hsrpp 2015 ijpp pain

Investigating the quality and safety ofdischarge prescriptions in mental healthhospitals: a prospective multicentre study

R.N. Keersa,b, S.D. Williamsa,c,J.J. Vattakatucheryd,e, P. Brownf, J. Millerg,L. Prescottd and D.M. Ashcrofta,b

aManchester Pharmacy School, University of Manchester, bNIHR PrimaryCare Patient Safety Translational Research Centre, Manchester,cUniversity Hospital of South Manchester NHS Foundation Trust,d5 Boroughs Partnership NHS Foundation Trust, Warrington, eUniversityof Liverpool, fManchester Mental Health and Social Care NHS Trust andgGreater Manchester West Mental Health NHS Foundation [email protected]

While research suggests that miscommunication and medica-tion errors associated with discharge communications placepatients at risk of harm,[1] the frequency and nature of problemsassociated with writing discharge prescriptions in UK mentalhealth hospitals have not been fully explored. This study aimedto investigate the safety and quality of inpatient discharge pre-scriptions in three mental health NHS trusts in the North Westof England.

Trained pharmacy teams at each study site prospectivelyscreened discharge prescriptions for acute adult and elderlyinpatients over a 6-week period between February–March2014. Any errors in prescribing, clerical details (e.g. demo-graphic information) and communication of medicationsstopped (that were prescribed prior to admission) were rec-orded on a standardised form. A multidisciplinary panelassessed each prescribing error (PE) to confirm its presence,nature and potential severity. Those PEs having potentially sig-nificant, serious or life-threatening consequences for patientswere considered clinically relevant.[2] Logistic regressionanalyses were conducted using STATA® v13 software todetermine the potential predictors of identified errors. Thestudy was approved by the University of Manchester ResearchEthics Committee (Ref. 13279) and audit committees at eachstudy site.

A total of 274 discharge prescriptions were screened duringthe study period, containing 1456 newly prescribed or omittedmedication items. One fifth (20.8%) of discharges containingprescribed medications were affected by at least one PE, withan error rate per prescribed item of 5.08% (95% CI 4.07–6.33%). Nearly three quarters (73%) of PEs were considered tohave potentially clinically relevant consequences for patients.At whole discharge level, increasing polypharmacy was asso-

ciated with an elevated risk of PEs (compared with 0–5 items,6–10 items OR 2.50 (95% CI 1.18–5.30) p = 0.017, 11+ itemsOR 8.04 (95% CI 2.64–24.52) p < 0.01). General PractitionerTrainees and Core/Specialist Trainees (grouped together) weresignificantly more likely to make PEs when compared withtheir Foundation Year colleagues (whole discharge level OR3.62 (95% CI 1.17–11.18) p = 0.025, individual medicationitem level OR 2.80 (95% CI 1.12–6.95) p = 0.027). One ormore clerical errors affected 71.9% (95% CI 66.3–76.9%) ofdischarge prescriptions, and most frequently involved failingto or incorrectly specifying whether the patient’s GP or hospi-tal services should continue prescribing medications (43.4%).In total, 67.2% of prescriptions requiring communication ofmedications stopped during admission contained errors, andmost commonly involved central nervous system class medi-cines (83.8%).

Errors associated with mental health hospital discharge pre-scriptions are a common and important threat to patient safety.Particular emphasis should be placed on correct documenta-tion of clerical information and communication of changes toestablished medications as part of improvement efforts, asthese errors each affected more than two thirds of discharges inthis study. The finding that GP Trainee and Core/SpecialistTrainee prescribers appeared more likely to make PEs on dis-charge prescriptions than their junior colleagues also meritsfurther investigation. A potential limitation of this study wasthat the number of medication communication change errorsmay have been underestimated as they were limited to medi-cines taken prior to admission that were stopped during inpa-tient stay.

1. Kripalani S, LeFevre F, Phillips CO et al. Deficits in com-munication and information transfer between hospital-based and primary care physicians. JAMA 2007; 297: 831–841.

2. Keers RN, Williams SD, Vattakatuchery JJ et al. Preva-lence, nature and predictors of prescribing errors in mentalhealth hospitals: a prospective multicentre study. BMJOpen 2014; 4:e006084.

Identifying the challenges of maintaininga good safety culture in communitypharmacy using the Manchester PatientSafety Assessment Framework

C.E.L. Thomas, D.M. Ashcroft, D. Parker andD.L. Phipps

[email protected]

Safety culture has become firmly established within orga-nisational research and practice as a means for explai-ning and predicting safety-related behaviour (1). The aimof this study is to gain a deeper understandingof safety culture and how it is maintained in communitypharmacy.

Four community pharmacies, a total of 23 participants wererecruited between May and July 2014 in Greater Manchester.

bs_bs_banner

2 International Journal of Pharmacy Practice 2015; Supplement 1

International Journal of

Pharmacy Practice

IJPP 2015, Supplement 1: 2–27© 2015 The Authors.IJPP © 2015Royal Pharmaceutical Society

© 2015 The Authors. IJPP © 2015 Royal Pharmaceutical Society International Journal of Pharmacy Practice, 23 (Suppl. S1), pp. 2–27

Participants were recruited through local contacts and pharma-ceutical committees. In each pharmacy, focus groups were heldin which participants assessed their pharmacy’s safety cultureusing the Manchester Patient Safety Assessment Framework(MaPSaF) and then discussed the outcomes of their assessmentas a team. MaPSaF is based on the five-level model of safetyculture developed by Parker and Hudson (2). At the highestlevel of safety culture, risk management is seen as an integralpart of behaviour in the workplace. Researchers visited phar-macies on monthly follow-up visits for four months to observeand discuss their progress following the initial focus group.Verbatim transcripts from the focus groups were subjected totemplate analysis. This involved identifying a priori themesfrom the literature as well as themes emerging from the data tocreate a template that was then applied to each transcript.Ethical approval was obtained from the University of Man-chester Research Ethics Committee.

Seven common factors were identified as challenges tomaintaining safety culture in community pharmacy: firstly,though error reporting was common, knowing how to learnfrom errors was often perceived as challenging; following ruleswas identified as integral, however they were sometimesshown to be difficult to follow due to lack of flexibility; latentfactors, mainly differences in support offered by area manage-ment, led to variation in safety culture. Thirdly, the intensepressure to complete a large amount of tasks simultaneouslywithin limited time made maintaining a good safety culturesometimes challenging.

Communication was noted as a challenge, particularly forpharmacies that had varying shifts, especially when a lack ofwritten communication existed among the team. Task manage-ment decisions such as delegation and prioritisation were high-lighted as needing to be constantly balanced. Finally, keepingthe importance of patient safety in mind at all times was seen asfundamental to maintain a good safety culture. Each factor wasshown to vary between pharmacies, for example, flexibility ofrules was not as problematic for the independent pharmacywhen compared with the medium-sized chain.

Overall participants felt that being able to discuss the pharm-acy’s safety culture as a team to be a valuable experience. Ourfindings suggest a range of factors that need to be managedwithin a community pharmacy setting in order to maintain agood safety culture. We have found MaPSaF to be a helpful dis-cussion tool and a useful framework for assessing safetyculture in community pharmacy.

1. Choudhry RM, Fang D, Mohamed S. The nature of safetyculture: A survey of the state-of-the-art. Safety Science2007; 45(10): 993–1012.

2. Parker D, Hudson P. Understanding your culture. Man-chester: Shell International Exploration and Production.2001.

Modelling unplanned readmission risk inpatients admitted to an acutemedical unit

R. Millera,b, P. McCaffreya, C. Darcya, N. Blacka,A. Friela and K.Burnettb

aAltnagelvin Area Hospital, L’Derry and bUniversity of Ulster, [email protected]

Integrated Medicines Management (IMM) is now anaccepted standard of high quality pharmaceutical care onhospital wards [1]. However, patients admitted to the AcuteMedical Unit (AMU) are often in the hospital for too short atime to enable full implementation of IMM to fully addresstheir pharmaceutical care needs. This study aimed to developa model identifying patients admitted to AMU at higher riskof unplanned readmissions (suggestive of requiring baselinepreventative intensive pharmaceutical care); based on resultsthe team aims to then develop new care pathways for thesepatients during their stay in AMU and post-discharge fromhospital.

Both research ethical and trust governance permission wassought and granted for this study. Retrospective medical,demographic and socioeconomic data, together with 12 monthunplanned readmissions information, were collected for 500patients via computer held and chart records for patients(aged ≥ 40 years) with non-elective admissions to the AMU ofa 481-bed hospital, over a six-month period. Terminallyill patients in receipt of palliative care were excluded. Datawere coded and entered directly into the Statistical Packagefor the Social Sciences® version 20. The potential relationshipbetween dichotomised patient variables at baseline (affectingat least 5% of the population) and ≥1 unplanned readmissionsover 12 months was initially explored using chi-squared analy-sis with p ≤ 0.25 indicative of requiring further exploration [2].These variables were individually entered into logistic regres-sion models. The odds ratio (95% CI, Expβ) for each variablewas recorded. Those variables with a chi-squared p ≤ 0.05, andstatistically significant (p ≤ 0.05) logistic regression 95% CI,Exp β lower limit ≥1 were selected and entered into a backwardstepwise multivariate logistic regression to produce the finalmodel.

The 500 patients (249 female and 251 male, aged65.0 ± 14.5 years) included in data collection stayed in hospi-tal for an average of 7.3 days (range = 1–118 days); their 12month nonelective readmission rate was 29.6%. Forty-sevenvariables had a chi-squared p ≤ 0.25, with 16 of these alsohaving potentially significant predictive properties whenentered individually into logistic regression models. The finalpredictive readmission risk model included: patient drinksmore than 25 units of alcohol per day (95% CI, Exp β = 2.038–12.926, p = 0.001); patient uses a compliance aid (95% CI,Exp β = 1.948–16.951, p = 0.002); and patient has a history ofrespiratory disease (95% CI, Exp β = 1.107–3.208, p = 0.020).This three-variable model had 58.2% specificity, 69.3% sensi-tivity and 65.1% accuracy when the cut point was set to 0.4.The Hosmer & Lemeshow goodness-of-fit test for this modelyielded an acceptable p value of 0.667.

Oral abstracts 3


The team were aware of issues with patients with alcoholdependence being repeatedly readmitted to the AMU; the finalmodel confirms this anecdotal observation. The low figure interms of specificity suggests further data and additional vari-ables may need to be collected and explored. The model alsorequires validation via application to a further independentdata-set. Future work will address this model refinement andvalidation in order to produce a meaningful and workable pre-dictive model.

1. Scullin C, Hogg A, Luo R, Scott MG, McElnay JC.Integrated medicines management – can routine implemen-tation improve quality? J Eval Clin Prac 2012; 18(4):807–815.

2. Hosmer DW, Lemeshow S. Applied Logistic Regression.Wiley series in Probability and Mathematical Statistics.NewYork: John Wiley & Sons, 1989.

Where do they go? Destinationunknown: an exploratory study of thedisposal of transdermal drug patches inthe private healthcare sector (UK)

L. Breena, H. Zamanb, A. Mahmoodb, W. Nabib,F. Mansooralib, Z. Patelb, M. Aminb andA. Nasimb

aBradford University School of Management, University of Bradford, UKand bBradford School of Pharmacy, School of Life Sciences, University ofBradford, [email protected]

The effective disposal of medication and more specificallyaccidental exposure to fentanyl via transdermal patches hasrecently been highlighted in two key documents [1, 2]. Whilethe volume of unused medicines cost the NHS over £300million every year [1], the volume of transdermal patch wasteis unknown. There is a need for greater pharmacy interventionin the effective disposal of medicines to resolve issues such ashospital (re)-admissions, stockpiling leading to patient self–prescribing/dosing, and land and water pollution. The aim ofthis study was to examine transdermal patch disposal systemsand practice amongst private sector care providers in the UK.This was part of a larger study focusing on transdermal patchapplication.

A semi-structured qualitative questionnaire was posted torespondents qualified to administer transdermal patches in carehomes, hospice and private hospitals. This was initially pilotedon three care home respondents. Post-alteration based on thepilot feedback, a final questionnaire with 21 questions wasdeployed focusing on: patch application and administration,training and advice, disposal methods and policies plus associ-ated training and the impact of action taken. An audit was notconsidered appropriate at this stage until the scale of the dis-posal issue was known. Data were recorded from both openand closed form questions (Likert scale and yes/no answers).Descriptive statistics were used to analyse the data using SPSS.

Ethical approval was obtained from the University of BradfordEthics Committee.

A total of 488 questionnaires were deployed to staff in 34healthcare settings (6 private hospitals, 7 hospices and 21 carehomes). Sixty-one questionnaires (12.3%) were returned: 40from hospices (65.6%), 10 from care homes (16.4%), 8 fromindependent hospitals (13.1%) and 3 from home carers (4.9%).When commenting on drug patch disposal 17% of respondentsasserted that they always returned this waste to pharmacywhile 54% of respondents stated that they either rarely or neverdid this. 74% of respondents stated that medicines disposal waspart of their job but paradoxically only 30% of respondentsstated that they had undergone training for this. Access andknowledge of governing policy was fragmented with 7 out of61 respondents saying they had no knowledge of appropriatepolicies and only 11 respondents reported offering guidance onpatch disposal to patients.

While this study did not aim to investigate drug patch dis-posal in relation to any one institutional type, the outputs indi-cate that there are common issues such as lack of guidance andtraining in the private hospitals, hospices and care homesexamined. Where there should be uniformity and consistencyin practice across the sector the results indicate that there aremixed practices and disjointed knowledge concerning the dis-posal of transdermal patches and this is worrying. In conclu-sion, the issue of medicines disposal generally continues toexist on the improvement agenda of healthcare professionals inthis era of austerity. To inform clear guidance and policy it isproposed that a multi-institutional in-depth audit of transder-mal patch disposal practice be undertaken under the supervi-sion of trained pharmacy staff.

1. The Leeds Teaching Hospitals NHS Trust. Help Us toReduce Waste. http://www.bringyourmedicinesleeds.nhs.uk/help-us-to-reduce-waste/ (accessed 08 December2014).

2. Medicines Health Regulatory Agency. TransdermalFentanyl ‘Patches’: Reminder of Potential for Life-Threatening Harm from Accidental Exposure, Particularlyin Children. http://www.mhra.gov.uk/Safetyinformation/DrugSafetyUpdate/CON432900 (accessed 22 September2014).

We push, pump and drip-drop, but whatare our patients getting? Observation ofintravenous medication preparation andadministration practices

M. Ronan and C. Kirke

Pharmacy Department, Tallaght Hospital, [email protected]

Medication administration via the intravenous (IV) route isknown to be hazardous1, with high preparation and administra-tion error rates noted in observational studies2. This studyaimed to determine:



1. if IV medicines are prepared and administered in accord-ance with hospital guidelines and

2. the proportion of the prescribed dose received by patientswith current infusion practices.

Robert Gordon University Research Ethics Committee(REC) approval and hospital REC Chairman’s approval weregranted.

A paper-based data collection form based on reported obser-vational studies2,3 categories was piloted, adjusted and final-ised. A single pharmacist observed a convenience sample of IVdrug preparation and administration meeting study inclusioncriteria during a two-hour data collection period per day on 14adult medical/surgical in-patient wards over six weeks in Feb-ruary and March 2014. Participants were observed once onlyand gave informed consent prior to observation. Data (partici-pant demographics, preparation and administration details)were recorded on the ward and observations subsequentlychecked against hospital guidelines. Data were analysed usingSPSS and descriptive statistics compiled.

