estimate (expressed as a proportion) of caseload relating

to gambling. There was also a section that assessed clinicians’ knowledge about and attitude

towards gambling and mental health, and another section that focused on role legitimacy (i.e.,

whether or not clinicians believe that gambling is within the scope of their role responsibilities).



The final section identified clinicians’ current screening, assessment, treatment and referral

practices and their level of confidence in undertaking these tasks with the patient population.

The questionnaire comprised mostly of Likert scales, where clinicians were required to respond to

statements like: ‘Problem gambling and mental illness commonly occur together’ with response

options ranging from ‘strongly disagree’ to ‘strongly agree’. Other items required dichotomous

responses, for example: ‘Do you know where to refer clients with a gambling problem’, ‘yes or no’.

Several items assessed the frequency of clinicians’ current practices, for example: ‘How often do

you screen for problem gambling?’ where response options included ‘never, rarely, sometimes,

often, or almost always’. Other items assessed level of comfort or confidence in responding to

problem gambling, for example: ‘How confident are you in assessing problem gambling?’, where

clinicians could answer: not confident, somewhat confident, moderately confident or very confident.

There were also open ended questions where clinicians could describe their training history and

reasons for not screening patients for problem gambling.

Following piloting of the three-page questionnaire and amendments, an online version was made

available via Qualtrics software. The survey took approximately 15 minutes to complete and was

completed anonymously.

Procedures

The research team identified services representative of community-delivered private and public

service provision, including two services serving large metropolitan populations, two large private

services and two regional services and based on existing networks or former collaborations

identified suitable youth, primary health and support services. The lead PI then contacted each of

the sites’ directors to explain the nature of the research and invite them each to participate. All nine

services and 14 sites agreed to participate in the study.

The research team undertook a series of briefings at the services (e.g. at departmental and team

meetings, and at clinical review meetings) to introduce the project to clinicians and to explain how

the findings could benefit clinicians’ practice and the service. The research team worked with

practice managers to identify clinicians best placed to assist with the project and to develop the

best approaches to promote the survey at each service.

The survey was administered to clinicians in one of two ways, depending on local needs: either via

a hard copy distributed during staff meetings and collected by the research team, or online, via an

email with an invitation sent from the research team via the practice manager to participate and a

link to the survey on Qualtrics. The research team worked closely with practice managers and

team leaders to monitor and enhance completion of the survey, send reminder emails (when

response rate was <50 per cent) and undertook follow-up visits and phone calls where necessary.

Clinicians were given the opportunity to enter a draw to win a Samsung Tablet as an incentive to

complete the survey, and details were detached from their responses to protect confidentiality.

The study was given approval by the Eastern Health Human Research Ethics committee, approval

reference number: LR120/1314 and additional ethical review was undertaken and approval

granted by sites not directly covered by the Eastern Health Human Research Ethics committee.

The research protocol and Eastern Health ethics application was reviewed and approved by

governing bodies, head offices or CEOs at other mental health services without formal ethics

committees. Data collection took place between September 2014 and March 2015.



Table 8. Structure of the clinicians' survey

Section Number of items

Type of questions

Topic

Purpose

1 8

Forced choice

Demographics

To describe the demographic characteristics of the sample

2 3

Likert scale

Knowledge

To assess clinicians’ current knowledge of mental health and gambling

3 21

Forced choice

Likert scale

Multiple choice

Screening and assessment

To identify clinicians’ current practice in screening and assessment for problem gambling, including establishing:

whether or not clinicians ask patients about gambling, or screen for gambling problems;

method of screening for gambling problems in patients;

level of comfort around asking patients about gambling;

level of confidence in their ability to detect problem gambling with patients;

level of confidence in assessing for problem gambling with their patients;

actions clinicians currently take when they identify a patient experiencing problem gambling; and

the mental illnesses among which problem gambling is most commonly observed by clinicians.

4 7

Likert scale

Multiple choice

Open ended

Referral practices

To identify clinicians’ current referral practices, including:

how often patients are referred to other services for gambling problems;

if clinicians know where to refer patients to if they have a gambling problem, and, if so, where they refer patient to; and

confidence in making a referral to other services if a patient has a gambling problem.

5 6

Likert scale

Multiple choice

Treatment

To identify clinicians’ current treatment practices with patients who have gambling problems, including:

how often clinicians treats patients with gambling problems;

what treatment for problem gambling is offered at the services; and

confidence in treating gambling problems.

Data analysis

Data were pooled from 14 different sites across nine different mental health services. Statistical

analyses were conducted using Statistical Package for the Social Sciences (SPSS; version 22.0,

IBM Inc., Chicago).

Descriptive statistics were generated (means, standard deviations, percentages) to explore

clinician characteristics (i.e., gender, age, occupation, type and location of mental health service at

which the clinician works). Analyses were performed to identify clinician and service practices

around screening, referral and treatment for problem gambling, including the proportion of

clinicians with training in problem gambling, the proportion of patients in the clinicians’ caseload

affected by problem gambling, and the proportions of clinicians and services routinely screening,

referring and treating patients for gambling problems. The proportions expressed in tables and

figures exclude missing data which represented fewer than five cases on any item. The raw data

including missing responses are displayed in the corresponding tables in the appendices.



Associations between problem gambling represented by categorical data (i.e., level of clinicians’

confidence in screening and referring patients for problem gambling) and past training in problem

gambling were explored using Pearson chi-square tests. When expected frequency criteria were

not met due to small cell sample size, adjacent categories were collapsed into smaller categories

(where appropriate) in order to fulfil the necessary Pearson chi-square requirements and to gain

statistical power, using Fisher exact tests where appropriate. For example, Likert scale items (e.g.

‘How often do you ask patients about their gambling?’) were simplified and collapsed into two

categories (the response options ‘never’ was grouped as ‘no screen’; and rarely, sometimes, often

or always response options grouped as ‘screening’). Differences in mean scores on the sub-scales

(e.g. total score on items assessing knowledge) for prior training status, gender etc. were

examined using independent samples t-tests. Differences in the estimates of the mean proportion

of the clinicians’ caseloads affected by gambling, by healthcare profession, was explored using

ANOVA. Dependent on sample distribution, one-way analyses of variance or non-parametric

equivalents for continuous variables were used to compare ratings across sites or subgroups of

respondents. For a breakdown of responses to individual items see Tables 24–40 in Appendix 2.

Results

Please note that detailed data tables arising from the clinicians’ survey can be found in

Appendix 2.

Demographic data

Table 9 displays the demographic characteristics of the clinicians who participated in the survey.

Of the 311 mental health professionals who participated in the clinician’s survey, the majority were

female (72.7 per cent), with a mean age of 40.1 years (SD = 11.1 years). The most commonly

reported professions were nurse (25.7 per cent), medical (20.9 per cent), and support worker (18.0

per cent). Overall, the majority of clinicians were employed at public mental health services for

adults (65.3 per cent), followed by mental health community support services (14.5 per cent), and

public mental health services for children and adolescents (9.6 per cent).

Table 9. Demographic characteristics of the clinicians who participated in the survey

Total sample N = 311

Mean age (range)

(SD)

40.1 years (20–67 years)

(11.1 years)

Gender

Male

Female

27.3%

72.7%

Profession

Nurse

Doctors

Support Worker

Social Worker

Occupational Therapist

Psychologist

Other

Missing

25.7%

20.9%

18.0%

11.9%

8.7%

8.0%

5.8%

1.0%




Type of service

Public mental health service – adult

Mental health community support service

Public mental health service – child/adolescent

Primary health care

Private mental health service

65.3%

14.5%

9.6%

7.4%

3.2%

Mean practice duration years

(range)

12.08 years

(<1 year – 40 years)

Previous training in problem gambling

No

Yes

86.8%

12.2%

Estimated proportion of clinician caseload with gambling problems (range)

10.24%

(0 – 65%)

Overall, clinicians had many years of practice experience, with a mean of 12.08 years (SD = 10.45

years). Despite being highly experienced at delivering services to patients with mental health

problems, only a small proportion of clinicians (12.2 per cent; 38 out of 311) had received previous

training in problem gambling and its impact on this population. The nature of training reported was

highly variable, with several clinicians reporting one-off workshops typically lasting less than a day,

while for others training ranged from as little as a single webinar, through clinical discussions,

supervision or in-house training to academic units on masters courses and accredited college

training. There was also a significant difference in past training rates across professions (2 =

16.54, p<0.05), with psychologists (23.7 per cent), nurses (23.7 per cent) and doctors (21.1 per

cent), more likely to have received training compared to social workers, support workers (both 13.2

per cent), occupational therapists (5.3 per cent) and those in the ‘other’ profession category (0 per

cent) as shown in Figure 1.

Figure 1. Proportion of clinicians with previous training in problem gambling by profession

Clinicians were asked to estimate the proportion of their caseload involving problem gambling, and

the mean estimate was 10.2 per cent (SD = 11.15 per cent), ranging from 0 to 65 per cent of total

caseloads (see Table 9). Further analyses revealed that 19.8 per cent of clinicians reported no

patients with problem gambling in their caseloads, while 32.5 per cent reported a prevalence rate

Psychologists24%

Nurses24%

Doctors21%

Social workers13%

Support workers

13%

Occupational therapists

5%



of 1–9 per cent, 39 per cent reported a prevalence rate of 10–25 per cent, 7.8 per cent reported a

prevalence rate of 26–50 per cent, and 1.3 per cent reported more than 50 per cent of their

caseloads involved patients with problem gambling (see Figure 2). Male clinicians were

significantly more likely to report a higher proportion of patients in their caseload experiencing

problem gambling (M = 12.86, SD = 13.37) than were female clinicians (M = 9.25, SD = 10.05),

t(306) = 2.24, p<0.05, 95 per cent CI [0.42, 6.78], which may reflect the higher male/female ratio of

their caseloads. Nurses, social workers and support workers reported a higher mean proportion of

their patients experiencing problem gambling than psychologists, doctors and occupational

therapists, though this was not statistically significant (F = 2.0, p = 0.06).

Figure 2. Clinicians' estimates of the proportion of their caseloads involving patients with gambling problems

Comorbidity of problem gambling and mental health illness

Clinicians were asked to identify the most common mental health disorders seen in their practice,

where comorbid problem gambling is also observed. As shown in Figure 3, clinicians reported that,

in their experience, problem gambling most commonly occurs with alcohol use disorder (70.7 per

cent), followed by mania/bipolar disorder (55.3 per cent), then drug use disorder (44.7 per cent),

personality disorders (40.5 per cent) and major depression (40.2 per cent).

No problem gambling(19.8%)

1-9%(32.5%)

10-25% (39.0%)

26-50% (7.80%)

>50%(1.3%)



Figure 3. Mental health illnesses clinicians associate with comorbid problem gambling

Knowledge about gambling and mental illness

The level of clinicians’ current knowledge of the bi-directional relationship between problem

gambling and mental illness was assessed by their response to four items (see Table 10 for

questions on this scale). Higher total scores indicate higher levels of knowledge about gambling in

the context of mental illness. Overall, the mean total score for was 13.6 (SD = 1.95) out of a

maximum possible score of 20, indicating that clinicians had a reasonable level of knowledge.

Clinicians who had received previous training in problems gambling had a significantly higher

mean total knowledge score (M = 14.7, SD = 1.65) than did clinicians who had not received

previous training (M = 13.45, SD = 1.94), t(306) = 3.66, p<0.001.

A closer look at individual items (as shown in Table 10), indicated that the majority of clinicians

(64.6 per cent) demonstrated a good understanding of the comorbidity of problem gambling and

mental illness, as evidenced by agreeing or strongly agreeing with the statement: ‘Problem

gambling and mental illness commonly occur together’. Most clinicians also had a good

understanding of the impact of problem gambling on the severity of a patient’s mental illness, with

95.2 per cent of the sample agreeing or strongly agreeing with the statement: ‘Problem gambling

can worsen a client’s mental illness’. Fewer clinicians (46.7 per cent) reported understanding what

causes and/or maintains problem gambling issues.

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Alcohol use disorder

Mania/Bipolar disorder

Drug use disorder

Personality disorders

Major depression

Generalised anxiety

Psychosis

Post-traumatic stress disorder

Social phobia

Other

Panic/Agoraphobia

Eating disorder



Table 10. Clinicians' responses on the current knowledge about gambling and mental

illness survey


Problem gambling and mental illness commonly occur together

Strongly agree

Agree

Uncertain

Disagree

Strongly disagree

11.9%

52.7%

28.3%

7.1%

0%

Problem gambling can worsen a client’s mental illness

Strongly agree

Agree

Uncertain

Disagree

Strongly disagree

41.3%

53.9%

4.5%

0.3%

0%

I understand what causes and/or maintains problem gambling

Strongly agree

Agree

Uncertain

Disagree

Strongly disagree

3.2%

43.5%

39.0%

12.9%

1.3%

I am aware of what screening and assessment tools are available to detect problem gambling

Strongly agree

Agree

Uncertain

Disagree

Strongly disagree

0.3%

10.3%

21.9%

48.8%

19.0%

Current screening and assessment practices

Clinicians’ current screening and assessment practices with their patients for problem gambling

were explored using a 21-item questionnaire. Figure 4 displays the frequency that clinicians ask

their patients about gambling at all (i.e., not just problem gambling). While 87.3 per cent of

clinicians reported asking patients about their gambling behaviour, only a minority reported asking

‘almost always’ (4.2 per cent) or ‘often’ (17.2 per cent). Just over a third of clinicians (38.0 per cent)

reported ‘sometimes’ asking their patients about gambling. Forty per cent of clinicians reported

‘never’ (12.5 per cent) or ‘rarely’ (27.9 per cent) asking their patients about their gambling

behaviour.



Figure 4. Frequency that clinicians ask patients about gambling

Figure 5 displays the frequencies that clinicians screen their patients for problem gambling. Again,

while overall 77.0 per cent of clinicians report screening patients for problem gambling, only a

minority of report doing it almost always (3.9 per cent) or often (11.3 per cent). Although just under

a third of clinicians (29.1 per cent) reported sometimes screening for problem gambling, the

majority reported screening only rarely (32.7 per cent) or never (22.8 per cent). In addition, those

with former training were not significantly more likely to screen at least sometimes than clinicians

without prior training (2 = 1.57, p = 0.21).

Figure 5. Frequency that clinicians screen patients for problem gambling

When clinicians who do screen for problem gambling at least rarely (n = 238) were asked the

method used to detect problem gambling. Of the 260 who responded to this item, the vast majority

(91.5 per cent) reported that this was achieved through ‘informal discussion’, 6.5 per cent reported

‘using set questions included in the service’s intake assessment’ and only 1.9 per cent reported

using a standardised or formal gambling screening instrument (see Figure 6).

Almost always4.2%

Often17.2%

Sometimes38.0%

Rarely27.9%

Never12.7%

Almost always(3.9%) Often

(11.3%)

Sometimes(28.9%)

Rarely(32.5%)

Never (22.8%)

Missing(0.6%)



Figure 6. Methods clinicians used to identify problem gambling

Several key themes emerged from the reasons clinicians gave for not screening patients for

problem gambling (see Table 11). The main themes identified relate to systemic issues (e.g.

screening not a part of standard intake assessment, or not an organisation requirement),

knowledge base and training issues (e.g. lack of knowledge of how to screen for gambling

problems), conflicting treatment priorities (e.g. other issues that patients present with are a higher

priority), and role issues (e.g. screening for gambling is outside the scope of a clinician’s role)

Table 11. Key themes arising from the reasons clinicians gave for not screening for problem

gambling

Theme Example

Outside of role responsibilities ‘Role is medical intake’

‘My scope of practice does not include this role’

Lack of knowledge about and training in appropriate problem gambling screens

‘Unfamiliar with screening’

‘… lack of knowledge of how to screen and approach gambling problems’

Problem gambling screening not an organisational requirement

‘Not required to’

‘The organisation requires screening for health and AOD use but not gambling’

Problem gambling screening not a part of standard intake assessment

‘…because not a part of standard assessment…’

‘It is not part of our general screening tool’

Pursued only when client mentions issues

‘Patients have not given indication that gambling is a problem’

‘…only if client mentions it’

Conflicting treatment priorities ‘From my own personal experience, many of my consumers are more concerned with how they can best budget their Centrelink payments to cover bills, accommodation, food, petrol/transport and education costs.’

Clinicians were also asked to indicate how comfortable they were with asking patients about their

gambling behaviours. Overall, the majority of clinicians (82.6 per cent) reported that they felt

comfortable asking patients about gambling, 39.2 per cent reported being very comfortable and

43.4 per cent reported being somewhat comfortable asking patients about their gambling

behaviours. Only a minority of clinicians reported being somewhat uncomfortable (12.5 per cent) or

very uncomfortable (4.8 per cent) about asking patients about their gambling behaviours.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Informal discussion Set questions in service’s intake

assessment

Standardised orformal gambling

screening instrument

% o

f c

lin

icia

ns



Just over a third of clinicians (38.9 per cent) reported feeling confident about detecting and/or

screening for problem gambling with their patients (4.5 per cent very confident, 34.4 per cent

moderately confident; see Figure 7). This means that with the majority of clinicians (61.1 per cent)

reported they were not confident (only somewhat confident 41.4 per cent, not confident 19.4 per

cent). Further analysis revealed that the proportion of clinicians confident in detecting or screening

for problem gambling with their patients was significantly greater among those who had received

prior training (2 = 6.35, p< 0.05).

Only 35.3 per cent of clinicians reported being confident in assessing their patients for problem

gambling (4.5 per cent very confident and 30.8 per cent moderately confident). This means that the

majority of clinicians (64.7 per cent) were not confident (only somewhat confident 40.3 per cent,

not confident at all 24.4 per cent). Further analysis revealed that the proportion of clinicians

confident in assessing their patients for problem gambling was significantly greater among those

who had received prior training (2 = 12.0, p<0.01).

Figure 7. Clinicians' level of confidence in detecting or screening patients for problem gambling

Clinicians were also asked about their current practices when they identify a patient experiencing

problem gambling. Figure 8 displays what clinicians currently do when patients with gambling

problems are identified. It should be noted that with the exception of the ‘nothing’ and ‘never

identified a problem gambler’ responses, responses are not mutually exclusive responses (i.e.,

clinicians could endorse more than one action), so these total more than 100 per cent. The most

commonly endorsed responses were referral to external gambling treatment provider (66.6 per

cent), followed by address financial or social consequences (44.7 per cent), and conduct further

assessment (38.6 per cent). Encouragingly, only 1.0 per cent of clinicians reported that the ‘do

nothing’ when they identified patients experiencing problem gambling.

Very confident

(4.5%)

Moderately confident(34.4%)

Somewhat confident(41.2%)

Not confident(19.3%)

Missing(0.6%)



Figure 8. Clinicians' current responses when patients experiencing problem gambling are identified

Clinicians’ attitudes towards problem gambling

Clinicians attitudes towards problem gambling, how it relates to mental health, and screening and

assessment of their patients for problem gambling was explored using an 11 item questionnaire,

with each item utilising a five-point Likert scale (strongly disagree, disagree, uncertain, agree,

strongly agree). The minimum score possible on this inventory is -22 with a maximum score of 22.

Negative scores may be indicative of a reluctance to deal with gambling issues. Table 12 displays

the items on this questionnaire and the proportion of clinicians who reported each type of response

on the Likert scale.

Overall, the mean score on this scale was 8.54 (SD = 4.93), with scores ranging from -13 to 20.

This suggests that overall clinicians were generally willing to address gambling issues with their

patients, when they arise. There was a trend towards more positive attitudes in relation to

responding to gambling among clinicians who had received prior training (M = 9.92 (SD = 4.4)

versus M = 8.38 (SD = 4.96), t(305) = 1.81 p = 0.07). Further analysis also revealed that there was

a difference between clinicians with prior training and those without in responses to the item

‘Detecting problem gambling does not require a formal screen; it can just be addressed if a client

mentions it’. In this case, a larger proportion of clinicians with previous training disagreed or

strongly disagreed with this statement, although this trend failed to reach statistical significance.

There were no other significant differences found between clinician groups in responses on this

scale.

Returning to results for the whole sample, the majority of clinicians (85.4 per cent) disagreed or

strongly disagreed with the statement: ‘There is no point conducting gambling screening as my

service does not treat problem gamblers’. The majority of clinicians also agreed that gambling is a

clinical disorder (78.7 per cent), as evidenced by disagreeing or strongly disagreeing to the

statement: ‘Gambling disorder is not really a clinical disorder’), and that standardised screening

tools did not necessarily only have to be used when patients mention gambling themselves (65.7

per cent, as evidenced by disagreeing or strongly disagreeing to the statement ‘Use of

standardised screening tools is only necessary if a client mentions gambling’). Clinicians also

disagreed that there were too many more important issues to focus on with this population to

0% 10% 20% 30% 40% 50% 60% 70%

Refer to external treatment provider

Address financial or social consequences

Conduct further assessment

Provide psychological treatment

Never identified a problem gambler

Other

Provide pharmacotherapy

Nothing

% of clinicians



warrant screening for gambling problems, with the majority (74.6 per cent) disagreeing or strongly

disagreeing with this statement. The majority of clinicians (79.2 per cent) also disagreed or strongly

disagreed with the statement: ‘Problem gambling does not co-occur with mental health problems

often enough to bother screening’ and the statement ‘Screening/assessment and referral for

problem gambling is not part of my job’. Just over half of clinicians (54.4 per cent) disagreed or

strongly disagreed with the statement ‘There is not enough time to conduct problem gambling

screening or assessment in my workplace’. The majority of clinicians agreed with the statements:

‘It is important to identify gambling problems among mental health clients’ (90.0 per cent) and most

(72.9 per cent) agreed with the statement ‘A brief problem gambling screen would be a useful part

of my routine clinical practice’. Finally, the majority (56.0 per cent) of clinicians disagreed with the

statement ‘patients accessing mental health treatment do not want to be screened for gambling

problems’.

.



Table 12. Clinicians' current attitudes towards problem gambling

RS

Item

(%)

Strongly

agree

Agree Uncertain Disagree Strongly

disagree

* There is no point conducting gambling screening as my service does not treat problem gamblers

3

1.0%

12

3.9%

30

9.7%

178

57.6%

86

27.8%

* Gambling is not really a clinical disorder 5

1.6%

14

4.5%

47

15.2%

183

59.0%

61

19.7%

* Detecting problem gambling does not require a formal screen; it can just be addressed if a client mentions it

2

0.6%

39

12.6%

65

21.0%

171

55.3%

32

10.4%

* Use of standardised screening tools is only necessary if a client mentions gambling

1

0.3%

39

12.7%

80

26.1%

166

51.4%

21

6.8%

* People accessing mental health treatment do not want to be screened for gambling problems

0

0%

28

9.1%

108

35.0%

153

49.5%

20

6.5%

* There are too many more important issues to screen for than problem gambling 1

0.3%

32

10.4%

45

14.6%

184

59.7%

46

14.9%

* Problem gambling does not co-occur with mental health problems often enough to bother screening

0

0%

8

2.6%

56

18.2%

180

58.4%

64

20.8%

* There is not enough time to conduct problem gambling screening or assessment in my workplace

7

2.3%

58

18.9%

75

24.4%

147

47.9%

20

6.5%

* Screening/assessment and referral for problem gambling is not part of my job 5

1.6%

18

5.8%

41

13.3%

186

60.4%

58

18.8%

It is important to identify gambling problems among mental health clients 84

27.1%

195

62.9%

16

5.2%

8

2.6%

7

2.3%

A brief problem gambling screen would be a useful part of my routine clinical practice

55

17.7%

171

55.2%

53

17.1%

21

6.8%

10

3.2%

Scoring note: Strongly agree = 2, agree = 1, uncertain = 0, disagree = -1, strongly disagree = -2

* in the RS column denotes items that are reverse scored: strongly agree = -2, agree = -1, uncertain = 0, disagree = 1, strongly disagree = 2

Score is calculated by summing score on each item. A negative total score may be indicative of reluctance to deal with problem gambling in patients



Current referral practices

Clinicians were asked a number of questions to establish their current referral practices when they

do identify a patient experiencing problem gambling. Figure 9 displays the proportion of clinicians

who report referring patients almost always, often, sometimes, rarely or never.

Figure 9. Frequency clinicians report referring patients experiencing problem gambling to external services

As shown in Figure 9, only 40.1 per cent of clinicians reported that they refer clients with a

gambling problem to external treatment services, at least ‘sometimes’.

When clinicians were asked if they knew where they could refer patients who are experiencing

problem gambling, 80.5 per cent responded with ‘yes’ (19.5 per cent responded with ‘no’). Figure

10 describes where clinicians are likely to actually refer patients when problem gambling is

identified. The majority of clinicians reported referring patients to Gambler’s Help services,

delivered either face-to-face (49.2 per cent) or over the telephone (49.5 per cent). Clinicians also

reported referring to online services (33.4 per cent), to financial counselling services (33.8 per

cent) and to Gamblers Anonymous or peer support (22.2 per cent). Only a small proportion of

clinicians reported referring patients to private psychologists or psychiatrists (12.5 per cent), or to a

specialist, hospital-based gambling service (9.6 per cent), or to private addiction/gambling

therapists (7.1 per cent) and 5.8 per cent referred to other services.

Almost always2.9%

Often8.5%

Sometimes28.7%

Rarely41.7%

Never18.2%



Figure 10. Services to which clinicians report referring patients experiencing problem gambling

Clinicians were also asked to rate how confident they were to refer a patient experiencing problem

gambling to an outside treatment service. Clinicians rated their confidence on a four-point Likert

scale, ranging from ‘not confident’ to ‘very confident’. As shown in Figure 11, only a minority of

clinicians (12.9 per cent) reporting feeling ‘very confident’ about referring patients to services for

problem gambling, with the majority (39.2 per cent) reporting that they felt only ‘somewhat

confident’.

Figure 11. Clinicians' level of confidence in referring patients to outside services for problem gambling

0% 10% 20% 30% 40% 50% 60% 70%

Gamblers Helpline

Gamblers Help Services (face to face)

Financial Counselling

Gambling Help Online

Gamblers Anonymous/Peer Support

Private psychologists/psychiatrists

Specialist Hospital-Based Gambling Service

Private addiction/gambling therapists

Other

% of clinicians

Very confident12.9%

Moderately confident

26.5%

Somewhat confident

39.2%

Not confident21.4%



When clinicians were asked if they thought it was important to refer patients who were

experiencing gambling problems to specialist gambling agencies for further treatment, the majority

(90.0 per cent, see Figure 12) ‘agreed’ or ‘strongly agreed’ that it was important to do so. Figure 13

displays clinicians’ opinions about the capacity of specialist gambling services to deal with patients

with a mental illness. Just over half (52.3 per cent) reported they were uncertain about whether

Gambler’s Help services were equipped to manage comorbid clients (see Figure 13).

Figure 12. Clinicians' opinions about the importance of referring patients to outside agencies when problem

gambling is identified

Figure 13. Clinicians' response to the statement ‘Gambling Help Services are not equipped to deal with clients with

mental illness’

Current treatment practices

Clinicians were asked a number of questions to establish the current treatment practices of both

themselves and the services at which they work. When asked how often they personally treat

Strongly agree30.1%

Agree59.9%

Uncertain9.4%

Disagree/ Strongly disagree

0.6%

Strongly agree0.6%

Agree5.8%

Uncertain52.4%

Disagree34.4%

Strongly disagree

6.8%



patients with gambling problems, the majority of clinicians (66.0 per cent) reported that they did so

rarely (42.5 per cent) or never (23.5 per cent, see Figure 14).

Figure 14. Frequency clinicians report treating patients for problem gambling

When asked about the treatment that they, and/or the service in which they work, offer to patients

experiencing problem gambling, the majority of clinicians (44.7 per cent) reported that counselling

was offered, followed by assessment (40.5 per cent, see Figure 15). Less than 20 per cent of

clinicians reported that they, or their service, offered financial counselling (16.4 per cent), other

treatments (15.1 per cent), medication (11.6 per cent), or peer support (10.3 per cent) to patients

with comorbid problem gambling. Less than 10 per cent of clinicians reported providing financial

aid or relief to patients. Finally, 19.9 per cent of clinicians reported that they, or their service,

offered no treatment for patients who are identified as experiencing problem gambling.

Figure 15. Type of treatment provided by clinicians and/or their service or practice

Clinicians were also asked to rate their level of confidence in treating patients with comorbid

problem gambling. As shown in Figure 16, the majority of clinicians (79.6 per cent) reported that

they were ‘not confident’ (48.3 per cent) or only ‘somewhat confident’ (31.3 per cent) in treating

Almost always0.3%

Often5.2%

Sometimes28.2%

Rarely42.7%

Never23.6%

0% 10% 20% 30% 40% 50% 60% 70%

Counselling

Assessment

None

Financial counselling

Other

Medication

Peer support

Financial Aid/Relief

% of clinicians



problem gambling when it is identified. Further analysis revealed that there was a difference

between clinicians with prior training and those without in responses to the item ‘How confident are

you in treating a client’s problem gambling?’ In this case, a larger proportion of clinicians with

previous training had at least some degree of confidence (76.3 per cent) in their capacity to treat

gambling problems compared to those without previous training (48.0 per cent), (2 = 10.59,

p<0.01).

Figure 16. Clinicians' level of confidence in treating patients for problem gambling

Another question checked clinicians’ understanding of the external Gambler’s Help service system

and the programs available within it. Overall, most clinicians (78.7 per cent) reported that they had

a poor understanding of the Gambler’s Help service system, with just over half (54.5 per cent) of

clinicians ‘disagreeing’ or ‘strongly disagreeing’, almost a quarter (24.2 per cent) uncertain, and

only 21.3 per cent ‘agreeing’ or ‘strongly agreeing’ with the statement, as shown in Figure 17.

Figure 17. Clinicians' level of agreement to the statement: ‘I have a good understanding of the Gamblers' Help

service system and the programs available’

When asked if they understood the types of treatments with empirical evidence of their

effectiveness for problem gambling, most clinicians did not (77.7 per cent), with around half (50.0

per cent) disagreeing or strongly disagreeing, almost a third (27.7 per cent) uncertain, and 22.3 per

cent agreeing or strongly agreeing (see Figure 18).

Very confident

2.6%

Moderately confident

17.8%

Somewhat confident

31.3%

Not confident

48.4%

Strongly agree1.6%

Agree19.7%

Uncertain24.2%

Disagree42.6%

Strongly disagree

11.9%



Figure 18. Clinicians' understanding of the types of treatments effective for problem gambling

Further analysis revealed that there was a difference between clinicians with prior training and

those without in responses to the item ‘I understand the types of treatment that have proven helpful

for problem gambling’, with a significantly larger proportion of clinicians with previous training

agreeing with the above statement (47.7 per cent) than those without previous training (17.8 per

cent) (2 = 17.20, p<0.001).

Finally, when clinicians were asked if they thought mental health and problem gambling clinicians

could work effectively together to support patients with comorbid diagnoses, over 80 per cent of

clinicians agreed (27.5 per cent) or strongly agreed (56.6 per cent) that this was possible (see

Figure 19).

Figure 19. Clinicians' level of agreement to the statement: ‘Mental health and problem gambling clinicians can work

effectively together to support clients’

Discussion

Summary and interpretation of findings

Encouragingly, the clinician survey indicated that mental health clinicians recognise the importance

of responding to gambling issues, yet despite being a highly experienced workforce with an

Strongly agree1.3%

Agree21.0%

Uncertain27.7%

Disagree36.1%

Strongly disagree

13.9%

Strongly agree27.6%

Agree56.6%

Uncertain12.6%

Disagree1.9%

Strongly disagree

1.3%



estimated 10 per cent of their caseload affected by problem gambling, there were low levels of

screening and low confidence in responding to problem gambling. This may reflect the low rates of

previous training in problem gambling (only 12 per cent of clinicians reported having received this)

and limitations in the nature, extent and quality of training, especially as problem gambling is rarely

covered as a topic in clinical graduate training programs. However, since more than one-third

reported that 10-25 per cent of their caseload was affected by gambling, addressing this training

gap is a clear priority.

Overall, clinicians’ knowledge was reasonably high, although almost one in five agreed with the

statement that ‘gambling is not a clinical disorder’. Clinicians without previous training in problem

gambling had lower scores on the items assessing knowledge items and demonstrated more

negative attitudes (i.e. were more likely to disagree with items concerning role legitimacy) than did

clinicians with prior training. Although three-quarters of the sample agreed that a brief screening

tool to detect problem gambling would be useful in their practice, only a minority (10 per cent) were

aware of available tools. These findings further highlight the importance of providing specialist

training in problem gambling to support clinicians’ willingness to respond to gambling issues as

part of their practice. Recognition that clinicians have a significant role to play in the identification

and management of problem gambling supports the findings of earlier studies (Corney, 2011;

Sanju & Gerada, 2011; Sullivan et al., 2000; Sullivan et al., 2007; Temcheff et al., 2014; Tolchard

et al., 2007). Research suggests that shame, denial and stigma are common reasons why people

with gambling problems are reluctant to seek treatment (Suurvali, Cordingley, Hodgins, &

Cunningham, 2009) and there is an estimated five-year latent period between the development of

the problem and professional help-seeking (Tavares, Zilberman, Beites, & Gentil, 2001). Mental

health clinicians are already highly skilled in discussing sensitive issues in stigmatised populations,

and as such, are ideally placed to explore their patients’ gambling behaviour.

Overall, current screening and assessment practices were limited, particularly given the estimated

prevalence of problem gambling among their caseloads. Again, this could be attributable to a lack

of previous training in problem gambling, or a broader systemic issue, whereby services have yet

to develop clear policies around screening and assessment of problem gambling. Although, the

vast majority (82 per cent) were at least ‘somewhat comfortable’ asking about problem gambling,

and 87.3 per cent reported that they did ask about gambling, a significant proportion (40 per cent)

either only asked rarely or never at all. Similarly, while 77 per cent said they screen for problem

gambling, only 4 per cent did so routinely (often or always) and a significant proportion (55 per

cent) reported doing so only rarely or never at all, and the majority reported low confidence in their

ability to detect gambling problems. This mirrors the findings of Achab et al. (2014), who found that

none of the 71 surveyed GPs in their Swiss study systematically screened for excessive gambling,

and is also consistent with the low rates of screening for problem gambling among clinicians

working in primary care and other health settings identified in previous research (Sanju & Gerada,

2011; S. A. Thomas et al., 2008; Tolchard et al., 2007). It is therefore likely that many patients with

gambling issues go undetected in mental health services, meaning that they are likely to miss

opportunities to receive early support or treatment that could benefit them.



The free-text comments on reasons for not screening suggested it was due to a lack of

training/knowledge on how to manage gambling issues, or low perceived need to undertake

screening. Addressing the training gap is critical because research suggests that service provider

training in other areas (e.g. substance use, mental health) is a determinant of whether or not they

screen for such issues (Lubman et al., 2008). The most common method for the identification of

gambling problems was via informal conversations, discussions or questions in sessions or intake

assessments. Only a minority reported using questions included in their service’s intake

assessment, or using a standardised/formal problem gambling screening tools reflecting the poor

knowledge of screening/assessment tools available. However overall, clinicians could see value in

screening for problem gambling and indicated that standardised screening tools should be used

even if patients do not mention problem gambling themselves.

With respect to actions taken when a patient with a gambling problem is identified, only 40 per cent

of clinicians reported referring patients to external services. While low referral rates may reflect the

absence of routine screening, it may also reflect that fact that only one in five (21 per cent)

reported understanding the Gambler’s Help service system and whether they are equipped to

manage comorbid clients. Few clinicians reported providing treatment for patients identified as

problem gamblers. In addition, low confidence in treatment provision was evident and this could be

attributed to poor knowledge of effective treatments for problem gambling, which is consistent with

the findings of Achab et al. (2014). In summary, these findings suggest mental health clinicians

may need to offer strategies and interventions that reduce gambling-related harm and offer

resources, support and referral options. This may require a greater focus on ongoing training and

education initiatives across the mental health sector.

Study limitations

There are a number of important limitations that must be considered when interpreting the findings

of the study. Although 311 clinicians completed the quantitative survey, around 30 per cent of the

workforce did not complete the survey (i.e. chose not to or were absent from work on days of data

collection). The findings may also not be representative of the broader mental health clinical

workforce, particularly those in regional and rural settings, as most participating services were

located in Melbourne. Since participation was voluntary, self-selection bias must be considered as

it is plausible that the clinicians most interested in gambling or with the greatest experience in

managing gambling problems were more willing to complete the survey. The knowledge, healthy

attitudes and proportion screening, assessing, referring and treating problem gamblers could

therefore all potentially be overestimates. Data were self-reported and not based on observed

behaviour. Similarly, it is impossible to eliminate social desirability effects whereby clinicians

respond to questions in a way in which they believe they ought to (i.e., in line with best-practice

rather than their own/actual practice), which could also serve to overestimate clinician responding

to problem gambling. However, it is likely that the ability to complete the survey anonymously and

knowledge that recruitment was overseen by an external agency should have countered these

biases to a certain degree. Despite these potential limitations, it is important to note the study’s

strengths. A large and diverse range of clinicians were sampled in terms of their profession and

level of seniority from a range of mental health services across a number of geographical

locations. As such, it is likely that the reported findings represent multiple perspectives and can be

applied to Victorian mental health services more broadly.



Study 2b: Barriers and facilitators to

responding to problem gambling

Aims

The aims of this chapter are to build on the findings of the workforce survey by examining in

greater detail, current practices in relation to problem gambling identification, referral and

management within the mental health workforce. Using in-depth interviews with mental health

clinicians and managers, this chapter investigates current practice in terms of screening. This

includes exploring the types of tools and questions that are currently administered, the feasibility of

routine screening, capacity to respond and workplace priorities. Of particular interest was

clinicians’ perceptions of patient responsiveness to screening and their willingness to disclose,

whether (as indicated in the literature review) issues such as shame and stigma arose, and

whether clinicians consider screening for problem gambling to undermine the development of the

therapeutic relationship. Finally, methods to aid implementation of screening in mental health

services were explored, including access to appropriate screening tools and clinical expertise, as

well as education and training needs.

Method

Sample characteristics

Semi-structured interviews were conducted with 30 clinicians (n = 17) and managers (n = 13) from

community mental health services in Victoria, Australia. The sample included 19 females and 11

males. Clinicians involved in a direct care role included registered nurses (n = 6), social workers (n

= 5), occupational therapists (n = 3), a case worker (n = 1) and a clinical psychologist (n = 1).

Twenty-four clinicians were recruited from metropolitan services and six were recruited from

regional areas across Victoria.