105 IV drug preparations and 66 IV drug administrations bynurses were observed.

Hospital guidelines were referred to in 40% (42/105) ofpreparations.

The volume or type of reconstitution or diluent fluid differedfrom guidelines in 21% of preparations (22/105).

An incorrect dose was prepared on 3 occasions but identifiedby the checking nurse and corrected.

73% (48/66) of administrations were outside recommendedrates. 65% of IV injections were more than 100% faster and36% of gravity infusions more than 50% slower than recom-mended (Table 1).

Table 1 Deviation from rate of infusion/injection recommended in hospi-tal guidelines

Deviation fromrecommendedrate

Gravity(n = 28)

Volumetricinfusionpump(n = 6)

Volumetricpumpwith safetysoftware(n = 6)

Slowinjection(n = 26)

No deviation 8 3 5 21–25% faster 41–25% slower 226–50% faster 1 226–50% slower 5 2 151–75% faster 2 1 151–75% slower 476–100% faster76–100% slower 1>100% faster 17>100% slower 5

Medication remained in the giving set and/or infusion bag atthe end of the infusion and was discarded, resulting in under-dosing in 95% of observed infusions (Table 2). This was in linewith hospital guidelines.

Table 2 Percentage of prescribed dose administered to patient

Percentage ofprescribed doseadministered

Gravity(n = 28)

Volumetricpump(n = 6)

Volumetricpump withsafetysoftware(n = 6)

Slowinjection(n = 26)

60–70% 270–80%80–90% 14 4 5 290–95% 10 2 195–100% 1 24Greater

than 100%1

Medication was prepared and administered in line with guide-lines in 21% of observations (8/66). Under-dosing due to medi-cation remaining in the line and/or bag occurred with 95% ofinfusions. Changes to hospital policy, practice and training arerequired to ensure the entire IV giving set is flushed to admin-ister of the full dose and to ensure administration at the recom-mended rate.

1. Berdot S, Sabatier B, Gillaizeau F, Caruba T, Prognon P,Durieux P. Evaluation of drug administration errors in ateaching hospital. BMC Health Services Research 2012;12(1):60.

2. Taxis K, Barber N. Ethnographic study of incidence andseverity of intravenous drug errors. BMJ 2003;326(7391):684.

3. Allan E, Barker K. Fundamentals of medication errorresearch. American Journal of Health-System Pharmacy1990; 47(3):555–571.

Nurse perceptions of medicineadministration in Parkinson’s disease

K. Lefteria, F. Liua and M. Browna,b

aUniversity of Hertfordshire and bMedPharm [email protected]

There are 376,250 older people living in care homes inEngland [1]. Drug therapy in this population is often challeng-ing due to altered age-related physiological functions, the pres-ence of several disease conditions, including Parkinson’sdisease (PD) and decline in cognitive capacity. PD brings aboutits own unique challenges as it is characterised by symptomsincluding tremor, rigidity and dysphagia and the complexity ofthe dosing regimen involves patients taking an average of 8.6tablets per day [2]. The aim of the research was to identifycurrent issues with the administration of medicines to PDpatients and to explore nurse perceptions on the developmentof appropriate drug delivery systems which could give tailoredsolutions to these issues.

Local nursing homes (206) were identified using yell.comand a sample of 32 was created using a random number genera-tor. Each home manager was telephoned and asked to nominate

Oral abstracts 5


a registered nurse from their staff to participate in a 60-minutein-depth semi-structured face to face interview with theresearch team. Details of the nurse’s experience of administer-ing medicines to PD patients and current methods used to over-come any issues experienced were explored. Current availableformulations were discussed and features of their “ideal” PDformulation considered. The interviews were recorded andtranscribed, coded and then analysed using inductive thematicanalysis to identify key themes. Ethics approval for this studywas obtained from the UH Research Ethics Committee.

All nursing homes contacted agreed to participate in thestudy with thirty one female and one male nurses being inter-viewed. The average duration of experience as a registerednurse was 15.5 years (range 0.5–40 years). The average sizenursing home in which they were employed was 46 patients(range 19–95). All interviewees had experience with adminis-tering medicines to PD patients and were currently caring for atleast two PD patients.

PD patients were mainly prescribed oral tablets with limiteduse of alternative formulations. Nurses reported that thiscaused issues for medicine administration and two-thirds hadresorted to crushing tablets to increase compliance. Of thosethat had not done this in practice the majority said that theywould consider doing this if necessary. Nurses considered thata lack of liquid formulations and the prescriber’s reluctance tochange to more costly alternatives were major contributingfactors to this problem. The inconvenient dosing regimens ofPD medicines were also reported as an issue.

Nurses considered an ideal solution to the issues would bethe availability of more liquid formulations or by avoiding theoral route completely e.g. a transdermal patch. Preferences fora reduction in dosing frequency and the need for new prepara-tions to be cost-effective were acknowledged.

Overall, the study has shown that current oral formulationsavailable for PD disease can cause issues and manipulation ofthese is required for administration. The alternatives arelimited and many are not being prescribed due to cost issues.Nurses consider either liquid formulations or drug delivery bythe transdermal route to be the most appropriate for PDpatients. However, development of new drug delivery systemswill need to consider both ease of administration to the patientand cost-effectiveness.

[1] The Adult Social Care Market and the Quality of Services,Care Quality Commission, November 2010, Figure 7:Places in homes for older people by population aged 65and over, 7.

[2] Grosset KA, Bone I, Grosset DG. Suboptimal medicationadherence in Parkinson’s disease. Mov Disord 2005;20(11): 1502–1507.

Recently registered pharmacytechnicians’ views on their education andtraining experiences in community andhospital pharmacy

S.D. Jee, E.I. Schafheutle and S. Willis

Manchester Pharmacy School, The University of [email protected]

Prior to registration, pharmacy technicians must completelevel-3 knowledge- and competence-based qualifications overtwo years, using a Further Education (FE) college, distanceprovider or approved NHS education centre (competencequalification), and complete work-based training typically inhospital or community pharmacy.1 Given the paucity ofresearch in the area, a programme of work was undertaken tobetter understand the quality of pharmacy technician educationand training. This paper aims to examine differences betweentrainee experiences in the two main sectors of community andhospital.

After piloting with six pharmacy technician trainees, acensus survey was administered by post and electronically(using Qualtrics (Provo, UT)) to 1457 recently registered (02/13–02/14) pharmacy technicians with four reminders. Thequestionnaire was informed by salient findings from qualita-tive interviews with education providers and employers.2 Itcontained five sections requesting views/information on:knowledge qualification, competence qualification, work-based training experience, open comments, and about you.Specific questions asked about content and support traineesreceived from education providers and employers, how thequalifications were funded, satisfaction levels with qualifica-tions and work experience, and time taken to complete these.Most questions used a forced-choice format, including Likert-type options. Comparative statistical analyses between sectorsconsisted of Mann–Whitney U, which compared mean ranks(MR), and chi-square tests using IBM SPSS v20. ManchesterUniversity ethics committee granted approval.

Six hundred and thirty-two usable responses were received(43.4%). The majority of respondents (550; 88%) were femaleand the average age was 35.26 ± 10.22. The majority (79.3%)of respondents were white British. Four hundred and seventy-five respondents (75.9%) trained in community and 133(21.3%) in hospital. Most of those that trained in community(93%) used distance providers for their knowledge and compe-tence qualification; most hospital trainees used an FE collegefor their knowledge qualification (78%) and an approved NHScentre for their competence qualification (57%). Statisticalanalyses showed a number of differences between respondentsthat trained in community and hospital. Those that trained incommunity were significantly more likely to contribute orprovide full funds for undertaking knowledge and competencequalifications than those in hospital (χ2 (4, N = 606) = 21.140,p < 0.001). Respondents that trained in hospital felt more sup-ported by their employing organisation (MR = 349 vs. 291,p = 0.001) and colleagues (MR = 345 vs. 292, p = 0.001);those in community felt more isolated (MR = 320 vs. 246,p < 0.001). Those that trained in hospital also felt more



satisfied, overall, with their experience in the workplace duringtraining (MR = 363 vs. 288, p < 0.001). Trainees in hospitalreceived more study time than those in community (MR = 387vs. 280, p < 0.001) and were more likely to complete qualifica-tions in two years or less, whereas those in community tooklonger (χ2 (3, N = 625) = 106.654, p < 0.001).

Findings illustrate the views of recently registered phar-macy technicians and the differences in experiences that existbetween those that trained in different sectors. It is importantthat the views of pharmacy technicians are heard whenmaking any revisions to education and training standardsand regulations in how pharmacy technician education andtraining is undertaken to ensure consistency across differentsettings.

1. General Pharmaceutical Council. Criteria for registrationas a pharmacy technician. 2013. London: General Pharma-ceutical Council.

2. Jee S, Willis S, Pritchard A, Schafheutle E. Insights into theprovision of Pharmacy Technician qualifications in GreatBritain. Presented at APTUK Annual Professional Confer-ence & Exhibition 2014: Patients First. 2014. AstonUniversity, Birmingham.

A pilot study of a multidisciplinary clinicalpain programme provided by the GoldCoast Medicare Local

M.A. Kinga, A. Sava and J. McSwanb

aGriffith University and bGold Coast Medicare Local and [email protected]

Chronic pain is experienced by almost 20% and interfereswith the daily activities of over 10% of adults living in Aus-tralia.1 Access to specialist multidisciplinary hospital basedpain management clinics is limited, resulting in long waitsfor all but the most severe cases. To assist patients who werelikely to have long waits to access hospital clinics, the GoldCoast Medicare Local, a primary care organisation that coor-dinates healthcare delivery and addresses local healthcarepriorities, piloted a multidisciplinary patient support, educa-tion and information programme. The programme aimed tohelp patients manage their pain with appropriate support inprimary care. The aim of this study was to evaluate theimpact of the programme.

Patients were identified from Gold Coast University Hos-pital wait lists for Persistent Pain, Orthopaedics, Rheumatol-ogy and Neurology, or referred by their General Practitioner.103 patients were telephoned to inform them of the pro-gramme, and then mailed further details and an invitation toenrol.

The programme included individual discussions withpatients to support them and guide them to relevant alliedhealth services, and ten evidence based education and informa-tion sessions. The group sessions lasted two hours and wereheld monthly commencing September 2013. Up to five addi-tional allied health services were also provided.

Demographic information was collected on those whoenrolled. The Pain Self-Efficacy Questionnaire,2 a measure of aperson’s ability to manage, or cope despite pain, was used toassess changes at six (February 2014) and ten (June 2014)months. Patients were also asked to comment on the pro-gramme. University ethics approval was granted.

The 48 patients that enrolled in the programme had anaverage age of 55 years, and pain duration of 12.9 years; 38were female. Two individual discussions were held with allparticipants and most attended at least half the education andinformation sessions (median = 5).

Improvements in Pain Self-Efficacy were observed between0 and 6 months, and again between 6 and 10 months(p < 0.001). The effect size for the change between 0 and 10months was 1.1, equating to a large clinically significantimprovement. Patients commented that they felt betterequipped, i.e. more motivated, knowledgeable and empow-ered, to self-manage their pain as a result of participating in theprogramme.

For patients experiencing chronic pain the ability toselfmanage their condition is a key outcome of therapy. Thefocus on self-management with support from health profes-sionals in primary care resulted in improvements that con-tinued throughout the life of the programme which were able tobe captured using the tools selected. While there was no controlgroup in this pilot, it is unlikely that a chronic pain conditionwould improve without intervention. Future research aims todetermine whether the improvements are maintained over timeand to evaluate the programme in a randomised controlled trial.A programme of education and information in combinationwith support from health professionals in primary care aidsself-management of chronic pain conditions and has the poten-tial to reduce hospital wait lists.

Thanks to Andrea Sanders, and Ruth Carey from the GoldCoast Medicare Local.

1. Blyth FM, March LM, Brnabic AJM, Jorm LR,WilliamsonM, Cousins MJ. Chronic pain in Australia: a prevalencestudy. Pain 2001; 89 (2–3): 127–134.

2. Nicholas MK. The pain self-efficacy questionnaire: takingpain into account. Eur J Pain 2007; 11 (2): 153–163.

Educational provision for improving theprescribing of junior doctors:a national survey

P.J. Lewis, S. Midgley, R. Quilliam andM.P. Tully

Manchester Pharmacy School, The University of Manchester, OxfordRoad, Manchester M13 9PL

[email protected]

Prescribing errors are a common problem in secondary careand can affect up to 50% of admissions [1]. Reducing harmfrom error is a priority for NHS hospital trusts and as suchmany interventions have been implemented locally in order toimprove prescribing. Such interventions include e-learning

Oral abstracts 7


packages, face-to-face teaching and prescribing assessmentamongst others. However, there is no obligation or nationalguidance on the delivery of such interventions across hospitaltrusts. This study aims to describe the current status of educa-tional provision on prescribing safety across acute trusts inEngland.

An online questionnaire was developed, piloted and emailedto the Chief Pharmacist and Director of Medical Education in164 teaching and non-teaching acute NHS hospital trusts inEngland. A reminder email was sent to non-respondents after athree-week period. The survey incorporated mainly closedquestions regarding types of educational interventions pro-vided within the hospital for foundation doctors and details oftheir delivery. Some open questions were included to obtainparticipants’ opinions of educational interventions and theirperceived effectiveness. Responses to closed questions wereentered into SPSS and analysed using descriptive statistics.Responses to open-ended questions were coded and similarlyanalysed.

Eighty-five hospital trusts responded to the questionnaire(RR = 51.8%). Nearly all trusts offered some form of educa-tional intervention (83/85) yet there was considerable variationin content and mode of delivery. Teaching sessions (small andlarge group face-to-face teaching, often delivered by pharma-cists) were the most commonly provided intervention (94% oftrusts, n = 78). E-learning was delivered in 70% (n = 58) oftrusts study guides in 10% (n = 8) and assessmentin 63% (n = 52). Three interventions together (E–learning,teaching sessions and assessment) was the most common com-bination provided (29%, n = 25). However, the content ofinterventions (e.g. completion of mock drug charts, practicedrug calculations, patient scenarios etc.) also greatly variedacross trusts. Some of those delivering educational interven-tions (20.5%, n = 17) also reported that they collected data ontheir effectiveness (e.g. via pharmacist intervention monitor-ing). Half (53%, n = 44) stated that they provided some form offeedback to doctors. This was commonly informal, ad hoc andinitiated only when concerns were raised. The UniversityEthics Committee judged that this work did not require ethicalapproval.

The findings demonstrate that there is no uniformity in theprovision of interventions to improve junior doctors prescrib-ing across NHS acute hospital trusts. This lack of uniformity islikely to continue in the absence of evidence to suggest whichapproach or combination of approaches work best. However,since few hospitals collect data on the effectiveness of theirinterventions, even local data on efficacy is lacking. There alsoappears to be little evidence of formal feedback to doctors.Future work should determine the effectiveness of educationalinterventions and identify practical approaches to formal feed-back provision.

1. Dornan T, Ashcroft D, Heathfield H, Lewis P, Miles J,Taylor D, Tully M, Wass V. An in depth investigation intocauses of prescribing errors by foundation trainees in rela-tion to their medical education. EQUIP study. 2009.London, General Medical Council.