Procedure

Practitioners were recruited from six large adult (n = 28) and child and youth mental health (n = 2)

services operating across 11 different sites. To ensure a wide range of experiences, recruitment

involved snowball sampling where service managers recruited practitioners from their own

agencies. Services included public community mental health, private mental health, emergency

and crisis support and mental health community support service. Thirty semi-structured individual

interviews were conducted face-to-face or by telephone. Interviews occurred over the months of

January and February 2015. The average duration of each interview was 31 minutes (range 16–57

minutes) and these were digitally recorded and transcribed verbatim. Clinicians were provided a

$40 store voucher for participating. The study was given approval by the Eastern Health Human

Research Ethics committee, approval reference number: LR120/1314 and additional ethical review

was undertaken and approval granted by sites not directly covered by the Eastern Health Human

Research Ethics committee. The research protocol and Eastern Health ethics application was

reviewed and approved by governing bodies, head offices or CEOs at other mental health services

without formal ethics committees. Data collection took place between January 2015 and

October 2015.



The interview schedule focused on identification and management of problem gambling in

community mental health services. Specific questions related to identification included:

How do clinicians usually become aware that a patient has gambling problems?

What do you/your service do to identify patients with gambling problems?

What sort of things impact on your/your service’s ability to identify gambling problems in

patients?

A series of questions were also asked about the management, treatment and referral of problem

gambling in their mental health setting. Indicative questions included:

What would you like your service to do to improve the way in which problem gambling is

identified?

What impacts on the ability of your service to manage gambling problems in patients?

Data analysis

NVivo 10 software was used to conduct the thematic analysis as described by Braun and Clarke

(2006). Transcripts were read and re-read and initial codes developed. Codes were grouped into

themes that were adjusted to capture new and emerging themes until the entirety of the dataset

was coded. A second researcher also coded one complete transcript and similar themes were

identified. Themes were then compared and contrasted with existing literature and finalised with

input from the wider research team.

Results

The thematic analysis identified 10 barriers and five facilitators to screening for problem gambling

in mental health services. As indicated on Table 13, barriers to screening were associated with

clinicians seeing gambling as a low priority, competing demands for clinician time and a lack of

access to tools in addition to client responsiveness and clinician access to resources and training.

Overall there was a growing awareness of problem gambling and the needs to screen, especially

as many clinicians reported that gambling could be an underlying issue or interfered with

treatments for other conditions. This chapter now describes in detail the identified themes.

Table 13. Summary of barriers and facilitators to screening for problem gambling

Barriers to screening

Gambling viewed as a low priority issue across the service system

Clinician over-burdened and limited time

Limited use or knowledge of appropriate tools

Clinician resistance to screening in general

Agency and clinician focus on risk rather than longer term issues

Prevalence of problem gambling perceived as low

Perception of burden of harm perceived as less than other mental health issues

Poor access to training or education

Low clinician confidence to respond impacts on willingness to screen



Barriers to screening

Patient embarrassment or shame hampers disclosure and a perceived willingness to engage in treatment

Facilitators to screening

Increased awareness of gambling prevalence in mental health services

Recognition of gambling as an underlying issue

Readiness for a brief tool

An openness for inclusion of gambling in mandatory screens and assessments

Demand for advice, training and provision of appropriate screening tool/s

Current practice for screening in mental health services

A small number of clinicians said that routine screening for problem gambling occurred, but far

more common was screening at the discretion of the clinician or because the patient disclosed a

problem. Current practice included an array of screening questions that for the most part had not

been derived from a validated screening tool.

Routine screening

Routine screening for problem gambling was sometimes conducted as part of intake or the initial

assessment. For some agencies, the inclusion of routine screening for problem gambling was a

recent development.

We do ask about it now in our intake screen, but that's only been in the

last six months. So we have an intake team now that I manage and we

refer out to many different mental health services and then they would

receive that information to say yes this person identifies as having

problems with gambling.

Male, Intake Manager

For some services there was not a formal tool, but rather a blank space that needed to be

completed, with little guidance about how the clinician should handle the issue. Screening did at

times occur even in the absence of a formal tool or question prompt. Some clinicians noted that

through conversation they sought to understand what was happening for the patient and this might

include gambling.

So there's a lot of that sort of assessment that goes on that's not

necessarily formal, it just happens on a conversational level…So it can

be a case of there's financial problems, not always too sure what it is, but

we suspect that there may be a gambling problem or a drug and alcohol

problem, or there's another issue. That you can often get in there.

Male, Nurse

One informant remarked that while clinicians will say that they screen for gambling the reality was

that screening was infrequent. For the most part however, screening was reported to be up to the

individual clinician’s discretion. This meant that it was sometimes ad-hoc and not always

completed. Previous knowledge and experience regarding problem gambling informed clinician’s

decisions to screen for gambling issues.



It's very ad hoc. It depends on the level of experience and knowledge

about problem gambling.

Female, Service Manager

I have one staff member who would screen. That just comes from their

own career.


Patient behaviours prompting screening

Often screening for problem gambling occurs only when a patient discloses a behaviour that could

occur because of a gambling problem (e.g. experiencing financial difficulties). The onus on the

patient to disclose meant that clinicians did not proactively seek to determine the presence of

problem gambling.

It may be a problem but they [the patient] either deny it's a problem, or

don't want it to be a problem, and so they will just skip over that question,

which as a result, so we do as well.

Female, Occupational Therapist

Yeah and I think regardless of the impact that gambling might be having

on someone, that they might not even bring it to the forefront. If they’re

not telling us that that’s a problem, then we don’t know about it.

Female, Team Leader

A reason for screening being initiated after certain behaviours were observed/reported was a

reported lack of physical signs that prompted the clinician to screen. Multiple clinicians noted that

with most addictive behaviours there were physical signs of intoxication or use such as dilated

pupils, sweating and slurring. Some clinicians commented that there were no physical signs of

gambling and others reported that there might be but that they were not skilled in being able to

identify the signs.

There were, however, other signs that there might be a gambling problem. One informant reported

that they had multiple patients whose finances were administered by the State Trustees. For these

patients there were repeated attempts to get money to gamble. As noted below, this informant

reflected that these attempts to source money were not that different to people with alcohol and

other drug issues.

There's warning signs that we look for I guess when you see that sort of

behaviour. I guess it's quite similar to substance misuse as well in the

ways that people go about trying to get money to either gamble with or

use drugs with. So I think there's a bit of a crossover.

Male, Social Worker

In addition, financial issues more broadly were a sign to screen or discuss further the issue of

gambling. Indeed, for some financial issues were the only prompt to screen for problem gambling.

Not all clinicians thought that financial issues were definitely an indicator of problem gambling. One

informant noted that financial problems were common among all of their patients.



I mean, I guess financial problems are common, but I don't know if it's a

good indicator, because I think a lot of our consumers would have

financial problems.


For the most part, however, patients prompted further questioning when there were requests for

assistance for groceries, bills or rent, difficulties with access to cash or not being able to afford

medications. For others, the trigger was changes in mood or other mental health conditions. For

example, some patients reported worsened depression when they lost money or conversely

improved mood after a win. If patients were presenting with issues that could not be explained by

other means, then gambling was thought as a possible explanation for the symptoms or issues

plaguing patients.

So I guess if we ruled out that the person's misusing substances and

then they constantly want food parcels or assistance … then I guess that

would raise some red flags.

Male, Social Worker

They're usually presenting either with a mood disorder or psychotic

presentation or risk issues. They will try to picture where this is coming

from. Is there a trigger for these kinds of presentations? Yeah, I guess

that mental state risk, drug and alcohol, kind of the rundown of where

they go. Then if you're exploring further [you may consider gambling].

Female, Nurse

Clinicians also reported that gambling was identified in the context of discussions about substance

use, either in terms of the relationship between substance use and gambling or through substance

use being ruled out as the source of a problem. As one informant explained, they only identified

the extent of a patient’s gambling behaviour as a result of talking about where the patient

consumed alcohol, which was a gaming venue. This informant described gambling as an

“undercurrent” whereas mental health and alcohol and other drugs (AOD) were the “visible

problems”.

Tools administered and questions asked

Very few clinicians reported that a standardised tool or set of screening or assessment tools were

administered. There was both limited knowledge of the tools and the names of available tools.

Multiple clinicians said that they did screen for problem gambling but when prompted for the types

of tools used it was identified that they did not actually included questions related to gambling.

Some agencies administered standardised tools for other mental health conditions (most often

AOD) and appended additional questions related to gambling onto these tools.

So they're doing the AOD ASSIST and then they just add gambling as an

add on to the end of that.


For the most part, questioning was associated with one of five domains. These domains were

direct questioning on gambling involvement, social activities involving gambling, financial

problems, harms and comorbidity with AOD issues.

Direct questioning related to gambling involvement most often involved any amount of time or

money spent gambling. It also included questioning on the time of day gambled, the location of



gambling (i.e., race track, club etc.) and the type of gambling. Questioning then often turned to

whether gambling had ever been a problem, issue or concern currently or in the past.

Has gambling been an issue for you in the past. Although sometimes I

might ask do you often gamble or go to the TAB or casinos, is that

something that you ever do, but generally it'd be more the question that I

mentioned before – has gambling been an issue in your life currently or

in the past.

Male, Occupational Therapist)

It’s just something around that issue, so do you have any issues with

gambling? Do you have any concerns about your gambling?

Female, Manager

I guess basically, do you ever gamble, being the ultimate – the first

question. Then if it's – and then a lot of people – because I guess buying

a Tattslotto ticket is a form of gambling and so differentiating between

that and problem gambling.


You might just put into conversation, do you ever gamble, or do you go to

the races, how do you spend your time? You can just ask more detailed

questions as you go. How do you feel if you lose money? How do you

feel if you win?

Female, Nurse

Some clinicians indicated that gambling-related screening was administered indirectly, using

financial situations or clinical judgement to determine whether an individual may be experiencing

problems with their gambling.

One informant reported asking about gambling as a recreational activity. They believed this

approach helped reduce the stigma attached to admitting involvement in gambling. This included

understanding what the patient did in their spare time and the sorts of recreational activities that

they enjoyed. This approach perhaps involves less judgement and is more about establishing the

context in understanding what patients do with their time and money.

I guess apart from essential expenditure what other sorts of things are

you spending on and maybe just run through a few possibilities, like do

you use any money on any substances? Do you use any – [do you] go to

the pokies? What do you do for recreation, for fun, that sort of thing?

Female, Social Worker

Gambling screening was frequently linked to questions on financial management. This was raised

especially when there were indicators of not having enough money for medications or treatment. It

was also in relation to getting money in order to buy something else (i.e., alcohol). For others,

questions were related to obtaining money to gamble.

If money comes up as an important thing, how do you get money? It's a

case of do you ever take any risks? Do you ever gamble? Do you need

to borrow money? How do you get it?

Male, Nurse



Do you have any problems with finances? Is that as a result, if there are

financial issues; is that a result of gambling? Then exploring that further

with them there as to what, how often and that sort of thing.

Female, Nurse

Some questions were related to the harms of gambling including going without food, shelter, legal

and forensic concerns, harms to others and employment difficulties. There was also awareness

that gambling could become an issue during treatment. This would become apparent because of

financial difficulties or requests for material support.

Feasibility of routine screening at a service and systems level

At the systems level, multiple clinicians commented that gambling was not a priority issue.

Frequently, clinicians provided rankings as to the top and bottom issues that the agency and

individual clinicians thought should take priority. Rankings were related to agency targets and the

amount of funding attached to identifying or managing problem gambling.

Two clinicians mentioned new legal requirements to screen for issues that had not been

mandatory in the past. The new Mental Health Act required a great deal more paper work, forms

and time and energy. Problem gambling was not part of the new Mental Health Act. This meant

that problem gambling was a low priority and even less time was available for additional screening

items.

Look there's so much going on that they're asked to do with the let's talk

stuff, with the recovery, the new Mental Health Act. There's an awful lot

that they have to look at, so they probably see that as probably quite low

even though it does affect people's mental health state and can make

people worse. It's not seen as a priority because my organisation hasn’t

seen it as a priority.

Female, Nurse Manager

At the agency level, multiple clinicians indicated that gambling was not perceived as an important

issue. Some clinicians commented that at an agency level there was a greater emphasis on

physical health problems over mental health issues.

I actually think the key issue is that gambling is still not seen as a priority

in the health area. There are so many other things that are seen as

important, particularly population level things like diabetes management,

obesity, dental problems, physiotherapy, the age sort of stuff. All of that

and chronic disease, there is so much work that needs to be done on that

and it’s such a massive area that gambling gets pushed to the side as

another one of those oh yeah, they’ve got gambling issues, just like they

might have something else that they do too much of.

Female, Manager

This was in part due to a focus on risks and symptoms, and as indicated by one informant, a focus

on a traditional medical model. Clinicians stated that disorders that were perceived to be

associated with greater risk were screened first. High-risk conditions identified included

schizophrenia, drug-induced psychosis, bipolar affective disorder, psychosis, suicide, aggression,

forensic history and AOD. Multiple clinicians commented that there had been a greater focus on



screening and the management of mental health conditions including AOD issues. However,

gambling had not been part of that discussion.

Clinicians reported that this meant that there was no momentum for gambling screening and

identification in as much as limited support for training and education in problem gambling or even

general awareness that it was a problem that should be addressed.

Individual clinician-level barriers to screening

Individual clinician-level barriers to screening were multiple and inter-connecting. These included

over-burdened and limited time, a lack of knowledge or use of screening and assessment tools

and conversely, a reluctance for more tools and a resistance to screening in general. Gambling

was also not perceived as a common problem and there was a perception that it was not as

harmful as other disorders. There was also reported inadequate access to training and education

resources leading to low confidence in identifying and managing problem gambling. Despite these

barriers to screening there was also recognition that gambling was often an underlying issue that

should at some time be treated.

Over-burdened and time

Multiple clinicians commented on difficulty in managing the scope of their current role and

responsibilities. This was primarily related to workload issues and the amount of paperwork that

was required (e.g. forms to complete). For these clinicians, there was a reluctance to increase their

workload further.

Feeling over-burdened and that there was a lack of time was associated with role responsibility.

Some clinicians indicated that gambling was not their core business and it was therefore not their

responsibility to screen, identify or manage the problem.

The other issues aren't our issues. It sounds horrible but they're not.

Once they're mentally stable, mentally well, they need to go out the door

because we have clients coming.


Limited use or knowledge of screening tools

A common theme was a lack of access, awareness or knowledge of appropriate screening tools.

Multiple clinicians said that, to their knowledge, there was no tool available in their agency to

screen for problem gambling.

I wouldn't know what question specifically to ask.


Sorry, in actual fact we do have a screen called ASSIST which is the

drug and alcohol screen. But I can't actually remember off the top of my

head if it even mentions anything about gambling.

Male, Nurse

I would say for me it'd be lack of knowledge around particular screening

tools about it.

Male, Occupational Therapist



Reluctance for more tools

Some clinicians perceived the current screening and assessment process as a burden, onerous or

overwhelming. There was recognition for the need for appropriate screening for problem gambling

but there was also reluctance for more tools. Part of the reason was the number of tools that are

administered but their results not used and that assessments covered mental health as well as

physical health.

It's like assessment after assessment. I think clinicians are also a bit

annoyed with having more and more assessments that are required. We

go through physical health now. We have this ridiculously long

questionnaire on physical heath and it covers head to toe.

Male, Nurse

In addition, other clinicians reported that clinicians and patients were annoyed at the extensive and

long screening and assessment tools. The main issue was, however, that another screening tool

would create more paperwork.

Resistance to screening in general

For some clinicians, there was a preference for identifying issues through conversation rather than

a standardised screening tool. This style of questioning was developed through experience. When

approaching sensitive topics the clinician would first describe what it was and then ask questions

about the patient’s experience. This included developing their own personal preferences in terms

of the types of questions as well as the way in which questions were asked.

A resistance to screening was not just associated with problem gambling. Multiple clinicians noted

that they did not screen for AOD use even though there was a screening tool and that it was part of

the role. For these clinicians there was a preference for discussing the problem as part of a

conversation rather than formal screening. Another informant reported that they preferred to work

with the material presented by the patient. For this informant, screening was viewed as not

relevant and interfering with providing support or guidance to the patient.

… they've just come in for the relationship break-up because they're

feeling a bit down and sad and need a bit of support for a couple of

sessions. It's really hard to then go and say, how much do you drink?

How much do you smoke? How much do you – do you problem gamble?

Are you in a domestic violence situation? Are you using any illicit drugs?


Clinician focus on immediate risk

Multiple clinicians reported that there was a focus on minimising actual risk or harm. Clinicians

working in emergency departments reported that their focus was very much on risk and symptoms

rather than the underlying causes. Sometimes this was not intentional, but rather a reflection of

being swept up in a crisis or other acute situation.

It's not something I asked during the first assessment. I was focusing on

trying to check out whether he was about to kill himself or somebody else

because both of those seemed to be quite high on the risk list of

possibilities.

Male, Service Manager



Clinicians commented that a focus on symptoms when combined with fluctuating caseloads meant

that screening for non-urgent conditions was not conducted. Related to immediate risk was the

threat of harm to self or other. For those working in emergency and crisis support services,

gambling was viewed as something that was of minimal importance because it was not life

threatening. There was also a perception that gambling was not an acute issue but rather a longer-

term concern.

But that initial crisis kind of stuff, usually because they're presenting in

crisis around risk issues, we're not usually focusing so much on the long-

term issues around whether it would be gambling and that sort of stuff

that I guess is less workable in an instant.

Female, Nurse

Gambling not perceived as a common problem

Clinicians differed markedly in their perception of the prevalence of problem gambling in mental

health services. This ranged from a perception that problem gambling was very rare through to it

being very common. This was in part explained by clinicians being in differing roles and services,

with clinicians from youth services reported that problem gambling was rare whereas those from

AOD services reporting that it was more common. Quite often, estimates of prevalence were

based on current or past caseloads.

Multiple clinicians acknowledged that estimates of problem gambling in mental health were likely to

be under-reported. This was because of a lack of focus on problem gambling and inadequate

screening or assessment, but inadequate screening was in part due to a perception that gambling

was not a common problem. Clinicians who thought gambling was not a problem prioritised other

potential issues including AOD screening.

It's not a standard question. Do you have a gambling problem? Whereas

I would be more proactive in seeking out information say about drugs and

alcohol because it's a lot more common.


Priority placed on identifying disorders associated with greater harm

In addition to service level priorities, it was reported that clinicians also ranked issues according to

harm and or perceived need. One informant stated that AOD issues were higher on the list of

priorities because of their interaction with medication.

Look I think clinicians would say, yeah okay, it is probably something we

need to ask clients about. But it's not high on our list of priorities. I would

say drug and alcohol use is higher, deemed to be higher on the list of

priorities because of the interaction that it has with various medications

that a person might be on and that sort of thing.


There was also a perception that gambling was not as harmful as other behaviours and that while

it might be associated with spending money or small losses it did not cause harm. For example,

one informant compared gambling harms to excessive shopping.



It’s like oh yeah, so they shop. It’s something that happens. They’re not

really in the framework of going oh, that could be an issue. So the other

things take so much priority because they have the massive impact. If

someone's got a chronic disease it is life threatening ultimately and they

wouldn't see gambling in that same scope.


While gambling was not viewed as associated with risk of harm for some there was a perception

that it should be addressed, but in a service that focuses on longer-term issues. Related to harms

and risk was the perception that gambling was not a health issue. One informant noted that

gambling had been part of their screen, but had been removed for this reason.

One of the earlier drafts of our health prompt actually included problem

gambling as one of the sort of 20 questions, one minute health check.

One of the questions was a similar sort of thing around gambling and do

you have issues with gambling. It was taken out I think because there

was feedback that it was a bit too confusing because it was supposed to

be a health prompt or a health screen, why are we asking about

gambling. That's too confusing. I think maybe some people felt it was

intrusive.


The view that gambling was not a mental health issue was not widely endorsed. However, one

informant reported that their patients also did not make the connection between gambling and

mental health. It was not clear whether the clinician also doubted the connection between

gambling and mental health.

Inadequate access to training or education

Education was an issue not just in terms of broad access to knowledge and information about

gambling but also at an agency level. For some clinicians, there was a relationship between not

feeling comfortable asking about problem gambling and not having the skills or knowledge to

undertake screening.

Multiple clinicians said that they would have difficulty identifying and assessing for problem

gambling because they had been provided none or very limited training opportunities.

We've not had training, put it that way.


I just don't think we get enough information about gambling and how to

assess for it.




Confidence to respond to detection of a problem

There were concerns that clinicians were not expert at identifying problem gambling. It was

reported that the lack of screening exasperated the problem, making it very difficult to accurately

detect.

We might see it amongst patients. I don't know if we will be any better at

picking that up mind you than a general member of the population.


I don't think we're very good at identifying, really, anything to be honest. It

tends to probably more fall down onto the clinician asking the questions

rather than having a more formal intake.

Male, Nurse

Similarly another informant noted that they could identify problem gambling with the use of a

screening tool but were not sure how to screen for the underlying issues.

I can identify it, if someone admits to it. But I guess it's probably

underlying stuff as well that would be classed as a concern that I'm

probably not as aware of how to thoroughly screen that.


Part of the problem was the perception that identifying problem gambling would raise a whole

range of issues that the clinician was not equipped to manage. When problem gambling was

identified there were concerns that the clinician would not know how to respond. This was in part

related to confidence and having training in identification and management of problem gambling.

Recognition of gambling as an underlying issue

Gambling may not have been viewed as a critical issue for many clinicians but there were frequent

reports that gambling might be an underlying issue for other mental health concerns that eventually

would need attention. This informant noted that while gambling was the lowest priority it was

interrelated with mental and physical health, accommodation needs and AOD issues.

I think that financial or any kind of stress is really detrimental to people's

mental state. So I think it is very important that people are asking about

it. It is something that is being pushed more and more at them at the

moment, that we recognise it as an issue. But it often loses – it's probably

not the main focus. I think if I was just to rate them, people would –

primarily concerned about a person’s mental state, their physical health,

their accommodation, drug and alcohol use and then gambling

sometimes is left to last really. But they all interrelate so it's kind of

important.

Male, Social Worker)

Some clinicians commented that gambling threatened to undermine the effectiveness of treatment

approaches for other disorders. This became an issue when money that had been allocated to

medication, housing, transport or other essentials had been spent on gambling.



According to one informant, gambling may have caused a range of negative moods including

depression, anxiety and also suicidal attempts. Others suggested that negative mood states were

related to gambling. One informant described gambling as a coping mechanism for low mood.

So it is a coping strategy that he does too when his anxiety kicks in or

when he's of low mood and he is kind of looking for that rush as a way of

compensating for that.


Patient responsiveness to screening

There were multiple issues raised associated with shame and stigma that impacted on clinician

and patient willingness to screen and be screened. These included a willingness to disclose a

gambling problem and varying levels of readiness to change. There was shame, stigma and

embarrassment attached with talking about problem gambling and also in the act of help-seeking.

Some clinicians noted the impact of the development of the therapeutic relationship on patient

responsiveness to screening and also a preference for working with the same clinician once a

problem was disclosed.

Willingness to disclose a gambling problem

There was a range of views as to the willingness of patients to disclose gambling problems. Some

clinicians reported that they knew that some of their patients had a problem with gambling but that

it was not always disclosed. Although some patients either had screened positive to problem

gambling or had disclosed that they gambled a large amount they were not ready to do something

about their gambling. For some this impacted on their willingness to disclose the extent of their

problem. Others indicated that gambling was not normally disclosed at the first interview because

the person was embarrassed and preferred to conceal their gambling.

Well it won't of course naturally come up. The clients that I had in

particular were embarrassed about it and tried to conceal it. So initially

you're doing your assessment, you're diagnosing, you're getting a sense

of other contributing factors like drugs, alcohol, chronic disease issues,

and you ask the question about do you drink? How much do you drink

every day? Do you gamble? So you're often get the no, no, I don't

gamble.


Multiple clinicians stated that while patients initially denied that there was a problem, follow-up

questioning identified that gambling was an issue. This perhaps indicates that casual questioning

about gambling activities was not an effective way of identifying problem gambling and that further

questioning was required.

Conversely, other clinicians reported that they did not experience any difficulties with patients not

disclosing gambling issues. Clinicians that reported that gambling was disclosed tended to ask

about gambling as part of a holistic approach to screening and assessment. That is, a

comprehensive assessment was undertaken that included problem gambling.

One informant noted the importance of asking about gambling at the initial interview as part of a

suite of issues. They reported that if this was not done gambling was unlikely to be raised as part

of a normal conversation.



I guess part of what we sell is what we're doing is we're trying to support

them holistically. We're not just focusing on someone's mental health. So

during that screen you'll say, we're trying to get a picture of everything.

So that initial assessment quite often they'll open up and say, yeah I've

got an issue with either alcohol or gambling that you normally wouldn't

have in general conversation.

Female, Nurse

Readiness for change

A positive screen for problem gambling does not necessarily mean that a patient will want to work

on gambling issues. Multiple clinicians identified the relevance of readiness to change or the stage

of change model to explain patient disclosure of problem gambling. One occupational therapist

reported, however, that although their patient was ready to change their gambling at intake, their

readiness had reduced by the time they attended an appointment.

They'll identify that it's an issue and over the phone a lot of the time

anyway they'll identify that it's been a problem and they want to manage

it, but then when it comes to talking face to face they're a little bit more

reluctant to discuss things.


Although some clinicians reported that some patients preferred to work with just one clinician, this

did not mean that they were ready or able to work on the problem. One informant noted that some

patients did not want to identify with having more than one problem.

Sometimes they don't want the double-whammy and say oh well you've

got gambling as well then.


Some clinicians noted that there were no issues with disclosure if the patient had sought help for

problem gambling. This informant reported that if the patient had a problem with gambling but it

was not the primary reason for seeking help then the patient was unlikely to disclose.

Sometimes we pick up key issues that we think are key issues and the

client says I don’t want to deal with that. I’m here because I want to deal

with x, so then you really have to respect what it is the client’s there for.


A critical issue when formal tools were not administered at the point of screening or assessment

was that it made it more difficult for patients to acknowledge or indicate that they had a gambling

problem. This informant noted that identifying the problem early for many is related to a sense of

relief that the problem was no longer hidden.

So it's just a ticked box amongst many other things, but it does give

scope for that service to then be aware of that and support the person

around their gambling issues.

Male, Nurse



Our experience is that it tends to be a sense of relief once it is flagged

and out in the open.


Shame and stigma attached to problem gambling

Clinicians reported that gamblers experienced high levels of shame and stigma about gambling,

mental health issues more broadly and also about help-seeking. In terms of problem gambling,

multiple clinicians suggested that there was stigma associated with gambling because it was

potentially more of a hidden issue than disorders associated with AOD use. One informant

suggested that there was shame attached to problem gambling because the behaviour was viewed

as ‘stupid’.

There’s a lot of shame attached to it and there’s also a lot of thinking that

it is something stupid that they’ve done. I’ve had clients tell me they

would rather have a drug and alcohol problem, because you can actually

see they have a problem.


Indeed, comparisons were regularly drawn between gambling and AOD issues in terms of admitting that there was a problem. One informant noted that patients would rather disclose that they were under the influence of alcohol than say that they had been gambling.

Interestingly, the second informant reported that patients were not comfortable at all in disclosing

their frequency of gambling consumption and perhaps this type of questioning in some way

contributes towards patient reluctance to disclose.

One social worker reported that they didn’t have any patients that wanted to discuss their gambling

behaviours. This was because their patients were guarded about their gambling. The informant

noted that their patients were open about AOD use, but secretive about their gambling behaviours.

This informant attributed the difference to greater acceptance in the community of AOD issues

rather than gambling problems.

Other clinicians reported that they have never had any issues with patient disclosure even though

the problem was potentially embarrassing. These clinicians stated that asking about problem

gambling directly was helpful. Another informant reported that they did not think there was as much

stigma attached to problem gambling as there was with other mental health issues.

I personally don't believe that there's such a stigma attached to people

with gambling issues as there is to people with drug and alcohol issues. I

don't think it's a touchy subject. I think there are other areas that would

be much more sensitive to approach than gambling.

Male, Nurse

Impact on the therapeutic relationship

There was a view put forward by multiple clinicians that screening for problem gambling at the first

point of contact might not yield an honest response. The main reason provided was that sufficient

time had not passed for rapport or trust to be developed. Some clinicians suggested that patients

were reluctant to admit to a gambling problem or would under-report the extent of the problem until

trust was established.



However, as noted by one informant, if the patient had not been screened for problem gambling

then they were unlikely to disclose in early sessions. This informant reported that instead

gambling, as well as AOD issues, might be disclosed over the longer term.

Often at intake because they're new to the service and they're new to

you, just meeting you the first time, they may not feel comfortable talking

about it. They're happy to talk about the things that we already know, but

as far as alcohol, drug use, gambling or any other sorts of things, that

they can actually come out slowly over time.


Perhaps the development of trust and rapport was related to the possible impacts of disclosure. If

problem gambling was disclosed and harms identified then it was possible that other people would

be notified or that the patient would be put on an administration order that usually meant someone

else managing their finances.

A preference to work with the existing clinician

Clinicians reported that for many patients that identified problem gambling there was a preference

to work with the existing clinician. This was because rapport had already been developed and it

was easier or more convenient to not have to repeat their story. There were also concerns about

involving yet another clinician.

It was unclear whether this was because of embarrassment or a desire to in some way contain

knowledge of the problem. For this patient, there was a desire for the clinician to handle the problem

and a preference not to involve others.

When we ask do you think we can have a conversation now that you've

said that you do have a problem, is it worthwhile having a conversation

with family or your GP or whatever it may be, or how about going to

Gambling Help Services to see if you can get a bit more support, they'll

be a bit wary of that. Can't you do it? Do I have to tell the world?


Management of problem gambling

The management of problem gambling was discussed in terms of further assessment, treatment

approaches and internal and external referral.

Further assessment for problem gambling

Clinicians frequently described screening and assessment interchangeably. This depended on

their role in the organisation and also how the service was structured in terms of having separate

intake or screening to those who conducted assessment or case management. Where assessment

was described, clinicians for those most part suggested that formal screens for problem gambling

were not administered; rather it was a conversation or discussion.

One occupational therapist reported that if they asked about gambling it would be during case

planning and as part of screening for addictive behaviours. Others said that gambling would be

investigated further as part of a treatment and recovery plan.



If I was going to ask about it, it would be when we're doing a treatment

and recovery plan, because one of the subheadings there is gambling

and addictive behaviours.


The treatment and recovery plan has addictive behaviours as well. I think

it's got drug and alcohol, one question and then I think gambling is

actually separate to that. If they identify any needs – so when you talk to

a client and go through that, that it comes up then automatically.


Assessment was viewed as an activity that was often ongoing and did not always involve re-

administration of standard tools. Multiple clinicians reported that assessment for problem gambling

did not occur in the early sessions. Rather these clinicians stated that AOD assessment occurred

in the fifth week and gambling could be the same. This informant suggested that prompts needed

to be in place to provide structure to the timing of assessment especially when there were multiple

assessments that were required.

If it's drug and alcohol, then make sure by the fifth week, that you've had

some discussion with the client around is there an issue there – Same

thing with gambling.


There was a concern that where there was not formal screening or assessment or good processes

in place for follow-up, gambling could be missed. One informant noted that as the number of

assessments increased the chances of missing important issues such as gambling also increased.

Similarly, when screening does not occur, patients were at risk of progressing through the service

without ever being asked about problem gambling.

Treatment approaches

Treatment approaches in responding to problem gambling once it had been identified varied.

Approaches included minimal or opportunistic interventions, motivational interviewing and cognitive

behavioural approaches as well as case management. Minimal or opportunistic approaches

included having a conversation about gambling and understanding what the patient wanted to do

about the problem.

Clinicians then provided brief advice or information on strategies to change the cycle of gambling,

harm minimisation (including budgeting how much could be spent each week on gambling),

seeking support and relapse prevention. One informant noted that their AOD staff were skilled at

motivational approaches and could help develop strategies around delaying gambling.

Staff are skilled AOD staff, so insofar as the stages of change and

relapse prevention and they would be able to use those strategies

around delaying.


Interestingly, there was very limited discussion on how to approach the problem if the person was

ambivalent to change. One informant stated that they explored the pros and cons of gambling

(rather than change).



Well, you just talk about the pros and cons, I guess. But I find that if you

push somebody towards, that's when they push back. That's not smart.

But just being open and talking about the good bits and the bad bits

sometimes is enough to get people to rethink. Or if they just consider that

it might not be the best thing.

Female, Nurse

Beyond minimal interventions, treatment approaches were for the most part based on what

clinicians thought might be an appropriate treatment or what they had heard other clinicians

discussing. There were multiple attempts to describe Cognitive Behaviour Therapy (CBT)

approaches that mostly focused on the identification of triggers (and not challenging or correction).

One informant noted that psychologists in their agency may administer CBT for depression

associated with gambling, but not specifically for problem gambling.

We've got psychologists that – but I don't know. None of them are

experts in treatment of gambling problems. The odd referral that goes to

them would be like the depressed anxious person, and they might

provide CBT.


An issue raised by multiple clinicians was a lack of training on administering CBT for problem

gambling. One informant reported that more training was required on the delivery of longer-term

treatments that made a difference longer-term to the gambling problem. Co-case management

with another clinician in the same service was mentioned by just one informant as a helpful model

for ongoing management of problem gambling. In this case, co-case management was with a

specialist service and was perceived as helpful to the mental health worker as well as the patient:

This informant reported that co-case management was preferable because there was shame and

stigma associated with seeking help from a gambling service.

Also with Gambler's Help as well I feel confident in linking people with

those services, but directly providing – well a lot of clients don't want to

be linked with those services and whether it's because of shame from

gambling or not actually feeling that they're at a point to work towards it.

As a case manager then knowing that this issue exists, I don't know a lot

other than linking with other services where to go from there.


Referral approaches

When and how referrals were made was largely dependent on the role of the informant. For

example, case managers routinely referred for most issues and gambling was no exception. Even

if there was capacity within the agency there was a preference for patients to seek targeted help

outside of the agency.

We'd still prefer to direct somebody to Gambler's Help, rather than onsite,

because our service operates more of a supportive role for consumers to

engage with services externally.


Some case managers also had experience in gambling counselling and if this was the case they

were more likely to treat than refer. Referrals were made for treatment where there were not



appropriately skilled clinicians available in the service or when the patient indicated a preference

for help outside of the current service.

Clinicians reported a range of referral sources. Some agencies had close relationships with

specialist gambling services, either because they were embedded in the agency or located close

by. For others, however, referrals were ad-hoc and were the result of an internet search or asking

others in their agency if they knew someone. Some clinicians referred to the Gamblers Helpline as

a means of accessing some information but beyond this many clinicians applied an ad-hoc

approach to finding someone that could assist.

For some clinicians, seeking information online was an exercise in collaboration with the patient. It

also led to other referrals such as 12-step programs and financial assistance.

So what we did was we looked online together. We found him a

counsellor over at Knox so that he could actually then talk about why he

was gambling. We did get him into a Gambling Anonymous… counselling

too because it had effects on his family life because it does have a

knock-on effect. We also then had to help with financial.


Referral was also made if the person was deemed at risk. This extract demonstrates referral to

State Trustees if the family finances were under threat.

Referral also meant that some clinicians did not immediately respond to a gambling problem. As

described below, the informant first needed to learn about available referral options, before then

taking these options to the patients before a referral is even made. This highlights inexperience in

the provision of a response to gambling, with this informant wanting to first gather appropriate

knowledge about potential options, before then having the discussion with the patient about how

the patient wanted to respond. In this case they had access to “gambling team members”, knew

where to look for gambling help services, and approached the conversation with the patient by

exploring how the patient wanted to proceed.

Methods to aid implementation of routine screening

Multiple strategies were reported that could aid the implementation of routine screening. This

included the inclusion of problem gambling in mandatory screens or assessment and providing

clinicians with appropriate tools and resources and access to expertise. Almost unanimously there

was a call for education and training in screening, assessment and the management of problem

gambling.

Inclusion in mandatory screens or assessment

In general there was agreement that clinicians and patients were required to complete a large

volume of screening tools across a broad range of health areas. However, there was also

acknowledgement that gambling was an issue that should be included in intake and screening

tools.



Supposedly we’re meant to be having one admission form coming in at

some stage with everything in it. So hopefully within that eight-page

document or something ridiculous, it will actually have something about

gambling.

Female, Team Leader

Clinicians placed importance on routine and standard processes for screening. It was reported that

processes were now in place for alcohol and to a lesser extent illicit drug screening. Processes

helped to ensure that screening occurred. Processes were important not just to ensure patients

were appropriately screened but also to ensure screening continued to occur when there was staff

turnover. The bottom line was that unless gambling was included in mandatory screening,

clinicians indicated that most clinicians would not ask about it.

If the question isn't on there to ask them do they have a gambling

problem then we're not going to ask them.


Processes were also important to prompt clinicians to screen. A common reason reported for not

screening was that they forgot or did not remember. Having an indicator or prompt to screen was

seen as a helpful addition.

Clinicians noted the importance of having gambling as part of routine screening so that clinicians

would also become accustomed to asking about gambling:

Well the intake is your first meeting so it would be in the report, but if you

had that question in the intake meeting and staff got used to asking that

question, then it would just become a normal question like we say with

the drugs and alcohol and the forensic and things like that. It would

become just second nature to ask it if it was in the assessment.


Even though clinicians identified that it was important that if screening for problem gambling was to

occur this needed to be in the context of mandatory assessment, this approach was not always

failsafe. One nurse suggested that inclusion in mandatory screens did not always mean that the

issue of gambling was identified in later follow-up assessment.

Where screening for problem gambling was included in initial intake there was a risk that if the

patient did not identify the problem, or if a problem develops, then it will not be later detected.

So, when a new person comes or is referred in it's one of the questions

that we actually ask on intake, but whether or not the case managers

then ask the question again, to actually find out if there is actually

anything happening. If the person has been with us for a long time, then

it's unlikely the question has been asked.

Male, Nurse

I think perhaps some of the ways that we would identify it is during the

initial intake assessment and I suppose it's often directly asked at that

point in time, but if people say no it doesn't really come up again.




Access to appropriate tools, resources and expertise

There was broad agreement that screening tools were infrequently used to screen for problem

gambling. Clinicians were generally positive towards access to appropriate tools and resources.

Multiple comments suggested direct questioning rather than indirectly eliciting information about

gambling was preferred.

I feel okay asking those questions, but perhaps at times it might be useful

having a structured screen on how it might be more useful, or otherwise I

might go about it in ways that it could be measured, other than just

simply asking.


Three clinicians commented that a screening tool similar to screens used for AOD would be

welcome. Multiple clinicians suggested that a brief screen was preferred especially given time

constraints as well as time required for additional gambling screening. A reason for utilising a brief

screen was to minimise the time required for existing screening as well as to streamline the

amount of screening.

Clinicians indicated that a screening tool that facilitated direct questions about gambling would be

helpful. In particular, those questions that were brief but quickly determined whether there was a

gambling problem were seen in a positive light.

Multiple clinicians reported that it would be helpful to have tools and resources to use once a

problem had been identified. This included information on the harms and impact of gambling,

strategies for change and where to seek further help.

So if we had packs where you had all this information about where to go

for Gamblers Anonymous and stuff and counselling and stuff like that that

you could give to your clients, then you're not opening a can of worms

and then just saying okay, bye, you can deal with it now.


Furthermore there was a call for resources that could be accessed by clinicians as well as patients.

These resources would ideally guide discussions and provide information to support the person in

seeking further resources and/or help for the gambling.

So potentially some techniques on how to bring that in and really get it

focused rather than just give someone a leaflet with a load of information,

which they're probably not going to read or understand. Information that's

accessible and also accessible to clinicians and staff, not just clients

would be helpful. I have a dreadful memory so I probably have to do

things all the time and I have to ask the same questions all the time. So

something that I could grab and look at quickly.

Male, Nurse

Having access to the right information was important. Multiple clinicians indicated resistance to

asking about problem gambling if they did not have tools and resources to respond if it were

identified. In addition to the person with the gambling problem, one informant noted it would be

helpful to have a carer information resource. This included how to support the person with the

gambling problem.



In addition to resources there was a call for increased access to expertise. However, this differed

according to the location and focus of the service. Those that were located alongside specialist

gambling services noted that they had access to expertise but expertise, in rural and other areas

was more difficult to source. One informant suggested up-skilling one person in their agency that

could provide advice and support as needed.

It wouldn't necessarily affect their caseload, it's just a sort of

extracurricular thing that workers would be able to be a champion of in

the team. So they might be the go-to person if a worker wants to ask a

question about it and they could keep any resources up to date and be

providing any updates in meetings about a particular issue or anything

like that. So it's just a sort of championing role rather than necessarily a

specialist worker.


This informant went on to explain that the role could provide education for community teams as

well as managers. Specifically on what problem gambling is and how screening and assessments

should be conducted. Overall clinicians indicated a need for more support for clinicians in how to

respond when problem gambling was identified.

Education and training

Training needs covered five main areas:

1. awareness of problem gambling

2. knowing the signs and how to introduce the issue

3. identification and application of appropriate screens

4. knowledge of how to assess

5. treatment/referral options.

Overall, there were comments that training needed to be resourced and to be made a priority it

needed to be funded.

A key issue was awareness of problem gambling. This included information on how to identify

problem gambling including the signs and symptoms.

I think more support for the clinicians when we do potentially identify the

problem gambling and also just more education on perhaps identifying

patterns that we can look for with our clients that would indicate there's

more of an issue that we're not picking up.

Male, Nurse

We're quite trained in drug and alcohol, we can pick if someone's

substance affected or, you know, we don't have the same skills with

gambling.

Female, Nurse



There were various levels of confidence in identifying problem gambling with some clinicians

indicating that a great deal of training would be required. Others, however, reported that if they had

access to appropriate tools then perhaps minimal training was needed.

The content of training requested also included how to raise the issue of problem gambling, how to

screen for problem gambling (asking the right questions) and then what to do about it.

You want to know well what can I do with it anyway, where can I go with

it and how can I help them.


Potentially some techniques on how to bring that in and really get it

focused rather than just give someone a leaflet with a load of information

which they're probably not going to read or understand.

Male, Nurse

I guess education for clinicians where an issue is identified so that

clinicians might feel better equipped to be able to either support that

person directly, or to know where to link them in.


Any kind of information around how to address it, where to go, what to

do, would be useful.

Female, Social Worker

Multiple clinicians mentioned the importance of making training specific and practical. One

informant stated that they had been to sessions that were general introductions to gambling.

What do I need to do to address it? Then, how do I get them the

specialist help if that's where they need to go? That's the information that

we would – that we take away as case managers.


I just don't know that they would be confident in order to offer some

support or strategies or direction. I think people would benefit much more

from further education about gambling.

Male, Nurse

The type of training requested was direct and to the point. There was more support for group

training rather than a webinar or audio-visual materials.

I probably would need more training on that. But if it's training, short and

sharp, nothing too detailed.


I think I find group learning better, like you get lots of online training,

mandatory training you have to do and it's a bit of a chore. I don't think

you benefit as much as an interactive kind of thing with other people,

where you have other people's experience and clinical experience in

presentations over their career.




Multiple clinicians requested an in-service training that provided information and resources. This

included resources that could be taken away and put into practice (rather than just information).

Training for supervisors was also reported as this was perceived as helpful in ensuring that

gambling was routinely discussed. Training of supervisors was also deemed helpful in training new

staff, especially in the administration of any gambling screening tools.

Multiple clinicians noted the importance of maintaining awareness of problem gambling in so much

as providing training booster sessions. This could be in team meetings or other periodic events.

One informant noted that follow-up was important to ensure that training is not just a one-off

session.

I think that maybe those that are run within the teams onsite can be

good. I don't know, I think it's the follow-up that creates the best learning

is, you know it's not just a one-off training session, this is going to be – so

we're going to follow up in six to 12 months.


I think once or twice a year having a reminder around what gambling is

all about is a good thing to institute in any area of mental health service,

because if we don't get refreshed, things drop off.


In addition, to keeping gambling front of mind, one informant noted that with staff turnover there

needed to be access to regular training. This included evaluation of whether training had an impact

on identification of problem gambling and whether it made a difference to overall patient outcomes.

Discussion

This chapter identified that current practice for problem gambling screening in mental health and

community services was for the most part ad-hoc or at the discretion of individual clinicians. Most

clinicians were not aware of screening tools or any standardised means of assessing problem

gambling. Instead when screening was conducted it included questions around gambling

engagement and to a lesser extent frequency, consumption or harm. Screening occasionally

occurred when patients first came into contact with the service but also during assessments

months later. Often screening only occurred if the patient disclosed or in some way indicated that

they had a gambling problem.

Regardless of whether gambling was included in the assessment tool or not, clinicians indicated

that general, descriptive questions about other areas of a patient’s life such as finances, social

activities, alcohol and drug use or mental health often led to a discussion about gambling.

Clinicians identified a number of ways in which they identified problem gambling, or became aware

that their patient was experiencing problems with discussion about finances the most common.

Discussions about substance use or recreational activities also alerted clinicians to a patient’s

potential problem with gambling.

These findings align with those from the Temcheff et al. (2014) study of Canadian youth mental

health workers who recognised the importance of addressing gambling issues and the role they

have to play in detection, but considered it secondary to more serious issues such as substance

use and aggressive/violent behaviour.



Current screening questions are likely not well suited to determining whether there was a gambling

problem. Given that the only clinicians that reported having received training in screening for

problem gambling were those that had previous experience in working with people with problem

gambling, it is possible that a lack of clinical expertise contributes towards reluctance for some

patients to disclose the issue. Indeed, several clinicians reported that after patients disclosed their

gambling, some did not want to discuss their problem.

There was inconsistency in the best time to screen that would yield an honest response to

questions associated with problem gambling. Some clinicians supported direct questioning at initial

screening and assessment even if the issue was not immediately worked on. For others there was

a concern that rapport needed to be developed and that it was only through an ongoing

relationship was the issue of problem gambling disclosed. It is unclear to what extent the lack of

standardised and appropriate screening tools played on these findings. Interestingly, similar

barriers to screening were observed, when attempts were first made to implement AOD screening

in mental health services (Lubman et al., 2008); and community health services (A. C. Thomas &

Staiger, 2012). Since then, there has been considerable investment in building mental health

clinician capacity in responding to AOD issues, such as through the Victorian Dual Diagnosis

Initiative as well as the Victorian Government’s ‘no wrong-door policy’. Such initiatives have seen

an increase in mental health workforce capacity to deal with AOD issues, as well as greater

utilisation of routine AOD screening across the sector.

Overall, clinicians reported that if gambling was not screened when a patient first came into contact

with a service that the chances of it being detected at a latter point in time was low. This was

because screening provided the initial flag for the problem and even though many services were

not able or willing to address the gambling, noting the problem in intake notes or other registration

documents meant that it could be addressed at a later point if this was what the patient wanted.

Multiple clinicians indicated concern however, that initial screening for gambling would result in

significant under-reporting due to the shame and stigma associated with the condition. For some

clinicians it was only through the development of trust and rapport and an ongoing relationship that

they felt the patient would be ready or comfortable in disclosing the issue.

This notion of gambling as an “underlying” or “undercurrent” condition was a recurrent theme.

There was a distinction made between the gambling as a hidden or unseen problem and

conditions such as alcohol and drugs that are associated with visible problems. This perhaps goes

some way towards explaining the common focus on AOD issues in mental health services. It

highlights that for some patients the causal issue may be the gambling which is often not assessed

or considered. As such, it is important that mental health workers consider the impact that problem

gambling can play in exacerbating mental health presentations.

Beyond screening, there was very limited knowledge of treatment approaches for problem

gambling. Some clinicians indicated that they would apply minimal or brief approaches such as

motivational interviewing that had been used with alcohol use disorders to problem gambling.

There was some evidence of discussion on how to minimise the risk of problem gambling, but for

the most part treatment involved referral to another professional either within the same service or

to an external specialist service such as Gambler’s Help.

The nature of acute mental health services means that workers operate according to need and

risk. This means that working with issues such as gambling over the longer term is not always

feasible. In addition, our sample indicated few psychologists provide screening in mental health

services in Victoria (just one identified in the current sample). Taken together this suggests, brief

interventions need to be offered that are evidence informed and can be easily implemented when



needed. This might include self-directed options (e.g. online programs) or single session

interventions that the clinician can offer when problem gambling is identified.

The findings indicate that assessment for gambling needs to be undertaken but that it need not be

during the initial assessment. Multiple clinicians stated that AOD was not assessed until five or six

weeks into treatment or during an episode of case management, and that this would be an

appropriate place to raise the issue of gambling. Indeed, if problem gambling was not recognised

during the initial assessment, it was identified that it may become apparent during subsequent

sessions, and there were a number of reasons for this:

the gambling was hidden or not seen as a priority due to other mental health concerns or

behaviours thought to pose a greater risk

gambling may come up as an issue once the patient’s mental health improves and the

patient has space to address other issues

gambling may come up in the context of other presenting issues impacting their mental

health, such as financial difficulties and demonstrated through related behaviours (e.g.

frequent requests for food parcels or financial assistance).

The assessment process also generated the opportunity for service providers to ask patients about

gambling. Assessment, as in the tools used to facilitate the collection of information about a

patient’s immediate and ongoing needs, took a variety of forms. Some services adopted a

structured approach with standardised tools (e.g. the ASSIST) embedded in the assessment while

others identified key domains such as ‘physical health’ and clinicians formulated their own

questions. Some assessment tools included a mixture of structured and unstructured sections.

Overall, the majority of clinicians indicated that gambling was not included in the assessment tool

used at their service, either as a direct question or as a prompt.

The majority of clinicians acknowledged the benefit of having a formal screen tool to identify

problem gambling, yet the inclusion of a new tool should take a number of factors into

consideration:

existing screening and assessment requirements

provision of specific training to facilitate use of the tool and identification process

provision of specific training to support appropriate responses to the identification of

problem gambling.

Furthermore, given gambling screening was a low priority due mostly to a perceived low

prevalence rate in mental health, screening tools need to be brief and quick to administer.

Some clinicians stated that they had not received training to identify problem gambling and at the

same time that they rarely had patients with problem gambling. Training needs covered five main

areas:

1. awareness of problem gambling including prevalence in mental health services

2. knowing the signs and how to introduce the issue especially in those that have not been

screened



3. identification and application of appropriate screens (and therefore training in what the

screens are, how to administer, what the results mean), and to a lesser extent

4. knowledge of how to assess, and

5. treatment/referral options, especially minimal or brief interventions.

Clinicians also indicated a preference for group over online training with practical examples that

they could relate to and with an opportunity to practise skills. This included access to brief and

accessible resources to help develop effective practice. Multiple clinicians noted the importance of

follow-up, booster sessions or other post-training contact to ensure that clinicians are supported

and that the training is implemented. Lastly, clinicians highlighted the importance of ensuring

training was evaluated to ensure it was appropriate and relevant to the target group.

Study limitations

These findings must be considered in light of a number of limitations. As with the clinician survey,

the findings of the qualitative interviews may be biased by the participation of clinicians who were

more interested in more experienced at responding to gambling, which would overestimate interest

and current responding. It was often difficult to gain a clear picture of the precise intake and

assessment processes operating at each service because of the language adopted, as terms such

as intake, screen, and assessment were often used interchangeably (during an interview) and

clarification was not always provided. Furthermore, there were often inconsistent reports on intake

processes among clinicians working in the same organisation/service, which suggests that some

clinicians were much more familiar with processes and screening/assessment tools than others.

Limitations of the qualitative component include a lack of triangulation with other data sources

(Mays & Pope, 2000). For instance, qualitative interviews with patients would have enabled us to

compare common themes and corroborate and strengthen our interpretation of the data. While a

form of analyst validation was undertaken to ensure that the interpretation of the data made sense,

we did not share our interpretations of the qualitative data with, nor ask for feedback on our

analysis from, clinicians themselves. Mays and Pope (2000) point out that this kind of respondent

validation can ensure that researchers interpretations correspond with participants. We attempted

to ensure a diverse sample of clinicians and managers and our analysis does pay attention to

differences in perspectives, which is another strategy for enhancing rigour in qualitative research

(Mays & Pope, 2000). However there are perspectives missing, including those of clinicians in

primary health care and child and adolescent mental health services.

Despite these study limitations, the breadth of perspectives represented provide important insights

into the barriers and facilitators of current responses to problem gambling within mental health

services. The interviews explored the experiences of a diverse range of health care professionals,

working at different levels in terms of leadership and case management and across a broad range

of mental health settings and geographical sites. Despite sampling a range of clinicians, consistent

themes emerged, proving further confidence in the findings. In addition to highlighting the marked

variability in current practice, the findings unveil some of the critical and practical challenges that

must be addressed (e.g. embedding gambling questions into intake assessments, comprehensive

clinician training), in order to meet the needs of this population.



Study 3a: Patient survey: Gambling behaviours

and prevalence of problem gambling

Aims

The aim of this study was to examine gambling participation and related issues and to estimate the

prevalence of problem gambling among individuals seeking treatment for mental health disorders

(henceforth referred to as patients) across a diverse range of services in Victoria using

standardised screening tools.

Specific research questions were:

1. What are the rates of gambling participation among patients attending mental health

services?

2. What gambling activities do patients attending mental health services engage in?

3. How frequently do patients with mental health problems gamble?

4. To what extent do patients with mental health problems experience gambling-related harm

(i.e., low-risk, moderate-risk and problem gambling based on categories of the PGSI?)

5. Are patients with mental health problems more likely to be experiencing gambling-related

harm than the general population?

6. Do patients attending mental services who are identified as problem gamblers report

poorer wellbeing?

7. Do patients attending mental services who are identified as problem gamblers report

higher rates of substance use?

8. Are there differences in gambling participation rates and gambling harm across subgroups

(e.g. mental health service type, age, gender, geographical location etc.)

9. What proportion patients attending mental services report being asked about their

gambling behaviour?

Method

Design and participants

Using a cross-sectional design, patients (N = 841) presenting to mental health services completed

a survey assessing their gambling behaviours. This patient sample represented 55 per cent of all

patients attending those services during the data collection period.



Measures

The online survey (see Appendix 3) captured demographic information (e.g. gender, ethnicity,

employment status etc.) and was comprised of standardised measures assessing problem

gambling, substance use and quality of life/wellbeing, including items with Likert scale and

dichotomous response options as well as numerical values (e.g. number of days, amount spent

etc.). A two-tier approach was used to assess gambling behaviour, whereby patients who indicated

that they had not participated in any form of gambling in the past year (a definition of gambling was

provided) skipped the entire section on gambling behaviours. Patients who reported any form of

gambling in the past 12 months proceeded to the gambling activity section, where they were

presented with a checklist of activities in accordance with activities captured in the Victorian

household prevalence survey (Hare, 2015). The online survey then asked questions about the

frequency (i.e., how many times per month patients participated in both land-based venues and

over the internet) and amount spent for the relevant gambling activities. Problem gambling was

assessed using the PGSI (Ferris & Wynne, 2001), alongside several other two–five-item screening

tool, the Lie/Bet, the two-, three-, four- and five-item BPGS, the PGSI-short form, NODS-CLIP,

NODS-CLIP2, BBGS, NODS-PERC (see Study 3b for a description of the problem gambling

screens).

Patients were asked a series of questions to identify if a mental health clinician had ever asked

them about their gambling at that service, and were also asked to rate on a 10-point scale how

important it is to reduce or stop gambling, how confident they were that they could stop or reduce

gambling (with 1 being “not at all important”, or “not at all confident”, and 10 being “very important”,

or “very confident”). Irrespective of their gambling status (i.e., both gamblers and non-gamblers), all

patients were asked to select from a list of service and support options (e.g. talking to a GP, call a

gambling helpline etc.) those that they would be likely to use if they were to experience gambling

problems. In the final section assessing health and wellbeing, all patients were asked to indicate

their lifetime and current mental health diagnoses from a list of mental health disorders, and

medications prescribed.

Quality of life was assessed using the three items from the Australian Treatment Outcome Profile

(henceforth the ATOPS; Ryan et al., 2014). Alcohol problems were assessed using the three-item

Alcohol Use Disorders Identification Test-Consumption (henceforth the AUDIT-C; Bush et al.,

1998), if they indicated that they had consumed any alcohol in the past year. Nicotine dependence

was assessed using the two-item Heavy of Smoking Index – HSI (Heatherton, Kozlowski, Frecker,

Rickert, & Robinson, 1989). To assess illicit drug use, patients first answered the single-item Drug

Use Screen (P. C. Smith, Schmidt, Allensworth-Davies, & Saitz, 2010), which asks if they have

used an illegal drug or prescription medication for non-medical reasons in the past 12 months.

Patients indicating that they had used were then asked how frequently they had used different drug

types (e.g. cannabis, cocaine amphetamines) with “never”, “once or twice”, “monthly”, “weekly”, or

“daily/almost daily” as response options. Patients were asked to report which drug gave them the

most concern, as well as reporting if they would like help cutting down or stopping use of any

substance. Finally, patients were asked how much they spend per fortnight on alcohol, cigarettes

and illicit drugs, if they had reported using these substances in the past year.

The survey was developed in collaboration with consumers of mental health services who advised

on the content and language. The survey was extensively piloted in multiple settings and

underwent several revisions and edits. The final online version of the survey took on average 15

minutes to complete and was accessed via a link and hosted by Qualtrics. A hard copy was also

made available in case patients were unable or unwilling to use the online version.



Procedures

The same mental health services that participated in the clinician survey agreed to participate in

the patient survey, with the exception of two small sites that were no longer able to facilitate patient

screening (see Study 2a, method section for a description of service and site selection

methodology). In total, eight separate mental health services and 12 individual sites/teams

participated, representing a broad mental health population in terms of the types of services

accessed (i.e., private, public, adult and youth), geographical locations and population

demographic. Initially the project co-ordinator met with directors of each site and discussed the

optimal and least disruptive approach to undertake data collection.

For most services (all but one), researchers were stationed in the waiting rooms of mental health

services and approached patients directly and invited them to participate in the survey.

Researchers worked closely with clinical and administrative staff to ensure they did not approach

patients who were acutely unwell (n = 165). Researchers explained the aim and nature of the

research, emphasising that being a gambler was not a pre-requisite for participation, and that all

responses were confidential and anonymous. If a patient expressed an interest in participating, and

was deemed able to provide informed consent (determined by the researcher using a standardised

approach), the patient was asked to indicate this on the online survey. Patients who consented to

participate then completed the survey using an iPad while waiting to see their clinician. The

researcher was available at all times to assist the patient, in survey completion if required (e.g. to

clarify the meaning of questions). If a patient was unable to use the iPad, the researcher entered

their responses that they provided verbally to the researcher and this was always conducted in a

quiet secluded area/room away from the main waiting room to maintain patient confidentiality.

Completed hard copies of the survey were returned immediately to the researcher and entered

manually into a SPSS database.

Completing the survey did not delay patients attending their appointments, and several surveys

were started immediately prior to their appointment and continued after (except in 14 cases when

the patient did not return to complete the survey). In the mental health community support service it

was necessary to adopt a different methodology for data collection, due to the primary approach to

provision of support involving staff engaging with patients in their homes or community. In this

instance, support workers in this service were trained in the administration of the survey and they

took the iPads to home visits so that the consumer could complete the survey in their own home,

with the support worker available to assist if necessary. All patients were offered a $10 store

voucher for the time involved in survey completion.

Data collection was undertaken by 10 researchers (all with psychology backgrounds), who were

trained by the project co-ordinator to ensure standardised practices. Data collection took place

between June 2015 and January 2016. The amount of time researchers spent collecting data at

each site varied, and was proportionate to the number of clinicians and patients at the site (i.e.

more time spent at sites with more clinicians) and was guided by the frequency with which patients

attended the service (e.g. weekly, fortnightly or monthly). Researchers were stationed in the

waiting room of services during opening clinic hours (full day) so that all available patients could be

approached. The study was given approval by the Eastern Health Human Research Ethics

committee, approval reference number: LR120/1314 and additional ethical review were undertaken

and approval granted by sites not directly covered by the Eastern Health Human Research Ethics

committee. The research protocol and Eastern Health ethics application were reviewed and

approved by governing bodies, head offices or CEOs at other mental health services without

formal ethics committees. Data collection took place between June 2015 and January 2016.



In total, 841 patients consented to and completed the survey, approximately 55 per cent of all

patients approached (see Figure 20).

Figure 20. Description of the patient survey sample

Data analysis

The prevalence of comorbid mental health and gambling problems is presented in the following

way:

1. overall prevalence from all screens across all services

2. prevalence by mental health service type

3. prevalence by specific mental health conditions; and

4. prevalence by specific subgroups (e.g. adult, youth, CALD).

All prevalence estimates for problem gambling are reported with exact binomial 95 per cent

confidence intervals. The statistical significance of observed differences in proportions was

examined using Pearson chi-squared or Fisher’s exact tests. Comparisons of demographic and

clinical characteristics were made between comorbid and non-comorbid sample groups in each

treatment population, using chi-square for categorical data and t-tests or non-parametric

equivalents for continuous data. Multiple logistic regressions were used on all cases with complete

data on age, gender, ethnicity and diagnosis to investigate factors independently associated with

comorbidity. Adjusted odds ratios with 95 per cent confidence intervals were obtained and

compared with the odds ratios from the univariate analysis. Interactions between age group,

gender, ethnicity, case-mix variables and site were also investigated.

To deal with outliers, the SPSS “Explore” function was used to identify the five most extreme

values for any variable. These values were classified as outliers, and were Windsorised to the next

highest value in each variable. The five extreme lowest values were not examined if they were

zero values.

The approach used to manage missing data varied according to the circumstances (e.g. how much

data was missing for a particular variable) and the variable. More specific information about how

Total sample

N=1528

Approached

n=1363

Agreed

n=855

Completed

n=841

Incomplete

n=14Declined

n=523Not approached

n=165



scores were calculated and how missing data was managed is described below for relevant

variables.

For responses to individual items see Tables (42–62).

Management of data

Gambling status

If a patient responded ‘yes’ to gambling and to having spent money on a certain gambling activity,

they were classified as a gambler. If a patient responded ‘yes’ to having spent money on a certain

gambling activity, but did not provide a response regarding their frequency of participation in or

expenditure on that activity, they were classified as a gambler, and missing data were maintained

as missing.

Frequency of gambling activity and expenditure on activity

In the initial dataset, patients’ frequency of participation in various gambling activities and

expenditure on each gambling activity was entered as per month, though this resulted in the

presence of decimal places in some responses. To overcome this issue, responses were multiplied

by 12, to reflect annual frequency of participation and expenditure. Prior to computing frequency of

participation in various gambling activities, the maximum frequency for gambling activity

participation was set to 60 (i.e., a maximum of two episodes per day). Any frequency value greater

than 60 was recoded to 60, and thus the yearly maximum for frequency of participation in a

gambling activity was 720 (i.e., 60 multiplied by 12).

Alcohol expenditure

As with frequency of participation in gambling activities and expenditure on gambling activity,

alcohol expenditure was recalculated to be annual expenditure.

Smoking expenditure

As with frequency of participation in gambling activities and expenditure on gambling activity,

smoking expenditure was recalculated to be annual expenditure.

The Heaviness of Smoking Index (HSI)

A new nicotine dependence category, “not dependent”, was created to capture the 71 patients that

were smokers but had values of 0 on the two scale items (i.e. smoked < 10 cigarettes a day, and

smoked first cigarette > 60 minutes after waking up).

Drug user status

If patients indicated having used illicit drugs or misused pharmaceutical drugs in the past 12

months, but did not indicate the type(s) of drug used, a 0 was manually entered for the number of

times they had used in the past 12 months. They were then classified as a non-user, and – in

alignment with how data for drug use variables were entered for patients who responded 0 for

number of times they had used in the past 12 months – all other variables regarding drug use were

given no value.



Management of missing data

Australian Treatment Outcomes Profiles (ATOPS)

ATOPS variable items one, two, and three had 11, 11, and 12 cases missing respectively. This

was true missing data, reflecting a minority of patients who completed an early version of the

survey that did not include the ATOPS, and were thus kept as missing.

Readiness to Change

Data for five patients were missing across the Readiness to Change variables assessing the

importance gamblers placed on changing their gambling behaviour and their level of confidence in

making that change. As these were all identified as being non-gamblers based on the “Gambler”

variable, missing data were manually re-entered as 0.

Total gambling expenditure in last month

When data regarding total gambling expenditure in the last month were missing for non-gamblers,

a 0 was manually entered to replace the missing data. When this data were missing for gamblers

(total of seven cases), data were maintained as true missing data.

Results

Please note that detailed data tables arising from the patients’ survey can be found in Appendix 4.

Participant characteristics

Overall sample size, gender, age and geographical region

The final sample consisted of 841 patients who completed the gambling section of the survey. Just

over half (50.9 per cent) the sample were male (n = 428), 48.3 per cent (n = 406) were female and

7 (0.8 per cent) reported “other” as their gender. The mean age was 38.1 years (SD = 13.3) with a

range of 16–95 years. Figure 21 shows the proportion of patients in each age category. The

majority of patients (n = 735; 87.4 per cent) reported residing in metropolitan areas at the time of

participation, while 9.6 per cent (n = 81) resided in regional areas and 3.0 per cent (n = 25) did not

specify.



Figure 21. Age categories of the total patient sample

Country of birth and other ethnic information

The majority of patients were born in Australia (n = 653; 77.6 per cent). Of those born in a country

other than Australia, 36.2 per cent (n = 68) of respondents were born in Europe, 30.3 per cent (n =

57) in the Western Pacific, 11.7 per cent (n = 22) in South-East Asia, 8.5 per cent (n = 16) in the

Eastern Mediterranean, 4.8 per cent (n = 9) in the Americas, 3.7 per cent (n = 7) in Africa, while

4.8 per cent (n = 9) did not provide an alternative country of birth. Patients born overseas had lived

in Australia for an average of 24.8 years (SD = 15.7 years, range 0–66 years).

The main language spoken at home by patients was English (n = 770; 91.6 per cent), followed by

Mandarin 1.4 per cent (n = 12), Arabic 1.0 per cent (n = 8), Greek 0.7 per cent (n = 6), Cantonese

0.6 per cent (n = 5), Italian 0.6 per cent (n = 5) and Vietnamese 0.5 per cent (n = 4). The remaining

3.7 per cent (n = 31) of patients reported speaking an unspecified ‘Other’ language.

Just over 10 per cent of patients (n = 87) identified as part of an ethnic minority group, with 2 per

cent (n = 17) identifying as Aboriginal or a Torres Strait Islander.

Marital status and accommodation

The majority of patients reported being single (never married, 64.9 per cent, n = 546), 19.7 per

cent (n = 166) reported being married or in a de-facto relationship, 14.3 per cent (n = 120) were

separated or divorced, and 1.1 per cent (n = 9) were widowed. In terms of accommodation, 35.3

per cent (n = 297) reported residing in a rental home, 22.1 per cent (n = 186) were living in their

parents’ or another family member’s home, 21.8 per cent (n = 183) were living in their own home,

and 9.5 per cent (n = 80) were living in public housing. Eleven per cent of patients reported

residing in more unstable forms of accommodation, with 6.4 per cent (n = 54) in supported

accommodation or transitional housing, 3.7 per cent (n = 31) in a boarding house, 0.6 per cent (n =

0%

5%

10%

15%

20%

25%

16-24 25-34 35-44 45-54 55-64 65-75+ Missing

% o

f p

ati

ents

Age group



5) in their friend’s home, 0.4 per cent (n = 3) had no usual residence or were homeless, and 0.1

per cent (n = 1) were residing in a caravan. The remaining 0.1 per cent (n = 1) did not provide their

type of accommodation.

In terms of household type, 28.7 per cent (n = 241) reported living in a single person household

with no children, 25.8 per cent (n = 214) were living in a group or shared household, 12.8 per cent

(n = 108) were living as a couple with children still living at home, 8.1 per cent (n = 68) were single

with children still at home (including joint custody), 7.7 per cent (n = 65) were living as a couple

with no children, 3.2 per cent (n = 27) were single with children not living at home, 3.0 per cent (n =

25) were living as a couple with children not living at home, and 10.7 per cent (n = 90) were living

in some other household arrangement.

Education, employment and benefits

Over 70 per cent of patients reported completing the final year of high school and above (25.0 per

cent (n = 210) Year 12, 20.6 per cent (n = 173), TAFE, 26.6 per cent (n = 223) University degree),

19.4 per cent (n = 163) reported completing to Year 10, 3.4 per cent (n = 29) reported completing

to primary school, 4.8 per cent (n = 40) reported completing some unspecified ‘other’ level of

education, while 0.4 per cent (n = 3) did not report their highest level of completed education.

Only a minority (29.2 per cent) of patients reported working with 12.4 per cent (n = 104) employed

full-time, 16.8 per cent (n = 141) part time, while 15.6 per cent (n = 131) were looking for work.

11.8 per cent (n = 99) reported that they were students, 3.6 per cent (n = 30) were volunteers, 3

per cent (n = 25) were retired, 1 per cent (n = 8) were occupied with household duties, 38.6 per

cent (n = 325) were receiving a Disability Support Pension, 5.5 per cent (n = 46) were receiving a

form of ‘Other’ pension, 6.4 per cent (n = 54) were of ‘Other’ employment status, and 0.8 per cent

(n = 7) were unsure of their employment status. Twenty per cent of patients (n = 170) had a

personal income of less than $500 per fortnight, 27.2 per cent (n = 229) between $500 and $799,

24.7 per cent (n = 208) between $800 and $1,299, 5.9 per cent (n = 50) between $1,300 and

$1,599, 7.5 per cent (n = 63) between $1,600 and $2,599, 3.8 per cent (n = 32) $2,600 or more,

and 10.6 per cent (n = 89) of patients were unsure.

Length of treatment at service

More than half (52.1 per cent; 438), of the patients reported that they had been receiving treatment

at the service for more than one year with 12.7 per cent (n = 107) for less than one year, 9.3 per

cent (n = 78) for less than six months, 10.5 per cent (n = 88) for less than three months, 15.3 per

cent (n = 129) for less than one month, with 0.1 per cent (n = 1) not reporting the length of their

treatment at the service.

Type of service

At the time of participation in the survey, the majority of patients (46.4 per cent; n = 390) were

attending a public adult mental health service, with 35.8 per cent (n = 301) attending an adult

private mental health clinic, 11.1 per cent (n = 93) a youth mental health service, and 6.8 per cent

(n = 57) a community mental health support service (see Figure 22).



Figure 22. Type of mental health service patients attended at the time of participation

Technology and internet use

The majority of patients (91.9 per cent) reported owning a mobile phone at the time of participation

and, of those, a large proportion (71.2 per cent) reported owning a smart phone. In total 754 (89.7

per cent) of the sample reported they access the internet. Sixty-five per cent of patients reported

they accessed the internet on their mobile phones, 55.2 per cent on a personal computer, 32.7 per

cent in their household dwelling, 24.0 per cent on a tablet (or similar device), 16.2 per cent

reported accessing the internet elsewhere in the community 15.1 per cent in their workplace, and

12.0 per cent in their place of study.

Mental health status

The majority of patients (92.7 per cent) reported that they had been diagnosed with a mental

health condition at some point in their lives (lifetime diagnosis), and 88.6 per cent reported being

given a mental health diagnosis in the last year (current diagnosis). Depression, anxiety, psychotic

disorder and bipolar disorder were the most commonly reported lifetime and current mental health

diagnoses (as shown in Figure 23). To see the rates of all diagnoses, see Table 42–44 in

Appendix 4.

Public mental health service

(Adult)46.4%

Public mental health service (Adolescent/

Youth)11.1%

Private mental health service

35.8%

Community support services

(outreach)6.8%



Figure 23. The four most commonly reported lifetime and current mental health diagnoses by patients

Many patients also reported having more than one current mental health diagnosis. Specifically,

56.4 per cent reported having two current diagnoses; 23.7 per cent reported three current

diagnoses; and 8.4 per cent reported having four or more current diagnoses (see Figure 24).

There were low rates of addictive disorders (around 10 per cent), with 2.2 per cent of patients

reporting a lifetime and 0.8 per cent a current gambling disorder. For anxiety disorders, the most

commonly reported was generalised anxiety disorder (77.7 per cent lifetime, 76.8 per cent current),

followed by panic disorder (42.4 per cent lifetime, 37.2 per cent current) and then social anxiety

disorder (39.7 per cent lifetime, 34.9 per cent current).

Figure 24. Proportion of patients with one or more current mental health diagnosis

0 20 40 60 80 100

Depression

Anxiety

Psychotic disorder

Bipolar disorder

% of patients

Me

nta

l h

ea

lth

co

nd

itio

n

Current

Lifetime

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

One Two Three Four or more

% o

f p

ati

en

ts

Proportion of patients with one or more current mental health diagnosis



Medication use

Over 80 per cent of patients reported being prescribed medication for their mental health illness in

the last 12 months. Overall, 46.6 per cent reported being prescribed antipsychotics, 45.2 per cent

antidepressants, 20.9 per cent mood stabilisers, 10.7 per cent sleeping tablets and 14.3 per cent

benzodiazepines. Only 1.9 per cent reported being on opioid substitution therapy, while 6.2 per

cent of patients reported taking some “other” type of prescription medication to treat their mental

health illness (see Figure 25).

Figure 25. Current medications patients report being prescribed for mental health conditions

Wellbeing and quality of life

Wellbeing was assessed using three items from the Australian Treatment Outcome Profile (ATOP).

Patients were asked to rate their psychological health status over the past four weeks (e.g. anxiety,

depression and problematic emotions and feelings, their physical health status over the past four

weeks (e.g. the extent of physical symptoms and how bothered they are by illness) and finally were

asked to rate their overall quality of life over the past four weeks (e.g. the extent to which they are

able to enjoy life or are satisfied with their living conditions). Patients were asked to provide a

score ranging from ’0’ representing a poor quality of life and ‘10’ representing a good quality of life.

Mean scores were 4.81 (SD = 2.63) for psychological health, 5.14 (SD = 2.44) for physical health

and 5.14 (SD = 2.55) for overall quality of life indicative of poor physical health and psychological

health and a poor overall quality of life. The distribution of scores across psychological health,

physical health, and overall quality of life items are displayed in Figure 26.

0% 10% 20% 30% 40% 50% 60%

Antipsychotics

Antidepressants

Mood stablisers

Benzodiazepines

Sleeping tablets

Other

Opioid substitution

Missing

% of patients

Me

dic

ati

on

typ

e



Figure 26. Distribution of scores across ATOP items assessing psychological health, physical health, and overall

quality of life

Alcohol use

The majority of patients (67.7 per cent) reported consuming alcohol at some time in the past year.

Of those patients, 16.9 per cent reported drinking two to three times per week, and 10.7 per cent

drinking 4 or more times a week (see Figure 27). The average weekly spend on alcohol among

alcohol drinkers (n = 499) was $43.85 (SD = $91.55, mode = $0, median = $15.00).

Figure 27. Frequency of patients’ self-reported alcohol consumption in the past year among patients who reported

any past-year alcohol use

0

10

20

30

40

50

60

70

80

90

100

1 2 3 4 5 6 7 8 9 10

% o

f p

ati

en

ts

Score

Psychological health

Physical health

Overall quality of life

Monthly or less

(44.1%)

2-4 times a month

(28.3%)

2-3 times a week (16.9%)

4+ times a week (10.7%)



Just over half of patients (50.6 per cent) who reported drinking in the last 12 months were doing so

above the NHMRC recommended guidelines for reducing health risks related to alcohol (i.e., > 2

standard drinks on each drinking occasion). Over 60 per cent of these patients reported binge

drinking in the past year (i.e., consuming six or more standard drinks in one drinking occasion,

National Health and Medical Research Council, 2009; see Figure 9), and 14.2 per cent of patients

who drank reported binge drinking at least weekly in the past year (National Health and Medical

Research Council, 2009; see Figure 10).

The mean AUDIT-C score among those who had consumed alcohol in the past year (n = 499) was

4.0 (SD = 2.8), with over half of those patients (55.2 per cent) identified as problem drinkers

according to the AUDIT-C criteria.

Tobacco use

Just under half of the sample (49.1 per cent) reported smoking in the past year, with just over a

third of smokers (36.6 per cent) smoking less than 10 cigarettes a day, another third (30 per cent)

smoking 11–20 cigarettes a day, almost a quarter (24.7 per cent) smoking 21–30 a day, and 8.7

per cent smoking 31 or more cigarettes a day (see Figure 28). The average weekly spend on

tobacco among smokers was $71.99 (SD = $59.27), Mode = $100.00, median = $60.00.

Figure 28. Number of self-reported cigarettes smoked on a typical day in the past year among smokers

The majority of patients who reported that they had smoked in the past year also reported smoking

their first cigarette within five minutes of waking (44.6 per cent), followed by 28.6 per cent of patients

who reported smoking their first cigarette within 6–30 minutes of waking. The mean score on the two

item FTND (Heatherton et al., 1989) for the 413 patients who reported smoking cigarettes in the past

year was 3.0 (SD = 1.9), with just over a quarter of patients who smoke (25.2 per cent) with high

dependence and just under a quarter (22.5 per cent) with moderate dependence.

Drug use

Almost a quarter (24.1 per cent, n = 203) of patients reported using an illicit drug or a prescription

medication for non-medical use (drug use used henceforth for both terms) in the past year. Among

these patients, the most commonly used illicit drug was cannabis (85.2 per cent), followed by

amphetamines (53.7 per cent) and then sedatives (37.4 per cent, see Figure 29). The average

weekly spend on illicit drugs among drug users was $120.22 (SD = 194.75, mode = $0,

median = $50.00).

Less than 10(36.6%)

11-20(30.0%)

21-30(24.7%)

31+(8.7%)



Figure 29. Specific drugs used among the total sample (N = 841) and among patients reporting drug use in the past

year (N = 203)

Poly-drug use was common among patients who reported past year drug use. Patients reported

using an average of 2.5 drugs (SD = 1.5 drugs) and up to a maximum of seven drugs.