Interventions to address potentiallyinappropriate prescribing in primary care:a systematic review of randomisedcontrol trials

B. Clyne, C. Fitzgerald, A. Quinlan, C. Hardy,R. Galvin, T. Fahey and S. M Smith

HRB Centre for Primary Care Research, Royal College of Surgeons [email protected]

Potentially inappropriate prescribing (PIP) is broadly definedas the use of medicines that introduce a greater risk of adversedrug-related events where a safer, as effective alternative isavailable. PIP can contribute to increased morbidity, adversedrug events and hospitalisations. PIP is common in olderpeople with an estimated prevalence of between 20 and 50% incommunity dwelling older adults, and the most important riskfactor of receiving a PIP is being on multiple drugs –polypharmacy. This systematic review aimed to assess theeffectiveness of interventions designed to reduce PIP inprimary care.

A systematic review of randomised controlled trials (RCTs)was conducted, adhering to PRISMA guidelines. PubMed,Embase, Scopus and the Cochrane library databases weresearched (June 2014). Search terms included inappropriateprescribing, inappropriate pharmacotherapy, RCT, andprimary care. Studies were included where: the population wascommunity dwelling older patients (≥65); the intervention tar-geted PIP as compared to usual care or other intervention; andthe outcome was change in PIP. Changes in PIP were measuredwith explicit (criterion-based, e.g. Beers or STOPP criteria) orimplicit (judgment-based) tools such as the Medicines Appro-priateness Index (MAI) which measures appropriatenessacross ten domains (e.g. indication, duration, cost), summed toprovide an overall score. Three reviewers independentlyassessed studies for eligibility, extracted data, and assessedmethodological quality using the Cochrane Collaboration’srisk of bias tool. As meta-analysis of results was not possibleowing to heterogeneity in interventions, a narrative synthesiswas carried out. No ethical approval was required.

Of 699 records reviewed, 14 RCTs met the inclusion criteria.A total of 157,334 patients were included across all studies.Prevalence of PIP ranged from 18% to 78% at baseline. Fiveintervention strategies were identified: pharmacist-led advice;multi-disciplinary team (MDT) meetings; multifaceted inter-ventions (combining two or more techniques) computerisedclinical decision support systems (CDSSs); and audit and feed-back. Pharmacist-led advice in four RCTs was associated withimprovement in PIP (mean MAI difference of 3.1). Significantimprovement in diminishing PIP was also found in one out ofthree MDT studies, two out of three multifaceted interventionsand the only audit and feedback study. Both CDSSs studieswere effective in reducing new PIP, particularly in relationto amitriptyline, but not in the discontinuation of existingPIP. The methodological quality of included studies was oftenpoor, particularly in reporting selection and attrition bias.There were small numbers of studies within each category and



there was considerable heterogeneity in the types of interven-tions grouped together and PIP assessment used.

This systematic review of RCTs has found that pharmacist-led interventions appear effective in decreasing PIP in primarycare, with other interventions such as MDTs, multifacetedapproaches and CDSSs showing promise. However, the het-erogeneity of the study interventions and the variable meth-odological quality of these RCTs mean that results requirefurther confirmation in ongoing RCTs. Furthermore, the clini-cal significance of reductions in MAI score is difficult to deter-mine. There is a need for better quality evidence upon which tobase approaches to improving prescribing for these vulnerablepatients.

Prevalence of potentially inappropriateprescribing (PIP) and potentiallyprescribing omissions (PPO) in older Irishadults: findings from a randomisedplacebo-controlled trial (TRUST)

D. O’Riordana,b, K. Walsha,b, S. Byrnea andP. Kearneyb

aPharmaceutical Care Research Group, School of Pharmacy, UniversityCollege Cork and bDepartment of Epidemiology and Public Health,University College [email protected]; [email protected]; [email protected];[email protected]

Potentially inappropriate prescribing (PIP) is a well docu-mented problem in older patients. It includes the practice ofadministering potentially adversely interacting medicines,medicines with relative and absolute contraindications andunnecessary medicines 1. The major adverse events associatedwith PIP include serious drug interactions, hospitalisations,morbidity and in some cases mortality1.

The aim of the study was to identify PIP and PPO’s using asubset of the Screening Tool of Older Persons Prescriptions/Screening Tool to Alert doctors to Right Treatment, (STOPP/START) criteria version 1 and 22 based on 258 communitydwelling Irish adults aged ≥65 years screened in an ongoingEuropean clinical trial.

Ethical approval was obtained for this study. A subset of thecriteria was used as information relating to drug strength, doseand duration of prescriptions in the TRUST dataset was notavailable. Therefore 40 PIP indicators and 14 PPO indicatorsfrom the STOPP/START version 1 criteria and 51 PIP indica-tors and 22 PPO indicators from version 2 were applied. PIP/PPO prevalence according to both sets of criteria wasestimated. Two pharmacists each reviewed half of the dataindependently. Each reviewer then randomly sampled 10% ofthe other reviewer’s data for consistency. Finally, all the datawas cleaned, this involved a three step cycle of screening, diag-nosing and editing any suspected data irregularities.

Data on 258 patients screened in the study were analysed.The mean age (± SD) of the patients was 73.2 ± 5.1 years, 133(51.6%) were female. The most common morbidity recordedwas hypertension (53.9%). The median number of drugs con-

sumed was 4 (IQR = 4). The overall prevalence of PIP was14.7% (n = 38) considering all 40 STOPP version 1 criteriawhile the overall prevalence of PPO was 16.7% (n = 43) con-sidering 14 START version 1 criteria. The overall prevalence ofPIP and PPO using version 2 was 15.9% (n = 41) and 20.5%(n = 53) respectively. There was a significant associationbetween PIP (in both STOPP/START version 1 and 2) andpolypharmacy when adjusted for age, sex and multimorbidity(adjusted OR 4.39, p = 0.001, 95% CI 1.77–10.87 and adjustedOR 3.27, p = 0.009, 95% CI 1.35–7.93 respectively). The asso-ciation between PPO (in both STOPP/START version 1 and 2)and polypharmacy when adjusted for age sex andmultimorbidity (adjusted OR 0.49, p = 0.118, 95% CI 0.20–1.20 and adjusted OR 1.07, p = 0.874, 95% CI 0.48–2.35respectively) was not significant.

These findings indicate that PIP and PPO’s are prevalent inolder adults screened for a randomised control trial (RCT)using a subset of the STOPP/START version 1 and 2criteria. Screening tools such as STOPP/START have thepotential to identify inappropriate prescribing at the point ofrandomising patients to clinical trials. The use of a structuredpharmacist review of medication (SPRM) supported by a com-puterised decision support system (CDSS) and STOPP/START version 2 is currently underway.

1. Lindley CM, Tully MP, Paramsothy V, Tallis RC. Inappro-priate medication is a major cause of adverse drug reactionsin elderly patients. Age and Ageing 1992; 21: 294–300.

2. O’Mahony D, O’Sullivan D, Byrne S, O’Connor MN, RyanC, Gallagher P. STOPP/START criteria for potentiallyinappropriate prescribing in older people: version 2. Ageand Ageing 2014; 0: 1–6.

Prescribing appropriate polypharmacy forolder people in primary care: a qualitativestudy of general practitioners’perceptions and experiences

C.A. Cadogana, C. Ryana, G.J. Gormleyb,P. Passmorec, J. Francisd, N. Kersee andC Hughesa

aSchool of Pharmacy, Queen’s University Belfast, bDepartment ofGeneral Practice, Queen’s University Belfast, cCentre for Public Health,Queen’s University Belfast, dSchool of Health Sciences, City UniversityLondon and eSchool of Population Health, University of [email protected]

Ensuring appropriate combinations of medications in olderpopulations with multimorbidity is an ongoing and everincreasing challenge. Evidence to support the effectiveness ofinterventions to improve appropriate polypharmacy in olderpeople is weak.[1] It has been proposed that future interventionsshould adopt a theory-based approach and involve key stake-holders, such as healthcare professionals (HCPs), during inter-vention design.[1] Theory-based qualitative interviews usingthe Theoretical Domains Framework (TDF) can be used toidentify specific barriers and facilitators to achieving desired

Oral abstracts 9


changes in HCPs’ clinical practice.[2] The aim of this study wasto explore general practitioners’ (GPs) approaches to prescrib-ing polypharmacy to older patients and to identify key theoreti-cal domains that acted as barriers and facilitators to theprescribing of appropriate polypharmacy. This study formspart of an ongoing mixed methods project seeking to developan intervention to improve appropriate polypharmacy for olderpeople in primary care. The intervention will specifically targetidentified theoretical domains.

A purposive sample of up to two general practices (urban,rural) from each Health and Social Care Trust area (n = 5) inNorthern Ireland was recruited into the study by the NorthernIreland Clinical Research Network. Semi-structured inter-views were conducted with recruited GPs (up to two per prac-tice) using a TDF-based topic guide. The topic guide (pilotedwith two academic GPs) explored GPs’ views on the term‘polypharmacy’. Questions covering each TDF domain wereused to explore GPs’ perceptions of barriers and facilitatorsto ensuring the prescribing of appropriate polypharmacyfor older people. A clinical scenario of inappropriatepolypharmacy was included to stimulate discussion. Data wererecorded and transcribed verbatim. Transcripts were indepen-dently checked for accuracy. Following data saturation, datawere analysed using both content analysis and the frameworkmethod. Ethical approval was granted by the Office ofResearch Ethics Committees Northern Ireland.

Fourteen GPs participated in the study. GPs’ definitions ofthe term ‘polypharmacy’ varied, with a primary focus onnumbers of drugs prescribed. ‘Knowledge’, ‘Skills’, ‘Beliefsabout consequences’, ‘Social influences’ and ‘Social/professional role and identity’ were identified as key theoreti-cal domains that facilitated the prescribing of appropriatepolypharmacy to older patients. For example, GPs reportedthat their clinical knowledge and skills facilitated them inmaking the necessary prescribing changes to ensure that olderpeople receive appropriate polypharmacy. ‘Environmentalcontext and resources’ was the main theoretical domain thatprevented GPs from prescribing appropriate polypharmacy toolder people in primary care, particularly time needed toaddress polypharmacy.

The study findings show that GPs believe they have the requi-site knowledge and skills to ensure older people receive appro-priate polypharmacy. However, time pressures and resourcelimitations are currently preventing GPs from optimising olderpeople’s prescriptions more frequently in routine clinical prac-tice. This analysis will be integrated with other project compo-nents (semi-structured interviews of community pharmacists,focus groups involving older patients receiving polypharmacy).Synthesis of all data will be used to develop an intervention toassistGPsandcommunitypharmacists inachievingappropriatepolypharmacy in older people in primary care.

This work was supported by The Dunhill Medical Trust[grant number: R298/0513].

1. Patterson SM, Cadogan CA, Kerse N, Cardwell CR,Bradley MC, Ryan C, Hughes C. Interventions to improvethe appropriate use of polypharmacy for older people.Cochrane Database Syst Rev 2014; 10: CD008165.

2. Michie S, Johnston M, Abraham C, Lawton R, Parker D,Walker A. Making psychological theory useful for imple-

menting evidence based practice: a consensus approach.Qual Saf Health Care 2005; 14: 26–33.

Inpatient electronic prescribing – howinvolved are inpatients withtheir medication?

S. Jheetaa, C. Lubranta, S. Garfielda,b andB. D. Franklina,b

aCentre for Medication Safety and Service Quality, Imperial CollegeHealthcare NHS Trust, London, UK and bThe Department of Practice andPolicy, UCL School of Pharmacy, Mezzanine Floor, BMA House, London,[email protected]

It has been suggested that greater patient involvement inmedication may improve medication safety. However, it isnot known to what extent electronic prescribing (EP) maysupport inpatient involvement in UK hospitals. EP systemscould potentially create a barrier to patient involvement ifpatients have limited access to their medication records. Con-versely, EP could facilitate the production of patient-specificinterfaces which could be used to support increased patientinvolvement.

We aimed to 1) identify whether commercial EP systemsavailable in England have features which support inpatientinteraction with their electronic medication record; and 2)explore the extent to which inpatients are supported to interactwith EP systems in a sample of English hospitals.

First, websites of commercial EP systems used inEngland[1,2] were searched to identify any features thatsupport inpatient access or interaction. Second, thirteenhospital trusts in England with well-established inpatientEP systems were identified[1,2]; in each trust, a relevant phar-macist was emailed in July 2014 and invited to answer ques-tions about inpatient interactions with their EP system. Basedon pilot work, we developed open questions, includingbrief prompts, to explore if and how inpatients view details oftheir prescribed medication or use the system when self-administering medication. Recipients were invited torespond by email or telephone depending on their preference.Key points from the data were collated, tabulated forcomparison and summarised descriptively. Ethical approvalwas not required as the work was considered a serviceevaluation.

Of fourteen commercial EP systems identified, only onewebsite referred to possible inpatient involvement; this wasvia an ‘interactive patient console’ which could provideaccess to medical information, educational materials andcommunication with healthcare professionals. It requiredadditional hardware such as televisions, microphones andkeyboards.

A representative of nine (69%) trusts responded to oursurvey, most of whom used commercial EP systems. Nonereported that inpatients were able to access or interact directlywith their electronic medication record, although two reportedthat their system had a ‘patient friendly’ medication record



screen which could be shown to the patient by a healthcare pro-fessional if requested. Seven reported that printouts of inpa-tient medication were possible if requested by the patient butthese were not routinely used. Self-administration of medica-tion was reported to be rarely used; four trusts provided a papermedication record to any self-administering patients aspatients were unable to access their electronic medicationrecord.

This exploratory study suggests that inpatient EP systems donot seem designed to accommodate patient interaction. Thereis potential for inpatients to view ‘patient-friendly’ screenswith some systems although these are not routinely used. Thismay partly be due to the requirement for additional hardware tosupport more widespread use. Although there is a shift towardspaperless systems, currently the most common way to providepatients with information about their prescribed medication isthrough paper printouts. A limitation is that email responseswere more limited. EP suppliers and healthcare providersshould consider how inpatients could best be involved withtheir medication with EP systems.

1. Ahmed Z, McLeod MC, Barber N, Jacklin A, Franklin BD.The use and functionality of electronic prescribing systemsin English acute NHS Trusts: a cross-sectional survey.PLoS ONE 2013; 8(11): e80378. doi:10.1371/journal.pone.0080378

2. Cresswell K, Coleman J, Slee A, Williams R, Sheikh A etal. Investigating and learning lessons from early experi-ences of implementing ePrescribing systems into NHS hos-pitals: a questionnaire study. PLoS ONE 2013; 8(1):e53369. doi:10.1371/journal.pone.0053369

Exploring the perceptions andexperiences of people who use and thosethat provide a shared care clozapineservice

C. Sowerby and D.A. Taylor

University of [email protected]

No Health without Mental Health 2011[1] and DeliveringBetter Mental Health Outcomes for People ofAllAges 2011[2]emphasise that: care should be driven by patients, there shouldbe choice in how patients obtain their care, and care should beindividualised and recovery focused in order to improvepatients’ independence.

Clozapine clinics, managed within secondary care mentalhealth trusts, are commonly used to supply clozapine to thosein the community. A clozapine shared care service is an alter-native option, where people taking clozapine obtain it fromtheir GP and community pharmacy; who are supported by sec-ondary care mental health colleagues.