Figure 30 displays the frequencies with which patients with past year drug use reported using

specific drugs. Of the patients who reported past year drug use, 25.1 per cent reported using

cannabis, 6.4 per cent using sedatives, 6.0 per cent using amphetamines, and 3.5 per cent using

opioids on a daily or almost daily basis. Eighteen per cent of patients reported using hallucinogens,

12.8 per cent reported using cocaine, and 8.9 per cent reported using inhalants at least once or

twice in the past year.

Seventy-one per cent of patients who reported drug use in the past year identified a primary drug

of concern (henceforth referred to as PDOC). The three most commonly reported PDOCs were

cannabis (27.2 per cent), amphetamines (20.2 per cent) and alcohol (16.2 per cent).

0 10 20 30 40 50 60 70 80 90 100

Cannabis

Amphetamines

Sedatives

Hallucinogens

Opioids

Cocaine

Inhalants

% used in the past year

Dru

g

All patients

Patients reportingpast year drug use



Figure 30. Frequency of drug use in the past year

Overall gambling behaviour

Of the 841 patients, 348 reported that they had gambled in the past year (i.e., reported that they

“had gambled at any point/time in the last 12 months where gambling includes wagering on a race

or event, buying a lottery ticket, playing keno or playing cards at home as well as playing the

pokies or betting on sports”). The overall rate of participation in any gambling activity in the total

sample was 41.4 per cent [95 per cent CI = 38.1–44.7], which is substantially lower than the 61.6

per cent [95 per cent CI = 59.1 per cent – 64.0 per cent] (excluding raffles) reported in the general

population survey undertaken in 2014 by the VRGF4 (see Figure 38).

A significantly larger proportion of males (n = 198, 46.3 per cent) reported that they had

participated in some type of gambling activity in the past year than did female patients (n = 148,

36.5 per cent; (2(1, n = 834) = 8.26, p= 0.01), note that gender analyses excludes the seven

patients who identified their gender as ‘Other’).

Of the patients who reported participating in gambling activities in the past year (henceforth

gamblers), half reported playing pokies or electronic gaming (50.3 per cent) or Lotto, Powerball or the

Pools (49.7 per cent), and just over a quarter reported betting on horse or greyhound racing (26.1 per

cent) or buying scratch tickets (25.3 per cent). Figure 31 and Figure 32 respectively display the

proportion of all patients and of gamblers who reported engaging in each specific type of gambling

activity. In terms of gender differences, among gamblers gambling activities more common among

female patients were scratch tickets, Lotto, Powerball or Pools, and pokies or electronic gaming,

while all other activities were more common among male patients (see Figure 32).

4 Note differences in gambling participation rates may reflect differences in the demographic characteristics of the two samples and at the different time-points at which data were collected.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

% o

f p

ati

en

ts

Drugs

Daily or almost daily

Weekly

Monthly

Once or twice

Never



Figure 31. Proportion of patients who reported engaging in specific gambling activities in the past year

Figure 32. Proportion of gamblers who reported engaging in specific gambling activities in the past year

19.0%

19.2%

6.9%

12.6%

2.5%

1.0%

2.0%

2.0%

1.2%

0.2%

22.7%

22.0%

14.5%

8.6%

8.9%

8.6%

4.0%

3.0%

1.6%

0.9%

20.8%

20.6%

10.8%

10.5%

5.7%

5.0%

3.0%

2.5%

1.4%

0.6%

0% 20% 40% 60% 80% 100%

Pokies or electronic gambling

Lotto, Powerball or Pools

Betting on horse or greyhound racing

Scratch tickets

Casino table games

Betting on sports

Keno

Informal private betting

Bingo

Betting on events

% of patients

Ga

mb

lin

g a

cti

vit

y

All patients

All malepatients

All femalepatients

52.0%

52.7%

18.9%

34.5%

6.8%

2.7%

5.4%

5.4%

3.4%

0.7%

49.0%

47.5%

31.3%

18.7%

19.2%

18.7%

8.6%

6.6%

3.5%

2.0%

50.3%

49.7%

26.1%

25.3%

13.8%

12.1%

7.2%

6.0%

3.4%

1.4%

0% 20% 40% 60% 80% 100%


Lotto, Powerball or Pools


Scratch tickets

Casino table games

Betting on sports

Keno


Bingo

Betting on events

% of patients

Ga

mb

lin

g a

cti

vit

y

All gamblers

Malegamblers

Femalegamblers



Venue type

The majority of gamblers reported attending physical venues (i.e., as opposed to gambling online)

to participate in gambling activities in the past year. In terms of the frequency of gambling based

on mean visits, the top three activities gamblers reported participating in (defined by spending any

money on that activity) at physical venues were: betting on casino table games (M = 70.4 times,

SD = 189.9 times, range = 0–720 times), followed by playing Keno (M = 70.2 times, SD = 161.8

times, range = 0–720 times) and then playing Lotto, Powerball or the Pools (M = 51.6 times, SD =

119.2, range = 0–720, see blue columns in Figure 33).5 However, when the medians6 were

examined playing pokies or electronic gambling (median = 24) and Lotto, Powerball or the Pools

(median = 24) were the most frequently played activities. The top three activities gamblers reported

participating in online were: betting on sports (M = 26.4 times, SD = 65.6 times, range = 0–396

times), followed by betting on horse or greyhound racing (M = 21.0 times, SD = 92.3 times, range =

0–600 times), and then playing pokies or electronic gambling (M = 14.1 times, SD = 80.7 times,

range = 0–720 times, see orange columns in Figure 33). Finally, betting on events took place

exclusively online.

When venue and online gambling were combined (see Figure 34), the median scores indicated

that the top three most frequently played activities were playing pokies or electronic gambling and

Lotto, Powerball or the Pools (both with a median of 24), whereby all other activities (except betting

on events) had a median of 12.

5 Note that the frequency of gambling participation on each activity is a mean score among those who engaged in that gambling activity which could be distorted by the low number of gamblers in certain activities (e.g. casino table games). 6 The median is lower than the mean because participants were asked how many times per month had they spent money on each gambling activity (if they had indicated that they had gambled on that activity in the past year). For example for someone gambling only once a year on an activity, the number of times per month would be less than one, hence the many resulting ‘zeros’ brings down the median score.



Figure 33. Frequency of gamblers' participation in specific gambling activities at physical venues and online in the

past year (n = 348)

Figure 34. Frequency of gamblers' participation in specific gambling activities (at venue or online) in the past year

(n = 348)

0 20 40 60 80

Casino table games

Keno

Lotto, Powerball or the Pools



Betting on sports

Scratch tickets

Bingo

Betting on events

Informal private games

Mean number of times

Ga

mb

lin

g a

cti

vit

y

Online

Physical venue

0 20 40 60 80 100

Casino table games

Keno




Betting on sports

Scratch tickets

Bingo

Betting on events

Informal private games

number of times

Ga

mb

lin

g a

cti

vit

y

Mean

Median



Average spend on gambling in the past year

All monetary amounts described in this section are Australian dollars. On average, the total

amount that gamblers reported spending on any gambling activities in the past month was $118.85

(SD = 317.12, range = $0–$3,000). The median spend was $20.00, and the mode among those

who had gambled in the past month was $50.00.

The total amount of money gamblers spent on each gambling activity at each type of venue over

the past year was estimated, when:

patients indicated they had gambled (any activity) in the past year, and

they had participated in a particular gambling activity (e.g. playing Keno) in the past year,

and

the amount they spent in the past month on a particular gambling activity at a specific

venue (e.g. spent $340 playing Keno at a venue in the past month) was reported.

Gamblers’ past year average spend was then calculated by multiplying the monthly spend of valid

cases by 12.

Overall, gamblers spent more money participating in gambling activities located at physical venues

than they did participating online. The three activities at physical venues that gamblers spent the

most on playing in the past 12 months were: betting on casino table games (M = $1,823.45, SD =

$2,930.12, range = $0–$12,000.00), followed by playing pokies or electronic gambling (M =

$1,679.10, SD = $2,838.12, range = $0–$12,000.00), and then playing Keno (M = $1,297.67, SD =

$3,414.36, range = $0–$12,000.00).

The average annual spend on online gambling activities was more modest, for the most part. The

three activities that gamblers reported spending the most playing online in the last 12 months

were: betting on horse or greyhound racing (M = $1,725.43, SD = $2,062.58, range = $12.00–

$7,200.00), followed by scratch tickets (M = $1,440.00, SD = $1,539.87, range = $120.00–

$3,600.00), and then playing pokies or electronic gambling (M = $1,222.00, SD = $1,706.57, range

= $6.00–$6,000.00, see Figure 35). Gamblers did, however, spend more participating online than

at physical venues for three gambling activities in the past year: scratch tickets ($1,440.00 online

vs. $271.34 at a venue), betting on horse or greyhound racing ($1,725.43 online vs. $1,274.11 at a

venue), and betting on events ($330.00 online vs. $0 at a venue).



Figure 35. Gamblers' spend on specific gambling activities at venues and online in the past year

Problem gambling

Prevalence

Problem gambling behaviour was assessed using the nine-item Problem Gambling Severity Index

(henceforth the PGSI; Ferris & Wynne, 2001). The PGSI divides gambling behaviour into four risk

categories: non-problem gambling, low-risk gambling, moderate-risk gambling and problem

gambling (henceforth referred to as gambling risk status).

As shown in Figure 36, the majority (n = 493; 58.6 per cent of total sample) were non-gamblers,

and so it can be assumed that they were not experiencing any gambling problems at the time of

the survey, though it is possible that they could have experienced gambling-related difficulties prior

to the last 12 months. Overall, the mean PGSI total score among the gamblers (n = 348 patients)

was 3.2 (SD = 5.1, range = 0–27), just inside of the moderate-risk range. Of the total sample (N =

841), 19.6 per cent (n = 165) had a PGSI score in the non-problem gambling range, 7.1 per cent (n

= 60) had a PGSI score in the low-risk range, 8.3 per cent (n = 70) had a PGSI score in the

moderate-risk range risk range, and 6.3 per cent (n = 53) were identified as problem gamblers.

$0 $500 $1,000 $1,500 $2,000

Casino table games


Keno


Betting on sports

Bingo


Scratch tickets

Betting on events

Mean amount spent

Ga

mb

lin

g a

cti

vit

y

Online

Physical venue



Figure 36. Gambling risk across the whole sample

There was also a significant relationship between patients’ gender and gambling harm category

(2(4, n = 834) = 14.18, p<0.01). In this case, a higher proportion of gamblers with scores in any of

the gambling risk categories (i.e., low-risk gamblers, moderate-risk gamblers, or problem

gamblers) were male (see Figure 37).

Total Sample

N=841

Gambled in the past year

41.4% (n=348)

Non-problem gambler

19.6% (n=165)

Low risk gambler

7.1% (n=60)

Moderate risk gambler

8.3% (n=70)

Problem gambler

6.3% (n=53)

Non-gamblers

58.6% (n=493)



Figure 37. Gender breakdown of gambling categories

Figure 43 shows that despite lower rates of gambling participation, patients with mental health

problems experienced elevated rates of harm from gambling relative to the general population,

with a lower proportion falling in the non-problem (19.6 per cent vs. 57.6 per cent) and low-risk (7.1

per cent vs. 8.9 per cent) range, with three times as many falling in the moderate-risk range (8.3

per cent vs. 2.8 per cent), and with eight times as many falling in the problem gambling range (6.3

per cent vs. 0.8 per cent) on the PGSI7. It is, however, becoming increasingly more common to

consider gambling harm on a continuum. Indeed, a recent study found that “50 per cent, 34 per

cent, and 15 per cent of the total harm resulting from gambling in Victoria can be divided among

low-risk, moderate-risk and problem-gamblers, respectively” (Browne et al., 2016), with the

greatest harm experienced by low-risk gamblers. Adopting this approach with the current sample,

one in five patients with mental health problems (21.7 per cent) experienced at least some

gambling harm (i.e., identified as a low-risk, moderate-risk or problem gambler). Moreover, if a

patient with a mental health problem gambles, there is a strong possibility that they would be

experiencing at least some level of gambling harm, given that 52.5 per cent of gamblers fell into

the low-risk, moderate-risk or problem gambling harm category.

7 Note differences in rates of gambling harm by PGSI category may reflect differences in the demographic characteristics of the two samples and different time-points at which data were collected.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Non-gamblers Non-problemgamblers

Low riskgamblers

Moderate riskgamblers

Problemgamblers

% o

f p

atie

nts

Gambling category

Female

Male



Figure 38. Rates of gambling participation and harm in current patient sample and general Victorian population

In terms of levels of harms, 52.5 per cent of gamblers in the current sample were experiencing

gambling harm at the time of participation (see Figure 39).

Figure 39. Level of harm past year gamblers report currently experiencing

29.9%

57.6%

8.9%

2.8%0.8%

58.6%

19.6%

7.1% 8.3%6.3%

0%

10%

20%

30%

40%

50%

60%

70%


Low risk gamblers Moderate riskgamblers

Problem Gamblers

% o

f p

ati

en

ts

Gambling risk category

General Victorian Population

Mental Health PopulationSample

47.4%

17.2%20.1%

15.2%

-10%

0%

10%

20%

30%

40%

50%

60%

70%

Non-problem gamblers Low risk gamblers Moderate riskgamblers

Problem gamblers

% o

f g

am

ble

rs

GambIing risk category

≈ 8 times ≈ 3 times

21.7% of the current patient

sample experiencing gambling

harm

52.5% of gamblers

experiencing gambling harm

I_



Table 14 shows the number and percentage (with 95 per cent confidence intervals) of patients

falling in each of the gambling categories, highlighting that up to 10 per cent of the mental health

sample are moderate-risk gamblers and that up to 8 per cent of the mental health population are

problem gamblers.

Table 14. Proportion of patients in each of the gambling categories

n % [95% CI]

Non-gamblers 493 58.6 [55.3 – 61.9]

Non-problem gamblers 165 19.6 [16.9 – 22.3]

Low-risk gamblers 60 7.1 [5.4 – 8.8]

Moderate-risk gamblers 70 8.3 [6.4 – 10.2]

Problem gamblers 53 6.3 [4.6 – 7.9]

Characteristics of problem gamblers

Age

One-way ANOVA revealed a significant difference in mean age across gambling category (F = 2.8,

p<0.05) and post hoc tests indicated that this was due to non-gamblers being significantly younger

(mean age = 37.1) than non-problem gamblers (mean age = 40.5).However, in terms of gambling

participation and gambling harm across the age groups, a larger proportion of problem gamblers in

the 35–44 years age group were experiencing gambling-related harm (see black columns in Figure

40). Furthermore, gamblers in the 65+ age group had the highest rates of participation, but

experienced the lowest rates of gambling harm (see blue and orange columns in Figure 40).

Finally, a greater proportion of gamblers in the 35–44 year age category were experiencing some

level of gambling harm (see orange column in Figure 40).

Figure 40. Age breakdown of gambling category

0%

10%

20%

30%

40%

50%

60%

16-24 25-34 35-44 45-54 55-64 65+

% o

f p

ati

en

ts

Age group

Gambling participation (any)

Experiencing harm (any)

Problem gamblers



Country of birth and other ethnic information

While there was no there was no significant association between country of birth and gambling

category (2(4, N = 841) = 6.30, p = 0.18), a higher proportion of patients in all categories were

born in Australia, as opposed to countries overseas.

Furthermore, while the association between the main language patients reported speaking at

home (i.e., English vs. another language) and gambling harm category was not significant for

patients who reported gambling participation (2(3, n = 348) = 4.14, p = 0.25), the proportion of

patients who reported speaking English as their primary language reduced as gambling risk

increased.

Finally, there were no significant associations between identification with an ethnic minority group,

or as Aboriginal or Torres Strait Islander and participation in gambling activities either (any).

Indeed, only a minority of gamblers reported identifying as part of an ethnic minority group (9.2 per

cent), or as Aboriginal or Torres Strait Islander (1.7 per cent).

Marital status

There was no significant relationship between marital status and gambling category (2(12, N =

841) = 16.17, p = 0.18), however the proportion of never married (single) patients increased as

gambling risk increased (see Figure 41).

Figure 41. Marital status and gambling category

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Non-gambler Non-problemgambler

Low riskgambler

Moderate riskgambler

ProblemGambler

% o

f p

ati

en

ts

Gambling category

Married /de facto

Separated/divorced

Never married

Widowed



Education, employment and benefits

No significant relationships were found between highest level of education achieved and gambling

risk categories (2(15, n = 348) = 11.73, p = 0.70). The majority of gamblers reported completing

high school to at least Year 10 level. Approximately 69 per cent of gamblers reported completing

Year 12 and above (i.e., TAFE qualification or University degree).

There was a significant relationship between employment status and gambling harm category for

gamblers (2(3, n = 347) = 8.47, p<0.05). Specifically, a higher proportion of moderate-risk and

problem gamblers reported that they were currently neither employed nor studying (see Figure 42).

Figure 42. Current employment status and gambling harm category

Service type

Differences were found in the proportion of patients presenting with scores in the moderate-risk

and problem gambling ranges on the PGSI across service types. Specifically, higher rates of

moderate-risk and problem gambling were observed in mental health community support services

(outreach) vs. specialist public mental health services (14.0 per cent vs. 6.4 per cent; (2(2, n =

841) = 5.1, p = 0.08), in public vs. private services (7.0 per cent vs. 5.0 per cent 2 = per cent;

(2(2, n = 841) = 2.51, p = 0.29) and in adult vs. youth services (7.0 per cent vs. 1.1 per cent, (2(2,

n = 841) = 9.8, p<0.01) though only the latter reached statistical significance (see Figure 43).

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Non-problemgambler

Low risk gambler Moderate riskgambler

Problem gambler

% o

f g

am

ble

rs


Employed orstudying

Unemployed



Figure 43. Proportion of gamblers presenting to mental health services with moderate-risk or problem gambling

Mental health status

Similar to the whole patient sample, the majority of gamblers (92.8 per cent) reported that they had

been diagnosed with a mental health condition at some point in their lives (lifetime diagnosis), and

87.6 per cent reported a mental health condition diagnosis in the last year (current diagnosis).

Depression, anxiety, psychotic disorder and bipolar disorder were again the most commonly

reported current mental health diagnoses by gamblers (see Figure 44).



Figure 44. Current mental health diagnoses and gambling harm category

Furthermore, there were few differences in the proportion of gamblers who reported having any

mental health disorder, anxiety, or depression, though the proportion reporting having a psychotic

disorder were higher among those experiencing gambling harm. Logistic regression analyses

showed that there were certain mental health disorders where the odds of experiencing gambling

harm and problem gambling were higher. Specifically, the odds of a patient with a drug use

disorder experiencing gambling harm was 3.6 times greater than patients without a drug use

disorder, and the odds of a patient with psychotic disorder experiencing gambling harm was 2.39

times greater than patients without a psychotic disorder (see Table 15). When gender and age

were added as covariates, the odds of a patient with borderline personality disorder experiencing

gambling harm was 3.05 times greater than patients without borderline personality disorder (95 per

cent CI = 1.14–8.14, p<0.05). The odds of patients experiencing problem gambling were also

significantly greater for patients who reported being currently diagnosed with drug use disorder and

borderline personality disorder. Specifically, the odds of a patient diagnosed with a drug use

disorder experiencing problem gambling was 3.41 times greater than those without a diagnosis of

a drug use disorder, and the odds of a patient diagnosed with borderline personality disorder

experiencing problem gambling was 2.59 times greater than patients without a borderline

personality disorder (see Table 15). These findings were maintained when controlling for age and

gender.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Non-problemgambler

Low risk gambler Moderate riskgambler

Problem gambler

% o

f g

am

ble

rs


Any

Depression

Anxiety

Psychotic disorder



Table 15. Odds of gambling harm (any) gambling among gamblers with current mental

health diagnoses

Odds of any gambling harm Odds of problem gambling

Disorder Odds ratio 95% CI for Odds Ratio

Odds ratio 95% CI for Odds Ratio

Drug use disorder 3.56** 1.40–9.06 3.41** 1.46–7.92

Psychotic disorder 2.39*** 1.48–3.86 1.56 0.84–2.89

Alcohol use disorder 2.31 0.87–6.13 1.26 0.40–3.90

Borderline personality disorder 2.10 0.88–4.98 2.59* 1.06–6.35

Bipolar disorder 1.03 0.60–1.78 1.83 0.92–3.61

Anxiety disorder (any) 0.96 0.61–1.51 1.02 0.55–1.88

Depression 0.95 0.60–1.51 0.76 0.41–1.40

Eating disorder 0.58 0.10–3.52 0.00 N/A

Other disorder (unspecified) 0.45 0.20–1.02 0.38 0.09–1.65

*p<0.05, **p<0.01, ***p<0.001

Furthermore, as shown in Table 16, the proportion of patients falling into the moderate-risk and

problem gambling categories increased with the number of current mental health diagnoses

patients reported currently having. For example 11.3 per cent of patients with four or more mental

health disorders were problem gamblers compared to only 6.6 per cent of those with one or more

mental health disorders.

Wellbeing and quality of life

Mean scores on the three scales of the ATOP were calculated for each of the gambling categories.

Figure 45 displays mean scores on the ATOP scales for each gambling category.



Table 16. Number of mental health diagnoses and proportion experiencing gambling harm

Non-gambler Non-problem gambler Low-risk gambler Moderate-risk gambler Problem gambler

n % n % n % n % n %

At least one mental health disorder (n = 762)

456 58.9 143 19.2 50 6.7 64 8.6 49 6.6

At least two mental health disorders (n = 474)

275 58.0 86 18.1 38 8.0 40 8.4 35 7.4

At least three mental health disorders (n = 199)

112 56.3 31 15.6 15 7.5 20 10.1 21 10.6

At least four mental health disorders (n = 71)

40 56.3 9 12.7 5 7.0 9 12.7 8 11.3

*Note: numbers in this table sum to greater than that of total sample (N = 841), as categories capturing number of diagnoses are not mutually exclusive.



Figure 45. Mean scores on ATOP scales for each gambling category

A significant difference was found between groups for means scores on the physical health scale

(F(4, 823) = 2.83, p<0.05), whereby problem gamblers and moderate-risk gamblers had the lowest

scores. However, no significant differences between gambling categories were found for mean

scores on the psychological health scale (F(4, 823) = 0.82, p = 0.51) (see Figure 45b) or on the

overall quality of life scale (F(4, 823) = 1.04, p = 0.39) (see Figure 45c).

Alcohol use

While there were no significant differences between groups with regards to problem drinking (as

defined by scores the AUDIT-C (2(4, N = 841) = 8.28, p = 0.08), a larger proportion of low-risk

gamblers (53.3 per cent) were classed as problem drinkers (as indicated by the AUDIT-C).

Tobacco use

There was a significant relationship between tobacco use and gambling category (2(4, N = 841) =

28.44, p<0.001). Figure 46 shows the proportion of smokers in each gambling category.

Specifically, a higher proportion of moderate-risk and problem gamblers reported smoking tobacco

in the past year.

Drug use

There was a significant relationship between past year illicit drug use (any drug) and gambling

category (2(4, N = 841) = 21.16, p<0.001), with a greater proportion of moderate-risk gamblers

and problem gamblers reporting that they had used illicit drugs in the past year (see Figure 47).

0

1

2

3

4

5

6


Low riskgambler


Problemgambler

Mean

sco

re

Gambling category

A. Physical health

0

1

2

3

4

5

6


Low riskgambler


Problemgambler

Mean

sco

re

Gambling category

B. Psychological health

0

1

2

3

4

5

6


Low riskgambler


Problemgambler

Mean

sco

re

Gambling category

C. Overall quality of life



Figure 46. Proportion of smokers in each of the gambling categories

Figure 47. Past year drug use status by gambling category

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%


Low riskgamblers


Problemgamblers

% o

f p

ati

en

ts

Gambling category

Non-smoker

Smoker

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%


Low riskgamblers


Problemgamblers

% o

f p

ati

en

ts

Gambling category

Non-user

Drug user



Gambling behaviour

Participation in specific activities

Rates of participation in specific types of gambling activities for gamblers in each gambling harm

category are shown in Table 17.

The top three gambling activities of patients in each gambling harm category were:

Problem gamblers: Pokies or electronic gaming (81.1 per cent), Lotto, Powerball or the

Pools (45.3 per cent), and scratch tickets (32.1 per cent).

Moderate-risk gamblers: Pokies or electronic gaming (70.0 per cent), Lotto, Powerball or

the Pools (44.3 per cent), and betting on horse or greyhound racing (31.4 per cent).

Low-risk gamblers: Pokies or electronic gaming (45.0 per cent), Lotto, Powerball or the

Pools (44.3 per cent), and scratch tickets (35.0 per cent).

Non-problem gamblers: Lotto, Powerball or the Pools (55.8 per cent), Pokies or

electronic gaming (33.9 per cent), and betting on horse or greyhound racing (24.8 per

cent).

There was little variation between gambling categories in terms of activities that patients

participated in, with pokies or electronic gaming the most common in all groups, except for non-

problem gamblers. Lotto, Powerball or the Pools were the most common for non-problem

gamblers.

There was, however, some variation between the non-problem and problem gambler risk

categories in terms of how likely they were to participate in some gambling activities. Specifically,

compared to non-problem gamblers, problem gamblers were significantly more likely to participate

in:

Pokies or electronic gaming – OR = 8.54, p<0.001

Casino table games – OR = 2.83, p<0.05

Keno – OR = 5.25, p<0.01

Bingo – OR = 6.65, p>0.05

There was also variation in the top three activities of reported participation across all gambling risk

categories between patients in this sample and participants from the general population. In the

case of the latter, the top three gambling activities reported by participants in each gambling harm

category were:

Problem gamblers from the general population: Lotto, Powerball or the Pools (67.4 per

cent), Pokies or electronic gaming (66.5 per cent) and betting on racing (52.5 per cent).

Moderate-risk gamblers from the general population: Lotto, Powerball or the Pools

(80.6 per cent), Pokies or electronic gaming (58.4 per cent), and raffles, sweeps and

competitions (53.0 per cent).



Low-risk gamblers from the general population: Lotto, Powerball or the Pools (68.9 per

cent), raffles, sweeps and competitions (57.8 per cent) and betting on racing (43.5 per

cent).

Non-problem gamblers from the general population: Lotto, Powerball or the Pools

(65.9 per cent), raffles, sweeps and competitions (60.8 per cent) and betting on racing

(26.8 per cent).



Table 17. Participation in specific gambling activity by gambling category

Keno Bingo Casino table games

Pokies or electronic gaming Horse or greyhound racing

Scratch tickets

Sports Lotto, Powerball, or the pools


Events

n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%)

Non-problem gamblers

(n = 165)

8

(4.8%)

2

(1.2%)

14

(8.5%)

56

(33.9%)

41

(24.8%)

36

(21.8%)

15

(9.1%)

92

(55.8%)

11

(6.7%)

3

(1.8%)

Low-risk gamblers

(n = 60)

2

(3.3%)

2

(3.3%)

11

(18.3%)

27

(45.0%)

14

(23.3%)

21

(35.0%)

5

(8.3%)

26

(43.3%)

6

(10.0%)

1

(1.7%)

Moderate-risk gamblers

(n = 70)

5

(7.1%)

4

(5.7%)

12

(17.1%)

49

(70.0%)

22

(31.4%)

14

(20.0%)

17

(24.3%)

31

(44.3%)

3

(4.3%)

1

(1.4%)

Problem gamblers

(n = 53)

10

(18.9%)

4

(7.5%)

11

(20.8%)

43

(81.1%)

14

(26.4%)

17

(32.1%)

5

(9.4%)

24

(45.3%)

1

(1.9%)

0

(0.0%)



Frequency of gambling participation in top three gambling activities

Figure 48, Figure 49 and Figure 50 show the average number of times patients in this sample and

participants from the general population reported participating in the top three gambling activities,

broken down by gambling harm category.

Figure 48. Average number of times in the past year patients reported playing Pokies or electronic gambling

Figure 49. Average number of times in the past year patients reported betting on horse of greyhound racing

0

20

40

60

80

100

120

140

160

Non-problemgamblers


Problem gamblers

Me

an n

um

be

r o

f ti

me

s p

ast

year


Mental HealthSample

GeneralVictorianPopulation

0

50

100

150

200

250

Non-problemgamblers


Problem gamblers

Me

an n

um

be

r o

f ti

me

s p

ast

year


Mental HealthSample




Figure 50. Average number of times in the past year patients reported playing Lotto, Powerball or the Pools

Overall, and with the exception of playing pokies or electronic gambling, on average gamblers

attending mental health services reported betting on horse or greyhound racing (see Figure 49) or

playing Lotto, Powerball and the pools (see Figure 50) less often than did participants from the

general population. On average, problem gamblers in the mental health population reported

playing the pokies or electronic gaming more times in the past year than did participants in the

general population (see blue columns in Figure 48).

Average monthly spend by gambling harm category

Average monthly spend was calculated for all gamblers in each gambling harm category, and is

displayed in Figure 51. Compared to non-problem gamblers, problem gamblers spent

approximately eleven times more money per month on gambling activities, moderate-risk gamblers

spent approximately three times more money per month on gambling activities, and low-risk

gamblers spent approximately 30 per cent more money per month on gambling activities.

0

10

20

30

40

50

60

70

80

Non-problemgamblers


Problem gamblers

Me

an n

um

be

r o

f ti

me

s p

ast

year


Mental HealthSample




Figure 51. Average monthly spend on gambling activity (any) for patients in each gambling harm category

Frequency of gambling participation

Beginning with the Pokies/EGMs, the mean PGSI score increased as the frequency of play

increased, and there was a significantly higher mean PGSI scores observed among daily gamblers

(F = 22.11, p<0.001), which placed them in the problem range (see Figure 52). This finding

remained significant when analyses were restricted to male (F = 9.27, p<0.001) and female (F =

13.71, p<0.001) gamblers. A similar pattern was observed, with the frequency of betting on horse

or greyhound racing, whereby daily gamblers had a significantly higher PGSI score (F = 3.80,

p<0.01), also falling in the problem gambling range (see Figure 52). This finding remained

significant when analyses were restricted to male (F = 3.79, p<0.01) but not female (F = 1.20, p =

0.31) gamblers, possibly reflecting the low number of females who gambled regularly on this

activity. The pattern was very different for participation in Lotto, Powerball or the Pools and scratch

tickets, where mean scores on the PGSI for the total sample hovered around the moderate-risk

range (i.e., 2–7) regardless of how frequently gamblers reported engaging in the activity (See

Figure 52), with no significant differences observed when analyses were restricted by gender.

$39.12$50.32

$123.84

$439.79

$0.00

$50.00

$100.00

$150.00

$200.00

$250.00

$300.00

$350.00

$400.00

$450.00

$500.00

Non-problem gambler Low risk gambler Moderate risk gambler Problem gambler

Ave

rag

e m

on

thly

sp

en

d $

AU

Gambling category



Figure 52. Mean PGSI score as a function of frequency of participation in the top four gambling activities

Screening for gambling problems

Patients were asked a series of questions to establish whether or not they had been asked about,

or screened for, problem gambling since they had been attending the mental health service.

Overall, 42.9 per cent (n = 361) of all patients reported that they had been asked about gambling

or filled out a survey about gambling at some point since they first attended at the service (i.e.,

either straight away, or after some time).

The highest proportion of gamblers who reported being asked by a clinician about their gambling

behaviour when attending a mental health service were problem gamblers. Over 60 per cent (n =

35) of problem gamblers reported that a clinician asked them about their gambling, or that they

filled in a survey about gambling, at some point after they started attending the service. The

proportion of patients who reported being asked or screened for problem gambling decreased as

gambling risk decreased (see Figure 53).



Figure 53. Proportion of patients in each gambling category who reported being asked about or screened for their

gambling by a clinician

Help-seeking

Patients were asked to rate (from a list of options) how likely they would be to access particular

types of support services if they were to develop a gambling problem. Figure 54 displays the

proportion of patients in the sample who reported that they would be ‘likely’ or ‘very likely’ to use

each type of service if they developed a gambling problem. The most frequently endorsed service

was speaking to a mental health worker (66.9 per cent), followed by using self-help strategies (e.g.

budgeting, 66.4 per cent), self-exclusion from physical venues (58.2 per cent), and then speaking

to a gambling counsellor (56.7 per cent).

Figure 54. Types of help patients from the total sample were likely to seek if they develop a gambling problem

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Non-gamblers Non-problemgambler

Low riskgambler


Problemgambler

Total sample

% o

f p

ati

en

ts

Gambling category

No

Yes

0% 10% 20% 30% 40% 50% 60% 70% 80%

Gambling app

Self-help program

Online service

Support group

Gambling helpline

Family or friends

GP

Gambling counsellor

Self exclusion

Self-help strategies

Mental health worker



There were some variations by gambling harm category. For example, gamblers in all risk

categories consistently reported they would be likely to try self-help strategies (e.g. budgeting),

and/or speaking to a mental health worker or a gambling counsellor if they were to develop a

gambling problem. Only moderate-risk gamblers reported that they would be likely to try self-

exclusion from physical gambling venues (54.3 per cent); and low-risk gamblers reported that they

would talk to family or friends (51.7 per cent) if they were to develop a gambling problem. Table 18

displays the top three services gamblers in different risk categories reported they would be likely to

access if they developed a gambling problem in the future.

Table 18. Types of services gamblers in each risk category report they would be likely to

use if they develop a gambling problem

Non-problem gambler Low-risk gambler Moderate-risk gambler Problem gambler

Strategy 1 Speak to mental health worker

Speak to mental health worker

Use self-help strategies


(69.7%) (61.7%) (82.9%) (66.0%)

Strategy 2 Use self-help strategies




(64.8%) (58.3%) (74.3%) (64.2%)

Strategy 3 Speak to a gambling counsellor

Speak to a gambling counsellor, or family or friends

Self-exclusion from physical gambling venues

Speak to a gambling counsellor

(57.6%) (51.7%) (54.3%) (58.5%)

Readiness to change

Gamblers were also asked to answer two questions to establish how ‘ready’ they were to change

their gambling behaviour (henceforth readiness to change). These questions focused on the

importance gamblers placed on changing their gambling behaviour (How important is it for you to

reduce or stop gambling?), and their level of confidence in making that change (How confident are

you that you could reduce or stop gambling if you decided to?).

Beginning with the importance gamblers placed on reducing or stopping gambling. Figure 55a

displays mean scores on this question for gamblers in each risk category. A significant difference

in mean scores was found, F(3, 344) = 58.32, p<0.001, with increasing importance reported by

patients displaying increasing severity of gambling problems.

This pattern, however, was reversed for mean scores on the item that indicates gamblers’ level of

confidence in their ability to reduce, or stop, gambling if they decide to do so (see Figure 55b). In

this case, a significant difference was also found between gamblers in the different gambling risk

categories, F(3, 344) = 4.40, p<0.01, but this time problem gamblers had the lowest mean score of

any of the risk categories. This suggests that out of all the gamblers, problem gamblers had the

lowest confidence in their ability to reduce or stop gambling, should they decide to do so.



Figure 55. Mean scores on readiness to change items of gamblers in each risk category

Discussion

Summary and interpretation of findings

The first research question focused on establishing whether rates of gambling participation are

higher among patients attending mental health services than in the general population. Overall, the

rate of gambling participation (any activity) for the total sample was 41.4 per cent, which is

significantly lower than the 61.6 per cent reported in the most recent Victorian general population

survey (Hare, 2015).

The aim of the second research question was to identify the type of gambling activities patients

with mental health disorder participated in. The most common gambling activities among patients

were pokies or electronic gaming, closely followed by Lotto, Powerball or the Pools, and then

betting on horse or greyhound racing, and scratch tickets. These activities broadly align with the

most popular gambling activities in the general population (Hare, 2015).

The third research question identified the frequency of gambling in patients attending mental health

services; and the average spend on gambling activities. Most gamblers in the mental health

sample engaged in gambling activities on a monthly basis, and participated in gambling less

frequently over the past year relative to the general population (with the exception of the frequency

with which patients played pokies or electronic gaming among problem gamblers). Among

patients, the majority of gambling took place in physical venues and the average monthly spend

was $119, but the typical (i.e., mode) spend among those who had gambled in the past month was

$50 (note that comparable information for the general population is not available). The data also

suggest that patients who bet on pokies or EGM, and on horse/greyhound racing, but not Lottos,

Powerballs or Pools or scratch tickets, experienced significantly greater gambling-related harm.

This confirms literature indicating that EGMs are most commonly associated with problem

gambling, and are considered to be the most “addictive” form of gambling, with horse/greyhound

racing as one of the next most addictive activities (Dowling, Smith, & Thomas, 2005). In contrast,

lotteries and scratch tickets are generally considered to be “less addictive” as a result of their non-

continuous nature (Dowling et al., 2005).

The fourth and most pertinent research question explored patients’ risk of gambling harm and

prevalence of problem gambling. The findings were that 6.3 per cent of the patients were identified

by the PGSI as being ‘problem gamblers’, 8.3 per cent as ‘moderate-risk’ gamblers, 7.1 per cent as

‘low-risk’ gamblers and 19.6 per cent as ‘non-problem’ gamblers. Despite a lower rate of gambling

participation than the general population, a higher proportion of gamblers in the mental health



sample were in the moderate-risk (8.3 per cent vs. 2.8 per cent) and problem gambling (6.3 per

cent vs. 0.8 per cent) category of the PGSI. The rate of problem gambling was around eight times

higher than the general population, and the rate of moderate-risk gambling was around three times

higher than in the general population. Furthermore, 53 per cent of gamblers in the mental health

sample were experiencing at least some gambling-related harm (i.e., fell in the low-risk, moderate-

risk or problem gambling range on the PGSI). The rate of problem gambling in the current study

was higher than the 0.7 per cent to 4.4 per cent in non-disorder specific, psychiatric outpatient

services reported in the study’s literature review. The observed rate of moderate-risk and problem

gambling are even higher than the rates of problem gambling (5.8 per cent) and moderate-risk

gambling (6.4 per cent) reported among adults with psychotic disorders in outpatient services in

Victoria (Haydock et al., 2015). It is worth noting that while the small number of patients reporting a

current diagnosis of gambling disorder (i.e. less than 1 per cent) may seem paradoxical, it likely

reflects the fact that clinicians are not routinely screening or assessing gambling issues, and hence

patients were unlikely to have received such a formal diagnosis.

The fifth question aimed to establish whether patients with certain mental health disorders were

more likely to be experiencing gambling-related harm. Gamblers in all risk categories (i.e., non-

problem, low-risk, moderate-risk and problem gamblers) were highly likely to have a lifetime or

current mental health disorder, with depression, anxiety and psychotic disorders the most

commonly reported. Furthermore, gamblers currently diagnosed with drug use disorder were 3.6

times more likely to be experiencing some level of gambling harm (i.e., at least low-risk gamblers),

and those experiencing a psychotic disorder were 2.4 times more likely to be experiencing some

kind of gambling harm. This supports the findings reported in the literature review that people with

psychotic disorders are more likely to report problem and/or pathological gambling (Aragay et al.,

2012; Haydock et al., 2015). Gamblers with a current diagnosis of drug use disorder were 3.4

times more likely to be problem gamblers, and patients diagnosed with borderline personality

disorder were 2.6 times more likely to be problem gamblers.

The sixth research question focused on identifying whether problem gamblers reported poorer

wellbeing than non-gamblers or non-problem gamblers. In contrast to expectations, problem

gamblers did not report lower psychological health or poorer overall quality of life than non-problem

gamblers. This contradicts earlier research where problem gamblers reported a poorer quality of

life than non-problem gamblers (Manning et al., 2012; McCormack & Griffiths, 2011). It is possible

however, that the wellbeing of patients with mental health problems is sufficiently low that a

gambling problem is unlikely to reduce wellbeing further. Alternatively it may be that the ATOP

single-item scales were insufficiently sensitive to detect differences between gambler and non-

gamblers, and that a more robust measure of wellbeing such as the WHO-QOL-BREF

(Skevington, Lotfy, & O'Connell, 2004). Moderate-risk gamblers (but not problem gamblers)

reported a significantly lower mean score on the physical health item than all other gambling

category groups. This aligns with others research which suggests individuals experiencing

gambling-related harm report poorer physical health than those who do not experience gambling-

related harm (Subramaniam, Abdin, Vaingankar, Wong, & Chong, 2015).

The seventh research question was whether problem gamblers reported high rates of substance

use. Rates of alcohol, tobacco and illicit drug use among patients were generally higher than those

reported in the general population; however, these rates were consistent with the rates reported in

other studies involving mental health populations (see for example Croton, 2007; Zimmermann,

Lubman, & Cox, 2012). For alcohol, 40 per cent of problem gamblers in the current study screened

positive for problem drinking, but it was the low-risk gamblers that had the highest rates of problem

drinking (around 55 per cent). For tobacco use, problem gamblers had the highest rates of

smoking. Overall, illicit drug use was reported by one-quarter of the sample, and both moderate-



risk and problem gamblers reported the highest rates of illicit drug use in the past year. These

findings are consistent with the high rates of comorbid substance use (i.e., alcohol, tobacco and

drug use disorders) previously reported among problem gamblers in both community (Lorains et

al., 2011) and treatment settings (Dowling, Cowlishaw, et al., 2015).

The eighth research question was to determine whether there were demographic differences in

gambling participation and harm across gambling risk categories. As expected, both gambling

participation and moderate-risk and problem gambling were more common among male than

female participants. This supports earlier research reporting higher rates of male problem

gamblers in the community (Bonnaire et al., 2016; Castrén et al., 2013) and higher rates of male

moderate-risk and problem gamblers seeking treatment in mental health services (Haydock et al.,

2015). Participants in the 65+ age category had the highest rates of gambling participation of all

age categories, which could reflect increased leisure opportunities among retirees. Increasing

gambling rates among older adults has been noted in the US, Canada, and Australia (Tse, Hong,

Wang, & Cunningham-Williams, 2012). Having more free time and disposable incomes, facing life

transitions such as retirement, reduced opportunities to socialise, spousal loss, and chronic illness

may increase gambling participation (McNeilly & Burke, 2002).There may also be features of

gambling venues that attract this population, such as discounted hot meals, chartered

transportation, and attractions that cater to various ethnic/cultural minority groups (Dyall, Tse, &

Kingi, 2009). Additionally, this population could have an elevated risk of experiencing gambling-

harm due to age-related issues such as loneliness, isolation from society, depression, fixed

incomes, reduced cognitive capacity and vulnerability to retirement-related inactivity, with a recent

Victorian study finding that loneliness was a predictor of problem gambling for both men and

women (Botterill, Gill, McLaren, & Gomez, 2016). In the current study however, this age group had

the lowest levels of gambling-related harm and none were identified as a problem gambler.

There were no significant differences in gambling category by marital status or by education level,

identifying as belonging to a minority group or as an Aboriginal or Torres Strait Islander, or

speaking a non-English language as a main language. Unemployment rates however, rose

significantly as gambling harm increased, with unemployment associated with gambling problems

in previous research (Bonnaire et al., 2016; Hing, Russell, Tolchard, & Nower, 2016; Sproston,

Hing, & Palankay, 2012). With regards to differences across service types, higher rates of

moderate-risk and problem gambling were observed in mental health community support service

versus public mental health settings, in public versus private settings, and in adult versus youth

services. It is possible that the lower rates observed in youth services reflected reduced access to

money, while those accessing private settings had more stable social and financial support, and/or

experienced less social disadvantage than those accessing public mental health services.

Similarly, the community support service typically provides support to patients with enduring or

disabling mental illness, and who are more likely to be socially disadvantaged, which could explain

the elevated rates of gambling harm among patients attending this service.

The ninth research question concerned whether patients had been asked about their gambling by

a clinician at the mental health service they were attending. While only 43 per cent of all patients

reported being asked about their gambling, the proportion who had been asked increased as

gambling harm category increased.

The final research question explored the types of help patients would be likely to engage with if

they developed a gambling problem. Although gamblers reported that they would try a wide range

of strategies, several strategies were endorsed more frequently in every gambling harm category.

These included; speak to a mental health worker, speak to a gambling counsellor, and use self-



help strategies. A higher proportion of moderate-risk and problem gamblers reporting that they

would use self-help strategies.

Study limitations

Although a large number of patients were surveyed (N = 841), this represented only 55 per cent of

the total available sample. This participation rate with a convenience sample was not unexpected

given the complex and vulnerable nature of the mental health sample being surveyed. It is possible

that patients under-reported their gambling behaviour because of demand characteristics (i.e.

being reluctant to report gambling behaviour given that they were receiving treatment for a mental

health issue and many (40 per cent) were receiving disability support pensions as their main form

of income). It is also possible that participants under-reported their gambling behaviours, spend

and activity due social desirability effects.

The prevalence estimate is likely to be conservative as patients who were acutely unwell, unstable

or with multiple morbidities did not participate, and previous research has found such patients

experience greater gambling-related harm (Haydock et al., 2015). Indeed, clinical staff encouraged

researchers to avoid inviting the most chronic, complex or disabled patients to participate in the

survey, because of behavioural risk issues. In fact, our own findings provide further evidence of

higher rates of moderate-risk gambling and problem gambling among those with more severe

(multiple) mental health disorders and those with psychotic disorders. Prevalence estimates for

gambling risk may therefore have been elevated further, had clients with more complex needs

been better represented in the study. The clinic-based method for recruitment of patients

accessing care via public adult community mental health services may also have not included the

more complex cohort of patients who do not access clinic-based care. Due to poor engagement

with the service, risks of behaviours of concern, or transience, a number of patients only receive

outreach care, so would not have had the opportunity to participate. The prevalence of gambling

participation and gambling harm could also be underestimated because the PGSI was only

completed by patients who had gambled in the past 12 months. As such it is possible that patients

classified as non-gamblers in the present study might have gambled and/or experienced gambling-

related difficulties prior to the year assessed in the survey, but this would not have been captured

due to the 12 month time-frame utilised in the survey for gambling participation and the PGSI.

A further consideration related to the existing gambling focus within the included mental health

services. Only one of the 12 sites received funding for gambling-specific work. They employed a

0.2EFT psychiatrist and a 0.8EFT allied health professional, although their primary focus was on

providing assessment, brief intervention and consultation for patients and staff from Victorian

Gambler’s Help services, much of which occurred offsite. This site (from which 68 participants

were recruited) had the fifth highest gambling participation rate and second highest rate of problem

gambling (11.8 per cent), though there were two other services with similar problem gambling

rates. As such, it is unlikely that the inclusion of this service as a recruitment site would have

inflated the gambling rates significantly.

Another limitation was that the survey was entirely reliant on patient self-report, including estimates

of gambling behaviours, which could potentially be affected by recall bias and attribution. Concerns

with the accuracy of self-report expenditure and other gambling behaviours have been noted in the

literature (R. Volberg, Gerstein, Christiansen, & Baldridge, 2001) . In terms of being able to

generalise, a strength of the study was its inclusion of a broad mental health sample across

multiple treatment settings, although recruitment was focused on community/out-patient settings

where patients are generally more stable and hence the estimate of gambling-related harm may be

lower than if patients of residential services had been included. Although a variety of services



where sampled across Victoria, the sample was largely homogenous in terms of its demographic

characteristics, with the majority (77 per cent) born in Australia, with English as their first language

(92 per cent), and with only 12 per cent identifying as belonging to an ethnic minority group or as

Aboriginal and/or Torres Strait Islander. While this demographic reflected the population attending

those services, overall rates of participation and gambling harm may have differed had a more

diverse population been specifically targeted for recruitment. Finally, the findings highlight

increased rates of moderate-risk and problem gambling relative to previous Victorian general

population data, however it is important to note that these differences may reflect the demographic

profile of many participants, with certain characteristics particularly associated with gambling

behaviours (e.g. a larger proportion of single and unemployed participants in the current study).

Differences in the prevalence of gambling harm may also reflect the different methodologies and

time-points in which the two studies were undertaken, and any direct comparisons should be

restricted to demographically matched samples.



Study 3b: Psychometric testing of gambling

screens

Aims

The aim of this study is to compare the classification accuracy of multiple brief problem gambling

screening instruments for use within mental health services.

Method

Measures

The classification accuracy of 10 brief screening instruments (two to five items) and the 23 single

items that make up these screening instruments (Table 20) were compared to the Problem

Gambling Severity Index (PGSI) as a reference standard. The 10 brief screening instruments

included:

Lie/Bet Questionnaire

Brief Problem Gambling Screen (BPGS) [two-, three-, four- and five-item versions]

PGSI Short Form, National Opinion Research Center Diagnostic Screen – Loss of Control,

Lying and Preoccupation Screen (NODS-CLiP)

NODS-CLiP2

Brief Biosocial Gambling Screen (BBGS)

National Opinion Research Center Diagnostic Screen – Preoccupation, Escape, Chasing

and Risked Relationships screen (NODS-PERC).

The PGSI and brief screening instruments were administered only to participants who reported

participation on any gambling activity in the previous 12 months, with non-endorsement responses

imputed for non-gamblers. The items of these screening instruments and their psychometric

properties are described in detail in Study 1. Each brief screening instrument and their respective

single items was converted to a 12-month timeframe. With the exception of the PGSI items, all

items were dichotomously scored: (0) No, (1) Yes. The original item response format of the PGSI

was employed [(0) Never, (1) Sometimes, (2) Most of the time, (3) Almost always. In the

classification accuracy calculations of the single PGSI items, these response formats were

recoded into dichotomous responses [(0) No, (1) Yes], whereby any endorsement (i.e.,

Sometimes, Most of the time, and Almost always) was scored as a positive endorsement.

Classification accuracy coefficients

Multiple classification accuracy coefficients were computed, including:

sensitivity (the true positive rate, i.e., the proportion of positive test results among those

with the disorder). Sensitivity is the measure used to report how effective a brief screening

instrument is in identifying people with gambling problems. Higher sensitivities indicate



more effectiveness in identifying people with gambling problems. For example, a screening

instrument with a sensitivity of 0.800 against the problem gambling cut-off on the nine-item

PGSI (PGSI scores ≥ 8) is interpreted to mean that the instrument correctly detected 80

per cent of patients classified as problem gamblers on the PGSI.

specificity (the true negative rate, i.e., the proportion of negative test results among those

without the disorder). Specificity is the measure used to report how effective a test is in

identifying people without gambling problems. Higher specificities indicate more

effectiveness in identifying people without gambling problems. For example, a screening

instrument with a specificity of 0.800 against the problem gambling cut-off on the nine-item

PGSI (PGSI scores ≥ 8) is interpreted to mean that the instrument correctly detected 80

per cent of patients NOT classified as problem gamblers on the PGSI.

overall diagnostic accuracy (the total number of true positives and true negatives divided

by the total sample size). Diagnostic accuracy combines sensitivity and specificity data to

indicate how well a brief screening instrument correctly identifies or rules out gambling

problems. For example, a screening instrument with an overall diagnostic accuracy of

0.800 against the problem gambling cut-off on the nine-item PGSI (PGSI scores ≥ 8) is

interpreted to mean that the instrument correctly discriminated between problem gambling

and non-problem gambling for 80 per cent of patients.

Table 19 provides a simple description of the way in which these indices of classification accuracy

are calculated (Baratloo, Hosseini, Negida, & El Ashal, 2015; Parikh, Mathai, Parikh, Sekhar, &

Thomas, 2008) . It groups individuals into one of four categories. The two columns divide people

into groups according to whether they do or do not have a gambling problem. The two rows divide

people according to whether they have a positive or negative test result using a brief screening

instrument.

Table 19 Description of how indices of classification accuracy are calculated.

Screening test results Actually have a gambling problem according to the PGSI?

Yes No

Positive True positive (TP) False positive (FP)

Negative False negative (FN) True negative (TN)

Where:

TP = the number of people with gambling problems and with a positive screening test

FP = the number of people without gambling problems and with a positive screening test

FN = the number of people with gambling problems and a negative screening test

TN = the number of people without gambling problems and with a negative screening test

Using this information:

Sensitivity = [TP/(TP+FN)]

Specificity = [TN/(TN+FP)]



Overall diagnostic accuracy = [(TP+TN)/(TP+TN+FP+FN)]

These indices are generally multiplied by 100 when they are interpreted as they are expressed as

a proportion of the population under study.

Reference standard

The classification accuracy analyses employed the PGSI as the reference standard against which

the single items and brief screening instruments were compared. Although there is no “gold

standard” laboratory or biological test for diagnosing gambling disorder (American Psychiatric

Association, 2013), the widespread use across population and clinical samples and the established

psychometric properties of the PGSI makes it a reasonable reference standard to serve as a proxy

for a “gold standard”.

Because mental health services may have different needs in relation to the severity of gambling

problem they want to identify, multiple scores on the PGSI were employed as the reference

standard against which the brief screening instruments were compared:

the problem gambling cut-off score (PGSI scores ≥ 8);

the original moderate-risk gambling cut-off score (PGSI scores ≥ 3); and

the low-risk gambling cut-off score (PGSI scores ≥ 1).

The first criterion for the selection of an appropriate brief screening instrument in mental health

service populations was satisfactory classification accuracy, as indicated by sensitivity, specificity,

and diagnostic accuracies equal to or greater than 0.80 (DiStefano & Morgan, 2011; Glascoe,

2005). As recommended by R. A. Volberg et al. (2011), the second criterion for the selection of an

appropriate brief screening instrument in these settings was high sensitivity as the aim in these

settings is to capture problem gamblers, even at the expense of including relatively large numbers

of at-risk or non-problem gamblers.



Table 20. Single items employed as one-item screening instrumentsa

Screen and item number Question

PGSI item 1 b Have you bet more than you could really afford to lose?

PGSI item 2 b Have you needed to gamble with larger amounts of money to get the same feeling of excitement?

PGSI item 3 b When you gambled, did you go back another day to try to win back the money you lost?

PGSI item 4 b Have you borrowed money or sold anything to get money to gamble?

PGSI item 5 b Have you felt that you might have a problem with gambling?

PGSI item 6 b Has gambling caused you any health problems, including stress or anxiety?

PGSI item 7 b Have people criticized your betting or told you that you had a gambling problem, regardless of whether or not you thought it was true?

PGSI item 8 b Has your gambling caused any financial problems for you or your household?

PGSI item 9 b Have you felt guilty about the way you gamble or what happens when you gamble?

NODS-CLiP item 1 (also NODS-PERC item 1 and NODS-CLiP2 item 2 c,d,g

Have there been periods lasting two weeks or longer when you spent a lot of time thinking about your gambling experiences or planning out future gambling ventures or bets?

NODS-CLiP item 2/NODS-CLiP2 item 1 c,g Have you tried to stop, cut down, or control your gambling?

NODS-CLiP item 3 (also NODS-CLiP2 item 3) c,g Have you lied to family members, friends, or others about how much you gamble or how much money you lost on gambling?

NODS-PERC item 2 (also NODS-CLiP2 item 5) c,d,g Have you gambled as a way to escape from personal problems?

NODS-PERC item 3 (also NODS-CLiP2 item 4) c,d,g Has there been a period when, if you lost money gambling one day, you would return another day to get even?

NODS-PERC item 4 d,g Has your gambling caused serious or repeated problems in your relationships with any of your family members or friends?

Lie/Bet item 1 g Have you felt the need to bet more and more money?

Lie/Bet item 2 g Have you had to lie to people important to you about how much you gambled?

BBGS item 1 e,g Have you become restless, irritable, or anxious when trying to stop and (or) cut down on gambling?

BBGS item 2 e,g Have you tried to keep your family or friends from knowing how much you gambled?



Screen and item number Question

BBGS item 3 e,g Did you have such financial trouble as a result of gambling that you can to get help with living expenses from family, friends, or welfare?

BPGS (PPGM10) item f,g Would you say you have been preoccupied with gambling?

BPGS (SOGS 4) item f,g Have you often gambled longer, with more money or more frequently than you intended to?

BPGS (PPGM8C) item f,g Have you made attempts to either cut down, control or stop gambling?

a All items scored over the previous 12 months

b PGSI: Problem Gambling Severity Index. Original PGSI item response format was recoded into dichotomous responses [(0) No, (1) Yes], whereby any endorsement (i.e., Sometimes, Most of the time, and Almost always) was scored as a positive endorsement.

c NODS-CLiP/NODS-CLiP2: National Opinion Research Center Diagnostic Screen for Gambling Disorders – Loss of Control, Lying and Preoccupation screen; NODS-CLiP2 adds Chasing and Escape to the three NODS-CLiP items.

d NODS-PERC: National Opinion Research Center Diagnostic Screen for Gambling Disorders – Preoccupation, Escape, Chasing and Risked Relationships screen

e BBGS: Brief Biosocial Gambling Screen

f BPGS: Brief Problem Gambling Screen (PPGM: Problem and Pathological Gambling Measure; SOGS: South Oaks Gambling Screen)

g Scored dichotomously: (0) No, (1) Yes



Results

The classification accuracy coefficients of each of the single items and brief screening instruments

using the PGSI score of 8 or higher (problem gambling cut-off score) are displayed in Table 21.

Several single items (PGSI item 1, PGSI item 3, PGSI item 5, PGSI item 6, PGSI item 8, PGSI

item 9, NODS-CLiP item 2, NODS-PERC item 3, BBGS item 2, and BPGS [PPGM item 8C] item)

and all of the brief screening instruments displayed adequate sensitivity, specificity, and overall

diagnostic accuracy compared to the problem gambling cut-off score on the PGSI. The PGSI Short

Form displayed the highest overall diagnostic accuracy (0.976), but the three-item, four-item and

five-item BPGS displayed the highest sensitivities (1.000). Of the single items, the PGSI Item 8

displayed the highest overall diagnostic accuracy (0.964), but the PGSI item 1 displayed the

highest sensitivity (0.908). The two-item brief screening instruments displayed comparable overall

diagnostic accuracy (0.935 for the Lie/Bet Questionnaire, 0.933 for the two-item BPGS), although

the two-item BPGS displayed slightly higher sensitivity (0.868) than the Lie-Bet Questionnaire

(0.811). Of the three-item instruments, the PGSI Short Form displayed the highest overall

diagnostic accuracy (0.976), although the three-item BPGS displayed the highest sensitivity

(1.000). Of the four-item instruments, the NODS-PERC (0.897) displayed a slightly higher overall

diagnostic accuracy than the four-item BPGS (0.875), but the four-item BPGS displayed slightly

higher sensitivity (1.000) than the NODS-PERC (0.981). Finally, both five-item instruments

displayed comparable diagnostic accuracies (0.864 for the NODS-CLiP2, 0.874 for the five-item

BPGS), but the five-item BPGS (1.000) displayed a slightly higher sensitivity than the NODS-CLiP2

(0.981).


using the PGSI score of 3 or higher (moderate-risk gambling cut-off score) are displayed in Table

22. Two single items (PGSI item 1 and BPGS [PPGM item 8C] item) and several brief screening

instruments (NODS-CLiP, three-item BPGS, NODS-PERC, four-item BPGS, NODS-CLiP2, and

five-item BPGS) displayed adequate sensitivity, specificity, and overall diagnostic accuracy

compared to the original moderate-risk gambling cut-off score on the PGSI. The PGSI item 1

(0.949), NODS-CLiP (0.942), three-item BBGS (0.942), and five-item BPGS (0.941) displayed the

highest diagnostic accuracies, but the five-item BPGS displayed the highest sensitivity (0.943). Of

the single items, the PGSI item 1 displayed the highest overall diagnostic accuracy (0.949) and

sensitivity (0.837). Neither of the two-item instruments displayed an adequate sensitivity. Of the

three-item instruments, the NODS-CLiP and three-item BPGS displayed comparable diagnostic

accuracies (0.942), but the three-item BPGS (0.927) displayed a slightly higher sensitivity than the

NODS-CLiP (0.902). Of the four-item instruments, the NODS-PERC and the four-item BPGS

displayed comparable diagnostic accuracies (0.939), but the four-item BPGS (0.935) displayed a

higher sensitivity than the NODS-PERC (0.854). Finally, of the five-item instruments, five-item

BPGS (0.941) displayed a slightly higher overall diagnostic accuracy than the NODS-CLiP2

(0.926); and also displayed a slightly higher sensitivity (0.943) than the NODS-CLiP2 (0.919).


using the PGSI score of 1 or higher (low-risk gambling cut-off score) are displayed in Table 23.

Only the five-item BPGS displayed adequate sensitivity, specificity, and overall diagnostic

accuracy compared to the low-risk gambling cut-off score on the PGSI.



Table 21. Classification accuracy of the single items and brief screening instruments using

PGSI score of 8+ as reference standarda,b,c

Screening measure Sensitivity Specificity Overall diagnostic accuracy

Single items

PGSI item 1 d 0.962 0.905 0.908

PGSI item 2 d 0.792 0.957 0.946

PGSI item 3 d 0.906 0.947 0.944

PGSI item 4 d 0.623 0.980 0.957

PGSI item 5 d 0.906 0.953 0.950

PGSI item 6 d 0.925 0.945 0.944

PGSI item 7 d 0.642 0.970 0.949

PGSI item 8 d 0.849 0.972 0.964

PGSI item 9 d 0.925 0.909 0.910

NODS-CLiP item 1e,i 0.604 0.962 0.939

NODS-CLiP item 2e,i 0.868 0.895 0.893

NODS-CLiP item 3e,i 0.673 0.968 0.950

NODS-PERC item 2f,i 0.792 0.918 0.910

NODS-PERC item 3f,i 0.868 0.954 0.949

NODS-PERC item 4f,i 0.642 0.978 0.957

Lie/Bet item 1 i 0.660 0.954 0.936

Lie/Bet item 2 i 0.660 0.971 0.951

BBGS item 1 g,i 0.698 0.964 0.948

BBGS item 2 g,i 0.830 0.956 0.948

BBGS item 3 g,i 0.623 0.981 0.958

BPGS (PPGM10) item h,i

0.585 0.973 0.949

BPGS (SOGS 4) item h,i

0.755 0.942 0.930

BPGS (PPGM8C) item h,i

0.925 0.895 0.897

Two-item screening instruments

Lie/Bet i 0.811 0.943 0.935

BPGS (two-item) h,i 0.868 0.938 0.933

Three-item screening instruments

PGSI Short form d 0.906 0.981 0.976

NODS-CLiP e,i 0.981 0.878 0.885

BBGS g,i 0.962 0.935 0.937

BPGS (three-item) h,i 1.000 0.872 0.880

Four-item screening instruments

NODS-PERC f,i 0.981 0.891 0.897




BPGS (four-item) h,i 1.000 0.867 0.875

Five-item screening instruments

NODS-CLiP2 e,i 0.981 0.857 0.864

BPGS (five-item) h,i 1.000 0.865 0.874

a Total N = 841 (n = 53 PGSI 8+; n = 788 PGSI <8); base rate = 6.3%

b Sensitivity (the true positive rate, i.e., the proportion of positive test results among those with the disorder), specificity (the true negative rate, i.e., the proportion of negative test results among those without the disorder), and overall diagnostic accuracy coefficients (the total number of true positives and true negatives divided by the total sample size) ≥ 0.80 are bolded

c All items scored over the previous 12 months

d PGSI: Problem Gambling Severity Index. Original scoring retained for PGSI Short form. For single items, original PGSI item response format was recoded into dichotomous responses [(0) No, (1) Yes], whereby any endorsement (i.e., Sometimes, Most of the time, and Almost always) was scored as a positive endorsement.

e NODS-CLiP/NODS-CLiP2: National Opinion Research Center Diagnostic Screen for Gambling Disorders – Loss of Control, Lying and Preoccupation screen; NODS-CLiP2 adds Chasing and Escape to the three NODS-CLiP items.

f NODS-PERC: National Opinion Research Center Diagnostic Screen for Gambling Disorders – Preoccupation, Escape, Chasing and Risked Relationships screen

g BBGS: Brief Biosocial Gambling Screen

h BPGS: Brief Problem Gambling Screen (PPGM: Problem and Pathological Gambling Measure; SOGS: South Oaks Gambling Screen)

i Scored dichotomously: (0) No, (1) Yes


PGSI score of 3+ as reference standard a,b,c c


Single items

PGSI item 1 d 0.837 0.968 0.949

PGSI item 2 d 0.577 0.993 0.932

PGSI item 3 d 0.691 0.993 0.949

PGSI item 4 d 0.374 0.996 0.905

PGSI item 5 d 0.675 0.997 0.950

PGSI item 6 d 0.675 0.987 0.942

PGSI item 7 d 0.407 0.989 0.904

PGSI item 8 d 0.537 0.999 0.931

PGSI item 9 d 0.797 0.968 0.943

NODS-CLiP item 1 e,i 0.431 0.987 0.906

NODS-CLiP item 2 e,i 0.789 0.955 0.931

NODS-CLiP item 3 e,i 0.447 0.992 0.912

NODS-PERC item 2 f,i 0.675 0.967 0.924

NODS-PERC item 3 f,i 0.610 0.990 0.935

NODS-PERC item 4 f,i 0.390 0.996 0.907

Lie/Bet item 1 i 0.488 0.985 0.912




Lie/Bet item 2 i 0.415 0.990 0.906

BBGS item 1 g,i 0.496 0.994 0.922

BBGS item 2 g,i 0.561 0.986 0.924

BBGS item 3 g,i 0.366 0.996 0.904


0.374 0.992 0.901


0.626 0.987 0.935


0.821 0.957 0.937


Lie/Bet i 0.610 0.982 0.927

BPGS (two-item) h,i 0.675 0.983 0.938



NODS-CLiP e,i 0.902 0.948 0.942

BBGS g,i 0.740 0.985 0.949



NODS-PERC f,i 0.854 0.954 0.939

BPGS (four-item) h,i 0.935 0.940 0.939


NODS-CLiP2 e,i 0.919 0.928 0.926

BPGS (five-item) h,i 0.943 0.940 0.941













PGSI score of 1+ as reference standarda,b,c


Single items

PGSI item 1 d 0.689 1.000 0.932

PGSI item 2 d 0.415 1.000 0.873

PGSI item 3 d 0.492 1.000 0.889

PGSI item 4 d 0.268 1.000 0.841

PGSI item 5 d 0.464 1.000 0.883

PGSI item 6 d 0.503 1.000 0.892

PGSI item 7 d 0.317 1.000 0.851

PGSI item 8 d 0.366 1.000 0.862

PGSI item 9 d 0.661 1.000 0.926

NODS-CLiP item 1 e,i 0.328 0.997 0.851

NODS-CLiP item 2 e,i 0.639 0.982 0.907

NODS-CLiP item 3 e,i 0.328 0.998 0.853

NODS-PERC item 2 f,i 0.519 0.982 0.881

NODS-PERC item 3 f,i 0.437 0.997 0.875

NODS-PERC item 4 f,i 0.268 0.997 0.838

Lie/Bet item 1 i 0.355 0.991 0.853

Lie/Bet item 2 i 0.306 0.997 0.847

BBGS item 1 g,i 0.344 0.997 0.855

BBGS item 2 g,i 0.415 0.995 0.869

BBGS item 3 g,i 0.251 0.997 0.835


0.273 0.997 0.839


0.448 0.994 0.875


0.667 0.985 0.916


Lie/Bet i 0.448 0.991 0.873

BPGS (two-item) h,i 0.675 0.999 0.951



NODS-CLiP e,i 0.738 0.980 0.927

BBGS g,i 0.541 0.995 0.897



NODS-PERC f,i 0.689 0.982 0.918

BPGS (four-item) h,i 0.798 0.982 0.942





NODS-CLiP2 e,i 0.787 0.968 0.929

BPGS (five-item) h,i 0.803 0.982 0.943










Discussion

The five-item BPGS was the only measure to adequately detect low-risk gambling in these

services; this brief screening instrument detected 80.3 per cent of patients classified as low-risk,

moderate-risk, or problem gamblers on the PGSI. Moreover, this brief screening instrument

displayed the highest sensitivity in identifying moderate-risk gambling; this instrument detected

94.3 per cent of patients classified as moderate-risk or problem gamblers on the PGSI. Finally, this

screening instrument also detected 100 per cent of patients classified as problem gamblers on the

PGSI.

The five-item BPGS is recommended for services wanting to detect low-risk,

moderate-risk and problem gambling in their patients.

The five-item BPGS is comprised of the following items:

1. In the past 12 months, would you say you have been preoccupied with gambling?

2. In the past 12 months, have you needed to gamble with larger amounts of money to

get the same feeling of excitement?

3. In the past 12 months, have you often gambled longer, with more money or more

frequently than you intended to?

4. In the past 12 months, have you made attempts to either cut down, control or stop

gambling?



5. In the past 12 months, have you borrowed money or sold anything to get money to

gamble?

Each item employs dichotomous scoring: No = ‘0’, Yes = ‘1’. Scores range from 0 to 5, with

scores of 1 or more indicating problem gambling.

Several single items (PGSI item 1, PGSI item 3, PGSI item 5, PGSI item 6, PGSI item 8, PGSI

item 9, NODS-CLiP item 2, NODS-PERC item 3, BBGS item 2, and BPGS [PPGM item 8C] item)

and all of the brief screening instruments displayed acceptable classification accuracy in detecting

problem gamblers within mental health services. The different versions of the BPGS, however,

displayed the highest sensitivities, as they detected 100 per cent of patients classified as problem

gamblers on the PGSI. These screens also displayed comparable specificities and overall

diagnostic accuracies when using the problem gambling cut-off as the reference standard. The

longer versions of the BPGS therefore do not seem to provide any advantage over the three-item

version when the goal is to detect only problem gamblers in a service.

Two single items (PGSI item 1 and BPGS [PPGM item 8C] item) and several brief screening

instruments (NODS-CLiP, three-item BPGS, NODS-PERC, four-item BPGS, NODS-CLiP2, and

five-item BPGS) adequately detected moderate-risk gambling in these services. The three-item,

four-item, and five-item BPGS displayed superior sensitivities; they detected 92.7 per cent, 93.5

per cent, and 94.3 per cent of patients classified as moderate-risk or problem gamblers on the

PGSI, respectively. These three screening instruments also displayed relatively comparable

specificities and overall diagnostic accuracies when using the moderate-risk gambling cut-offs as

the reference standard. The longer versions of the BPGS therefore do not seem to provide much

advantage over the three-item version when the goal is to detect moderate-risk and problem

gamblers in a service.

The three-item BPGS is recommended for services wanting to detect moderate-risk

or problem gambling (but not low-risk gambling) in their patients.

The three-item BPGS is comprised of the following items:

1. In the past 12 months, have you needed to gamble with larger amounts of money to

get the same feeling of excitement?

2. In the past 12 months, have you often gambled longer, with more money or more


3. In the past 12 months, have you made attempts to either cut down, control or stop

gambling?

Each item employs dichotomous scoring: No = ‘0’, Yes = ‘1’. Scores range from 0 to 3, with

scores of 1 or more indicating problem gambling.

The selection of a brief screening instrument in mental health services, however, may be

determined by the need for a very brief screening instrument (Himelhoch et al., 2015). Two single

items, but no two-item screening instruments, displayed acceptable classification accuracy in

detecting both problem gambling and moderate-risk gambling. These were the PGSI item 1 (Have

you bet more than you could really afford to lose?) and the BPGS (PPGM item 8C) item (Have you

made attempts to either cut down, control or stop gambling?). Of these two items, however, the

PGSI item displayed superior overall diagnostic accuracy and sensitivity; this instrument accurately



detected 96.2 per cent of patients classified as problem gamblers on the PGSI and 83.7 per cent of

patients classified as moderate-risk or problem gamblers on the PGSI. While the first item of the

PGSI is a more efficient screening instrument in detecting both moderate-risk and problem

gamblers than the three-item BPGS, its derivation from the reference standard and inferior

sensitivity suggests that the three-item BPGS should be preferentially used if the service can

accommodate a slightly longer measure.

The first item of the PGSI is recommended for services wanting to identify

moderate-risk and problem gambling (but not low-risk gambling) in their patients

using a very brief (one–two item) screening instrument.

The first item of the PGSI is:

Thinking about the last 12 months, have you bet more than you could really afford to lose?

The item employs dichotomous scoring: No = ‘0’, Yes = ‘1’. Positive endorsement of the

item is indicative of problem gambling.

Study limitations

The findings of this study must be interpreted in terms of several limitations. First, the classification

accuracy was assessed against the nine-item PGSI which is the longer measure from which

several of the brief screening instruments are derived. This overlap between the reference

standard and the brief screening instruments may have inflated the classification accuracy

coefficients (Stinchfield & McCready, 2014). Indeed, the best-performing brief screening

instruments, such as the five-item BPGS three-item BPGS, as well as the first item of the PGSI,

each comprise items from the nine-item PGSI. This limitation cannot, however, be completely

eliminated as almost all of the available brief screening instruments were derived from DSM-IV

criteria or from existing instruments based on DSM-IV diagnostic criteria. Second, the PGSI and

brief screening instruments were only administered to individuals who reported participation in

gambling activities in the previous 12 months. Participants who had not gambled in the previous 12

months were then imputed with a score of zero on both the PGSI and the brief screening

instruments so that the classification accuracy coefficients were based on the entire sample of

patients attending mental health services.

This decision was made because the PGSI was designed to be administered to the adult

population who had gambled in the previous 12 months (Ferris & Wynne, 2001) and imputing

zeros for non-gamblers is standard practice in prevalence estimations of problem gambling (ACIL

Allen Consulting, The Social Research Centre, & The Problem Gambling Research and Treatment

Centre, 2014; Davidson & Rodgers, 2010; Davidson, Rodgers, Taylor-Rodgers, Suomi, & Lucas,

2015; Dowling, Youssef, et al., 2016; Hare, 2015; Queensland Government, 2012; Social

Research Centre, 2013; Sproston et al., 2012). Moreover, there is some evidence to suggest that

only a very small proportion of the population (0.4 per cent) positively endorse an item on the PGSI

despite not having gambled in the previous 12 months (The Allen Consulting Group, Problem

Gambling Research and Treatment Centre, & the Social Research Centre, 2011). Regardless, the

inclusion of a significant proportion of participants who were artificially administered a negative

endorsement on both the brief screening instruments and the reference standard may have

inflated the classification accuracy coefficients presented in this study. Third, the timeframe for all

brief screening instruments that were developed as lifetime measures (e.g. the Lie/Bet, NODS-

CLiP, and NODS-PERC) were converted to 12 month timeframes. While this limits the

generalisability to other results relating to these screening instruments, this conversion was



deemed important so that there was consistency in the timeframe across the screening

instruments evaluated in the study and because instruments screening for problem gambling

across the previous 12 months are more likely to adequately discriminate between individuals

experiencing current problem gambling and those in remission or recovery (Problem Gambling

Research and Treatment Centre, 2011; Stinchfield et al., 2007; Stinchfield & McCready, 2014).

Finally, the data are based on participant self-report, whereby the validity of the data may not be

accurate if participants wished to conceal the severity of their gambling. This limitation was,

however, addressed by reassuring participants about the confidentiality of their responses and that

their responses would not affect their treatment in the mental health service.



Discussion and recommendations

In this three-year Victorian multi-component study, the overarching aims were to: i) identify current

clinician responses to problem gambling in mental health services; ii) identify the prevalence of

problem gambling, its relationship with comorbid psychiatric disorders, and iii) explore suitable

problem gambling screening instruments. These three key research questions were addressed by

adopting a multi-method approach across five discrete research activities.

The first activity entailed an extensive review of the national and international literature on the

prevalence of co-occurring gambling and mental health problems; screening, assessment, and

management of problem gambling, psychological and pharmacological treatment interventions; and

service models and effective partnerships. Key findings were that the rates of help-seeking for

gambling problems in the Victorian population are currently low, and since gambling problems

commonly co-occur with mental health problems that require treatment, mental health professionals

have a critical role to play in the identification and early management of gambling problems.

Moreover, since problem gambling may compromise treatment engagement and outcomes, timely

identification and early intervention is paramount. The findings of the literature review highlighted the

importance of identifying problem gambling through routine screening in mental health settings, with

a view to generalist first-level gambling interventions conducted within the service or appropriate

referral to specialist gambling services for management of the gambling problem. However, there is

currently no validated brief screening tools that are recommended for use in mental health

populations. A further gap identified in the literature review is the paucity of evidence to guide

psychological and pharmacological treatment approaches for different subpopulations of problem

gamblers based on psychiatric comorbidity. Similarly, there is limited empirical knowledge around

models of care and tailored treatment approaches for this population, however, universal screening,

risk assessment and diagnosis, supportive therapies, prevention, psychoeducation and cross-

sectoral collaboration are desirable features of an effective service response.

The second activity aimed to assess the problem-gambling-related knowledge, attitudes and

practices of mental health clinicians, with a focus on role legitimacy, screening, assessment, referral

and treatment of patients with problem gambling within mental health settings. From a broad range

of mental health services (youth, adult, public, private etc.) and sites (metropolitan, regional etc.),

311 clinicians completed an anonymous survey assessing these domains. Clinicians estimated that

one in ten patients are affected by gambling problems. Encouragingly, most clinicians reported that

they were knowledgeable about problem gambling, and an overwhelming majority considered

screening, assessment and referral to be a core part of their role. It is a concern however, that only 4

per cent routinely (often or always) ask their patients about their gambling and that only 2 per cent

use a standardised/formal screening tool to detect problem gambling. It is a concern that a

significant proportion (40 per cent) rarely or never ask about problem gambling and that 55 per cent

rarely or never screen for problem gambling. Clinicians’ confidence in their ability to detect problem

gambling was poor as was managing gambling problems, with less than one-quarter feeling

knowledgeable about effective treatments and only one-third reporting that they could actively treat

the gambling problem. While concerning, these results are unsurprising given that only a minority

(12 per cent) had ever received any form of training in problem gambling. Although two-thirds of

clinicians reported that they refer patients with gambling problems to external treatment providers,

few reported actually doing so, and the data suggest that this may reflect concerns that specialist

services are not equipped to deal with mental health issues or a lack of understanding of what

gamblers help services offer. Nonetheless greater levels of knowledge, confidence and responding

to this population observed among clinicians with prior training provide a compelling justification for

greater investment in workforce training.