The aim of this study was to explore the perceptions andexperiences of people receiving/delivering this service tobetter understand its effectiveness and acceptability.

Semi-structured interviews and focus group methodologywere used to explore perceptions and experiences of studyparticipants.

Potential professional participant groups included clozapineservice users (CSU), general practitioners; community psychi-atric nurses, community and hospital pharmacy staff andresponsible clinicians. Ethical approval was granted in July2013. A phenomenological analytical approach was adoptedusing Interpretative Phenomenological Analysis (IPA) toenable exploration of how the phenomenon of a shared careclozapine service had been experienced by study participants,and what this phenomenon meant to the individual.

A total of 38 participants were recruited; including 32healthcare professionals (HCPs) and 6 CSUs. There were 13HCPinterviews, 1 patient interview and 6 focus groups, result-ing in 20 transcripts. As the primary concern of IPAis with thedetailed account of individual experience, there is benefit froma concentrated focus on a smaller number of cases. All partici-pant groups were similarly represented in number. Transcriptswere analysed within and across participant groups as an itera-tive process and resulted in four superordinate themes sharedacross the participant groups studied. These were: ClozapineProcess, Shared Care, The Provision of Care andMultiprofessional Relationships.

Knowledge of the clozapine process, had a positive effecton the experience of shared care clozapine for HCPs andCSUs, supporting a sense of confidence. The clozapineprocess itself caused anxiety for HCPs and CSUs alike, due tothe consequences of a delayed clozapine supply.Multiprofessional relationships with HCPs and CSUs devel-oped over time and were facilitated through communication.These relationships enabled HCPs to feel valued and includedwithin the MDT. Shared care enabled CSUs to take ownershipof their health and develop independence through realisation oftheir capability and a change in relationship between CSUs,HCPs and their medicines. Barriers related to the uptake ofShared care were identified as clozapine-associated stigma,changes in ownership of responsibility and professionals’ levelof involvement in CSUs’ care. HCPs and CSUs perceivedclozapine as different in the provision of care, because of theprocess of supply and its beneficial effects on symptoms. BothCSUs and HCPs agreed the provision of care for physicalhealth was just as important as mental health.

Results demonstrate that shared care clozapine empowerspeople who take clozapine to gain ownership of their healthand supports multi-professional relationships.

1. Department of Health. No Health Without MentalHealth : A Cross Government Outcomes Strategy, 2011.

2. Department of Health. Delivering Better Mental HealthOutcomes for People of All Ages, 2011.

Oral abstracts 11


Illness and medication beliefs amongelderly Kuwaiti patients with type 2diabetes in primary health care setting inKuwait: a qualitative study

M. Alsaleh, M. Weiss and H.E. Family


Kuwait is ranked as one of the top ten countries in the world inregard to the incidence of adult diabetes mellitus (DM).[1]

Although there are health services in Kuwait for DM it isimportant to understand the beliefs that impact on the health-related behaviours of Kuwaiti diabetics. This will help identifythe triggers that prompt people to seek medical advice for theircondition, to understand how people take their medications,and how/where they obtain medical information. This researchaimed to explore the beliefs of older DM patients’ about theircondition and use of medicines in Kuwait.

Ethical approval was obtained from Kuwaiti Ministry ofHealth Ethics Committee to conduct semi-structured inter-views with type two DM patients. A topic guide, developedfrom the objectives of the study was used to guide the inter-views. Participants were recruited from the primary healthcare diabetic clinics pharmacy by the researcher (MA). Asampling frame was employed in the recruitment process,whereby every fifth patient who met the inclusion criteriawas invited to participate in the study. The inclusion criteriawere: type two DM for >6 months, living in Kuwait, age 50and over, and receiving oral hypoglycaemic medications (butnot insulin). Informed consent was obtained from all partici-pants. Interviews were audio-recorded and conducted inArabic. Interviews were transcribed verbatim and analysedusing thematic analysis.[2]

Interviews were conducted with 15 patients (7 males) andlasted between 20 and 40 minutes. Nine themes emergedfrom the data: causes of diabetes, behaviour toward medica-tion, attitude to control diabetes, views about medicine, viewsabout doctors and pharmacists, medication reminders,sources of information and support, social gatherings for menor women and social behaviour related to social gatherings.This paper reports the strongest theme “behaviour towardsmedication.” Within this theme participants describedwanting only the best treatment which related to evaluatingdiabetic medications according to their trade name, prefer-ences for medicines originating from western over middle-eastern countries and the original medicines packaging overhospital packaging: ‘honestly, if the Glucophage wasn’t madein Germany I wouldn’t consider to take it’. Participants’ nar-ratives revealed that patients trust the doctor’s but not thepharmacist’s medicines advice. Patients also describedhoarding and stock piling medicines after the Iraqi invasion,during which time there had been no access to medicines, aswell as sharing DM medicine between husband and wife oramongst neighbours and friends.

This is the first qualitative study to explore the social andcultural beliefs of diabetic patients’ illness and medicine-taking behaviour in Kuwait. However, it is recognised that this

is a small study which cannot be generalised to the wider popu-lation. Future work will include investigating whether theseviews are prevalent amongst a wider diabetic population. Anunderstanding of how the illness and medication beliefs heldby Kuwaiti DM patients influences their health behaviours willbe helpful in identifying ways of shaping their behaviourthrough targeted interventions to promote a healthier lifestyleand improved DM management.

1. International Diabetes Federation. IDF Diabetes Atlas, 6thedn. Brussels, Belgium: International Diabetes Federation,2013.

2. Braun V, Clarke V. Using thematic analysis in psychology.Qualitative Research in Psychology 2006; 3(2):77–101.

Investigating the impact of medicineinformation services on patientoutcomes: a systematic review

N. Alkhaldi, D. Wright and D. Bhattacharya

University of East [email protected]

A 2002 systematic review of the impact of medicine informa-tion (MI) services on patient outcomes, comprising six studies,found limited evidence and recommended that more robuststudies are required to demonstrate their effectiveness (1). Infinancially constrained healthcare systems it is becomingincreasingly important to demonstrate the value of differenthealthcare services. It may therefore be the absence of rigorousevidence that is associated with the reported reduction in MIservices in the USA over recent years (2). This study aims toupdate the previously performed systematic review and collatethe evidence for the effects of MI services on patient outcomes.

Published and unpublished studies were identified throughelectronic and manual searches. The following databases weresearched from inception to December 2013: CINAHL,EMBASE, MEDLINE and the Centre of Reviews and Dis-semination. The primary author(s) were contacted throughemail to identified unpublished studies. After scoping searcheswere conducted, the search strategy combining the terms medi-cine information, drug information and dial access was used incombination with truncations, wild cards ($), adjacent searchoptions (adj2), and other relevant Boolean operators permittedby the database. Studies reported in English, of any design,assessing the impact of MI services in secondary care andreporting patient outcomes were included. Independentscreening of titles and abstracts was performed by two review-ers using a bespoke abstract screening tool. Disagreementswere resolved by consensus with referral to a third reviewer ifnecessary. The Cochrane risk of bias tool was used to assess thequality of included studies. Quality appraisal was indepen-dently performed by two reviewers; studies were not excludedbased on quality.

Ten relevant published articles were identified of which fourwere from the UK, three from USA, two from Canada and onefrom Norway. Single-centre studies accounted for eight andtwo were multi-centre. All studies were descriptive rather than



experimental and no comparison groups were used. Five wereprospective and five were retrospective in design. In moststudies the impact of MI services on patients’ outcome(s) wascaptured through using survey methods. Three studies used amulti-professional expert panel comprising clinical pharma-cists and doctors in order to identify and evaluate the patientoutcomes of the MI service. A positive impact on patient out-comes was reported for 40% to 76% of MI service users. Onelarge scale multi-centre observational study reported 44 fewerdeaths per hospital per year in the 1029 hospitals with an MIservice when taking into account other related factors. To date,no randomised controlled trials have been performed to dem-onstrate the effectiveness of the provision of MI services.

The research methodologies used to investigate the impactof MI services on patient outcome(s) tend to rely on users sub-jectively stating whether patients benefited from the advice orreview panels predicting patient outcomes as a result of theadvice provided. With MI centres disappearing within somehospitals in the USA, this may provide an opportunity forcluster randomised controlled trials to more accurately deter-mine the effect of MI services on patient outcomes.

1. Hands S, Stephens M, Brown D. A systematic review of theclinical and economic impact of drug information serviceson patient outcome. Pharm World Sci 2002; 24: 132–138.

2. Rosenberg JM, Koumis T, Nathan JP et al. Current status ofpharmacist-operated drug information centers in the UnitedStates. Am J Health Syst-Pharm 2004; 61: 2023–2032.

Economic evaluation of asoftware-supported structuredpharmacist medication review inhospitalised older patients

J. Gallaghera, S. McCarthya, D. O’Sullivana,D. O’Mahonyb, P. Gillespiec, N. Woodsd andS. Byrnea

aSchool of Pharmacy, University College Cork, bDepartment of GeriatricMedicine, Cork University Hospital, cSchool of Business and Economics,National University of Ireland, Galway and dCentre for Policy Studies,University College Cork, [email protected]

A recent randomised controlled trial (RCT) conducted in anIrish hospital applied a structured pharmacist review of medi-cation (SPRM) supported by computerised decision supportsoftware (CDSS). This novel intervention was predominantlybased on the application of the STOPP (Screening Tool ofOlder Persons Potentially Inappropriate Prescriptions)/START (Screening Tool to Alert Doctors to the Right Treat-ment) criteria to patient medication records. The aim of thispaper was to examine the cost-effectiveness of pharmacistsapplying the SPRM/CDSS intervention on elderly hospitalinpatients in comparison with usual pharmaceutical care.

This evaluation is a trial based cost-effectiveness analysis.Full details of the study method are detailed elsewhere[1].Patients were recruited and cluster randomised (according to

admitting consultant) to either intervention or control armsbetween June 2011 and June 2012. Time horizon for this evalu-ation was confined to patient discharge or 10 day follow-up andperspective was that of the healthcare provider. Cost data wereinformed by a combination of individual patient trial data, pub-lished healthcare salaries and Irish diagnosis-related groupdata for inpatient stays. Incremental analysis was undertakenusing a multi-level mixed effect regression model. Uncertaintywas addressed using 95% confidence intervals and cost-effectiveness acceptability curves. The probability of the inter-vention being cost-effective was measured using hypotheticalthreshold levels which the healthcare provider would bewilling to pay for the prevention of an adverse drug reaction(ADR) [2]. Ethical approval was obtained from all relevantbodies.

Application of SPRM/CDSS in hospitalised older patientsis the dominant strategy. It reduces the likelihood of a patientexperiencing an ADR, in addition having a reduced cost incomparison with usual care (see Table 1). Application tointervention group of patients cost €19 000; this was pre-dominantly composed of the cost of pharmacist time (€40 perpatient).

Table 1 Incremental cost-effectiveness analysis

Incremental analysisIntervention (n = 361)versus Control (n = 376)

Mean patient cost for hospitalstay – €

(Standard deviation)

13242 (15530) 15465 (19310)

Incremental cost: total cost – €Difference in mean (95% CIs)

(P-value)

−815 (−3451, 1820) (0.544)

Incremental effect: ADR eventOdds ratio (95% CIs) (P-value)

0.655 (0.431, 0.994) (0.047)

Incremental effect: No. of ADRevents

Difference in mean (95% CIs)(P-value)

−0.064 (−0.135, 0.008) (0.081)

Threshold value (λ) per ADRaverted

Incremental cost-effectiveness analysis

(Probability that interventionis cost-effective)

€0 0.721€250 0.724€500 0.730€1000 0.737

Based on the evidence presented, SPRM/CDSS is likely to bedetermined to be cost-effective, even at a €0 willingness to pay.However, this cost-effectiveness analysis was based on the pre-vention of adverse drug reactions rather than a more recog-nised outcome e.g. QALY. Hence, there is still an element ofuncertainty regarding the cost-effectiveness status of theintervention.

1. O’Sullivan D et al. A randomised controlled trial to assessprevention of adverse drug reactions in hospitalised olderpatients using a software-supported structured pharmacist

Oral abstracts 13


intervention. Clinical Pharmacology and Therapeutics [inpress].

2. Gillespie P et al. The cost-effectiveness of a structured edu-cation pulmonary rehabilitation programme for chronicobstructive pulmonary disease in primary care: thePRINCE cluster randomised trial. BMJ Open 2013; 3(11):e003479.

Feasibility evaluation of an electronicmethod for documentation of clinicalpharmacy interventions and activities insecondary care

J. Pattersona, A. Campbella andS. Cunninghamb

aPharmacy Department, Ulster Hospital, Dundonald, Belfast BT16 1RHand bSchool of Pharmacy & Life Sciences, Robert Gordon University,[email protected]

The positive contribution and role of the clinical pharmacist insecondary care are well established in the UK1. The issue ofhow to effectively benchmark clinical pharmacy services inNorthern Ireland (NI) has long been debated. Existing localpaper-based documentation methods could not provide thedata required. An alternative computer-based system, ‘Elec-tronic Pharmacist Intervention Clinical System’ (epics©),developed and supported locally in NI, was selected for evalu-ation to standardize reporting.

To evaluate the feasibility of using epics© to gather data forbenchmarking clinical pharmacy activity and outcomes.

The setting was a teaching hospital in NI. Quantitative inter-vention and workload activity data were collected for clinicalpharmacy services provided to the Coronary Care Unit (CCU).The software developer provided training, and epics© wastrialled during January 2014. Data were collected daily by theprincipal researcher using epics© for a 2-month period(February–March 2014). Each intervention was assigned aclinical significance grade, based on the Eadon scale2, toprovide qualitative data on patient outcome. Ethical reviewwas carried out by University Ethics Panel and local NHSapproval was sought but deemed unnecessary.

During the 2-month period, 1508 interventions were docu-mented on CCU as shown in Table 1.

Table 1 Summary of clinical pharmacist intervention data for CCU,February–March 2014

February 2014n (% of total)

March 2014n (% of total)

Totaln (% of total)

General datafor CCU

No. of admissions 145 142 287No. of discharges 79 104 183Total interventions 727 781 1508Clinical significance

grade –Eadon scale

n (% of total) n (% of total) n (% of total)

1 0 (0%) 0 (0%) 0 (0%)2 0 (0%) 13 (1.7%) 13 (0.9%)3 144 (19.8%) 285 (36.5%) 429 (28.4%)4 580 (79.8%) 481 (61.6%) 1061 (70.4%)5 3 (0.4%) 2 (0.3%) 5 (0.3%)6 0 (0%) 0 (0%) 0 (0%)epics© category n (% of total) n (% of total) n (% of total)Admission 222 (30.5%) 212 (27.2%) 434 (28.8%)Allergy 6 (0.8%) 10 (1.3%) 16 (1.1%)Information 42 (5.8%) 50 (6.4%) 92 (6.1%)Incorrect 27 (3.7%) 36 (4.6%) 63 (4.2%)Missing 49 (6.8%) 56 (7.2%) 105 (7.0%)Discharge 158 (21.7%) 146 (18.7%) 304 (20.1%)Other 223 (30.7%) 271 (34.7%) 494 (32.7%)epics© stage n (% of total) n (% of total) n (% of total)Admission 404 (55.6%) 422 (54.0%) 826 (54.8%)Inpatient 119 (16.4%) 133 (17.0%) 252 (16.7%)Transfer 0 (0%) 6 (0.9%) 6 (0.4%)Discharge 204 (28.0%) 220 (28.2%) 424 (28.1%)

Med Rec onadmission

109 103 212

% of patients withMed Rec onadmission

75% 72.5% 74%

Med Rec on discharge 73 65 138% of patients with

Med Rec ondischarge

92.4% 62.5% 75.4%

Average meannumber ofinterventionsper patient

4.6 5.4 5.0

Analysis of the characteristics and range of data for all rec-orded interventions (n = 1508) showed most interventionsoccurred at admission (n = 826) or discharge (n = 424) stageswith drug history (n = 222) and medicines reconciliation atdischarge (n = 138) being most common, respectively. Otherinterventions included: kardex review (n = 158); laboratorytest reviewed (n = 189); patient education (n = 79); laboratorytest required (n = 67), and missing frequency (n = 39). Themajority (70%) were graded as having improved patient care.Epics© showed that 84% of time was spent on patient-focusedwork activities.

The epics© software enabled daily collection of clinicalpharmacy intervention and workload activity data that could beused for benchmarking service provision. It provided an exten-



sive database of quantitative and qualitative information andhas the potential to meet local and regional demands for clini-cal pharmacy data. Epics© is not integrated with all othersystems and in the future any pharmacy software should,ideally, be part of a single electronic patient record.

1. National Institute for Health and Clinical Excellence(2007). Technical Patient Safety Solutions for MedicinesReconciliation on Admission of Adults to Hospital.PSG001. www.nice.org.uk (Online).

2. Eadon, H. Assessing the quality of ward pharmacists’ inter-ventions. The International Journal of Pharmacy Practice1992; 1: 145–147.

General public expectations of acommunity pharmacy led weightmanagement service offered duringsmoking cessation

A.E. Weidmanna, B. Addisona, A. Browna andJ. Youngb

aRobert Gordon University, Aberdeen and bNorthumbria University,Newcastle upon [email protected]

The World Health Organization estimates obesity prevalencewill double to 41% of the Scottish population by 2030.1

The government has placed a strong emphasis on communitypharmacies to provide a wide range of easy-to-access weightmanagement services to the general public.2 A programme ofstudies by this research team, establishing views, attitudesand awareness of pharmacists, counter assistants and thegeneral public towards weight management services cur-rently delivered by UK community pharmacies, has identi-fied a desire for a person-centred nutrition programme, basedon the structure of, and used in conjunction with, thesuccessfully implemented nicotine replacement therapy(NRT) scheme. The aim of this research was to explore theexpectations of the general public towards such a tailoredservice.

A cross-sectional survey with a sampling frame of 1500 ran-domly selected members of the Scottish general public aged 18years and over (obtained from the electoral register) was con-ducted during June–August 2013. Questionnaires, with a cov-ering letter describing the study’s aim and assurance ofconfidentiality, were mailed to all participants. Non-responders were mailed up to two reminder questionnaires at 4weekly intervals. The questionnaire comprised informationon: awareness of services available from community pharma-cies, respondents’ attitudes towards and expectations of weightmanagement services provided in the context of NRT, demo-graphic data and a final open question inviting furthercomment. Five-point Likert scales, semantic differentials andopen/closed questions were used as response options. Data wasanalysed using descriptive statistics and cross-tabulations inSPSS (vs 21). Content analysis was performed the open ques-tion responses. The study was approved by the Ethical Review

Panel of the School of Pharmacy and Life Sciences at RobertGordon University, Aberdeen.

A total of 255 questionnaires were returned (response rateof 17%). Demographics: The majority of respondents weremale (48.2%; n = 123), aged 18–29 years (47.1%; n = 115)and in full-time employment (60.8%; n = 155). The majorityof respondents were ex-smokers (34.9%; n = 89) or smokers(18%; n = 46). 67.1% (n = 171) classed themselves as havinga normal body weight with 24.4% (n = 59) being overweightor obese. The majority of respondents (87.5%; n = 223) werenot currently actively participating in a weight loss pro-gramme. Awareness of services available from communitypharmacies: Only 2% (n = 5) and 5.9% (n = 15) had previ-ously made use of a pharmacy or GP led NRT service. Themajority agreed or strongly agreed (51.7%; n = 132) that theywere unaware of any weight management services offered bycommunity pharmacies. Attitudes towards and expectationsof weight management services provided in the context ofNRT:73.9% (n = 34) of smokers and 78.6% (n = 70) ofex-smokers consider there to be a need for a pharmacy-basedweight loss service specifically tailored to people who wouldlike to stop smoking. 72.2% (n = 164) of respondents wouldconsider using such a service.

Findings of this study illustrate a somewhat mixed picture ofthe general public’s expectations regarding weight manage-ment support services and specifically for patients wishingto stop smoking. These warrant further investigation. Weacknowledge potential study limitations of bias, size and trans-ferability to other areas in the UK.

1. The Scottish Government (2010). Preventing Overweightand Obesity in Scotland. A Route Map Towards HealthyWeight. http://www.scotland.gov.uk/Resource/Doc/302783/0094795.pdf (accessed October 2013).

2. Scottish Public Health Network. SOAR2 Project WorkingGroup. Scottish Obesity Action Resource – Update. Sep-tember 2012.

Ebola virus disease: understanding howrelevant health information isdisseminated in Nigeria

O.P. Adigwea, J. Alfab and A.B. Umorub

aHealth Policy Research and Development Unit, National Assembly,Abuja, Nigeria and bDepartment of Pharmaceutical Services NationalAssembly, Abuja, Nigeria

The high mortality rate and rapid spread associated with theEbola virus disease (Ebola) justify its classification as an inter-national health emergency[1]. This is in addition to the fact thatcurrently, no cure exists for the disease. As prevention remainsa key factor in controlling the spread of the disease, providingaccurate and reliable health information in a timely manner isof utmost importance. This study aimed at understanding howhealth information relating to the Ebola was disseminated inNigeria.

Across-sectional survey was designed to collect the relevantdata from a convenient sample in a government establishment

Oral abstracts 15


that interfaces with international visitors and the Nigerianpublic. Face and content validation was undertaken by anexpert panel. Piloting did not result in any changes. Thedata were collected between July and August 2014 and ana-lysed using SPSS (version 17). Research ethics and govern-ance approval were obtained from the National Assemblymanagement.

An 80% response rate was achieved (561/700). Majorityknew of Ebola (93%; 520/561) and other important informa-tion regarding the disease, for example, that it is viral (91%;473/520). Knowledge in other areas was, however, inaccurateor inadequate. For instance only 57% (296/520) knew that itdid not have a cure and close to half (46%; 239/520) felt thateveryone infected with the virus would certainly die within21 days. Respondents accessed information about Ebola fromvarious sources with television and radio emerging as themost relevant, and posters, the least.

Figure 1 Sources on information on Ebola

Healthcare decisions made in emergencies such as Ebola areunderpinned by information that individuals have access to. Itis therefore important that health information provision is safeand efficient [2].Our study identified electronic and print mediaas the most efficient means of disseminating relevant informa-tion. This is particularly important in resource scarce settings.Reliance on social media and religious leaders for informationon Ebola is worrisome as the quality of information provided isunknown. Evidence emerging from this study can thereforehelp relevant bodies improve the provision of health informa-tion, as well as ensure that the information is valid and reliable.

1. WHO (2014). Statement on the Meeting of the InternationalHealth Regulations Emergency Committee regarding the2014 Ebola Outbreak in West Africa [Online]. http://www.who.int/mediacentre/news/statements/2014/ebola-20140808/en/

2. Adigwe OP. Health literacy: patient involvement andengagement with healthcare. Reply to Raynor, DK. Healthliteracy. British Medical Journal 2012; 344:e2188[Online]. http://www.bmj.com/content/344/bmj.e2188/rr/587175

Outcomes from the Irish nationalhepatitis C prospective treatment registry

E. Graya, A. O’Learya,b, C. Walsha,c, C. Bergina,d

and S Norrisa,d

aIrish Hepatitis C Outcomes Research Network, St. James’ Hospital,Dublin 8, bNational Centre for Pharmacoeconomics, St James’ Hospital,Dublin 8, cDepartment of Mathematics, University of Dublin, TrinityCollege and dSt James’ Hospital, Dublin [email protected]

The Irish Hepatitis C Outcomes and Research network(ICORN) Treatment Registry is a prospective outcomes studythat collects real-world clinical and economic outcomes forpatients treated with direct acting antiviral agents (DAA)(telaprevir and boceprevir), when added to a dual regimen ofpegylated interferon/ribavirin for patients with Genotype 1hepatitis C (HCV).

The aim of the study is to determine sustained virologicalresponse (SVR) rates, eligibility for response-guided therapy(RGT), discontinuation rates, tolerability and total costs oftreatment.

The national ICORN HCV treatment registry is a web-basedtool hosted on an electronic platform developed by the DublinCentre for Clinical Research in conjunction with ICORN.Ethical approval for the study was obtained from the St.James’s Hospital/Tallaght Research Ethics Committee.Patients selected for treatment are consented for participationin the study and clinical data are captured at each hospital siteusing ICORN data collection sheets for the duration of treat-ment, and during post-treatment follow-up to outcome assess-ment at SVR 12 weeks and SVR 24 weeks. Patient variablescollected are determined by study outcomes and include bothcategorical and continuous variables, stratified according tobaseline demographics, HCV characteristics, and on-treatmentmedication requirements, tolerability, patient admissions,laboratory and diagnostic data. Data are subsequently enteredinto the registry prior to report generation, systematic qualitycontrol procedures and data analysis. Analysis is undertakenusing SPSS Version 21®.

A total of n = 279 patients are registered (June 2012–August2014) across 7 hospitals. The cohort is predominantly male(72%) with a median age of 45 (range 18–72) and the majorityare Irish born (71%). 70% are treatment naive and 29% are cir-rhotic. Genotype 1, 1a and 1b account for 27%, 45% and 28%of the cohort respectively. Telaprevir is the DAA of choice(68%). At baseline, 56% of patients satisfy the criteria forRGT. Outcome data are complete for n = 136 patients. TheSVR rate was 51%. Discontinuation of therapy due to treat-ment futility rules, adverse events and intolerance occurred inn = 62 (46%) patients. Safety data has been analysed for acohort of n = 110. 41% of the cohort required an interventionfor the management of anaemia, 30% (n = 32) of patients hadneutrophil levels consistent with grade 3/4 neutropenia(neutrophils < 0.75 x109/L) and grade 3/4 thrombocytopenia,as evidenced by platelets below 50 x 109/L, was reported in18.7% (n = 20) of the cohort. For n = 74 who completed treat-ment, 81% (n = 68) achieved an SVR24, indicating a relapserate of 19%.



The efficacy rates, of 65–85% obtained in phase III clinicaltrials are higher then that seen in the real world setting 1–2. Thisis due to higher discontinuation rates, which is attributed to amore frequent occurrence of virological treatment failures andadverse events, a direct outcome of greater patient heterogene-ity than seen in clinical trials.

1. Trembling PM, Tanwar S, Dusheiko GM. Boceprevir: anoral protease inhibitor for the treatment of chronic HCVinfection. Expert Rev Anti Infect Ther 2012; 10(3): 269–279.

2. Chang MH, Gordon LA, Fung HB. Boceprevir: a proteaseinhibitor for the treatment of hepatitis C. Clin Ther 2012;34(10): 2021–2038.

The impact of contextual factors on theprescribing of psychoactive drugs witholder people: an analysis of treatmentculture in nursing homes using anethnographic approach

C. Shawa, C. Hughesa and B. McCormackb

aQueen’s University, Belfast and bQueen Margaret University, [email protected]

Prescribing of psychoactive medications for older residents innursing homes has been a cause for concern and such medica-tions have been described as ‘chemical restraints’[1]. One factorwhich may influence the prescribing of these medicines istreatment culture, defined as the way in which prescribing ofpsychoactive medication is undertaken[2]. Nursing homes havebeen defined as resident-centred (least likely to use psychoac-tive medication), traditional (most likely) or ambiguous interms of treatment culture[2]. The aim of this research was toexplore treatment culture in nursing homes in respect of theprescribing of psychoactive medications, using an ethno-graphic approach.

A cross-sectional study carried out in 2008 used an adaptedtreatment culture questionnaire in Northern Ireland and NewZealand to determine the treatment culture of 179 nursinghomes[2]. Following on from this, six nursing homes wererecruited into the first phase of a three-phase study. The currentphase (Phase 2) recruited three homes from the original six,one in each category of treatment culture. Data were collectedvia observations, in-depth interviews with nursing home staffand examination of residents’ nursing notes (demographicsand prescribing details, focusing on psychoactive medicines).Observations of the day-to-day working activities of all con-senting nursing home staff and GPs were undertaken over aperiod of approximately 100 hours at each site, on differentdays, and shifts. In-depth interviews sought to explore andexplain observational data. Data were analysed using Schein’sFramework and thematic analysis. Ethical approval wasgranted by the relevant ethics committee.

Observations took place during 94 hours in a traditionalhome, 104 hours in an ambiguous home and 101 hoursin a resident-centred home. The traditional home showed

strict routine, little interaction between staff and residentsand the administration of psychoactive drugs without a non-pharmacological intervention. In the ambiguous and resident-centred homes, there was a greater focus on resident choice,individualised care and interaction between staff and residents,with more emphasis on non-pharmacological treatments suchas distraction and one-on-one attention. In-depth interviewstook place with 15 participants across all homes: three manag-ers, six nurses and six care assistants. Interview data was con-sistent with these findings; a nurse from a traditional homeemphasised the need for ‘specific routines’ while a care assis-tant from a resident centred home placed importance on resi-dent ‘preference and choice’. Ten sets of residents’ nursingnotes were examined. The traditional nursing home had ahigher incidence of psychoactive prescribing, while ambigu-ous and resident-centred homes had less prescribing of thesemedications. The resident-centred home had a higher inci-dence of ‘when required’psychoactive medications as opposedto daily administration in the traditional home.

Initial analysis has shown evidence of behaviours which areconsistent with the treatment culture category to which eachhome had been assigned. Modification of these behaviours,although challenging, may be achieved through staff educationand training. This may lead to a shift towards a more flexible,resident-centred culture and a reduction in the prescribing anduse of psychoactive medication.

1. Inouye S, Marcantonio E, Metzger E. Doing damage indelirium: the hazards of antipsychotic treatment in elderlypeople. Lancet Psychiatry 2014; 1: 312–315.

2. Hughes CM, Donnelly A, Moyes SA, Peri K, Scahill S,Chen C, McCormack BG, Kerse N. ‘The way we do thingsaround here’: an international comparison of treatmentculture in nursing homes. JAMDA 2012; 13: 360–367.

Involving lay researchers in datacollection: a descriptive study

S. Garfielda,b, S. Jheetaa, A. Bischlerc,A. Jacklina, C. Nortona, F. Hussona andB.D. Franklina,b

aCentre for Medication Safety and Service Quality, Imperial CollegeHealthcare NHS Trust, London, UK, bThe Department of Practice andPolicy, UCL School of Pharmacy, Mezzanine Floor, BMA House, London,UK and cPharmacy Department, Chelsea and Westminster HealthcareNHS Foundation Trust, London, [email protected]

It is widely recognised that patients and the public should beinvolved in research in a meaningful, rather than tokenistic,way[1,2]. A recent systematic review[2] found that lay peoplehave been mostly involved in agenda setting and protocoldevelopment. It was much less common for lay people to beinvolved in execution of the research, such as carrying outdata collection.