The third activity was an exploration of the barriers and facilitators to responding to problem

gambling, through 30 in-depth qualitative interviews with clinicians drawn from mental health

services. The findings revealed marked disparity in how clinicians currently ask about and respond

to gambling and suggest there is a need to address barriers to screening. This included a lack of

knowledge and skills, concerns about validity of disclosure and potential impact on rapport, a

perception that problem gambling has a low burden of disease, an absence of knowledge of

effective interventions, and limited access to specialist services. Interestingly, many of the same

barriers were identified in relation to screening for alcohol and drug issues in mental health services

over a decade ago, providing opportunities to examine which initiatives were most effective in

addressing these barriers in the current service system and their relevance for responding to

gambling. Facilitators to screening that emerged in the interviews included awareness of the

prevalence of gambling problems in mental health populations, recognition of problem gambling as

an underlying issue, a readiness to use screening tools as part of assessment processes and the

availability of advice/training in how to respond when a gambling problem is identified. Beyond

screening, there was very limited knowledge of treatment approaches (with some clinicians offering

brief interventions such as motivational interviewing) and a tendency to consider referral to another

professional or service for treatment of the problem.

The fourth research activity entailed a survey of 841 patients attending eight separate mental health

services across 12 individual sites. In the majority of services, patients completed an online survey in

the reception of the mental health service while waiting for their appointment. In one service that

operated a primarily outreach model of care, case managers of the service were trained in

conducting the surveys with their patients. The survey examined gambling and related behaviours

and permitted a point-prevalence estimate of gambling harm. While gambling participation rates

were much lower than they are in the Victoria adult general population, over half of those who had

gambled in the past year were experiencing gambling-related harm, with 8 per cent identified as

moderate risk and an additional 6 per cent identified as problem gamblers on the PGSI. The most

common gambling activities were pokies or electronic gaming, closely followed by Lotto, Powerball

or the Pools, followed by betting on horses or greyhounds, and then scratch tickets. Most patients

that gambled engaged in gambling activities on a monthly basis, spending on average $119 per

month. Moderate-risk gamblers reported poorer levels of physical wellbeing, and both moderate-risk

and problem gamblers were more likely to report tobacco smoking and illicit drug use. Male patients

and those diagnosed with psychosis, drug use disorders and borderline personality disorder were

more likely to experience harm from gambling. Further evidence of inadequate screening practices

emerged from the patient survey, in that only 43 per cent reported having been asked about their

gambling since attending the mental health service. The discrepancy between reported rates of

screening among clinicians (where 77 per cent reported screening) and patients could potentially be

explained by patients’ perception of being screened (i.e. not recognising that it could have taken the

form of a discussion initiated by the clinician or the patient). However, encouragingly, the majority

(two-thirds) of participants indicated that speaking to their mental health worker or using self-help

strategies were the preferred means for seeking help if they were to experience a gambling problem.

The fifth activity aimed to determine the optimal brief screening tools in mental health settings. Using

patient survey data, the psychometric properties of 10 brief problem gambling screening instruments

(two to five items) and the 23 single items that comprise those screening instruments were

examined. The sensitivity, specificity and overall classification accuracy were compared. The

findings were that several single items and all of the brief screening instruments displayed adequate

sensitivity, specificity, and overall diagnostic accuracy in detecting people at risk of developing

gambling problems. The psychometric properties of item 1 of the PSGI (“In the last 12 months, have

you bet more than you could really afford to lose?”) indicated that this single question accurately

detected 96 per cent of patients with mental health problems classified as problem gamblers on the



full PGSI and 84 per cent of patients classified as moderate-risk or problem gamblers on the full

PGSI. However, if a mental health service wishes to identify patients at risk of gambling-related harm

(i.e., at low or moderate risk), the recommended screening instrument for detecting low-risk,

moderate-risk and problem gambling is the five-item BPGS which accurately detects 80 per cent of

low-risk, 94 per cent of moderate-risk and 100 per cent of problem gamblers.

Overall, the findings from this project provide important insights into current mental health service

responses to problem gambling, and opportunities for improvement. However, it is important to

recognise a number of limitations related to the project design. Firstly, there is the issue of sampling

bias, where participants in the clinician survey and interviews may have been those with the greatest

interest in gambling or have the greatest experience in problem gambling, such that the findings may

overestimate current clinician responding to the issue. Along this theme, social desirability effects,

whereby clinicians respond to questions in a way in which they believe they ought to, i.e., in line with

best practice rather than their own/actual practice may have compounded this overestimate of

current responding. In the patient survey, identified rates of gambling participation and harm may be

an underestimate because the most severely unwell (e.g. acutely psychotic) patients are less likely

to have participated in the survey. Since we know from the literature that gambling harm is elevated

in such presentations (Haydock et al., 2015), it is feasible that the landscape is in fact bleaker and

that gambling harm is greater than indicated and clinician responding more modest than indicated by

the current data. This issue also challenges how representative the findings are of the broader

mental health population and how generalisable the results are to settings serving more acutely

unwell populations. Finally, for both clinicians and patients all responses were subjective self-report

which could potentially be affected by recall bias and attribution, with no objective/corroborative

measures of their gambling behaviours or clinical practice.

Despite these limitations, there are a number of important strengths to the project that increase our

confidence in the research findings. These include the adoption of multiple robust methodologies to

address the key research questions, the large number of clinicians and patients recruited, the

breadth of mental health services involved, the number of different service sites across Victoria from

which participants were drawn, and the expertise and insights of the research team and clinical

reference panel.

Clinical reference panel

The project findings highlight the gaps and opportunities for improved responding to problem

gambling within mental health settings. To help develop practical approaches for improving service

responses, a clinical reference panel was convened to provide feedback and guidance on the

preliminary findings and how they could inform the development of recommendations for best

practice in responding to problem gambling in the mental health sector. This half-day event was

attended by a diverse range of clinicians (N = 32) working at the frontline, as well as team leaders

and service managers in the mental health and gambling sectors, consumers, VRGF senior staff and

Department of Health and Human Services representatives. Findings from the five studies were

presented to the panel, and then using a RAPID Process Improvement Methodology, panel

members examined strengths and weakness of current practices, and priorities for action. Echoing

the findings of the clinician survey and interviews, there was a clear consensus that there are many

areas for improvement in terms of responding to problem gambling within mental health settings.

Priorities for action identified were



reducing stigma

raising consumer and clinician awareness of problem gambling

addressing the lack of routine gambling screening and assessment

improving the lack of problem gambling lived experience/knowledge in training or service

provision

addressing fragmentation of mental health and gambling services and funding models.

In terms of solutions, panel members identified five low-cost/effort initiatives that could immediately

improve mental health service response to problem gambling:

1. implementing a validated brief screen for problem gambling (the single item PGSI as a

minimum) at intake

2. raising consumer awareness that gambling problems and mental health issues commonly

occur, and help is available through the service

3. identifying champions/role models for good clinical practice around problem gambling

(who keep gambling on the agenda and build capacity within a team)

4. sharing of existing education packages around problem gambling and clinical approaches

5. educating staff about the Gambler’s Help system and development of local referral

pathways.

In terms of medium-cost/effort initiatives, the panel identified three key initiatives:

1. investment in clinical supervision and support, and clearly defined models of care for

responding to gambling within mental health settings

2. specialist training for clinicians in screening, assessment, brief intervention and treatment

(where relevant) of problem gambling

3. development of partnerships and integrated working between mental health and gambling

services with established memorandum of understanding (MOU) to promote

accountability.

Conclusion

Consistent with previous research, the report’s findings highlight major gaps in current responding

to problem gambling in Victorian mental health services. Given the elevated rates of problem

gambling observed in patients attending mental health services, and the finding that one in two

gamblers are experiencing gambling-related harm, there are significant opportunities for

prevention, early identification and intervention for a population at elevated risk.

Encouragingly, clinicians are broadly positive towards working with patients with problem gambling

and consider it part of their role. However, only a minority have received any form of training, and

they tend to see gambling as a secondary or non-critical issue. In addition, the lack of a

standardised screening process or tool within mental health services, and concerns among



clinicians that asking about gambling too early would increase stigma, means that identifying and

responding is both discretional and ad-hoc. Low rates of detection (and subsequent referral) is

further exacerbated by clinicians’ low confidence in treating problem gambling, as well as their

poor understanding of the Gambler’s Help system and concerns about the capacity of Gambler’s

Help services to manage patients with mental illness.

In terms of supporting earlier identification of problem gambling within mental health services, the

findings indicate that a one- to three-item problem gambling screen is an effective method, and this

could be easily added to existing intake processes. The adoption of a brief screening tool into

routine clinical practice was widely supported by clinicians across all stages of the project;

however, the need for this to be part of a broader system response was also identified. This should

include comprehensive training in the assessment and management of problem gambling (given

the breadth of evidence-based treatment and support options now available), as well as improved

intersectorial partnerships and referral pathways. Together, these approaches will ensure that

patients with mental health and gambling issues receive timely and appropriate intervention to

optimise their recovery and wellbeing.

Implications of research findings

This multi-component study provided consistent findings relating to the identification and

management of patients with gambling problems within mental health services. The findings

highlighted gaps in current service provision and opportunities for improved responding to the needs

of this population. The implications of this work, informed by discussion with the clinical reference

panel and targeted at services and the service system, are listed below together with

recommendations for future research.

1. Implications for the service system

(a) Raising consumer and carer awareness that the risk of gambling-related harm is

greater among those with mental health issues. The research revealed that among

patients with mental health problems, gamblers were more likely to be experiencing

gambling-harm than not. Furthermore, rates of problem gambling in this population

were eight times higher and rates of moderate-risk gambling three times higher than

they were in the general adult population. Clinicians estimated that one in every 10

patients on their caseload was affected by gambling. These findings support earlier

research studies included in the literature review which point towards elevated rates of

gambling-related harm for individuals with mental health disorders. One mechanism

for increasing consumer or carer awareness could be through the development of

targeted health promotion materials or resources (and related campaigns), ensuring

these are readily available in primary health, AOD and mental health services.

(b) Reducing the stigma associated with problem gambling and instilling hope that

recovery is possible for patients, families and the broader community. Both the

literature review and the qualitative clinician interviews highlighted stigma as a

common barrier to seeking help for a gambling problem. One potential mechanism for

reducing stigma among patients with a mental health problem could be to establish a

platform for the dissemination of gambling recovery success stories, for example

through peer-led initiatives that normalise the experience of gambling problems and

help-seeking behaviours and promote the reality of recovery success. One related

recommendation from the clinical reference panel was that the contribution of ‘lived-

experience’ is not tokenistic, but an integral part of any training or awareness-raising



program, to help communicate from a patient perspective why addressing gambling is

important and what has worked to help individual patients.

(c) Improving intersectorial partnerships and establishing joint policy and funding

initiatives to enhance the integration of care for patients with gambling and

mental illness. One of the key outputs of the clinical reference panel was identifying

the need to harness opportunities for joint-working between commissioners of mental

health and gambling services, moving away from fragmented, siloed operations and

towards shared joint strategic planning. As proposed in a recent paper on policy and

service delivery approaches for patients with comorbid problem gambling and mental

health issues, cross-sector integration is likely to require incentives and long-term

government commitment and support (Martyres & Townshend, 2016). The Victorian

Responsible Gambling Foundation (VRGF) and Victorian Department of Health and

Human Services (DHHS) should consider forming a joint working group to examine

opportunities for joint policy and funding initiatives to enhance the integration of

service delivery.

(d) Increasing clinical skills in screening, assessment and referral pathways. In

support of the findings from the literature review, the clinician survey and qualitative

interviews evidenced low rates of screening, assessment and referral. Over half the

clinicians reported rarely or never screening for problem gambling. Similarly, less than

half of the patients surveyed had been asked about their gambling problem. In the

qualitative interviews, clinicians indicated a preference for group over online training,

and a preference for training with practical examples (e.g. vignette-based training) they

could relate to and with opportunities to practise skills. Multiple clinicians noted the

importance of follow-up, booster sessions or other post-training support and the need

for training to be evaluated. Specialist training must aim to increase knowledge,

confidence and capacity to recognise the signs of problem gambling, and include

guidance on how to introduce the issue of gambling, how to identify and apply

appropriate screens, how to administer the screens and interpret scores,

recommended responses for different gambling risk severity and referral options. An

existing resource that could be adapted for mental health clinicians is the ‘Slots and

Shots’ guideline produced by Rowe, White, Long, Roche, and Orr (2015).

(e) Increasing confidence and capacity to deliver treatments targeting problem

gambling. The clinician survey and qualitative interviews demonstrated that few

clinicians were confident in managing or treating a gambling problem, but that rates of

confidence and general willingness to respond were higher among clinicians who had

previously received training in problem gambling. With evidence from the literature that

integrated CBT (which simultaneously addresses cognitions and behaviours that result

in psychological distress and problematic gambling behaviour) is an effective

treatment option, there are opportunities to upskill key clinicians within mental health

services to provide this treatment option where indicated, especially where they are

already providing intensive psychological support to patients with multiple morbidities.

Training could be offered to relevant services across the breadth of potential

intervention options, including harm reduction and self-help strategies, brief

intervention or CBT (for services with capacity), as well as joint treatment planning with

Gambler’s Help agencies for services without treatment capacity.



(f) Ensuring Gambler’s Help services are equipped to meet the needs of patients

with mental illness. The clinician survey and qualitative interviews identified concerns

regarding the capacity of Gambler’s Help services to meet the needs of patients with

mental illness. It is recommended that capacity is assessed and training provided if

necessary. This is pertinent since the patient survey indicated that patients with the

most complex needs (those with multiple disorders, drug use disorder, psychosis and

borderline personality disorder) are more likely to be experiencing gambling-related

harm. Gambler’s Help services should identify staff who are skilled in managing

patients with comorbid gambling problems and mental illness, and be offered relevant

training and supervision to support this work.

2. Implications for services

(a) Improving processes for early identification of patients with gambling problems.

The finding in the literature review that mental health disorders typically predate the

onset of problem gambling highlights the importance of targeting mental health

populations. The literature review, clinician survey and patient survey data all point to

the need to implement and standardise processes to identify problem gambling. The

qualitative interviews indicated that on the rare occasions when routine screening does

happen, clear protocols have been established. As highlighted through the qualitative

research with clinicians, mental health services should, as a minimum, consider

embedding a screening tool within the service’s intake process (as most have done

with alcohol and drugs). The findings from the study examining the psychometric

properties of problem gambling screening tools indicated that a range of screen

instruments performed adequately in mental health service populations. A single item

screen (PGSI-item 1) or a three-item screen (BPGS) most effectively detected

moderate-risk or problem gambling, and a five-item screen (BPGS) most effectively

detected low-risk, moderate-risk or problem gambling within a mental health

population.

(b) Raising the profile of problem gambling within mental health services. While

patients reported that mental health workers were the preferred method for seeking

support if they experienced a gambling problem, clinicians had low confidence in

managing problem gambling and it was typically seen as a ‘low priority’ within a

patient’s treatment plan. It is therefore important to communicate to mental health

clinicians that problem gambling can serve as a stressor that impacts on a person’s

mental state, increasing the risk of relapse (and potential acute/crisis service use) as

well as their ability to adhere to treatment (e.g. significant financial loss impacts

housing stability, ability to pay for medication/supportive therapy). One potential

method (that has been effective in addressing substance use within mental health

services) is to identify a ‘champion’, portfolio holder or working group, so that gambling

is given a higher priority and treatment and referral options are actively promoted

within the service.

(c) Supporting staff training opportunities and clinical supervision in managing

problem gambling. The clinician survey revealed that only 12 per cent of the

workforce had received at least some specialist training in problem gambling,

highlighting opportunities for further professional development. The clinical reference

panel argued that one of the strengths of existing shared-care initiatives was the

opportunity for joint participation in professional development (e.g. forums supporting

reciprocal arrangements, where mental health clinicians provide consultation and



training to gambling counsellors and gambling counsellors provide consultation and

training to mental health clinicians in their respective areas of expertise), and this is a

useful model to consider. Services should support staff opportunities for ongoing

professional development on topics relating to identifying and responding to problem

gambling and provide a mechanism for facilitating clinical supervision related to

problem gambling.

(d) Developing working partnerships between mental health and local Gambler’s

Help services. The findings from the literature review on models of care and outputs

of the clinical reference panel highlight the need to invest in gambling–mental health

shared-care partnerships. This is particularly important as the clinician survey

indicated a poor understanding of what Gambler’s Help services offer and few

clinicians were regularly referring patients to these services. This could be achieved by

establishing service agreements or memoranda of understanding, with shared care

models defining clear referral pathways between Gambler’s Help and mental health

services. This could include the establishment of local processes to facilitate

collaboration (e.g. streamlined referral processes, joint assessment, shared-care plans

and in-reach work), as well as in-service training about the Gambler’s Help system and

local processes.

3. Recommendations for future research

(a) Determining rates and patterns of gambling harm across all mental health

populations. While this study surveyed a diverse range of community mental health

services, the findings cannot be generalised to other mental health settings (e.g. acute

inpatient services and public mental health outreach services) where gambling harm

may differ. In addition, the high rates of gambling harm among patients with substance

use problems identified in the literature and patient survey highlights the importance of

a more robust examination of gambling harm within AOD treatment settings. Finally,

given the high rate of gambling harm among patients with mental health problems,

further research is needed to understand why this population is at such risk and the

most effective strategies to minimise harm.

(b) Determining effective models of care for patients with comorbid problem

gambling and mental illness. The literature review highlighted the dearth of empirical

knowledge about models of care for gambling in mental health settings. However,

drawing on the more extensive AOD literature, current evidence suggests that models

should comprise multidisciplinary teams, continuous treatment teams, an integrated

treatment philosophy, and stage-wise or tailored treatment approaches. A core

minimum set of model features includes universal screening across all mental health

practitioners and services, risk assessment and diagnosis of symptoms, supportive

therapies, prevention and psychoeducation, collaboration with other services and/or

GPs, and supportive policies and procedures. The literature review also draws

attention to the limitations of integrated treatment model evaluations to date. Further

research is needed to determine the impact and cost-effectiveness of service models,

at both a patient/use and service level through rigorous evaluation of existing models

as well as pilot/feasibility studies of integrated care.



(c) Developing effective treatments for individuals with comorbid gambling and

mental health disorders. The literature review of treatment approaches concluded

that there is a paucity of evidence on which to base treatment recommendations for

different subpopulations of problem gamblers based on their psychiatric comorbidity

(as comorbid patients have typically been excluded from treatment trials for gambling

problems). This gap in evidence on treatment effectiveness was also identified as a

key priority area by members of the clinical reference panel. Research demonstrating

a clear relationship between comorbidity and poorer treatment outcomes remains

limited. Further evidence of this relationship may encourage clinicians to screen and

respond to gambling problems. This could potentially be achieved through investment

in systems that support the monitoring of outcomes of patients with comorbid issues

and greater focus on trials of integrated psychosocial interventions for these

populations.



References

Abbott, M., Bellringer, M., Hodgins, D., Palmer Du Preez, K., Landon, J., Sullivan, S., & Feigin, V. (2012). Effectiveness of problem gambling brief telephone interventions: A randomised controlled trial. Wellington: Ministry of Health.

Achab, S., Chatton, A., Khan, R., Thorens, G., Penzenstadler, L., Zullino, D., & Khazaal, Y. (2014). Early detection of pathological gambling: Betting on GPs' beliefs and attitudes. BioMed Research International, 2014, 360585. doi:10.1155/2014/360585

ACIL Allen Consulting, The Social Research Centre, & The Problem Gambling Research and Treatment Centre. (2014). Third social and economic impact study of gambling in Tasmania: Volume 2, 2013 Tasmanian Gambling Prevalence Survey. Melbourne: ACIL Allen Consulting.

American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders (5th ed.). Arlington, VA: American Psychiatric Association.

Aragay, N., Roca, A., Garcia, B., Marqueta, C., Guijarro, S., Delgado, L., . . . Valles, V. (2012). Pathological gambling in a psychiatric sample. Comprehensive Psychiatry, 53(1), 9–14. doi:10.1016/j.comppsych.2011.02.004

Australian Institute for Primary Care. (2009). Comorbidity Treatment Service Model evaluation – Final report. Bundoora, VIC: Australian Government Deapartment of Health.

Baratloo, A., Hosseini, M., Negida, A., & El Ashal, G. (2015). Part 1: simple definition and calculation of accuracy, sensitivity and specificity. EMERGENCY–An Academic Emergency Medicine Journal, 3(2), 48–49.

Bartley, C. A., & Bloch, M. H. (2013). Meta-analysis: Pharmacological treatment of pathological gambling. Expert review of neurotherapeutics, 13(8), 887–894.

Ben-Tovim, D., Esterman, A., Tolchard, B., & Battersby, M. (2001). The Victorian Gambling Screen. Melbourne: Victorian Research Panel.

Berkson, J. (1946). Limitations of the application of fourfold table analysis to hospital data. Biometrics Bulletin, 2(3), 47–53.

Biddle, D., Hawthorne, G., Forbes, D., & Coman, G. (2005). Problem gambling in Australian PTSD treatment-seeking veterans. Journal of Traumatic Stress, 18(6), 759–767. doi:10.1002/jts.20084

Billi, R., Stone, C. A., Marden, P., & Yeung, K. (2014). The Victorian gambling study: A longitudinal study of gambling and health in Victoria 2008–2012. Victoria, Australia: Victorian Responsible Gambling Foundation.

Blaszczynski, A., Ladouceur, R., & Moodie, C. (2009). The Sydney Laval Universities Gambling Screen: Preliminary data. Addiction Research and Theory, 16(4), 401–411. doi:10.1080/16066350701699031

Blix, O., & Eek, U. (2005). Long-term treatment for patients with severe mental illness and substance abuse. Heroin Addiction and Related Clinical Problems, 7(2), 11–18.

Bonnaire, C., Kovess-Masfety, V., Guignard, R., Richard, J. B., du Roscoat, E., & Beck, F. (2016). Gambling type, substance abuse, health and psychosocial correlates of male and female problem gamblers in a nationally representative French sample. Journal of Gambling Studies. doi:10.1007/s10899-016-9628-4

Botterill, E., Gill, P. R., McLaren, S., & Gomez, R. (2016). Marital status and problem gambling among Australian older adults: The mediating role of loneliness. Journal of Gambling Studies, 32(3), 1027–1038. doi:10.1007/s10899-015-9575-5

Boyle, P., & Kroon, H. (2006). Integrated dual disorder treatment comparing facilitators and challenges of implementation for Ohio and the Netherlands. International Journal of Mental Health, 35(2), 70–88. doi:10.2753/imh0020-7411350205

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. doi:10.1191/1478088706qp063oa



Brett, E. I., Weinstock, J., Burton, S., Wenzel, K. R., Weber, S., & Moran, S. (2014). Do the DSM-5 diagnostic revisions affect the psychometric properties of the Brief Biosocial Gambling Screen? International Gambling Studies, 14(3), 447–456. doi:10.1080/14459795.2014.931449

Brown, M., Oldenhof, E., Allen, J. S., & Dowling, N. A. (2016). An empirical study of personality disorders among treatment-seeking problem gamblers. Journal of Gambling Studies, Epub, 1–22. doi:10.1007/s10899-016-9600-3

Bush, K., Kivlahan, D. R., McDonell, M. B., Fihn, S. D., & Bradley, K. A. (1998). The AUDIT alcohol consumption questions (AUDIT-C): An effective brief screening test for problem drinking. Archives of Internal Medicine, 158(16), 1789–1795.

Castrén, S., Basnet, S., Salonen, A. H., Pankakoski, M., Ronkainen, J. E., Alho, H., & Lahti, T. (2013). Factors associated with disordered gambling in Finland. Substance Abuse Treatment, Prevention, and Policy, 8(1), 1–10. doi:10.1186/1747-597x-8-24

Chou, K. L., & Afifi, T. O. (2011). Disordered (pathologic or problem) gambling and Axis I psychiatric disorders: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. American Journal of Epidemiology, 173(11), 1289–1297. doi:10.1093/aje/kwr017

Co-Exist NSW. (2008). Co-Exist NSW: Diversity Health Comorbidity Service. Parramatta, NSW: NSW Health.

Colasante, E., Gori, M., Bastiani, L., Siciliano, V., Giordani, P., Grassi, M., & Molinaro, S. (2013). An assessment of the psychometric properties of Italian version of CPGI. Journal of Gambling Studies, 29(4), 765–774. doi:10.1007/s10899-012-9331-z

Corney, R. (2011). Continuing professional development for general practitioners in the United Kingdom. Journal of Gambling Issues, 25, 108–112. doi:10.4309/jgi.2011.25.8

Cowlishaw, S., Merkouris, S., Chapman, A., & Radermacher, H. (2014). Pathological and problem gambling in substance use treatment: A systematic review and meta-analysis. Journal of Substance Abuse Treatment, 46(2), 98–105. doi:10.1016/j.jsat.2013.08.019

Crockford, D. N., & el-Guebaly, N. (1998). Psychiatric comorbidity in pathological gambling: A critical review. Canadian Journal of Psychiatry, 43(1), 43–50.

Croton, G. (2007). Screening for and assessment of co-occurring substance use and mental health disorders by alcohol and other drug and mental health services. Victoria: Victorian Dual Diagnosis Initiative Advisory Group.

Cunningham-Williams, R. M., Grucza, R. A., Cottler, L. B., Womack, S. B., Books, S. J., Przybeck, T. R., . . . Cloninger, C. R. (2005). Prevalence and predictors of pathological gambling: Results from the St. Louis personality, health and lifestyle (SLPHL) study. Journal of Psychiatric Research, 39(4), 377–390. doi:10.1016/j.jpsychires.2004.09.002

Cunningham, J. A., Hodgins, D. C., Bennett, K., Bennett, A., Talevski, M., Mackenzie, C. S., & Hendershot, C. S. (2016). Online interventions for problem gamblers with and without co-occurring mental health symptoms: Protocol for a randomized controlled trial. BMC Public Health, 16(1), 1.

Currie, S. R., Hodgins, D. C., & Casey, D. M. (2013). Validity of the Problem Gambling Severity Index interpretive categories. Journal of Gambling Studies, 29(2), 311–327. doi:10.1007/s10899-012-9300-6

Davidson, T., & Rodgers, B. (2010). 2009 survey of the nature and extent of gambling, and problem gambling, in the Australian Capital Territory: Final report. Canberra: The Australian Capital Territory Gambling and Racing Commission and The Australian National University.

Davidson, T., Rodgers, B., Taylor-Rodgers, E., Suomi, A., & Lucas, N. (2015). 2014 survey on gambling health and wellbeing in the ACT. Retrieved from Canberra:

de Castella, A., Bolding, P., Lee, A., Cosic, S., & Kulkarni, J. (2011). Problem gambling in people presenting to a public mental health service. Melbourne: Office of Gaming and Racing, Department of Justice.

Dell'Osso, B., & Hollander, E. (2005). The impact of comorbidity on the management of pathological gambling. CNS Spectrums, 10(8), 619–621. doi:10.1017/s109285290001957x



Derevensky, J. L., Gupta, R., & Winters, K. C. (2003). Prevalence rates of youth gambling problems: Are the current rates inflated? Journal of Gambling Studies, 19(4), 405–425.

DiStefano, C., & Morgan, G. (2011). Examining classification criteria: A comparison of three cut score methods. Psychological Assessment, 23(2), 354–363. doi:10.1037/a0021745

Donald, M., Dower, J., & Kavanagh, D. (2005). Integrated versus non-integrated management and care for clients with co-occurring mental health and substance use disorders: A qualitative systematic review of randomised controlled trials. Social Science and Medicine, 60(6), 1371–1383. doi:10.1016/j.socscimed.2004.06.052

Dowling, N. A., Cowlishaw, S., Jackson, A. C., Merkouris, S. S., Francis, K. L., & Christensen, D. R. (2014). The prevalence of comorbid personality disorders in treatment-seeking problem gamblers: A systematic review and meta-analysis. Journal of Personality Disorders, 1–20.

Dowling, N. A., Cowlishaw, S., Jackson, A. C., Merkouris, S. S., Francis, K. L., & Christensen, D. R. (2015). Prevalence of psychiatric co-morbidity in treatment-seeking problem gamblers: A systematic review and meta-analysis. Australian and New Zealand Journal of Psychiatry, 49(6), 519–539. doi:10.1177/0004867415575774

Dowling, N. A., Jackson, A. C., Suomi, A., Lavis, T., Thomas, S. A., Patford, J., . . . Abbott, M. (2014). Problem gambling and family violence: prevalence and patterns in treatment-seekers. Addictive Behaviors, 39(12), 1713–1717.

Dowling, N. A., Merkouris, S. S., Greenwood, C. J., Oldenhof, E., Toumourou, J. W., & Youssef, G. J. (2015). Risk and protective factors for youth problem gambling – A systematic review and meta-analysis of longitudinal studies. Clinical Psychological Review, Manuscript submitted for publication.

Dowling, N. A., Merkouris, S. S., & Lorains, F. K. (2016). Interventions for comorbid problem gambling and psychiatric disorders: Advancing a developing field of research. Addictive Behaviours, 58, 21–30. doi:10.1016/j.addbeh.2016.02.012

Dowling, N. A., Smith, D., & Thomas, T. (2005). Electronic gaming machines: Are they the 'crack-cocaine' of gambling? Addiction, 100(1), 33–45. doi:10.1111/j.1360-0443.2005.00962.x

Dowling, N. A., Youssef, G. J., Jackson, A. C., Pennay, D. W., Francis, K. L., Pennay, A., & Lubman, D. I. (2016). National estimates of Australian gambling prevalence: Findings from a dual-frame omnibus survey. Addiction, 111(3), 420–435. doi:10.1111/add.13176

Drake, R. E., Antosca, L. M., Noordsy, D. L., Bartels, S. J., & Osher, F. C. (1991). New Hampshire's specialized services for the dually diagnosed. New Directions for Mental Health Services, 1991(50), 57–67.

Drake, R. E., McHugo, G. J., Clark, R. E., Teague, G. B., Xie, H., Miles, K., & Ackerson, T. H. (1998). Assertive community treatment for patients with co-occuring severe mental illness and substance use disorder: A clinical trial. American Journal of Orthopsychiatry, 68(2), 201–215.

Drake, R. E., Mercer-McFadden, C., Mueser, K. T., McHugo, G. J., & Bond, G. R. (1998). Review of integrated mental health and substance abuse treatment for patients with dual disorders. Schizophrenia Bulletin, 24(4), 589–608.

Drake, R. E., O'Neal, E. L., & Wallach, M. A. (2008). A systematic review of psychosocial research on psychosocial interventions for people with co-occurring severe mental and substance use disorders. Journal of Substance Abuse Treatment, 34(1), 123–138. doi:10.1016/j.jsat.2007.01.011

Drake, R. E., Teague, G. B., & Warren, S. R. I. (1990). Dual diagnosis: The New Hampshire program. Addiction and Recovery, 10, 35.

Dussault, F., Brendgen, M., Vitaro, F., Wanner, B., & Tremblay, R. E. (2011). Longitudinal links between impulsivity, gambling problems and depressive symptoms: A transactional model from adolescence to early adulthood. Journal of Child Psychology and Psychiatry, 52(2), 130–138. doi:10.1111/j.1469-7610.2010.02313.x

Dyall, L., Tse, S., & Kingi, A. (2009). Cultural icons and marketing of gambling. International Journal of Mental Health and Addiction, 7(1), 84–96.



Echeburua, E., Gomez, M., & Freixa, M. (2011). Cognitive-behavioural treatment of pathological gambling in individuals with chronic schizophrenia: A pilot study. Behaviour research and therapy, 49(11), 808–814. doi:10.1016/j.brat.2011.08.009

Essock, S. M., Mueser, K. T., Drake, R. E., Covell, N. H., McHugo, G. J., Frisman, L. K., . . . Swain, K. (2006). Comparison of ACT and Standard Case Management for delivering integrated treatment for co-occurring disorders. Psychiatric Services, 57(2), 185–196. doi:doi:10.1176/appi.ps.57.2.185

Ferris, J., & Wynne, H. (2001). The Canadian problem gambling index. Ottawa: Canadian Centre on Substance Abuse.

Fong, T., Kalechstein, A., Bernhard, B., Rosenthal, R., & Rugle, L. (2008). A double-blind, placebo-controlled trial of olanzapine for the treatment of video poker pathological gamblers. Pharmacology, Biochemistry, and Behavior, 89(3), 298–303. doi:10.1016/j.pbb.2007.12.025

Gainsbury, S. M., Hing, N., & Suhonen, N. (2014). Professional help-seeking for gambling problems: Awareness, barriers and motivators for treatment. Journal of Gambling Studies, 30(2), 503–519. doi:10.1007/s10899-013-9373-x

Gainsbury, S. M., Russell, A., Hing, N., Wood, R., & Blaszczynski, A. (2013). The impact of internet gambling on gambling problems: A comparison of moderate-risk and problem Internet and non-Internet gamblers. Psychology of Addictive Behaviors, 27(4), 1092.

Gainsbury, S. M., Russell, A., Hing, N., Wood, R., Lubman, D. I., & Blaszczynski, A. (2014). The prevalence and determinants of problem gambling in Australia: Assessing the impact of interactive gambling and new technologies. Psychology of Addictive Behaviors, 28(3), 769.

Gambino, B. (2006). A comment on the utility of prevalence estimates of pathological gambling. Journal of Gambling Studies, 22(3), 321–328. doi:10.1007/s10899-006-9019-3

Gamblers Anonymous. Gamblers Anonymous 20 Questions. Los Angeles, CA: Author.

Gebauer, L., LaBrie, R., & Shaffer, H. (2010). Optimizing DSM-IV-TR classification accuracy: A brief biosocial screen for detecting current gambling disorders among gamblers in the general household population. Canadian Journal of Psychiatry, 55(2), 82–90.

Gerstein, D. R., Murphy, S. A., Toce, M. T., Hoffmann, J., Palmer, A., Johnson, R. A., . . . Sinclair, S. (1999). Gambling impact and behavior study: Report to the National Gambling Impact Study Commission. Chicago: National Opinion Research Center at the University of Chicago.

Glascoe, F. P. (2005). Screening for developmental and behavioral problems. Mental Retardation and Development Disabilities Research Reviews, 11(3), 173–179. doi:10.1002/mrdd.20068

Gooding, P., & Tarrier, N. (2009). A systematic review and meta-analysis of cognitive-behavioural interventions to reduce problem gambling: hedging our bets? Behaviour research and therapy, 47(7), 592–607.

Goodyear-Smith, F., Arroll, B., & Coupe, N. (2009). Asking for help is helpful: validation of a brief lifestyle and mood assessment tool in primary health care. Annals of Family Medicine, 7(3), 239–244. doi:10.1370/afm.962

Goodyear-Smith, F., Arroll, B., Kerse, N., Sullivan, S., Coupe, N., Tse, S., . . . Perese, L. (2006). Primary care patients reporting concerns about their gambling frequently have other co-occurring lifestyle and mental health issues. BMC Family Practice, 7, 25. doi:10.1186/1471-2296-7-25

Goodyear-Smith, F., Coupe, N. M., Arroll, B., Elley, C. R., Sullivan, S., & McGill, A. T. (2008). Case finding of lifestyle and mental health disorders in primary care: Validation of the 'CHAT' tool. British Journal of General Practice, 58(546), 26–31. doi:10.3399/bjgp08X263785

Gotestam, K. G., Johansson, A., Wenzel, H. G., & Simonsen, I. E. (2004). Validation of the lie/bet screen for pathological gambling on two normal population data sets. Psychological Reports, 95(3), 1009–1013.

Graham, H. L., Copello, A., Birchwood, M. J., Maslin, J., McGovern, D., Orford, J., & Georgiou, G. (2003). The Combined Psychosis and Substance Use (COMPASS) Programme: An integrated shared-care approach. In H. L. Graham, A. Copello, M. J. Birchwood, & K. T. Mueser (Eds.), Substance Misuse in Psychosis: Approaches to Treatment and Service Delivery (pp. 106). West Sussex, London: John Wiley & Sons Ltd.



Graham, H. L., Copello, A., Birchwood, M. J., Orford, J., McGovern, D., Maslin, J., & Georgiou, G. (2003). Cogntive-behavioural integrated treatment approach for psychosis and problem substance use. In H. L. Graham, A. Copello, M. J. Birchwood, & K. T. Mueser (Eds.), Substance Misuse in Psychosis: Approaches to Treatment and Service Delivery (pp. 181). West Sussex, London: John Wiley & Sons Ptd.

Graham, H. L., Copello, A., Birchwood, M. J., Orford, J., McGovern, D., Mueser, K. T., . . . Tobin, D. (2006). A preliminary evaluation of integrated treatment for co-existing substance use and severe mental health problems: Impact on teams and service users. Journal of Mental Health, 15(5), 577–591. doi:10.1080/09638230600902633

Grant, J. E., Kim, S. W., & Odlaug, B. L. (2007). N-acetyl cysteine, a glutamate-modulating agent, in the treatment of pathological gambling: A pilot study. Biological Psychiatry, 62(6), 652–657. doi:10.1016/j.biopsych.2006.11.021

Grant, J. E., Odlaug, B. L., Chamberlain, S. R., Potenza, M. N., Schreiber, L. R., Donahue, C. B., & Kim, S. W. (2014). A randomized, placebo-controlled trial of N-acetylcysteine plus imaginal desensitization for nicotine-dependent pathological gamblers. The Journal of Clinical Psychiatry, 75(1), 39–45. doi:10.4088/JCP.13m08411

Grant, J. E., & Potenza, M. N. (2006). Escitalopram treatment of pathological gambling with co-occurring anxiety: An open-label pilot study with double-blind discontinuation. Internation Clinical Psychopharmacology, 21(4), 203–209.

Grant, J. E., Steinberg, M. A., Kim, S. W., Rounsaville, B. J., & Potenza, M. N. (2004). Preliminary validity and reliability testing of a structured clinical interview for pathological gambling. Psychiatry Research, 128(1), 79–88. doi:10.1016/j.psychres.2004.05.006

Gross, J. J., & Jazaieri, H. (2014). Emotion, emotion regulation, and psychopathology: An affective science perspective. Clinical Psychological Science, 2(4), 387–401. doi:10.1177/2167702614536164

Hare, S. (2015). Study of gambling and health in Victoria: Findings from the Victorian Prevalence Study 2014. Melbourne: Victorian Responsible Gambling Foundation and Victorian Department of Justice and Regulation.