We here report on our experience in involving lay people indata collection as part of a study of inpatient involvementin medication safety. The first phase of the study involved

Oral abstracts 17


observations of doctors’ ward rounds, pharmacists’ ward visitsand nurses’ drug administration rounds, to determine if and howhealthcare professionals facilitate patient involvement in medi-cation safety. The study had a patient and clinical engagementgroup whose lay representatives suggested that lay peopleshould be involved in these observations in parallel with ourresearchers, a suggestion we adopted. The aim of the work pre-sentedherewas todescribe thebenefitsandchallengesofhavinglay people conduct these observations, in order to inform thefuture role of lay people conducting this type of research.

Observations took place between January and June 2014.NHS ethics approval was obtained for the main study but notrequired to conduct these additional interviews as lay observ-ers were not recruited via the NHS.

We conducted semi-structured interviews with the 3 laymembers involved in data collection and the 4 research teammembers involved in their recruitment, training or supporton the wards. The topic guide explored lay observers’ andresearchers’ views of the benefits and challenges of layinvolvement in the observations; it was informed by prelimi-nary discussions with our lay observers and researchers. Theinterviews were transcribed verbatim and coded openly, usingNVivo 8 for assistance. A second researcher independentlycoded 30% of interviews.

In addition, secondary qualitative analysis of the observa-tional data was carried out to identify the specific input of layobservers into the study findings.The lay members and researchers reported that lay membersadded value to the data by bringing new perspectives. Somechallenges were identified including the infrastructure notbeing in place to support this specialist lay research role, differ-ing paradigms of research governance held by the public andhealthcare professionals, and difficulties in recruitment of adiverse range of lay observers.

The secondary analysis demonstrated that five codes wereadded to the framework as a result of the lay observations.In this study, including observations by lay members addedvalue to the findings. There is a need to build infrastructure inNHS trusts to support this involvement.

This study was limited to interviews with the small numberof lay observers and researchers who were involved in oneresearch project. It is therefore unclear whether theoreticalsaturation was reached or how generalisable the findings are.However the full relevant population was interviewed and,given the scarcity of studies addressing this issue, these find-ings can help inform future lay involvement in data collection.

1. Snape D, Britten N, Froggat K, Gradingre F, Lobban F,Popay J, Wyatt K, Jacoby A. Exploring perceived barriers,drivers, impacts and the need for evaluation of publicinvolvement in health and social care research. A modifiedDelphi study. BMJ Open 2014; 4: e004943. doi:10.1136/bmjopen-2014-004943.

2. Domecq JP, Prutsky G, Elariyah T, Wang Z, Nabhan M,Shippee N, Brito JP, Boehmer K, Hasan R, Firwana B,Erwin P, Eton D, Sloan J, Montori V, Noon A, Dabrh AMA,Murad MH. Patient engagement in research: a systematicreview. BMC Health Services Research 2014; 14: 89. http://www.biomedcentral.com/1472-6963/14/89.

Patients’ medicines management afterhospital discharge – a socialnetwork analysis

B. Fylan Gwynn, A. Blenkinsopp, G. Armitageand D. Naylor

The University of [email protected]

Patients are at heightened risk of harm from their medicineswhen their care is transferred between providers, often becauseof poor information sharing and limited opportunities todiscuss managing their medicines with healthcare profession-als [1]. Patients may have a range of medicines contacts influ-encing how their discharge medicines are used. Using a socialnetworks theoretical framework, this research aims to describethe structure of patients’ personal and professional medicinesnetworks after they are discharged from hospital. It also aims tounderstand the functions provided by those networks.

The study was conducted with patients discharged from thecardiology wards of two acute hospital trusts. A quota sampleof 60 patients was constructed based on a range of demo-graphic variables: age, gender, deprivation and ethnicity. Semi-structured interviews were conducted six weeks after hospitaldischarge. The timing of the interview was designed to allowpatients to interact with healthcare professionals about theirmedicines and to order new supplies. The topic guide wasinformed by a literature review and was critically assessed by apatient representative. Interviews were conducted betweenJanuary and June 2014. Transcribed interview data were ana-lysed by one researcher using thematic analysis; text wascoded and codes were grouped together into themes describingthe structure and function of patients’ networks, and thecontent (what flows between network members) within them.NHS Research Ethics committee approval was granted.

Participants’ ages ranged from 35–80; 42 were male and 18were female and 9 were from ethnic minorities. Patients werefrom areas of high, medium and low deprivation. Patients hadbetween 1 and 15 individual medicines contacts in their net-works. Four types of contact were identified: healthcare pro-fessionals; healthcare support staff; personal contacts; andpersonal healthcare contacts. Some patients did not report GPsand community pharmacists to be members of their medicinesnetworks and others experienced isolation after they left hospi-tal. Content in the network included information and advice,and attitudes and shared experiences about medicines.Patients’ networks were multifunctional: they provided practi-cal and emotional support with medicines; provided, adjustedand monitored medicines; and offered education about medi-cines. Healthcare professionals performed duplicate functionsand some patients experienced safety incidents as a function oftheir networks, for example they failed to receive new orchanged medicines from their GP after their discharge. Somepatients had limited views of the care each professional shouldprovide, for example education and review by communitypharmacy. Patients perceived limited contact betweenmembers of their medicines networks.



Other research has shown that patients’ experiences of caretransition can be disorientating [2]. We found that patients’ net-works included professionals and personal contacts supportingpatients in their use of medicines after their discharge fromhospital. While some benefit from practical and emotionalsupport in the home, others experience limited personalsupport after leaving hospital. Organisation of services meantthat functions – such as educating patients – were duplicated,yet patients’ networks sometimes failed to safely optimise theirmedicines. Patients often do not experience care from profes-sionals, such as GPs, in the way they expect.

1. The Care Quality Commission. Managing Patients’ Medi-cines after Discharge from Hospital. London: The CareQuality Commission, 2009. CQC-039-500-ESP-102009.

2. Knight D, Thompson D, Mathie E, Dickinson A. ‘Seamlesscare? Just a list would have helped!’ Older people and theircarer’s experiences of support with medication on dis-charge home from hospital. Health Expect 2013; 16(3):277–291.

Are pharmacists ready for research?

R Kayyalia, D Berkoa and N Patelb

aKingston University and bUniversity of [email protected]

The pharmacist’s role has evolved over the years to suit an everchanging health system. The recent NHS restructure after theHealth and Social Care Act 2012[1] publication resulting inpharmacy services commissioning that depends on quality,efficiency and performance has made the involvement inresearch important. Pharmacy research is necessary to unlockpharmacists’ potential as a research base to provide evidence-based knowledge.[2] The aim of this study was to examine com-munity pharmacists (CPs) and hospital pharmacists (HPs)willingness to engage in research and what they would need inorder to incorporate it in their practise.

The study targeted all HPs in a large teaching hospital inLondon as well as randomly selected CPs within GreaterLondon (due to convenience). From 2250 pharmacies listed inGreater London, the minimum sample size calculated based on95% confidence interval and 5% margin error was 334. Struc-tured questionnaires, exploring (Partaking in Research,Research Ready, Barriers and Facilitators and Future Engage-ment) were either posted or hand delivered to CPs in February2014. No follow-up mailing was performed. An online ques-tionnaire was developed and sent to HPs. Consent (indicatedvia the returned questionnaires) for face-to-face interviews tobe undertaken in March 2014 was only obtained for eight CPs.Time did not permit interviews of HPs. The study wasapproved by Kingston University Ethics Committee.

A response rate of 20.4% (n = 51) was obtained from CPs.Fourteen out of 40 HPs responded (35%). The modal age rangeof responding CPs was 30–39 and they were equally distrib-uted by gender. The majority of HPs were females with amodal age range of 40–49. Thirty CPs (58.8%) and eleven(79%) HPs rated research as important in their practise. Audits

and service evaluations (mandatory research) were the mainresearch types undertaken. Engagement with clinical trials,applied health or clinical research was greater with HPs thanCPs (85.7% versus 33.3%). Barriers to undertaking researchwere time (most common), lack of training, lack of infrastruc-ture and support. The biggest incentive for research engage-ment was remuneration (76% (n = 39)) and time (85%(n = 12)) by CPs and HPs, respectively. CPs’ potential researchroles included raising public awareness (most common; 40%),recruiting patients or providing support. In contrast, HPs saw arole in leading and generating research and applying forfunding. The interviews highlighted that 5 out of the 8 CPswere not aware of the Research Ready tools and their potentialinvolvement in research.

HPs and CPs recognise the importance of research to theirpractice, but many are not aware of the push for involvement inresearch, or how they could go about doing this. Considera-tions for time, remuneration, training and support need to bemade in order for CPs, in particular, to be more activelyinvolved.

1. Health and Social Care Act (2012). http://www.legislation.gov.uk/ukpga/2012/7/contents/enacted (accessed Novem-ber 2013).

2. Royal Pharmaceutical Society. What Is Research Ready?http://www.rpharms.com/science-and-research/research-ready.asp (accessed November 2013).

Engaging the community pharmacysector in research: lessons learned from anational study of clinical productivity

D. Halsalla, S. Jacobsa, R. Elveya, F. Bradleya,T. Fegana, M. Hannb, E. Schafheutlea,A. Wagnerc and K. Hassella

aCentre for Pharmacy Workforce Studies, Manchester Pharmacy School,The University of Manchester, bCentre for Biostatistics, Institute ofPopulation Health, The University of Manchester and cNIHR, ClinicalResearch Network [email protected]

A major study of organisational factors associated with vari-ation in clinical productivity in community pharmacies inEngland was funded by the National Institute for HealthResearch (National Research Ethics Service approval [13/WM/0137]). This multi-method study, conducted across ninegeographical areas, involved a survey of community pharma-cies, analysis of activity data, a patient survey and stakeholderinterviews. It is not the aim of this paper to present researchfindings. Rather, in a study which touched upon commercialsensitivities, this paper aims to describe and makerecommendations for successful engagement with the commu-nity pharmacy sector to achieve full study support andparticipation.

No data were collected to address this aim, but the approachused to engage with the community pharmacy sector to encour-age study participation is described. Representatives of the

Oral abstracts 19


national negotiating committee, a community pharmacy multi-ple, and an independent pharmacy were approached and con-tributed to study design. On securing funding, this stakeholdergroup was expanded to include service commissioners andpatient representatives to provide ongoing advice on all aspectsof the study. Presentations were made to national and localpharmacy organisations (Company Chemists Association,Local Pharmaceutical Committees (LPCs) and Local Phar-macy Networks (LPNs)) to invite endorsement of the study.Individual owners, superintendent pharmacists and researchgovernance leads as gatekeepers to larger multiples were alsocontacted. Telephone calls, e-mails and one-to-one meetingsaimed to clarify the purpose and benefits of the study, addressconcerns and secure agreement to participate.

This study encountered reactions from the community phar-macy sector ranging from enthusiasm and encouragement todeep mistrust and negativity which stemmed partly from com-mercial sensitivities and concurrent changes in commissioningarrangements during 2013. Engagement of stakeholders inproject design and in an ongoing advisory capacity was invalu-able in addressing the concerns of the sector and ensuring thatearly misunderstandings were overcome through improvedcommunication. Early discussions, clarification and reassur-ance helped secure explicit endorsement from four of the nineLPCs approached, and four of the nine major pharmacy multi-ples agreed for their pharmacies to participate. Non-participation by five multiples necessitated expanding thenumber of geographical areas from five to nine to maintainsample size. Response rate to the pharmacy survey was low(34%) but sufficient to power the analysis. The process ofrecruitment of pharmacies to the second stage of the study(patient survey and stakeholder interviews) was greatlyenhanced by already established relationships and earlierlessons learned.

Engaging with the community pharmacy sector to secureparticipation in a major national study has presented severalchallenges and has achieved moderate success. While not allstakeholders endorsed the study, support from those who didhelped boost study participation. Careful consideration mustbe given to forms of communication and terminology used,and the political and commercial landscape in which commu-nity pharmacy operates must be anticipated when costingbids and planning timescales. Early engagement with phar-macy leaders and national bodies during study design canpromote support; face-to-face meetings with individualsuperintendent pharmacists and pharmacy owners remaincrucial for creating rapport.

A diary study of community pharmacists’mental workload

H.E. Family, M. Weiss and J. Sutton


Mental Workload (MWL) is a widely studied concept withinthe human factors and patient safety fields. MWL is measured

to identify how busy employees are, the complexity of tasksbeing carried out and whether additional tasks can be added totheir current workload without reducing task performance orsafety. [1] To date, the MWL of community pharmacists (CPs)practising in the United Kingdom (UK) has not been measured.This study aimed to measure CPs’ MWL throughout a day intheir usual practice.

Following receipt of university ethics approval a conveni-ence sample of 104 CPs who had participated in an earlierstudy were invited to participate. The 104 CPs were strategi-cally recruited through superintendent pharmacists and socialmedia to be demographically representative of UK CPs. FiftyCPs expressed interest in this follow-up study and were posteda diary containing questions on demographics, the enhancedand locally commissioned services offered and seven copies ofthe NASA Task Load Index (TLX).[2] The NASA-TLX is aself-report measure of MWL that measures six facets of MWL(time pressure, mental effort, mental demand, physicaldemand, task frustration and performance concern), each ratedon an 11 point scale (0–10). The mean of the scores on the sixitems provides an overall MWL score. Participants were askedto complete between two and seven MWL ratings over one dayand received a £5 highstreet voucher for their participation.MWL ratings were plotted against time of day. Bivariate corre-lations were produced between overall MWL and the demo-graphic characteristics of participants. Potentially significantrelationships were explored further with an independent t-test.

Forty CPs (19 men) returned their diaries (response rate of38%) and a total of 186 MWL ratings were made across thediaries, providing sufficient power to carry out the plannedanalyses. MWL ratings were plotted against time of day. Peaksin overall MWL were seen at 11 o’clock in the morning and fiveo’clock in the afternoon. Time pressure, mental and physicaldemand were the highest rated facets of MWL. Bivariate corre-lations revealed no relationship between participant sex, age,pharmacy experience, or the type of community pharmacy CPsworked in and overall MWL. However, the correlations indi-cated a relationship between overall MWL and whether the CPworked full time (>35 hours a week) or part-time (48 ratingswere from part-time CPs). An independent t-test showed thatpart-time CPs reported significantly lower overall MWL(mean = 4.39) compared with full-time CPs (mean = 5.14) (t(184) = 2.29, p < 0.05, r = 0.17).

These results suggest that MWL peaks for many CPs at spe-cific times of the day and that part-time CPs experience signifi-cantly less MWL compared with their full-time colleagues.However, the effect size of this difference was small. The diaryratings also suggest that time pressure, mental and physicaldemands are the primary sources of MWL that CPs experience.This initial study highlights the impact working patterns haveon MWL and has implications for the times of day that safetycritical tasks are carried out by CPs.

1. Wickens CD, Hollands JG, Banbury S, Parasuraman R.Mental workload, stress, and individual differences: cogni-tive and neuroergonomic perspectives. EngineeringPsychology and Human Performance (InternationalEdition), 4th edn. Upper Saddle River, NJ: Pearson, 2013:346–376.



2. Hart SG, Staveland LE. Development of the NASA-TLX(Task Load Index); results of empirical and theoreticalresearch. In: Hancock PA, Meskkati N, eds. Human MentalWorkload. Amsterdam, Netherlands: North-Esland, 1988.

An evaluation of compliance withmedicine-related criteria mandated in anational standard for patient dischargesummary information

B. Bourkea,b, C. Keanea and S. Cunninghamb

aSaint Vincent’s University Hospital, Dublin, Ireland and bSchool ofPharmacy & Life Sciences, Robert Gordon University, [email protected]

Transition of care between hospital and community is a vulner-able time for patient safety and medication management.1 Ithas been shown that failure to communicate or document stop-ping or withholding a medicine and failure to communicate ordocument omission of a pre-admission medicine are the mostcommon types of non-reconciliation on discharge from Irishhospitals.2 The discharge summary is an important tool in com-municating essential information about a patient’s hospitalisa-tion to their primary care physician. At present, a standardiseddischarge summary template is not available in Ireland.However, the Health Information and Quality Authority(HIQA) recently published the “National Standard for PatientDischarge Summary Information” outlining mandatory infor-mation that should be documented in a discharge summary.The aim of this study was to establish the current level of com-pliance with the medicine-related criteria in National Standardfor Patient Discharge Summary (PDS) Information by medicalteams in an Irish hospital.

A retrospective survey of 100 discharge episodes wascarried out at St Vincent’s University Hospital (SVUH),Ireland over a 4-week period. Patients discharged frommedical specialties were eligible for inclusion. Psychiatric andsurgical patients were excluded due to the limited number ofdischarges from psychiatry teams and the lack of changesmade to surgical patients’ medication regimens. Demographicand medication data were collected from multiple sources andcompared to information in the PDS and discharge prescrip-tion. Data was collected within one week of discharge and tenpercent of data collection forms were validated for data entry.Ethics approval was sought from RGU Research Ethics Com-mittee and SVUH Clinical Audit Committee. All the data wereentered into a database on SPSS® version 16.0 (SPSS Inc) and95% confidence intervals were calculated.

One hundred discharge episodes from 10 medical specialtiesincorporating 1011 medications were studied. Patient dis-charge summaries were available for all discharge episodes,but only 94 discharge prescriptions were available. Four dis-charge summaries did not have a “date completed” docu-mented but the majority (n = 74, 77.1%, ± 8.41) of theremaining 96 were completed on the day of discharge. Themajority of discharge summaries (n = 61, 61%, ± 9.56) andprescriptions (n = 51, 54%, ± 10.08) were completed by the

most junior member of the medical team. Details on the allergystatus were more likely to be documented on the prescriptionrather than the PDS. Discrepancies between the twodocuments occurred in 24.7% (n = 171, ± 3.21) of dischargemedicines. There were 256 medicines newly commenced, only10.9% (n = 28, ± 3.82) of which had the indication explicitlystated. The indication was implied for another 75.8% (n = 194,± 5.25), leaving 13.3% (n = 34, ± 4.16) drugs where theindication would have been unclear to the GP. Rationalefor changes to pre-admission medicines was also poorlydocumented.

Discharge documentation was not in full compliance withthe HIQA standard. Discharge summaries were completed in atimely manner, and prescriptions were found to be more accu-rate than the medication list in the discharge summary. Theelectronic system should be re-designed to prompt for theinformation required by the standard and to allow the auto-matic generation of the discharge prescription.

1. Tam VC, Knowles SR, Cornish PL, Fine N, Marchesano R,Etchells EE. Frequency, type and clinical importance ofmedication history errors at admission to hospital: a sys-tematic review. Canadian Medical Association Journal2005; 173(5): 510–515.

2. Grimes TC, Duggan CA, Delaney TP, Graham IM, ConlonKC, Deasy E et al. Medication details documented on hos-pital discharge: cross-sectional observational study offactors associated with medication non-reconciliation.British Journal of Clinical Pharmacology 2011; 71(3):449–457.

An exploration of the views of keystakeholders on generic drugsubstitution in Ireland:a qualitative study

P.J. McCaguea,b, M. Kantchevb, J. Careyb,A. Stauntonb and S. McCarthyb,c

aSchool of Pharmacy, Liverpool John Moores University, bSchool ofPharmacy, University College Cork, Cork and cCork University Hospital,Cork

Ireland is acknowledged to be a country with one of the lowestuse of generic medicines in Europe1. In an attempt to improvethe use of generic medicines in Ireland, in May 2013, TheHealth (Pricing and Medical Goods) Act 2013 was introduced.This legislation requires pharmacists to dispense specifiedgeneric medicines to patients regardless of the brand pre-scribed. The aim of this study was to ascertain the views ofgeneral practitioners (GPs), pharmacists and patients ongeneric medicines and the impact of the new governmentpolicy.

The study received full ethical approval from the ClinicalResearch Ethics Committee of the Cork Teaching Hospitals.Initially a convenience method of sampling was used to inviteparticipants to take part with subsequent participants recruitedby a ‘snowballing effect’. Interviews were carried out face-to-

Oral abstracts 21


face or by telephone depending on the availability and locationof the participant. An interview schedule of open-endedquestions was developed following a review of the literature.Interviews were recorded, transcribed verbatim and analysedusing thematic analysis. Each of the three stakeholder groupswere analysed separately.

Fourteen GPs, 17 pharmacists and 12 patients participated inthe interviews. Data saturation was achieved in all threegroups. A number of recurring themes were identified includ-ing willingness to accept generics, concerns, and successfulimplementation of the legislation into practice. GPs were gen-erally positive about the introduction of the legislation andappeared committed to the cost-effective use of generic medi-cines in their practice. Some older GPs had some concernsregarding quality of these products. The study identified anemerging group of young prescribers willing to use interna-tional non-proprietary names when prescribing. Althoughpharmacists themselves were positive about the introduction ofthis legislation, they believed that misconceptions held bypatients could be a barrier to successful change in practice. Anumber of pharmacists felt frustrated that the burden of imple-menting the legislation was being placed on them. Pharmacistsalso raised the issue of safety, expressing concern with respectto appropriateness of potential generic prescribing for drugswith a narrow therapeutic index. Both groups of healthcareprofessionals discussed strategies for improving the imple-mentation of this legislation. On the whole, patients held posi-tive views on generic medicines and recognised the cost-savingbenefit for both the health service and for patients who pay fortheir medication. The patients interviewed generally believedthat there was not enough information or education availableto them concerning generic medicines. They felt that the useof generics could cause confusion, particularly to elderlypatients.

This work has described, for the first-time, the views of keystakeholders on this major legislative change in Ireland. It isclear from our findings that improved educational interven-tions are required for both healthcare professionals andpatients in order to alleviate concerns or misconceptions theymay have. Policymakers now need to address this issue in orderto improve the uptake of generic medicines with both prescrib-ers and patients alike. Future work includes a questionnairestudy in order to obtain more generalisable results for thestakeholders described.

1. Brick A, Gorecki P, Nolan A (2013). Ireland: Pharmaceu-tical Prices, Prescribing Practices and Usage of Genericsin Comparative Context. [Online]. http://www.tara.tcd.ie/bitstream/handle/2262/67333/RS32.pdf?sequence=1&isAllowed=y (accessed September 2014).

An evaluation of NHS Health Checks incommunity pharmacy: changes incardiovascular risk factors over one yearin patients aged 40–74

S. Butterworth, S.C. Willis and P. Higginson


NHS Health Checks are a national cardiovascular risk assess-ment service offered since 2012 to people aged 40–74without recognised cardiovascular disease (CVD). TheHealth Check uses age, gender, weight, height, smoking,alcohol use, exercise, family history, blood pressure and cho-lesterol to calculate a 10-year CVD risk using QRisk2. Theaim of the Health Check is to identify a person’s CVD risk,communicate it in a way that is understood, and manage itappropriately. The study reported here undertaken in a singlerural community pharmacy evaluated all patients receivingthe service from when first commissioned until data collec-tion ended (December 2012–June 2014) against PublicHealth England (PHE) Standards and also investigated differ-ences in risk between patients. A sample of patients was fol-lowed up after one year by repeating the Health Check tomeasure change over time.

Eligible subjects were identified using pharmacy records orreferred by a nearby general practice (GP). Data collection wascompleted within the pharmacy with CVD risk calculatedusing HealthOptions software. Comparisons between sub-groups used Mann-Whitney test for unpaired data inStatsDirect. Follow-up Health Checks took place after oneyear with all patients who received an initial check invited totake part until a minimum sample of 50 was exceeded toincrease statistical power. Wilcoxon’s signed rank test forpaired data was used to determine change in risk over one year.Manchester University Research Ethics Committee grantedethical approval.

161 participants received a Health Check during the firstround of data collection with a full data set recorded for eachsubject meeting PHE Standard 3. Mean 10-year CVD riskwas 9.3% (95% CI 8.2–10.5). Risk in males was 12.9% com-pared with 7.1% in females (p = 0.0003). Seventy-eightpatients (48.4%) were referred to their GP for further assess-ment. Of these 59 (36.6%) had blood pressure >140/90 mmHg, 7 had raised total cholesterol, 23 (14.3%) a bodymass index >30 and 18 (11.1%) a QRisk of >20. Fifty-twopatients received a repeat Health Check, with QRisk2 foundto have increased over the year; female mean risk atincreased from 5.6% to 6.0% (p = 0.1134); for males riskincreased from11.0% to 11.8% (p = 0.0095).

Findings demonstrate that community pharmacy can effec-tively perform NHS Health Checks that meet PHE standards,despite concerns that non-GP providers are notable to do so(1). While the increase in QRisk2 over one yearappears disappointing, findings are consistent with the Rosehypothesis(2) for outcomes in low risk versus high risk popula-tions as the mean risk of patients receiving the service was9.3% and hence for most subjects additional treatment was



inappropriate. Despite this lack of reduction in risk, findingssuggest that community pharmacy is able to identify and referhigh risk patients – and, moreover, that a rural location is not abarrier to this. What remains unknown at this stage is the extentto which community pharmacy services can have a positiveimpact on longer term outcomes.

1. Research Works Ltd. Understanding the implementation ofNHS Health Checks. 2013.

2. Rose G. Sick individuals and sick populations. Int J ofEpidemiol 1985; 14(1): 32–38.

Improving the appropriateness ofprescribing in older patients: a systematicreview and meta-analysis of pharmacists’interventions in secondary care settings

K. Walsh, D. O’Riordan, P. Kearney,S. Timmons and S. Byrne

University College [email protected]

Older patients are at a much greater risk of potentially inappro-priate prescribing (PIP) than the general population as a conse-quence of polypharmacy. PIP has been found to increase therisk of adverse drug events (ADEs). SuchADEs can potentiallylead to increased morbidity, mortality and health resource uti-lisation.[1] Reduction of PIP is even more critical in personswith dementia who have a greater susceptibility to the deleteri-ous effects of potentially inappropriate medications.[2] Theobjectives of this review were to determine from the literaturewhether there was sufficient evidence to support the claim thatpharmacists’ interventions can improve the quality of prescrib-ing in older persons or persons with dementia who are admittedto hospital.

This systematic review and meta-analysis was undertaken incompliance with PRISMA guidelines. An electronic search ofthe literature was conducted using twelve databases frominception up to and including September 2014. The searchterms included synonyms and various combinations of the fol-lowing key words; “pharmacist”AND “inappropriate prescrib-ing” AND “elderly” OR “dementia” AND “hospital” AND“pharmaceutical care” AND “randomised controlled trial.”Study selection and risk of bias assessments were conductedby two independent reviewers. Data extraction were performedby one reviewer and verified by another. Consensus on all deci-sions was reached by discussion. Quantitative analysis wasperformed where appropriate.

A total of 1,752 records were found after duplications wereremoved. After two stages of review, five trials were included.One trial was excluded from quantitative analysis as it wasdeemed to be at a high risk of bias. In total, 1,164 patients wereincluded (589 in intervention and 575 in control arms respec-tively). The mean number of prescribed medications rangedfrom 7.7 to 8.7 in the intervention arms (mean = 8.1) and 7.3 to8.0 in the control arms (mean = 7.5). A key component of allincluded studies was inter-disciplinary teamwork. All trials

included patients with dementia however no study was foundwhichdealtwithdementiapatients specifically in thiscontext.Ameta-analysis using random-effects inverse variance methodsfound there was a statistically significant difference in the meanin favour of the pharmacists’ intervention when the summatedmedication appropriateness index data were analysed at dis-charge (n = 4, mean difference = −5.27, 95% CI: −8.44, −2.11,I2 = 93%). Overall these results should be viewed with somecaution as there were only a small number of included trials,there was considerable heterogeneity and the included studieswere assessed to have a moderate risk of bias.

The findings of this systematic review and meta-analysissupport the hypothesis that pharmacists can improve the appro-priateness of prescribing in older inpatients when they work aspart of a team. Hospital managers and policy-makers have akey role to play in encouraging pharmacists to join ward basedteams and to utilise their clinical skills in order to maximisepharmaceutical support of older people. More research isrequired to determine whether a pharmacist-led interventioncan reduce PIP and consequently improve outcomes forpersons with dementia.

1. Gallagher PF, Barry PJ, Ryan C, Hartigan I, O’Mahony D.Inappropriate prescribing in an acutely ill population ofelderly patients as determined by Beers’ Criteria. Age andAgeing 2008; 37(1): 96–101.

2. Brandt NJ, Turner T. Potentially inappropriate medicationsin older adults: a review of the 2012 Beers Criteria and theImplications in Persons with Dementia. Ment Health Clin2014; 4(4): 75. http://cpnp.org/resource/mhc/2014/07/potentially-inappropriate-medications-older-adults-review-2012-beers-criteria (accessed 25 September 2014).

Potentially inappropriate prescribing in amiddle-aged population: a cross-sectionalstudy in Northern Ireland using theEnhanced Prescribing Database

J.A. Coopera,b, F. Moriartyb, C. Ryana,S.M. Smithb, E. Wallaceb, K. Bennettc, C. Cahirc,D. Williamsd, M. Teelingc, T. Faheyb andC.M. Hughesa,b

aSchool of Pharmacy, Queen’s University Belfast, bHealth Research BoardCentre for Primary Care Research, Royal College of Surgeons in Ireland,cDepartment of Pharmacology & Therapeutics, Trinity Centre for HealthSciences and dDepartment of Geriatric and Stroke Medicine, RoyalCollege of Surgeons in [email protected]

Traditionally, the focus of potentially inappropriate prescrib-ing (PIP) has been on older people. However, there is evidencethat multimorbidity is also prevalent in middle-aged people(defined as 45–64 years)[1], but as yet, there has been little con-sideration of the associated risk of PIP in this age group.PROMPT (PRescribing Optimally in Middle-aged People’sTreatments) represents a set of 22 explicit prescribing criteriawhich may be applied to prescribing datasets (in the absence of

Oral abstracts 23


clinical information) to determine the prevalence of PIP inmiddle-aged people.[2] In this study, we aimed to investigate thelevel of PIP in Northern Ireland using data from the EnhancedPrescribing Database (EPD).

A retrospective cross-sectional study was conducted byapplying PROMPT to population-based data from the EPDbetween 01/01/2012 and 31/12/2012. Patients (aged 45–64years) were included if they had three months lead-in data priorto the study start date and continuous data collection during thestudy period. The prevalence for each PROMPT criterion, theoverall prevalence and corresponding 95% confidence inter-vals (CIs) were calculated. Logistic regression was used toinvestigate the association between PIP and gender (male/female), age group (45–49, 50–54, 55–59, 60–64 years) andpolypharmacy (≥4 different drugs classes). Ethical approvalwas granted by the National Research Ethics Service (NRES)Committee.