Harrison, M. L., Moore, K. A., Young, M. S., Flink, D., & Ochshorn, E. (2008). Implementing the comprehensive, continuous, integrated system of care model for individuals with co-occurring disorders: Preliminary findings from a residential facility serving homeless individuals. Journal of Dual Diagnosis, 4(3), 238–259. doi:10.1080/15504260802072305

Haydock, M., Cowlishaw, S., Harvey, C., & Castle, D. (2015). Prevalence and correlates of problem gambling in people with psychotic disorders. Comprehensive Psychiatry, 58, 122–129. doi:10.1016/j.comppsych.2015.01.003

Heatherton, T. F., Kozlowski, L. T., Frecker, R. C., Rickert, W., & Robinson, J. (1989). Measuring the Heaviness of Smoking: Using self-reported time to the first cigarette of the day and number of cigarettes smoked per day. British Journal of Addiction, 84(7), 791–800. doi:10.1111/j.1360-0443.1989.tb03059.x

Hides, L., Lubman, D. I., Elkins, K., Catania, L. S., & Rogers, N. (2007). Feasibility and acceptability of a mental health screening tool and training programme in the youth alcohol and other drug (AOD) sector. Drug and Alcohol Review, 26(5), 509–515. doi:10.1080/09595230701499126

Himelhoch, S. S., Miles-McLean, H., Medoff, D. R., Kreyenbuhl, J., Rugle, L., Bailey-Kloch, M., . . . Brownley, J. (2015). Evaluation of brief screens for gambling disorder in the substance use treatment setting. The American Journal on Addictions, 24(5), 460–466. doi:10.1111/ajad.12241

Hing, N., Russell, A., Tolchard, B., & Nower, L. (2016). Risk factors for gambling problems: An analysis by gender. Journal of Gambling Studies, 32(2), 511–534. doi:10.1007/s10899-015-9548-8

Hodgins, D. C., Peden, N., & Cassidy, E. (2005). The association between comorbidity and outcome in pathological gambling: A prospective follow-up of recent quitters. Journal of Gambling Studies, 21(3), 255–271. doi:10.1007/s10899-005-3099-3

Hollander, E., Pallanti, S., Allen, A., Sood, E., & Baldini-Rossi, N. (2005). Does sustained-release lithium reduce impulsive gambling and affective instability versus placebo in pathological gamblers with



bipolar spectrum disorders? The American Journal of Psychiatry, 162(1), 137–145. doi:10.1176/appi.ajp.162.1.137

Hollander, E., Sood, E., Pallanti, S., Baldini-Rossi, N., & Baker, B. (2005). Pharmacological treatments of pathological gambling. Journal of Gambling Studies, 21(1), 99–110. doi:10.1007/s10899-004-1932-8

Holtgraves, T. (2009). Evaluating the problem gambling severity index. Journal of Gambling Studies, 25(1), 105–120. doi:10.1007/s10899-008-9107-7

Jackson, A. C., Pennay, D., Dowling, N. A., Coles-Janess, B., & Christensen, D. R. (2014). Improving gambling survey research using dual-frame sampling of landline and mobile phone numbers. Journal of Gambling Studies, 30(2), 291–307. doi:10.1007/s10899-012-9353-6

Jazaieri, H., Urry, H. L., & Gross, J. J. (2013). Affective disturbance and psychopathology: An emotion regulation perspective. Journal of Experimental Psychopathology, 4(5), 584–599. doi:10.5127/jep.030312

Johnson, E. E., Hamer, R. M., & Nora, R. M. (1997). The lie/bet questionnaire for screening pathological gamblers. Psychological Reports, 80(1), 83–88.

Johnstone, M. J. (2001). Stigma, social justice and the rights of the mentally ill: Challenging the status quo. Aust N Z J Ment Health Nurs, 10(4), 200–209.

Kessler, R. C., Heeringa, S., Lakoma, M. D., Petukhova, M., Rupp, A. E., Schoenbaum, M., . . . Zaslavsky, A. M. (2008). Individual and societal effects of mental disorders on earnings in the United States: Results from the national comorbidity survey replication. American Journal of Psychiatry, 165(6), 703–711. doi:10.1176/appi.ajp.2008.08010126

Kessler, R. C., Hwang, I., LaBrie, R., Petukhova, M., Sampson, N. A., Winters, K. C., & Shaffer, H. J. (2008). DSM-IV pathological gambling in the National Comorbidity Survey Replication. Psychological Medicine, 38(9), 1351–1360. doi:10.1017/S0033291708002900

Kessler, R. C., & Üstün, T. B. (2004). The World Mental Health (WMH) Survey Initiative version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI). International Journal of Methods in Psychiatric Research, 13(2), 93–121. doi:10.1002/mpr.168

Kola, L. A., & Kruszynski, R. (2010). Adapting the integrated dual-disorder treatment model for addiction services. Alcoholism Treatment Quarterly, 28(4), 437–450. doi:10.1080/07347324.2010.511067

Korman, L., Collins, J., Littman-Sharp, N., Skinner, W., McMain, S., & Mercado, V. (2008). Randomized control trial of an integrated therapy for comorbid anger and gambling. Psychotherapy Research, 18(4), 454–465. doi:10.1080/10503300701858362

Kring, A. M., & Sloan, D. S. (2010). Emotion regulation and psychopathology. New York: Guildford Press.

Kring, A. M., & Werner, K. H. (2004). Emotion regulation and psychopathology The regulation of emotion Hove, UK: Psychology Press.

Ladd, G. T., & Petry, N. M. (2002). Disordered gambling among university-based medical and dental patients: A focus on Internet gambling. Psychology of Addictive Behaviors, 16(1), 76–79. doi:10.1037/0893-164x.16.1.76

Lepper, J., & Haden, B. (2013). Testing NLCLiP: Validation of estimates of rates of non-problematic and problematic gambling in a sample of British schoolchildren. Journal of Gambling Studies, 29(1), 15–27. doi:10.1007/s10899-012-9294-0

Lesieur, H. R., & Blume, S. B. (1987). The South Oaks Gambling Screen (SOGS): A new instrument for the identification of pathological gamblers. The American Journal of Psychiatry, 144(9), 1184–1188. doi:10.1176/ajp.144.9.1184

Lesieur, H. R., & Blume, S. B. (1991). Evaluation of patients treated for pathological gambling in a combined alcohol, substance abuse and pathological gambling treatment unit using the Addiction Severity Index. British Journal of Addiction, 86(8), 1017–1028.

Lesieur, H. R., & Blume, S. B. (1992). Modifying the Addiction Severity Index for use with pathological gamblers. The American Journal on Addictions, 1(3), 240–247. doi:10.3109/10550499209004026



Levens, S., Dyer, A.-M., Zubritsky, C., Knott, K., & Oslin, D. W. (2005). Gambling among older, primary-care patients. The American Journal of Geriatric Psychiatry, 13(1), 69–76.

Lorains, F. K., Cowlishaw, S., & Thomas, S. A. (2011). Prevalence of comorbid disorders in problem and pathological gambling: Systematic review and meta-analysis of population surveys. Addiction, 106(3), 490–498. doi:10.1111/j.1360-0443.2010.03300.x

Lubman, D. I., Hides, L., & Elkins, K. (2008). Developing integrated models of care within the youth Alcohol and Other Drug sector. Australasian Psychiatry, 16(5), 363–366. doi:10.1080/10398560802027294

Manning, V., Gomez, B., Guo, S., Low, Y. D., Koh, P. K., & Wong, K. E. (2012). An exploration of quality of life and its predictors in patients with addictive disorders: Gambling, alcohol and drugs. International Journal of Mental Health and Addiction, 10(4), 551–562. doi:10.1007/s11469-011-9349-y

Martyres, K., & Townshend, P. (2016). Addressing the needs of problem gamblers with co-morbid issues: Policy and service delivery approaches. Journal of Gambling Issues(33), xx–xx. doi:10.4309/jgi.2016.33.1

Mays, N., & Pope, C. (2000). Qualitative research in health care. Assessing quality in qualitative research. Bmj, 320(7226), 50–52.

McCambridge, J., & Cunningham, J. (2007). Against the odds: Should GPs have any involvement with gambling problems. British Journal of General Practice, 2007, 327.

McCormack, A., & Griffiths, M. (2011). The effects of problem gambling on quality of life and wellbeing: A qualitative comparison of online and offline problem gamblers. Gambling Research, 23(1), 63–81.

McElroy, S. L., Nelson, E. B., Welge, J. A., Nelson, E., Keck, P. E., & Hudson, J. I. (2008). Olanzapine in the treatment of pathological gambling: A negative randomized placebo-controlled trial. The Journal of Clinical Psychiatry 69(3), 433–440.

McMillen, J., & Wenzel, M. (2006). Measuring problem gambling: Assessment of three prevalence screens. International Gambling Studies, 6(2), 147–174. doi:10.1080/14459790600927845

McNeilly, D. P., & Burke, W. J. (2002). Disposable time and disposable income: Problem casino gambling behavior in older adults. Journal of Clinical Geropsychology, 8(2), 75–85.

Melbourne Health. (2016). Dual Diagnosis – SUMITT: Substance Use and Mental Illness Treatment Team. Melbourne: Author.

Merkouris, S. S., Rodda, S. N., Austin, D., Lubman, D. I., Harvey, P., Battersby, M., . . . Dowling, N. A. (In Press). GAMBLINGLESS. FOR LIFE study protocol: A pragmatic randomised trial of an online cognitive-behavioural program for disordered gambling. BMJ Open, In Press.

Miller, H. (2014). Complex lives: Co-occuring conditions of problem gambling. North Melbourne: Victorian Responsible Gambling Foundation.

Minkoff, K., & Cline, C. A. (2004). Changing the world: The design and implementation of comprehensive continuous integrated systems of care for individuals with co-occurring disorders. Psychiatric Clinics of North America, 27(4), 727–743. doi:10.1016/j.psc.2004.07.003

Mission Australia. (2011). Triple Care Farm: A safe place for change 1989–2009: Mission Australia.

Monisse-Redman, M. (2015). Collaborative practice development to improve clinical outcomes for adolescents with co-occurring mental illness, substance use and impulse control disorders. International Journal of Child and Adolescent Health, 8(2), 109–117.

Morasco, B. J., Vom Eigen, K. A., & Petry, N. M. (2006). Severity of gambling is associated with physical and emotional health in urban primary care patients. General Hospital Psychiatry, 28(2), 94–100. doi:10.1016/j.genhosppsych.2005.09.004

Moskowitz, J. A. (1980). Lithium and lady luck: Use of lithium carbonate in compulsive gambling. New York State Journal of Medicine, 80(5), 785–788.

Mueser, K. T., Drake, R. E., & Miles, K. M. (1997). The course and treatment of substance use disorder in persons with severe mental illness. Treatment of Drug-Dependent Individuals with Comorbid Mental Disorders, NIDA Research Monograph, 172, 86–109.



Mueser, K. T., Noordsy, D. L., Drake, R. E., & Fox, L. (2003). Integrated treatment for dual disorders: A guide to effective practice. New York: Guilford Press

Najavits, L. M. (2011). Treatments for PTSD and pathological gambling: What do patients want? Journal of Gambling Studies, 27(2), 229–241. doi:10.1007/s10899-010-9198-9

Najavits, L. M., Smylie, D., Johnson, K., Lung, J., Gallop, R. J., & Classen, C. C. (2013). Seeking safety therapy for pathological gambling and PTSD: A pilot outcome study. Journal of Psychoactive Drugs, 45(1), 10–16. doi:10.1080/02791072.2013.763557

National Health and Medical Research Council. (2009). Australian guidelines to reduce health risks from drinking alcohol. Canberra: Commonwealth of Australia.

Neal, P. N., Delfabbro, P. H., & O'Neil, M. G. (2005). Problem gambling and harm: Towards a national definition. Melbourne: Office of Gaming and Racing, Victorian Government Department of Justice.

Nicolato, R., Romano-Silva, M. A., Correa, H., Salgado, J. V., & Teixeira, A. L. (2007). Lithium and topiramate association in the treatment of comorbid pathological gambling and bipolar disorder. Australian and New Zealand Journal of Psychiatry, 41(7), 628. doi:10.1080/00048670701400040

NSW Health. (2015). Effective models of care for comorbid mental illness and illicit substance use – Evidence check review. North Sydney, NSW: NSW Ministry of Health.

Overton, S. L., & Medina, S. L. (2008). The stigma of mental illness. The stigma of mental illness, 86(2), 143–151. doi:10.1002/j.1556-6678.2008.tb00491.x

Parhami, I., Mojtabai, R., Rosenthal, R. J., Afifi, T. O., & Fong, T. W. (2014). Gambling and the onset of comorbid mental disorders: A longitudinal study evaluating severity and specific symptoms. Journal of Psychiatric Practice, 20(3), 207–219. doi:10.1097/01.pra.0000450320.98988.7c

Parikh, R., Mathai, A., Parikh, S., Sekhar, G. C., & Thomas, R. (2008). Understanding and using sensitivity, specificity and predictive values. Indian journal of ophthalmology, 56(1), 45.

Pasternak, A. V., & Fleming, M. F. (1999). Prevalence of gambling disorders in a primary care setting. Archives of Family Medicine, 8(Nov/Dec), 515–520.

Petry, N. M. (2002). A comparison of young, middle-aged, and older adult treatment-seeking pathological gamblers. Gerontologist, 42(1), 92–99.

Petry, N. M., Rash, C. J., & Alessi, S. M. (2016). A randomized controlled trial of brief interventions for problem gambling in substance abuse treatment patients. Journal of Consulting and Clinical Psychology. doi:10.1037/ccp0000127

Pietrzak, R. H., & Petry, N. M. (2005). Antisocial personality disorder is associated with increased severity of gambling, medical, drug and psychiatric problems among treatment-seeking pathological gamblers. Addiction, 100(8), 1183–1193. doi:10.1111/j.1360-0443.2005.01151.x

Pilver, C. E., Libby, D. J., Hoff, R. A., & Potenza, M. N. (2013). Problem gambling severity and the incidence of Axis I psychopathology among older adults in the general population. Journal of Psychiatric Research, 47(4), 534–541. doi:10.1016/j.jpsychires.2012.12.013

Potenza, M. N., & Chambers, R. (2001). Schizophrenia and pathological gambling. The American Journal of Psychiatry, 158(3), 497–508.

Problem Gambling Research and Treatment Centre. (2011). Guideline for Screening, Assessment and Treatment of Problem Gambling. Melbourne: Monash University.

Productivity Commission. (1999). Australia’s Gambling Industries, Report no. 10. Retrieved from Canberra:

Queensland Government. (2012). Queensland Household Gambling Survey 2011–12. Queensland: Department of Justice and Attorney-General.

Quilty, L. C., Watson, C., Robinson, J. J., Toneatto, T., & Bagby, R. M. (2011). The prevalence and course of pathological gambling in the mood disorders. Journal of Gambling Studies, 27(2), 191–201. doi:10.1007/s10899-010-9199-8

Rockloff, M. J. (2012). Validation of the Consumption Screen for Problem Gambling (CSPG). Journal of Gambling Studies, 28(2), 207–216. doi:10.1007/s10899-011-9260-2



Rockloff, M. J., Ehrich, J., Themessl-Huber, M., & Evans, L. G. (2011). Validation of a one item screen for problem gambling. Journal of Gambling Studies, 27(4), 701–707. doi:10.1007/s10899-010-9232-y

Rockloff, M. J., & Schofield, G. (2004). Factor analysis of barriers to treatment for problem gambling. Journal of Gambling Studies, 20(2), 121–126. doi:10.1023/B:JOGS.0000022305.01606.da

Room, R., Turner, N. E., & Ialomiteanu, A. (1999). Community effects of the opening of the Niagra Casino. Addiction, 94(10), 1449–1466.

Rowan, M. S., & Galasso, C. S. (2000). Identifying office resource needs of Canadian physicians to help prevent, assess and treat patients with substance use and pathological gambling disorders. Journal of Addictive Diseases, 19(2), 43–58.

Rowe, C., White, M., Long, C., Roche, A., & Orr, K. (2015). Slots and Shots: A gambling resource for AOD workers. Retrieved from Melbourne, Victoria:

Ryan, A., Holmes, J., Hunt, V., Dunlop, A., Mammen, K., Holland, R., . . . Lintzeris, N. (2014). Validation and implementation of the Australian Treatment Outcomes Profile in specialist drug and alcohol settings. Drug and Alcohol Review, 33(1), 33–42. doi:10.1111/dar.12083

Sanju, G., & Gerada, C. (2011). Problem gamblers in primary care: Can GPs do more? British Journal of General Practice, 61(585), 248–249. doi:10.3399/bjgp11X567027

Sareen, J., Afifi, T. O., McMillan, K. A., & Asmundson, G. J. (2011). Relationship between household income and mental disorders: Findings from a population-based longitudinal study. Archives of General Psychiatry, 68(4), 419–427. doi:10.1001/archgenpsychiatry.2011.15

Schütz, C., Linden, I., Torchalla, I., Li, K., Al-Desouki, M., & Krausz, M. (2013). The Burnaby Treatment Center for Mental Health and Addiction, a novel integrated treatment program for patients with addiction and concurrent disorders: Results from a program evaluation. BMC Health Services Research, 13, 288.

Séguin, M., Robert, M., DiMambro, M., Lesage, A., Reidi, G., Roy, M., . . . Dutrisac, S. (2013). Gambling over the life course and treatment-seeking. International Gambling Studies, 13(3), 302–318. doi:10.1080/14459795.2013.812675

Shaffer, H. J., LaBrie, R., Scanlan, K. M., & Cummings, T. N. (1994). Pathological gambling among adolescents: Massachusetts Gambling Screen (MAGS). Journal of Gambling Studies, 1994(10), 339–362.

Shonin, E., Van Gordon, W., & Griffiths, M. D. (2014). Cognitive Behavioral Therapy (CBT) and Meditation Awareness Training (MAT) for the treatment of co-occurring schizophrenia and pathological gambling: A case study. International Journal of Mental Health and Addiction, 12(2), 181–196. doi:10.1007/s11469-013-9460-3

Skevington, S. M., Lotfy, M., & O'Connell, K. A. (2004). The World Health Organization's WHOQOL-BREF quality of life assessment: psychometric properties and results of the international field trial. A report from the WHOQOL group. Quality of Life Research, 13(2), 299–310.

Slutske, W. S. (2006). Natural recovery and treatment-seeking in pathological gambling: Results of two U.S. national surveys. The American Journal of Psychiatry, 163(2), 297–302. doi:10.1176/appi.ajp.163.2.297

Smith, N., Kitchenham, N., & Bowden-Jones, H. (2011). Pathological gambling and the treatment of psychosis with aripiprazole: Case reports. The British Journal of Psychiatry, 199(2), 158–159. doi:10.1192/bjp.bp.110.084996

Smith, P. C., Schmidt, S. M., Allensworth-Davies, D., & Saitz, R. (2010). A single-question screening test for drug use in primary care. Archives of Internal Medicine, 170(13), 1155–1160. doi:10.1001/archinternmed.2010.140

Social Research Centre. (2013). Gambling Prevalence in South Australia 2012. South Australia: Office for Problem Gambling, Department for Communities and Social Inclusion, State Government of South Australia.

Specker, S., Carlson, G., Edmonson, K., Johnson, P., & Marcotte, M. (1996). Psychopathology in pathological gamblers seeking treatment. Journal of Gambling Studies, 12(1), 67–81.



Sproston, K., Hing, N., & Palankay, C. (2012). Prevalence of gambling and problem gambling in New South Wales. Sydney: New South Wales Office of Liquor Gaming and Racing.

Staiger, P., Ricciardelli, L., McCabe, M., Young, G., Cross, W., Thomas, A., . . . McKinnon, B. (2008). Clients with a Dual Diagnosis: To what extent do they slip through the net? Melbourne: The beyondblue Victorian Centre of Excellence in Depression and Related Disorders.

Stein, L. I., & Test, M. A. (1980). Alternative to mental hospital treatment: I – Conceptual model, treatment program, and clinical evaluation. Archives of General Psychiatry, 37(4), 392–397.

Stinchfield, R. (2010). A critical review of adolescent problem gambling assessment instruments. International Journal of Adolescent Medicine and Health, 22(1), 77–93.

Stinchfield, R., Govoni, R., & Frisch, G. R. (2005). DSM-IV diagnostic criteria for pathological gambling: reliability, validity, and classification accuracy. The American Journal on Addictions, 14(1), 73–82. doi:10.1080/10550490590899871

Stinchfield, R., Govoni, R., & Frisch, G. R. (2007). A review of screening and assessment instruments for problem and pathological gambling In G. Smith, D. C. Hodgins, & R. J. Williams (Eds.), Research and measurement issues in gambling research (pp. 180–217). New York: Elsevier.

Stinchfield, R., Kushner, M. G., & Winters, K. C. (2005). Alcohol use and prior substance abuse treatment in relation to gambling problem severity and gambling treatment outcome. Journal of Gambling Studies, 21(3), 273–297. doi:10.1007/s10899-005-3100-1

Stinchfield, R., & McCready, J. (2014). Cross-validation and comparison of seven brief problem gambling screens. Ontario: Ontario Problem Gambling Research Centre.

Stuart, H. (2006). Mental illness and employment discrimination. Curr Opin Psychiatry, 19(5), 522–526. doi:10.1097/01.yco.0000238482.27270.5d

Subramaniam, M., Abdin, E., Vaingankar, J. A., Wong, K. E., & Chong, S. A. (2015). Comorbid physical and mental illnesses among pathological gamblers: Results from a population based study in Singapore. Psychiatry Research, 227(2–3), 198–205. doi:10.1016/j.psychres.2015.03.033

Sullivan, S. (1999). GPs take a punt with a brief gambling screen: Development of the Early Intervention Gambling Health Test (EIGHT Screen). Paper presented at the Culture and the gambling phenomenon. Proceedings of the 10th annual conference of the National Association of Gambling Studies Sydney.

Sullivan, S. (2011). Impulse control disorders in medical settings. In J. E. Grant & M. N. Potenza (Eds.), The Oxford Handbook of Impulse Control Disorders. US: Oxford University Press.

Sullivan, S., Arroll, B., Coster, G., Abbott, M., & Adams, P. (2000). Problem gamblers: Do GPs want to intervene? New Zealand Medical Journal, 113, 204–207.

Sullivan, S., McCormick, R., Lamont, M., & Penfold, A. (2007). Problem gambling: Patients affected by their own or another's gambling may approve of help from general practitioners. The New Zealand Medical Journal, 120, 1257.

Suurvali, H., Cordingley, J., Hodgins, D. C., & Cunningham, J. (2009). Barriers to seeking help for gambling problems: A Review of the empirical literature. Journal of Gambling Studies, 25(3), 407–424. doi:10.1007/s10899-009-9129-9

Tavares, H., Zilberman, M. L., Beites, F. J., & Gentil, V. (2001). Gender differences in gambling progression. Journal of Gambling Studies, 17(2), 151–159.

Temcheff, C. E., Derevensky, J. L., St-Pierre, R. A., Gupta, R., & Martin, I. (2014). Beliefs and attitudes of mental health professionals with respect to gambling and other high risk behaviors in schools. International Journal of Mental Health and Addiction, 12(6), 716–729.

The Allen Consulting Group, Problem Gambling Research and Treatment Centre, & the Social Research Centre. (2011). Social and economic impact study of gambling in Tasmania: Summary report. Hobart: Prepared for the Tasmanian Government Department of Treasury and Finance.

Thomas, A. C., & Staiger, P. K. (2012). Introducing mental health and substance use screening into a community-based health service in Australia: usefulness and implications for service change. Health and Social Care in the Community, 20(6), 635–644. doi:10.1111/j.1365-2524.2012.01079.x



Thomas, S. A., Jackson, A. C., Browning, C. J., & Piterman, L. (2010). A one-item screening tool for problem gambling for use in primary care settings. Monash University. Melbourne.

Thomas, S. A., Merkouris, S. S., Radermacher, H. L., Dowling, N. A., Misso, M. L., Anderson, C. J., & Jackson, A. C. (2011). Australian guideline for treatment of problem gambling: An abridged outline. The Medical Journal of Australia, 195(11), 664–665. doi:10.5694/mja11.11088

Thomas, S. A., Piterman, L., & Jackson, A. C. (2008). Problem gambling: What do general practitioners need to know and do about it? Medical journal of Australia, 189(3), 135–136.

Tiet, Q. Q., Finney, J. W., & Moos, R. H. (2008). Screening psychiatric patients for illicit drug use disorders and problems. Clinical Psychology Review, 28(4), 578–591. doi:10.1016/j.cpr.2007.08.002

Toce-Gerstein, M., Gerstein, D. R., & Volberg, R. A. (2009). The NODS-CLiP: A rapid screen for adult pathological and problem gambling. Journal of Gambling Studies, 25(4), 541–555. doi:10.1007/s10899-009-9135-y

Tolchard, B., Thomas, L., & Battersby, M. (2007). GPs and problem gambling: Can they help with identification and early intervention? Journal of Gambling Studies, 23(4), 499–506. doi:10.1007/s10899-007-9062-8

Toneatto, T., Brands, B., & Selby, P. (2009). A randomized, double-blind, placebo-controlled trial of naltrexone in the treatment of concurrent alcohol use disorder and pathological gambling. The American Journal on Addictions, 18(3), 219–225. doi:10.1080/10550490902787007

Tse, S., Hong, S.-I., Wang, C.-W., & Cunningham-Williams, R. M. (2012). Gambling behavior and problems among older adults: a systematic review of empirical studies. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 67(5), 639–652.

Victorian Government: Department of Human Services. (2007). Dual Diagnosis: Key directions and priorities for service development. Victoria: Author.

Volberg, R., Gerstein, D., Christiansen, E., & Baldridge, J. (2001). Assessing self‐reported expenditures on gambling. Managerial and Decision Economics, 22(1‐3), 77–96.

Volberg, R. A., Munck, I. M., & Petry, N. M. (2011). A quick and simple screening method for pathological and problem gamblers in addiction programs and practices. The American Journal on Addictions, 20(3), 220–227. doi:10.1111/j.1521-0391.2011.00118.x

Volberg, R. A., & Williams, R. J. (2011). Developing a brief problem gambling screen using clinically validated samples of at-risk, problem and pathological gamblers. Massacheuttes, U.S.A.: Gemini Research.

Volberg, R. A., & Williams, R. J. (2012). Developing a short form of the PGSI. Birmingham, U.K.: Gambling Commission.

Walker, M., Anjoul, F., Milton, S., & Shannon, K. (2006). A structured clinical interview for pathological gambling. Gambling Research: Journal of the National Association for Gambling Studies, 18(1), 39–68.

Waluk, O. R., Youssef, G. J., & Dowling, N. A. (2016). The relationship between problem gambling and attention deficit hyperactivity disorder. Journal of Gambling Studies, 32(2), 591–604. doi:10.1007/s10899-015-9564-8

Westphal, J. R., & Johnson, L. J. (2007). Multiple co-occurring behaviours among gamblers in treatment: Implications and assessment. International Gambling Studies, 7(1), 73–99. doi:10.1080/14459790601157905

Wieder, B. L., Lutz, W. J., & Boyle, P. (2006). Adapting integrated dual disorders treatment for inpatient settings. Journal of Dual Diagnosis, 2(1), 101–107. doi:10.1300/J374v02n01_12

Williams, R. J., & Volberg, R. A. (2010). Best practices in population assessment of problem gambling. . Guelph: Ontario Problem Gambling Research Centre.

Williams, R. J., Volberg, R. A., & Stevens, R. M. G. (2012). The popuation prevalence of problem gambling: Methodological influences, standardized rates, jurisdictional differences and worldwide trends. Ontario: Report prepared for the Ontario Problem Gambling Research Centre and the Ontario Ministry of Health and Long Term Care.



Winters, K. C., & Kushner, M. G. (2003). Treatment issues pertaining to pathological gamblers with a comorbid disorder. Journal of Gambling Studies, 19(3), 261–277.

Winters, K. C., Specker, S., & Stinchfield, R. (2002). Measuring pathological gambling with the Diagnostic Interview for Gambing Severity (DIGS). In J. J. Marotta, J. A. Cornelius, & W. R. Eadington (Eds.), The downside: Problem and pathological gambling (pp. 143–148). Reno, NV: University of Nevada Press.

Wood, K., & Anderson, J. (1995). The effect of hospital admissions of psychiatric case management involving general practictioners: Preliminary results. Australian and New Zealand Journal of Psychiatry, 29(2), 223–229.

Yakovenko, I., Quigley, L., Hemmelgarn, B. R., Hodgins, D. C., & Ronksley, P. (2015). The efficacy of motivational interviewing for disordered gambling: systematic review and meta-analysis. Addictive Behaviours, 43(11), 72–82. doi:10.1016/j.addbeh.2014.12.011

Young, M., Clark, C., Moore, K. A., & Barrett, B. (2009). Comparing two service delivery models for homeless individuals with complex behavioral health needs: Preliminary data From two SAMHSA treatment for homeless studies. Journal of Dual Diagnosis, 5(3–4), 287–304. doi:10.1080/15504260903359015

Zimmermann, A., Lubman, D. I., & Cox, M. (2012). Tobacco, caffeine, alcohol and illicit substance use among consumers of a national community managed mental health service. Mental Health and Substance Use, 5(4), 287–302. doi:10.1080/17523281.2012.703225



Appendix 1: Study 2a: Clinician survey

CLINICIAN GAMBLING QUESTIONNAIRE

Thank you for participating in this research. Please take some time to complete this questionnaire to give us an idea of your understanding of the relationship between mental health, problem gambling, screening and treatment. These questionnaires are confidential and anonymous and are intended for clinicians working in out-patient settings (if you work across in and out-patient settings, please respond to the questions as they relate to the out-patient setting).

GAMBLING AND MENTAL ILLNESS

To what extent do you agree with the following statements?

Strongly Agree

Agree Uncertain Disagree Strongly Disagree

9. Problem gambling and mental illness commonly occur together.

10. Problem gambling can worsen a client’s mental illness.

11. I understand what causes and/or maintains problem gambling issues.

DEMOGRAPHICS

1. Gender: Male Female

2. Age:_________ years

3. What is your profession? Medical Nurse Social Worker Psychologist OT

Other (please specify) _______________________

4. What type of service do you work at? Public Mental Health Service Private Mental Health Service Primary Health Care MHCSS/PDRSS AOD Service

5. What is the postcode of the service at which you work (primarily)? _ _ _ _ 6. How long have you been working clinically? _______ years

7. Have you had any previous training in problem gambling? No Yes (please specify:_______________________________________________)

8. What proportion of your caseload/practice do you estimate to have gambling problems? _ _ _%

SCREENING & ASSESSMENT

12. How often do you ask clients about their gambling? (Circle)

Never Rarely Sometimes Often Almost Always

13. How often do you screen for problem gambling? (Circle)




14. If you do screen/explore gambling problems, what method do you typically use? Informal discussion during appointment or interview

Using set questions in my services’ intake assessment

Using a standardised/formal gambling screen If so, which one? _____________________

15. If you don’t screen/explore gambling problems, why not?

_____________________________________________ ____________________________________________________________________________________________

16. How comfortable are you asking clients about their gambling behaviours?

Very comfortable Somewhat comfortable Somewhat uncomfortable Very uncomfortable

17. How confident are you in detecting/screening for problem gambling with your clients?

Very confident Moderately confident Somewhat confident Not confident

18. How confident are you in assessing for problem gambling with your clients?


To what extent do you agree with the following statements:

Strongly Agree


19. There is no point conducting gambling screening as my service does not treat problem gamblers.

20. Gambling is not really a clinical disorder.

21. Detecting problem gambling does not require a formal screen; it can just be addressed if a client mentions it.

22. Use of standardised screening tools is only necessary if a client mentions gambling.

23. People accessing mental health treatment do not want to be screened for gambling problems.

24. There are too many more important issues to screen for than problem gambling.

25. Problem gambling does not co-occur with mental health problems often enough to bother screening.

26. There is not enough time to conduct problem gambling screening or assessment in my workplace.

27. Screening/assessment and referral for problem gambling is not part of my job.

28. I am aware of what screening and assessment tools are available to me for detection of problem gambling.

29. It is important to identify gambling problems among mental health clients.

30. A brief problem gambling screen would be a useful part of my routine clinical practice.



31. When you identify a client with problem gambling, what do you do? (Please tick all that apply) Nothing Provide pharmacotherapy I have never identified a problem gambler Provide psychological treatment Conduct further assessment Address the financial or social consequences Refer to external gambling treatment provider Other (please specify): ____________________

32. In your experience, problem gambling is most commonly seen among which of the following psychiatric conditions? (Please tick the 4 most common)

Alcohol Use Disorder Drug Use Disorder Social Phobia Post-Traumatic Stress Disorder Eating Disorder Major Depression Generalized Anxiety Panic/Agoraphobia Psychosis Mania/bipolar Disorder Personality Disorders Other (please specify) ___

REFERRAL

33. How often do you refer clients with a gambling problem to other services for help with their gambling? (Circle)

Never Rarely Sometime

s Often

Almost Always

34. Do you know where you can refer clients with a gambling problem? Yes No

35. If you do refer, where do you refer them for further help (please tick all that apply)

Gambler’s Help services (face to face) Private addiction/gambling therapists Gambler’s Help Online Private psychologists/psychiatrists Gamblers Helpline Gamblers Anonymous/Peer support Alfred Hospital Gambling Service Financial Counselling Other (please specify) __________________________________

36. If you don’t refer clients identified as having a gambling problem, why not?

________________________________ ____________________________________________________________________________________________

37. How confident are you in referring clients for gambling treatment to appropriate services?


38. It is important to refer clients who experience problem gambling to specialist gambling agencies for further assistance. (Circle)

Strongly Agree

Agree Uncertain Disagre

e

Strongly Disagre

e

39. Gambling Help Services are not equipped to deal with clients with mental illness. (Circle)

Strongly Agree

Agree Uncertain Disagre

e

Strongly Disagre

e



THANK YOU FOR COMPLETING THIS QUESTIONNAIRE

Thank you so much for participating in this survey. If you would like to be entered in a draw to receive one of four Samsung Galaxy Tabs, please enter your email address below, remove this page from the questionnaire and give this page to the researcher present.

Email address: _______________________________________________________________________________

TREATMENT

40. How often do you (not your practice) treat clients with gambling problems? (Circle)


41. What treatments do you and/or your service/practice offer for problem gambling? (Please tick all that apply)

None Financial Counselling Assessment Counselling Peer support Financial Aid/Relief Medication Other (please specify __________________________________

42. How confident are you in treating a client’s gambling problem?

Very confident Moderately confident Somewhat confident

Not confident

To what extent do you agree with the following statements:

Strongly Agree


43. I have a good understanding about the Gambler’s Help service system and the programs available.

44. I understand the types of treatments that have proven helpful for problem gambling.

45. Mental health and problem gambling clinicians can effectively work together to support clients.

Any further comments:



Appendix 2: Study 2b: Detailed data tables of

clinician survey results

Table 24. Mental health illnesses clinicians commonly observe comorbid problem gambling

Total sample (N = 311)

In your experience, problem gambling is most commonly seen among which of the following psychiatric conditions?

n (%)

Alcohol use disorder 220 (70.7%)

Mania/Bipolar disorder 172 (55.3%)

Drug use disorder 139 (44.7%)

Personality disorders 126 (40.5%)

Major depression 125 (40.2%)

Generalised anxiety 77 (24.8%)

Psychosis 48 (15.4%)

Post-traumatic stress disorder 39 (12.5%)

Social phobia 23 (7.4%)

Other 18 (5.8%)

Panic/agoraphobia 4 (1.3%)

Eating disorder 3 (1.0%)

*Note that percentages do not total 100% because multiple responses could be selected

Table 25. Frequency clinicians ask patients about gambling, and screen for problem

gambling

Frequency

n (%)

Ask about gambling

(N = 311)

Screen for problem gambling

(N = 311)

Never 39 (12.5%) 71 (22.8%)

Rarely 86 (27.7%) 101 (32.5%)

Sometimes 117 (37.6%) 90 (28.9%)

Often 53 (17.0%) 35 (11.3%)

Almost always 13 (4.2%) 12 (3.9%)

Missing 3 (1.0%) 2 (0.6%)

Table 26. Methods or instruments clinicians use to screen for problem gambling

Screening method

(n = 238 clinicians who screen at least rarely)

n (%)

Informal discussion during appointment or interview

210 (88.2%)

Set questions in my service’s intake assessment 17 (7.1%)

Standardised/formal gambling screen 4 (1.7%)

Missing 7 (3.0%)



Table 27. Clinicians' level of comfort in asking patients about gambling behaviour


How comfortable are you asking clients about their gambling behaviours?

n (%)

Very uncomfortable 15 (4.8%)

Somewhat uncomfortable 39 (12.5%)

Somewhat comfortable 135 (43.4%)

Very comfortable 122 (39.2%)

Table 28. Clinicians' level of confidence in detecting, screening and assessing patients for

problem gambling

Level of confidence

n (%)

Detecting and screening for problem gambling

(N = 311)

Assessing patients for problem gambling

(N = 311)

Not confident 60 (19.3%) 75 (24.1%)

Somewhat confident 128 (41.2%) 124 (39.9%)

Moderately confident 107 (34.4%) 95 (30.5%)

Very confident 14 (4.5%) 14 (4.5%)

Missing 2 (0.6%) 3 (1.0%)

Table 29. Clinicians' current responses when patients experiencing problem gambling are

identified


When you identify a client with problem gambling, what do you do?

n (%)

Nothing 3 (1.0%)

Provide pharmacotherapy 9 (2.9%)

Other 230 (7.4%)

Never identified a problem gambler 50 (16.1%)

Provide psychological treatment 74 (23.8%)

Conduct further assessment 120 (38.6%)

Address financial or social consequences 139 (44.7%)

Refer to external gambling treatment provider 207 (66.6%)


Table 30. Frequency clinicians report referring patients experiencing problem gambling to

outside services


How often do you refer clients with a gambling problem to other services for help with their gambling?

n (%)

Never 56 (18.0%)

Rarely 128 (41.1%)

Sometimes 88 (28.3%)




Often 26 (8.4%)

Almost always 9 (2.9%)

Missing 4 (1.3%)

Table 31. Services to which clinicians report referring patients experiencing problem

gambling


If you do refer patients, where do you refer them to?

n (%)

Gamblers Helpline 154 (49.5%)

Gambler’s Help services (face to face) 153 (49.2%)

Financial Counselling 105 (33.8%)

Gambler’s Help Online 104 (33.4%)

Gamblers Anonymous/Peer support 69 (22.2%)

Private psychologists/psychiatrists 39 (12.5%)

Alfred Hospital Gambling Service 30 (9.6%)

Private addiction/gambling therapists 22 (7.1%)

Other 18 (5.8)


Table 32. Clinicians' level of confidence in referring patients to outside services for problem

gambling


How confident are you in referring clients for gambling treatment to appropriate services?

n (%)

Not confident 66 (21.2%)

Somewhat confident 121 (38.9%)

Moderately confident 82 (26.4%)

Very confident 40 (12.9%)

Missing 2 (0.6%)

Table 33. Clinicians' opinions about referring patients to outside agencies when problem

gambling is identified


It is important to refer clients who experience problem gambling to specialist gambling agencies for further assistance

n (%)

Strongly disagree 1 (0.3%)

Disagree 1 (0.3%)

Uncertain 29 (9.3%)

Agree 185 (59.5%)




Strongly agree 93 (30.0%)

Missing 2 (0.6%)

Table 34. Clinicians' opinion about gambling help services being equipped to deal with

patients with mental health illnesses


Gambling Help Services are not equipped to deal with clients with mental illness

n (%)


Disagree 106 (34.1%)

Uncertain 161 (51.8%)

Agree 18 (5.8%)


Missing 3 (1.0%)

Table 35. Frequency clinicians report treating patients for problem gambling


How often do you (not your practice) treat clients with gambling problems?

n (%)

Never 72 (23.2%)

Rarely 130 (41.8%)

Sometimes 86 (27.6%)

Often 17 (5.5%)

Almost always 1 (0.3%)

Missing 5 (1.6%)

Table 36. Type of treatment provided by clinicians and/or their service or practice


What treatment do you and/or your service/practice offer for problem gambling?

n (%)

Counselling 139 (44.7%)

Assessment 126 (40.5%)

None 62 (19.9%)

Financial counselling 51 (16.4%)

Other 47 (15.1%)

Medication 36 (11.6%)

Peer support 32 (10.3%)

Financial aid/relief 29 (9.3%)




Table 37. Clinicians' level of confidence in treating patients for problem gambling


How confident are you in treating a client’s gambling problem?

n (%)

Not confident 147 (47.3%)

Somewhat confident 95 (30.5%)

Moderately confident 54 (17.4%)

Very confident 8 (2.6%)

Missing 7 (2.2%)

Table 38. Clinicians' understanding of the external Gambler’s Help services and programs


I have a good understanding about the Gambler’s Help service system and the programs available

n (%)




Agree 61 (19.6%)


Missing 1 (0.3%)

Table 39. Clinicians' understanding of the types of treatments effective for problem

gambling


I understand the types of treatment that have proven helpful for problem gambling

n (%)




Agree 65 (20.9%)


Missing 1 (0.3%)

Table 40. Clinicians' level of agreement to the statement: Mental health and problem

gambling clinicians can work effectively together to support patients


Mental health and problem gambling clinicians can effectively work together to support clients

n (%)


Disagree 6 (1.9%)


Agree 175 (56.3%)


Missing 2 (0.6%)



Appendix 3: Study 3a: Patient survey

GAMBLING SURVEY

Please answer the following questions as well as you can to help us understand you and your needs. Identify the answers that best describe your situation. Please ask the researcher if you need some help.

SECTION 1: INFORMATION ABOUT YOU

The questions in this section ask you about you and your household. Your answers will give us an idea about what your current home and living situation is like.

1. What is your age?

2. How long have you been with the service?

Less than one month Less than three months Less than six months Less than one year More than one year

3. What is your post code?

4. What is your marital status? Married or de facto

Separated or divorced

Widowed

Never married (single)

5. What is your gender?

Male Female Other

6. Which of the following best describes your household?

Single person household (no children) Single with children still at home (including joint custody) Single with children not living at home Couple with no children Couple with children still at home Couple with children not living at home Group or shared household In some other arrangement

years



7. What type of accommodation do you live in? Own home Rental home Parents/family’s place Friend’s place Boarding house Caravan Public housing Shelter/refuge Supported accommodation/transitional housing No usual residence/homeless

8. What is your highest level of completed education? Primary Year 10 High School Year 12 High School TAFE University Other (please specify) ……………………………….

9. What is your current employment status? (tick all that apply)

In paid employment full time (35 hours/week or more) In paid employment part time/casual Primarily household duties Student Retired Looking for work Disability Support Pension Other Pension Unpaid voluntary worker Other Unsure

10. Could you please tell us your approximate fortnightly personal income before tax?

Less than $500 $500 – $799 $800 – $1,299 $1,300 – $1,599 $1,600 – $2,599 $2,600 or more Unsure

11. In what country were you born?

Australia Other (Please specify)

…..and If born overseas how many years have you lived in Australia?

years



12. What is the main language spoken at home? English Italian Greek Cantonese Arabic Mandarin Vietnamese Other (please specify)

13. Do you identify as Aboriginal or Torres Strait Islander? No Yes

14. Do you identify as part of an ethnic minority group? No Yes, please specify cultural or ethnic group

15. Do you own a mobile phone? No Yes Is it a smart phone (i.e. has internet access? ) No Yes

16. Where do you access the internet? (Please tick all that apply)

I access the internet ...

On my mobile phone On my tablet e.g. iPad On my personal computer In my household dwelling In my workplace At my place of study Elsewhere in the community

SECTION 2: INFORMATION ABOUT YOUR GAMBLING

The questions in this section relate to your gambling behaviour in the last 12 months. Some of the questions may not apply to you and may seem repetitive, but please try to be as accurate as possible.

1. Have you gambled at any point/time in the last 12 months? Gambling includes wagering on a race

or event, buying a lottery ticket, playing keno or playing cards at home – as well as playing the pokies or betting on sports.

Yes

No (Go to Question 16)



2. In the last 12 months, did you spend any money playing or betting on: (please tick all that apply)

YES

1. the pokies or electronic gaming machines?

2. horse or greyhound racing (including any bets at the Melbourne Cup, Spring Racing or on trackside virtual racing, but excluding all sweeps)?

3. casino table games (like blackjack, roulette and poker)?

4. sports (such as sports like AFL or cricket, but excluding fantasy sports)

5. events (such as election results, current affairs and TV shows)?

6. keno?

7. lotteries, Powerball or the Pools?

8. instant scratch tickets?

9. bingo?

10. informal private games (like playing cards at home)?

3. In the last 12 months, how many times per month have you spent any money playing or betting on

the pokies:

times per month

in a club, hotel or casino?

over the internet?




horse or greyhound racing:

times per month

at a racetrack, at an off-course venue

(e.g. TAB), or by telephone?

over the internet?


casino table games:

times per month

at a casino?

over the internet?


sports:

times per month

at a TAB, club, hotel, or casino?

over the internet?


events:

times per month

at a TAB, club, hotel, or casino?

over the internet?




keno:

times per month

at a club, hotel, newsagent or

Tattersall’s outlet?

over the internet?

9. In the last 12 months, how many times per month have you spent any money on lotteries, Powerball

or the Pools:

times per month

at a club, hotel, newsagent or

Tattersall’s outlet?

over the internet?


instant scratch tickets:

times per month

in a newsagent, or Tattersall’s outlet?

over the internet?


bingo:

times per month

in a club or hall?

over the internet?




informal private games (like playing cards at home)?

times per month

any location?

13. In the last 12 months, how much money, ON AVERAGE, did you spend betting on:

$ per month $ per month

(at venue) (over internet)

the pokies?

horse or greyhound racing?

casino table games?

sports?

events?

Keno?

lotteries, Powerball or the Pools?

instant scratch tickets?

bingo

informal private games (e.g. playing

cards at home?

14. In the last month how much money in total did you spend on gambling



15. The following questions relate to your gambling behaviour in the last 12 months. Thinking about

the past 12 months ...

Never Sometimes Most of the

time

Almost always

Have you bet more than you could really afford to lose?

Have you needed to gamble with large amounts of money to get the same feeling of excitement?

When you gambled, did you go back another day to try to win back the money you lost?

Have you borrowed money or sold anything to get money to gamble?

Have you felt that you might have a problem with gambling?

Has gambling caused you any health problems, including stress or anxiety?

Have people criticized your betting or told you that you had a gambling problem, regardless of whether or not you thought it was true?



Never Sometimes Most of the

time

Almost always

Has your gambling caused any financial problems for you or your household?

Have you felt guilty about the way you gamble or what happens when you gamble?

16. In the last 12 months:

No Yes

Have there been periods lasting 2 weeks or longer when you spent a

lot of time thinking about your gambling experiences or planning out

future gambling ventures or bets?

Have you tried to stop, cut down, or control your gambling?

Have you lied to family members, friends, or others about how much

you gamble or how much money you lost on gambling?

Have you gambled as a way to escape from personal problems?

Has there been a period when, if you lost money gambling one day,

you would return another day to get even?

Has your gambling caused serious or repeated problems in your

relationships with any of your family members or friends?

Have you felt the need to bet more and more money?

Have you had to lie to people important to you about how much you

gambled?

Have you become restless, irritable, or anxious when trying to stop

and (or) cut down on gambling?

Have you tried to keep your family or friends from knowing how

much you gambled?

Did you have such financial trouble as a result of gambling that you

had to get help with living expenses from family, friends, or welfare?



No Yes

Would you say you have been preoccupied with gambling?

Have you often gambled longer, with more money or more


Have you made attempts to either cut down, control or stop

gambling?

17. Please answer the following questions about whether you’ve been asked about gambling in this

service.

True

A mental health professional at this service asked me about gambling when I first

started attending the service.

A mental health professional at this service asked me about gambling after I’d

been attending the service for a while.

I was asked to fill out a survey about gambling when I first started attending the

service.

I was asked to fill out a survey about gambling after I’d been attending the

service for a while.

18. Changing your gambling ... On a scale of 1 to 10: (Please tick in the appropriate box)

1 2 3 4 5 6 7 8 9 10

How important is it for you to reduce or

stop gambling, (where 1 is not at all

important and 10 is very important?)

How confident are you that you could

reduce or stop gambling if you decided to,

(where 1 is not at all important and 10 is

very important?)

Where does reducing or stopping gambling

fit on your list of priorities, (where 1 is not

at all important and 10 is very important?)



19. If you were to develop an issue with your gambling, how likely is it you would ...

Very

Unlikely

Unlikely Neither

Likely

or

Unlikely

Likely Very

likely

Seek help from an online service

Telephone a gambling helpline

Talk to a gambling counsellor

face-to-face

Talk to a GP

Talk to mental health worker

Talk to family or friends about

the gambling

Use a gambling help app on your

mobile phone or tablet

Use a gambling self-help

program like an online program

or work book

Attend a support group for

gambling

Use strategies to help yourself,

like budgeting

Exclude yourself from a land

based or online gaming venue

SECTION 3: INFORMATION ABOUT YOUR HEALTH AND WELLBEING

The questions in this section ask you about your well being and your alcohol and drug use. Your

answers will help to give us a picture of how you’ve been feeling and what’s been happening for

you recently.

1. Have you ever been diagnosed with a mental health condition? (i.e. a problem that affects how a

person feels, thinks, behaves and interacts with other people)

Yes No (**Go to question 5)



2. Which of the following conditions have you EVER been diagnosed with? (Please tick all that apply)

Depression

Bipolar disorder or mania

Anxiety **If yes to go

Q.2b

A psychotic disorder, such as schizophrenia or

Schizoaffective disorder

An eating disorder, such as bulimia or anorexia


Alcohol abuse or dependence

Drug abuse or dependence

Gambling disorder

Other (please specify)

2b. Which of the following anxiety disorders have you EVER been diagnosed with? Panic disorder, panic attacks, or agoraphobia Generalised anxiety Obsessive compulsive disorder Social anxiety A phobia of a specific object Post-traumatic stress disorder (PTSD)

3. Which of the following conditions are you CURRENTLY diagnosed with (in the past year)?(Please

tick all that apply).

Depression

Bipolar disorder or mania

Anxiety **If yes to

go Q.3b

A psychotic disorder, such as schizophrenia or

Schizoaffective disorder

An eating disorder, such as bulimia or anorexia


Alcohol abuse or dependence

Drug abuse or dependence



Gambling disorder


3b. Which of the following anxiety disorders are you CURRENTLY diagnosed with?

Panic disorder, panic attacks, or agoraphobia Generalised anxiety Obsessive compulsive disorder Social anxiety A phobia of a specific object Post-traumatic stress disorder (PTSD)

4a. What medication types are you currently prescribed for your mental health condition?

I am not currently prescribed any medication for a mental health condition

Antidepressants e.g. venlafaxine (Effexor), fluoxetine (Prozac, Lovan), sertraline (Zoloft),

escitalopram (Lexapro), duloxetine (Cymbalta), mitazapine (Avanza)

Antipsychotics e.g. olanzapine (Zyprexa), resperidone (RisperDAL), clozapine (Clozaril, Clopine),

quetiapine (Seroquel), aripiprazole (Abilify), asenapine (Saphris)

Mood stabilisers e.g. sodium valproate (Eplilim), lithium (Lithicarb/Quilonum), carbomazepine

(Tegretol), lamotrigine (Lamictal)

Sleeping tablets e.g. zolpidem (Stilnox), zopiclone (Zimovane, Imovane), Temazepan (Temaze)

Benzodiazepines e.g. diazepam (Valium), alprazolam (Xanax), clonazepam (Rivotril)

Opioid substitution, e.g. buprenorphine (Suboxone), Methodone (Biodone).


4b Over the past 12 months how often have you taken your medication as prescribed?

Never

Rarely

Sometimes

Often

Always

4c Over the past 12 months have you shared these medications with other people?

Yes

No



5. How would you rate these aspects of your life where 0 = poor and 10 = good (please circle)? a. Psychological health?

0 1 2 3 4 5 6 7 8 9 10

Poor Fair Good Very Good Excellent

b. Physical health?

0 1 2 3 4 5 6 7 8 9 10


c. Overall quality of life

0 1 2 3 4 5 6 7 8 9 10


6. Please indicate the response that best describe your alcohol use in the past year.

a. How often do you have a drink containing alcohol? Never Monthly or less 2–4 times a month 2–3 times a week 4 or more times a week

b. How many standard drinks containing alcohol do you have on a typical ‘drinking day’? 1 or 2 3 or 4 5 or 6 7 to 9 10 or more

c. How often do you have six or more drinks on one occasion? Never Less than monthly Monthly Weekly Daily or almost daily

d. How much money (dollars$), on average, did you spend on alcohol per week in the past 12

months?



7. Please indicate the responses that best describe your smoking in the past 12 months. a. How often do smoke cigarettes? Never (skip to QUESTION 8) Monthly or less 2–4 times a month 2–3 times a week Daily or almost daily

b. How many cigarettes per day do you smoke? Less than 10 11–20 21–30 31 or more Unsure

c. How soon after you wake up do you smoke your first cigarette? Within 5 minutes Within 6–30 minutes Within 31–60 minutes After 60 minutes Unsure

d. How much money (dollars $), on average, do/did you spend on cigarettes per week?

8. How many times in the past 12 months have you used an illegal drug or used a prescription

medication for non-medical reasons? (Please write a number)

9. How many times in the past 12 months have you used these illegal drugs or prescription

medication for non-medical reasons?

Never Once or

twice Monthly Weekly Daily or

almost daily

Cannabis (marijuana, pot, grass,

hash, synthetic cannabis, etc.)

Cocaine (coke, crack, etc.)

Amphetamine type stimulants

(speed, meth, ice, diet pills,

ecstasy, etc.)

Inhalants (nitrous, glue, petrol,

paint thinner, etc.)



Never Once or

twice Monthly Weekly Daily or

almost daily

Sedatives or sleeping pills

(benzodiazepines, Xanax, Valium,

Serepax, etc.)

Hallucinogens (LSD, acid,

mushrooms, PCP, Special K, etc.)

Opioids (heroin, codeine,

morphine, methadone, etc.)

10. How much money (dollars $), on average, did you spend on these drugs per week in the past 12

months?

11. Which of these drugs gives you the MOST concern?

None of the above Alcohol Cannabis Cocaine Amphetamine type stimulants Inhalants Sedatives or sleeping pills Hallucinogens Opioids

12. Please indicate the responses that best describe your use of this drug.

Never/almost never

Sometimes Often Always/ almost always

Do/did you think your use of this

drug is/was out of control?

Does/did the prospect of missing a

fix or dose make you worried?

Do/did you worry about your use of

this drug?

Do/did you wish you could stop?



13. How difficult do/did you find it to stop or go without this drug?

Not difficult Quite difficult Very difficult Impossible Unsure

14. Would you like help to cut down or stop using this drug?

Yes No Unsure

THANK YOU VERY MUCH FOR YOUR TIME

PLEASE PASS THIS TO THE RESEARCHER TO RECEIVE YOUR VOUCHER



Appendix 4: Study 3a: Detailed data tables of

patient survey results

Table 41. Demographic characteristics of patient sample


Gender n (%)

Male 428 (50.9%)

Female 406 (48.3%)

Other 7 (0.8%)

Age group n (%)

16–24 169 (20.1%)

25–34 200 (23.8%)

35–44 188 (22.4%)

45–54 151 (18.0%)

55–64 106 (12.6%)

65–75+ 15 (1.8%)

Missing 12 (1.3%)

Country of birth n (%)

Australia 653 (77.6%)

Other 188 (22.4%)

Region of birth other than Australia n (%)

Europe 68 (36.2%)

Western Pacific 57 (30.3%)

South-East Asia 22 (11.7%)

Eastern Mediterranean 16 (8.5%)

Americas 9 (4.8%)

Africa 7 (3.7%)

Missing 9 (4.8%)

Language spoken at home n (%)*

English 770 (91.6%)

Unspecified 'Other' language 31 (3.7%)

Mandarin 12 (1.4%)

Arabic 8 (1.0%)

Greek 6 (0.7%)

Cantonese 5 (0.6%)

Italian 5 (0.6%)

Vietnamese 4 (0.5%)

Identify as Aboriginal and/or Torres Strait Islander n (%)

17 (2.0%)




Identify with an ethnic minority group n (%)

87 (10.3%)

Marital status n (%)

Single and never married 546 (64.9%)

Married or in a de facto relationship 166 (19.7%)

Separated or divorced 120 (14.3%)

Widowed 9 (1.1%)

Geographical region in which patients reside n (%)

Metropolitan 735 (87.4%)

Regional 81 (9.6%)

Missing 25 (3.0%)

Accommodation type n (%)

Rental home 297 (35.3%)

Parent's or family member's home 186 (22.1%)

Own home 183 (21.8%)

Public housing 80 (9.5%)

Supported accommodation or transitional housing 54 (6.4%)

Boarding house 31 (3.7%)

Friend's home 5 (0.6%)

No usual residence or homeless 3 (0.4%)

Caravan 1 (0.1%)

Missing 1 (0.1%)

Household type n (%)

Single person 241 (28.7%)

Group or shared 217 (25.8%)

Couple with children living at home 108 (12.8%)

Some other arrangement 90 (10.7%)

Single with children living at home 68 (8.1%)

Couple with no children 65 (7.7%)

Single without children living at home 27 (3.2%)

Couple without children living at home 25 (3.0%)

Highest level of education n (%)

University 223 (26.5%)

TAFE 173 (20.6%)

Year 12 210 (25.0%)

Year 10 163 (19.4%)

Primary 29 (3.4%)

Other 40 (4.7%)

Missing 3 (0.4%)




Employment n (%)

Full-time employment 104 (12.4%)

Part-time employment 141 (16.8%)

Household duties 8 (0.1%)

Student 99 (11.8%)

Retired 25 (3.0%)

Looking for employment 131 (15.6%)

Disability Support Pension 325 (38.6%)

Other Pension 46 (5.5%)

Volunteer 30 (3.6%)

Other 54 (6.4%)

Unsure 7 (0.8%)

Personal income per fortnight n (%)

Less than $500 170 (20.2%)

$500–$799 229 (27.2%)

$800–$1,299 208 (24.7%)

$1,330–$1,599 50 (6.0%)

$1,600–$2,599 63 (7.5%)

$2,600 or higher 32 (3.8%)

Unsure 89 (10.6%)

Type of service n (%)

Public Mental Health Service (Adult) 390 (46.4%)

Private Mental Health Service 303 (35.8%)

Public Mental Health Service (Child) 95 (11.1%)

Community Outreach Support Service 50 (6.7%)

Missing 3 (0.3%)

Length of treatment at this service n (%)

Less than one month 129 (15.3%)

Less than three months 88 (10.5%)

Less than six months 78 (9.3%)

Less than one year 107 (12.7%)

More than one year 438 (52.1%)

Missing 1 (0.1%)

Mobile phone ownership n (%)

773 (91.9%)

Smart phone ownership n (%)

599 (71.2%)

Method of accessing the Internet n (%)

Mobile phone 550 (65.4%)

Personal computer 464 (55.2%)




Household dwelling 275 (32.7%)

Tablet 202 (24.0%)

Elsewhere in the community 136 (16.2%)

Workplace 127 (15.1%)

Place of study 101 (12.0%)


Table 42. Lifetime mental health diagnoses self-reported by patients


Lifetime diagnoses (any) n = 780 (92.7%)

Specific lifetime diagnoses n (%)

Depression 540 (69.2%)

Anxiety 448 (57.4%)

Psychotic disorder 285 (36.5%)

Bipolar 181 (23.2%)

Drug abuse or dependence 107 (13.7%)

Borderline personality disorder 98 (12.6%)

Alcohol abuse or dependence 76 (9.7%)


Other disorders 59 (7.6%)

Gambling disorder 17 (2.2%)

More than two mental health disorders 547 (65.0%)

More than three mental health disorders 303 (36.0%)

More than four mental health disorders 150 (17.8%)

Specific lifetime anxiety diagnoses n (%)

Generalised anxiety disorder 348 (77.7%)

Panic disorder, panic attacks, agoraphobia 190 (42.4%)

Social anxiety 178 (39.7%)


Obsessive compulsive disorder 72 (16.1%)

Phobia 26 (5.8%)

Current diagnoses (any) in the past year n = 745 (88.6%)

Specific current diagnoses n (%)

Depression 460 (61.7%)

Anxiety 401 (53.8%)

Psychotic disorder 257 (34.5%)

Bipolar 144 (19.3%)

Borderline personality disorder 71 (9.5%)

Drug abuse or dependence 57 (7.7%)

Other disorders 52 (7.0%)




Alcohol abuse or dependence 42 (5.6%)


Gambling disorder 6 (0.8%)

More than two current mental health diagnoses 474 (88.6%)

More than three current mental health diagnoses 199 (56.4%)

More than four current mental health diagnoses 71 (23.7%)


Table 43. Specific current anxiety mental health diagnoses self-reported by patients


Specific current diagnoses n (%)

Generalised anxiety disorder 308 (76.8%)

Panic disorder, panic attacks, agoraphobia 149 (37.2%)

Social anxiety 140 (34.9%)


Obsessive compulsive disorder 47 (11.7%)

Phobia 20 (5.0%)


Table 44. Current medications patients report being prescribed for mental health conditions


Currently prescribed medication (past year) n = 684 (81.3%)

Antipsychotics 392 (46.6%)

Antidepressants 380 (45.2%)

Mood stabilisers 176 (20.9%)

Benzodiazepines 120 (14.3%)

Sleeping tablets 90 (10.7%)

Other medications 52 (6.2%)

Opioid substitution 16 (1.9%)


Table 45. Patients' self-reported compliance with currently prescribed medication for mental

health conditions


Frequency of taking medication as prescribed in the past year

n = 684

Never 3 (0.4%)

Rarely 4 (0.6%)

Sometimes 24 (3.5%)

Mostly 137 (20.0%)

Always 508 (74.3%)

Missing 8 (1.2%)



Table 46. Patients' self-reported participation in gambling activity in the past year


Participated in gambling activity (any) in the past year

n = 348 (41.4%)

Participation in specific gambling activity (past year)

n (%)

Playing pokies or electronic gaming 175 (50.3%)

Lotto, Powerball or the Pools 173 (49.7%)

Betting on horse racing or greyhounds 91 (26.1%)

Scratch tickets 88 (25.3%)

Betting on casino table games 48 (13.8%)

Betting on sports 42 (12.1%)

Keno 25 (7.2%)

Informal private betting 21 (6.0%)

Bingo 12 (3.4%)

Betting on events 5 (1.4%)

Frequency of participation in specific gambling activity: Physical venues

m times (SD)

Casino table games 70.4 (189.9)

Keno 70.2 (161.8)

Lotto, Powerball or the Pools 51.6 (119.2)

Pokies or electronic gambling 49.2 (114.6)

Betting on horse or greyhound racing 43.6 (128.7)

Betting on sports 39.2 (142.0)

Scratch tickets 34.2 (53.1)

Bingo 3.4 (10.4)

Betting on events 0 (0)

Frequency of participation in specific gambling activity at physical venue and online (past year)

Physical venue

m times (SD)

Online

m times (SD)

Casino table games 70.4 (189.9) 7.0 (35.0)

Keno 70.2 (161.8) 1.4 (7.2)

Lotto, Powerball or the Pools 51.6 (119.2) 7.8 (47.4)

Pokies or electronic gambling 49.2 (114.6) 14.1 (80.7)

Betting on horse or greyhound racing 43.6 (128.7) 21.0 (92.3)

Betting on sports 39.2 (142.0) 26.4 (65.6)

Scratch tickets 34.2 (53.1) 3.0 (19.9)

Bingo 3.4 (10.4) 1.0 (3.5)

Betting on events 0 (0) 9.6 (15.6)




Average spend on specific gambling activity at physical venue and online (past year)

Physical venue

m AUD (SD)

Online

m AUD (SD)

Casino table games $1, 823.45 ($2,930.12) $900.00 ($600.00)

Pokies or electronic gambling $1,679.10 ($2,838.12) $1,222.00 ($1,706.57)

Keno $1,297.67 ($3,414.36) $600.00 ($600.00)

Betting on horse or greyhound racing $1,274.11 ($2,481.15) $1,725.43 ($2,062.58)

Betting on sports $1,227.00 ($2,940.77) $833.20 ($884.61)

Bingo $714.00 ($780.92) $36.00 ($36.00)

Lotto, Powerball or the Pools $655.47 ($1,124.44) $717.55 ($1,238.63)

Scratch tickets $271.34 ($710.76) $1,440.00 ($1,539.87)

Betting on events $0 ($0) $330.00 ($127.28)

Table 47. Proportion of past year gamblers falling in to each risk category on the PGSI for

the whole sample


PGSI categories n (%) [95% CI]

Non-gamblers 493 (58.6%) [55.3–61.9]

Non-problem gamblers 165 (19.6%) [16.9–22.3]

Low-risk gamblers 60 (7.1%) [5.4–8.8]

Moderate-risk gamblers 70 (8.3%) [6.4–10.2]

Problem gamblers 53 (6.3%) [4.6–7.9]

Table 48. Gender breakdown of each of the gambling categories

Total sample (n = 834 patients who identified their gender as male or female)

PGSI categories Male

n (%)

Female

n (%)

Non-gamblers 230 (47.1%) 258 (52.9%)

Non-problem gamblers 83 (50.9%) 80 (49.1%)

Low-risk gamblers 35 (58.3%) 25 (41.7%)

Moderate-risk gamblers 47 (67.1%) 23 (32.9%)

Problem gamblers 33 (62.3%) 20 (37.7%)

Table 49. Age breakdown of gambling categories


Age group (n = 829 patients who indicated their age group)

Age 16–24

n (%)

Age 25–34

n (%)

Age 35–44

n (%)

Age 45–54

n (%)

Age 55–64

n (%)

Age 65+

n (%)

Non-gamblers (n = 489)

117

(23.9%)

118

(24.1%)

102

(20.9%)

82

(16.8%)

63

(12.9%)

7

(1.4%)




Age group (n = 829 patients who indicated their age group)

Age 16–24

n (%)

Age 25–34

n (%)

Age 35–44

n (%)

Age 45–54

n (%)

Age 55–64

n (%)

Age 65+

n (%)

Non-problem gamblers (n = 161)

27

(16.8%)

39

(24.2%)

32

(19.9%)

33

(20.5%)

23

(14.3%)

7

(4.3%)

Low-risk gamblers (n = 59)

13

(22.0%)

16

(27.1%)

12

(20.3%)

10

(16.9%)

7

(11.9%)

1

(1.7%)

Moderate-risk gamblers (n = 68)

9

(13.2%)

16

(23.5%)

23

(33.8%)

14

(20.6%)

6

(8.8%)

0

(0.0%)

Problem gamblers (n = 52)

3

(5.8%)

11

(21.2%)

19

(36.5%)

12

(23.1%)

7

(13.5%)

0

(0.0%)

Table 50. Country of birth and gambling categories


PGSI categories Born in Australia

n (%)

Born overseas

n (%)

Non-gamblers (n = 493) 369 (74.8%) 124 (25.2%)

Non-problem gamblers (n = 165) 138 (83.6%) 27 (16.4%)

Low-risk gamblers (n = 60) 49 (81.7%) 11 (18.3%)

Moderate-risk gamblers (n = 70) 55 (78.6%) 15 (21.4%)

Problem gamblers (n = 53) 42 (79.2%) 11 (20.8%)

Table 51. Language spoken at home and gambling categories


PGSI categories English speaking

n (%)

Non-English speaking

n (%)

Non-gamblers (n = 493) 439 (89.0%) 54 (11.0%)

Non-problem gamblers (n = 165) 161 (97.6%) 4 (2.4%)

Low-risk gamblers (n = 60) 56 (93.3%) 4 (6.7%)

Moderate-risk gamblers (n = 70) 65 (92.9%) 5 (7.1%)

Problem gamblers (n = 53) 49 (92.5%) 4 (7.5%)

Table 52. Ethnic identity and gambling category

Total sample (n = 833 patients who answered these questions)

PGSI categories Identified as part of an ethnic minority group

n (%)

Did not identify as part of an ethnic minority group

n (%)

Non-gamblers 55 (11.3%) 431 (88.7%)







Table 53. Aboriginal and Torres Strait Islander status and gambling category

Total sample (n = 840 patients who answered these questions)

PGSI categories Identified as part of Aboriginal or Torres Strait

Islander

n (%)

Did not identify as part of Aboriginal or Torres Strait

Islander

n (%)

Non-gamblers 11 (11.3%) 482 (88.7%)





Table 54. Marital status and gambling category


PGSI categories Married/

de facto

n (%)

Separated/

divorced

n (%)

Never married

n (%)

Widowed

n (%)

Non-gamblers (n = 493) 89 (18.1%) 70 (14.2%) 329 (66.7%) 5 (1.0%)

Non-problem gamblers (n = 165) 46 (27.9%) 29 (17.6%) 89 (53.9%) 1 (0.6%)

Low-risk gamblers (n = 60) 13 (21.7%) 6 (10.0%) 40 (66.7%) 1 (1.7%)

Moderate-risk gamblers (n = 70) 12 (17.1%) 8 (11.4%) 49 (70.0%) 1 (1.4%)

Problem gamblers (n = 53) 6 (11.3%) 7 (13.2%) 39 (73.6%) 1 (1.9%)

Table 55. Highest level of education achieved and gambling category


PGSI categories Primary

n (%)

Year 10

n (%)

Year 12

n (%)

TAFE

n (%)

University

n (%)

Other

n (%)

Non-gamblers (n = 490) 16 (3.3%) 84

(17.1%) 124

(25.3%) 100 (20.4%) 141 (28.8%) 25 (5.1%)

Non-problem gamblers (n = 165) 6 (3.6%) 29

(17.6%) 40

(24.2%) 37 (22.4%) 45 (27.3%) 8 (4.8%)

Low-risk gamblers (n = 60) 3 (5.0%) 14

(23.3%) 20

(33.3%) 11 (18.3%) 10 (16.7%) 2 (3.3%)

Moderate-risk gamblers (n = 70) 2 (2.9%) 19

(27.1%) 13

(18.6%) 16 (22.9%) 17 (24.3%) 3 (4.3%)

Problem gamblers (n = 53) 2 (3.8%) 17

(32.1%) 13

(24.5%) 9 (17.0%) 10 (18.9%) 2 (3.8%)

Table 56. Current employment status and gambling category


PGSI categories Employed or studying

n (%)

Neither employed nor studying

n (%)

Unsure

n (%)

Non-gamblers (n = 493) 184 (37.3%) 305 (61.9%) 4 (0.8%)

Non-problem gamblers (n = 165)

73 (44.2%) 92 (55.8%) 0 (0.0%)




PGSI categories Employed or studying

n (%)

Neither employed nor studying

n (%)

Unsure

n (%)

Low-risk gamblers (n = 60) 25 (41.7%) 35 (58.3%) 0 (0.0%)

Moderate-risk gamblers (n = 70) 20 (28.6%) 49 (70.0%) 1 (1.4%)

Problem gamblers (n = 53) 14 (26.4%) 39 (73.6%) 0 (0.0%)



Table 57. Lifetime mental health diagnoses and gambling category

Any Depression Anxiety Psychotic

disorder Eating disorder

Bipolar disorder

Borderline Personality disorder

Alcohol dependence disorder

Drug dependence disorder

Gambling disorder

Other (not specified)

Non-gambler

(n = 493)

457

(92.7%)

320

(64.9%)

265

(53.8%)

160

(32.5%)

44

(8.9%)

97

(19.7%)

60

(12.2%)

41

(8.3%)

63

(12.8%)

6

(1.2%)

29

(5.9%)

Non-problem gambler

(n = 165)

151

(91.5%)

105

(63.6%)

90

(54.5%)

42

(25.5%)

8

(4.8%)

36

(21.8%)

13

(7.9%)

12

(7.3%)

12

(7.3%)

0

(0.0%)

18

(10.9%)

Low-risk gambler

(n = 60)

55

(91.7%)

37

(61.7%)

33

(55.0%)

24

(40.0%)

3

(5.0%)

14

(23.3%)

8

(13.3%)

6

(10.0%)

6

(10.0%)

1

(1.7%)

5

(8.3%)

Moderate-risk gambler

(n = 70)

67

(95.7%

43

(61.4%)

31

(44.3%)

35

(50.0%)

8

(11.4%)

16

(22.9%)

8

(11.4%)

10

(14.3%)

13

(18.6%)

4

(5.7%)

6

(8.6%)

Problem gambler

(n = 53)

50

(94.3%)

35

(66.0%)

29

(54.7%)

24

(45.3%)

4

(7.5%)

18

(34.0%)

9

(17.0%)

7

(13.2%)

13

(24.5%)

6

(11.3%)

1

(1.9%)



Table 58. Current mental health diagnoses and gambling category

Any Depression Anxiety Psychotic

disorder Bipolar disorder

Eating disorder

Borderline Personality disorder

Alcohol dependence disorder

Drug dependence disorder

Gambling disorder

Other (not specified)

Non-gambler

(n = 493)

439

(89.0%)

280

(56.8%)

240

(48.7%)

141

(28.6%)

79

(16.0%)

24

(4.9%)

45

(9.1%)

21

(4.3%)

29

(5.9%)

3

(0.6%)

439

(89.0%)

Non-problem gambler

(n = 165)

143

(86.7%)

85

(51.5%)

76

(46.1%)

39

(23.6%)

30

(18.2%)

3

(1.8%)

8

(4.8%)

6

(3.6%)

6

(3.6%)

0

(0.0%)

143

(86.7%)

Low-risk gambler

(n = 60)

50

(83.3%)

32

(53.3%)

31

(51.7%)

21

(35.0%)

8

(13.3%)

0

(0.0%)

5

(8.3%)

5

(8.3%)

5

(8.3%)

0

(0.0%)

50

(83.3%)


(n = 70)

64

(91.4%)

37

(52.9%)

28

(40.0%)

33

(47.1%)

12

(17.1%)

2

(2.9%)

5

(7.1%)

6

(8.6%)

7

(10.0%)

0

(0.0%)

64

(91.4%)

Problem gambler

(n = 53)

49

(92.5%)

26

(49.1%)

26

(49.1%)

23

(43.4%)

15

(28.3%)

0

(0.0%)

8

(15.1%)

4

(7.5%)

10

(18.9%)

3

(5.7%)

49

(92.5%)



Table 59. Average monthly spend on gambling activity (any) for patients in each gambling

harm category

Average monthly spend on gambling activities $AU (n = 342 gamblers who reported monthly spend)

m SD Mode Median

Non-problem gamblers

(n = 162)

$39.12 $106.83 $20.00 $10.00

Low-risk gamblers

(n = 60)

$50.32 $79.50 $20.00 $11.00

Moderate-risk gamblers

(n = 68)

$123.83 $282.72 $50.00 $43.50

Problem gamblers

(n = 52)

$439.79 $672.11 $1000.00 $200.00



Table 60. Proportion of gamblers in each risk category who reported being asked about or screened for gambling problems by a clinician

n = 348 gamblers

Asked about gambling by clinician Screened for gambling problems by clinician

First started attending Later First started attending Later

Non-gambler

(n = 493) 87 (17.6%) 30 (6.1%) 43 (8.7%) 72 (14.6%)

Non-problem gambler

(n = 165) 35 (21.2%) 17 (10.3%) 16 (9.7%) 30 (18.2%)

Low-risk gambler

(n = 60) 16 (26.7%) 5 (8.3%) 8 (13.3%) 17 (28.3%)


(n = 70) 21 (30.0%) 16 (22.9%) 11 (15.7%) 14 (20.0%)

Problem gambler

(n = 53) 24 (45.3%) 17 (32.1%) 11 (20.8%) 16 (30.2%)

vicrgf

responsiblegambling


ResponsibleGambling

Our vision: A Victoria free from gambling-related harm

© Victorian Responsible Gambling Foundation, July 2017

This publication is licensed under a Creative Commons Attribution 3.0 Australia licence. The licence does not apply to any images, photographs, branding or logos.

This report has been peer reviewed by two independent researchers. For further information on the foundation’s review process of research reports, please see responsiblegambling.vic.gov.au.

This project was funded by the Victorian Responsible Gambling Foundation through the Grants for Gambling Research Program.

For information on the Victorian Responsible Gambling Foundation Research Program visit responsiblegambling.vic.gov.au.

DisclaimerThe opinions, findings and proposals contained in this report represent the views of the authors and do not necessarily represent the attitudes or opinions of the Victorian Responsible Gambling Foundation or the State of Victoria. No warranty is given as to the accuracy of the information. The Victorian Responsible Gambling Foundation specifically excludes any liability for any error or inaccuracy in, or omissions from, this document and any loss or damage that you or any other person may suffer.

Conflict of interest declarationThe authors declare no conflict of interest in relation to this report or project.

To cite this reportLubman, D, Manning, V, Dowling, N, Rodda, S, Lee, S, Garde, E, Merkouris, S & Volberg, R 2017, Problem gambling in people seeking treatment for mental illness, Victorian Responsible Gambling Foundation, Melbourne.

EnquiriesRosa Billi +61 3 9452 2625 [email protected]

Victorian Responsible Gambling FoundationLevel 6, 14–20 Blackwood StreetNorth MelbourneVictoria 3051

PO Box 2156Royal Melbourne HospitalVictoria 3050

Telephone: +61 3 9452 2600Facsimile: +61 3 9452 2660ABN: 72 253 301 291

See Inside gambling for the latest information, evidence and expert opinion on gambling issues

in Australia and overseas.

From betting on the races with Dad to professional punter: my journey

Teenage gambling: a business that’s only getting riskier

Helping people in their own language

Five minutes with...

Community educator, David Clark

Online magazine

insidegambling.com.au

Hear from:• public health experts • gambling sector professionals • people with firsthand experience

of gambling problems

Let us know what you’d like to read

If you have ideas for themes or stories for Inside gambling, we would love to hear from you.

Go to: insidegambling.com.au/contact-us

Subscribe to Inside gambling


CONNECT WITH US ON:vicrgf

responsiblegambling


ResponsibleGambling


July 2017


Problem gam

bling in people seeking treatment for m

ental illness


RESEARCH REPORT

July 2017

RESEARCH REPORT

problem gambling · problem gambling in people seeking treatment for mental illness lubman et al....

Documents