This study included 441,925 patients (49.4% female; 50.6%male). The overall prevalence of PIP in this population was21.9% (95% CI 21.8–22.0). During the study period, 13.9% ofpatients received one potentially inappropriate prescription(n = 61,395), 4.9% received two potentially inappropriate pre-scriptions (n = 21,519) and 3.1% received three or more poten-tially inappropriate prescriptions (n = 13,819). The mostcommon instances of PIP are shown in Table 1. Adjusting forage and gender, patients receiving ≥4 different drugs classeswere 12 times more likely to receive at least one potentiallyinappropriate prescription during the study period comparedwith patients receiving 0–3 different drugs classes (OR 12.18,95% CI 11.97–12.39, P < 0.01).

Table 1 Most common instances of PIP in middle-aged people using datafrom the EPD

Mostcommoninstancesof PIP

PROMPT criterion Patients(N = 441,925)(%)

1st Proton pump inhibitors (PPIs) should not beprescribed at doses above therecommended maintenance dosage forgreater than eight weeks.

33,809 (7.7)

2nd Strong opioids should not be prescribedwithout the co-prescribing of at least onebulk-forming or osmotic laxative.

31,110 (7.0)

3rd Benzodiazepines should not be usedlong-term (greater than four weeks).

17,698 (4.0)

4th First generation antihistamines should not beused as first-line agents for greater thanseven days.

11,098 (2.5)

5th Non-benzodiazepine hypnotics (zolpidem,zaleplon, zopiclone) should not be usedlong-term (greater than four weeks).

10,875 (2.5)

This study shows that during the study period, approximatelyone in five middle-aged adults received at least one potentiallyinappropriate prescription; with the risk of PIP increasing withpolypharmacy. These findings may help to inform interven-tions to improve prescribing practices in this age group.

1. Barnett K, Mercer SW, Norbury M, Watt G, Wyke S,Guthrie B. Epidemiology of multimorbidity and implica-tions for health care, research, and medical education: across-sectional study. Lancet 2012; 380: 37–43. 9.

2. Cooper JA, Ryan C, Smith SM, Wallace E, Bennett K, CahirC, Williams D, Teeling M, Fahey T, HughesCM (PROMPT Steering Group). The development of thePROMPT (PRescribing Optimally in Middle-agedPeople’s Treatments) criteria. (Article is under review).

Trends in polypharmacy and prescribingappropriateness from 1997 to 2012

F. Moriartya,b, C. Hardya, K. Bennetta,c, S. Smitha

and T. Faheya

aHRB Centre for Primary Care Research, Department of General Practice,Royal College of Surgeons in Ireland (RCSI), 123 St. Stephens Green,Dublin 2, Republic of Ireland, bHRB PhD Scholars Programme in HealthServices Research, Royal College of Surgeons in Ireland (RCSI), 123 St.Stephens Green, Dublin 2, Republic of Ireland and cDepartment ofPharmacology & Therapeutics, Trinity Centre for Health Sciences, StJames’s Hospital, Dublin 8, Republic of [email protected]

Medicines are one of the most common healthcare interven-tions and volumes prescribed have risen in recent years. It isunclear how this increase in prescriptions has impacted onpatients and to what extent this reflects increased use ofevidence-based treatments or rising inappropriate prescriptionof medicines. The aim of this study is to determine over a15-year period (1997–2012) in Ireland how prescribing pat-terns and rates of polypharmacy (being prescribed ≥5 regularmedications) have changed, and in older adults (aged ≥65years) to determine how the prevalence of potentially inappro-priate prescribing (PIP) has changed.

This repeated cross-sectional study using the Primary CareReimbursement Service administrative pharmacy claims data-base includes all people eligible for the General Medical Ser-vices scheme in the Eastern Health Board region of Ireland inthe years 1997, 2002, 2007 and 2012 (data pre-1997 wasunavailable). The number of regular (dispensed in ≥3 consecu-tive months) drug classes dispensed per participant during eachstudy year was determined. Using a subset of 30 criteriafrom the Screening Tool for Older Persons’ Prescriptions(STOPP)[1], the PIP prevalence in those aged ≥65 yearswas also calculated. Negative binomial regression modelswere fitted to determine, after adjusting for knownconfounders, the association between study year and the out-comes of polypharmacy (≥5 regular medications) and majorpolypharmacy (≥10 regular medications) while logistic regres-sion was used for the outcome of any PIP. Ethical approval wasnot required.

The study population ranged from 338,025 in 1997 to539,752 in 2012. There was a marked increase from 1997 to2012 in the proportion of participants with polypharmacy, par-ticularly among older age groups (8.3–30.2% and 17.8–60.4%for those aged 45–64 and ≥65 respectively). The incident rateratio for having polypharmacy in 2012 compared with 1997



was 4.16 (95% CI 3.23, 5.36) after adjusting for age group andgender, and for major polypharmacy was 10.53 (95% CI 8.58,12.91). The proportion of older study participants with PIPdecreased from 32.6% in 1997 to 28.6% in 2002 but has sincerisen to 32.8% and 37.3% in 2007 and 2012. Although pre-scribing of high doses of aspirin and digoxin has decreasedsince 1997, prevalence of other criteria increased substantially,such as long-term use of high dose proton pump inhibitorsfrom 0.8% to 23.8%. The odds ratio for the presence of any PIPversus none in 2012 compared with 1997 was 0.39 (95% CI0.39, 0.4) after adjusting for gender and polypharmacy levels.

This study demonstrates the trend of rising numbers ofmedicines prescribed in the older-aged population over time.The large increase in those taking ≥10 medicines is parti-cularly concerning from a health expenditure perspective anddue to the risk of adverse events in these complexpolypharmacy patients[2]. While STOPP criteria prevalencerose during the study, after controlling for increasingpolypharmacy levels, older individuals are now less likelyto have PIP. Although prescribing of some medicines hasimproved, a focus on PIP drug classes with high orincreasing prevalence is important to improve prescribingappropriateness.

1. Gallagher P, Ryan C, Byrne S, Kennedy J, O’Mahony D.STOPP (Screening Tool of Older Persons’ Prescriptions)and START (Screening Tool to Alert Doctors to RightTreatment): consensus validation. Int J Clin Pharm Ther2008; 46(2): 72–83.

2. Hajjar ER, Cafiero AC, Hanlon JT. Polypharmacy inelderly patients. Am J Geriatr Pharmacother 2007; 5(4):345–351.

Medicine-related services: do pharmacistssee things the same way as the public?

J. Krska, S.A. Corlett, S. Gammie, R.L. Loo andR.M. Rodgers

Medway School of Pharmacy, The Universities of Greenwich and [email protected]

For pharmacists to help optimise medicines, the services theyprovide must meet the expectations of people who use them.Relatively little is known about what the public expect andwant from pharmacy services or whether pharmacists’ percep-tions of expectations are accurate.

This work aimed to compare the perceptions of pharmacistsand the public on aspects of the Medicines Use Review (MUR)and New Medicines Service (NMS).

Questionnaires focusing on medicines-related services wereadapted from a previously validated instrument, using findingsfrom focus groups with pharmacists and the public. The ques-tionnaires were designed to overlap, enabling pharmacist andpublic views to be compared. Both were piloted before use.

The pharmacist questionnaire was distributed to all836 community pharmacies in Kent, Surrey and Sussex.

Non-responders received a second mailing and a telephone callwith further questionnaire, if requested. (1)

Interview-assisted questionnaires, a standard, cost-effectivemarket research method, were completed with 1000 partici-pants at High Street locations in ten towns across Kent. (2)

Chi-square tests were used to assess differences betweenviews (SPSS v20). University research ethics approval wasgranted.

The valid pharmacist response rate was 40.8% (341); 53.0%(179) were female and 65.4% (223) from large multiples.General public response rate was 47.2%; 52.6% were femaleand 60.5% used regular prescribed medicines.

Pharmacists considered that more people would know aboutthe MUR service than actually did: 38.7% thought at least halfthe public would have heard of it, whereas the actual figure was18.2%. The even lower lack of awareness of the NMS washowever anticipated: 93.2% of pharmacists thought very few/less than half the public would know about it, and in fact only8.6% did.

There were significant differences in the proportions ofpharmacists estimating and the public agreeing they would bewilling to make an appointment for a medicine-related service(47.2% versus 71.1%; p < 0.001) and in willingness to wait tosee the pharmacist: 56.0% of pharmacists thought peoplewould wait no more than 5 minutes, but only 20.3% of thepublic agreed with this waiting time, with 71.4% selecting10/15 minutes (p < 0.001). In contrast perceptions concerningdiscussion time with a pharmacist were more closely aligned,with just over half of both pharmacists and the public viewingno more than 15 minutes as acceptable.

There were differences in the reasons people might givewhen considering using a service (all p < 0.001). Fewer phar-macists (57.5%) than members of the public (76.0%) thoughtpeople would use a service ‘because they had problems withtheir medicines’. Most pharmacists thought the main reasonfor using the service would be ‘because the pharmacist asked’(93.0%) whereas fewer members of the public viewed this aslikely (68.7%). Over half the public (53.2%) viewed ‘wantingto help the pharmacist’ as a reason for using a service, com-pared with 22.0% of pharmacists.

Pharmacists’ and the public’s perceptions were not aligned.Pharmacists need greater awareness of what the public see asacceptable and desirable, particularly around waiting timesand appointments, as these are essential aspects of medicines-related services.

1. Rodgers RM, Loo R-L, Krska J. Public perception ofmedicines-related advisory services provided by pharma-cies. Int J Pharm Pract 2013; 21 (Suppl 2): 28–29.

2. Corlett SA, Dodds LJ, Gammie S. Do pharmacists thinkthat medicines advisory services improve patient adher-ence? Int J Pharm Pract 2014; 22 (Suppl 1): 34–35.

Oral abstracts 25


‘You just forget that the pharmacist isactually there . . .’: views of people withlong-term conditions (LTCs) on usingcommunity pharmacy forself-care support

O.J. Ogunbayoa, E.I. Schafheutlea, C. Cuttsb

and P.R. Noycea

aCentre for Pharmacy Workforce Studies (CPWS) and bCentre forPharmacy Postgraduate Education (CPPE), Manchester Pharmacy School,The University of Manchester M13 [email protected]

Self-care support by healthcare professionals (HCPs) is a dis-tinct holistic, patient-centred approach to managing LTCs.1

People with LTCs are regular users of community pharmacies,and dispensing and other services provide opportunities forself-care support. This study aimed to explore the views ofpeople with LTCs on self-care and how they utilise communitypharmacy for support.

Semi-structured, face-to-face interviews were conductedwith people with LTCs between May 2013 and June 2014. Par-ticipants were approached via a convenient sample of oneGeneral Practice and four community pharmacies in North-west England, and four pharmacists in two Health Boards inScotland. Forty-four participants in England and twenty-threein Scotland were identified purposively; fifteen in England andnine in Scotland were recruited and interviewed to achievemaximal variation2 of LTC type and demographics. Interviewtopics were developed from the literature and used as a frame-work for the analysis. Topics covered living with LTCs andundertaking self-care, sources of self-care support, and use ofcommunity pharmacy. All interviews were audio-taped, tran-scribed verbatim and analysed thematically. Interviews anddata analysis were undertaken simultaneously until data satu-ration was reached. NHS Research Ethics and R&D approvalswere obtained.

Participants’ mean age was 62 (range: 24–92) years and 16were females. Most (80%) participants had multiple LTCs.Participants had at least one of asthma/COPD (46%), diabetes(30%) and cardiovascular diseases (24%). The main themesthat emerged were structured using a predefined thematicframework based on the topic guide. Almost all participantsagreed that self-care was an integral part of daily living, andthat they were already engaging in self-care behaviours such asadhering to their medicines regimen. Many others suggestedthat they needed support in areas such as changing lifestyles,self-monitoring and access to self-care resources. Most partici-pants indicated that support networks such as family/carers andfriends were primary sources of information and support withself-care, although they also indicated that they relied onHCPs, particularly nurses in their GP practices. Almost allviewed community pharmacy as, primarily, the supplier oftheir prescribed medicines, and not as a resource where theycould get support for self-care of their LTCs. Additionally,many were unaware of the established community pharmacy’smedicines use services; the few who had experience of these

services did not view them as a resource to improve their self-care behaviours. While some participants recognised thesupport that community pharmacy provided to them such aswith minor ailments and some public health services (e.g. stopsmoking), most appeared reluctant to acknowledge commu-nity pharmacy’s role in supporting their LTCs needs, includingself-care.

While study participants with LTCs engaged in self-caresupported by their family, friends and some HCPs, they did notview community pharmacy as playing much of a role in themanagement and self-care support of their LTCs. The perspec-tives and needs of people with LTCs need to be better under-stood, so that community pharmacy services can be designedso as to contribute to their self-care needs.

1. Kennedy A, Rogers A, Bower P. Support for self carefor patients with chronic disease. BMJ 2007; 335(7627):968–970.

2. Creswell JW. Qualitative Inquiry and Research Design:Choosing among Five Approaches. Thousand Oaks, Cali-fornia: SAGE Publications, Inc., 2006.

Economic evaluation of a randomisedcontrolled trial of pharmacist-supervisedpatient self-testing of warfarin therapy

J. Gallaghera, S. McCarthya, N. Woodsb,F. Ryanc, S. O’Shead and S. Byrnea

aSchool of Pharmacy, University College Cork, bCentre for Policy Studies,University College Cork, cMcGee Pharma International and dDepartmentof Haematology, Cork University [email protected]

Management of oral anti-coagulation therapy (OAT) in a hos-pital based setting is an unnecessary burden on tertiarycare services. Alternative flexible systems of anti-coagulationmanagement are available. One option is pharmacist ledpatient self-testing (PST) of international normalised ratio(INR) levels. PST has demonstrated improvements in anti-coagulation control, but its cost-effectiveness is inconclusive.The aim of this study is to conduct a cost-effectiveness evalu-ation of an automated direct-to-patient expert system, enablingremote and effective management of patients on OAT.

We conducted an economic evaluation alongside arandomised controlled trial investigating a pharmacist led PSTmethod. Full results of clinical trial are published elsewhere [1].Trial was conducted in an Irish university teaching hospital.Long term anticoagulation patients (n = 162), were recruited toa 6-month cross over study between PST and routine care in ahospital based anti-coagulation clinic. Costs were informed byindividual patient resource use, published salary scales, datafrom suppliers of expert systems and in-house hospital data.The outcome measure of this economic evaluation was theincremental cost of patient management over a 6-monthperiod. Economic evaluation was from the healthcare payerperspective. Ethical approval was obtained from all respon-sible committees.



On a per patient basis over a 6-month period, PST resulted inan incremental cost of €59.08 in comparison with routine care.Patients achieved a significantly higher time in therapeuticrange (TTR) during the PST arm in comparison with routinecare. Overall cost of managing a patient through pharmacistsupervised PST for a 6-month period is €226.45. Additionalanalysis of strategies from a societal perspective indicated thatPST was the dominant strategy.

Table 1 Cost-effectiveness of 6 months of PST versus anticoagulationmanagement service (usual care)

Patient self-testing Anticoagulationmanagementservice

Mean % TTR (95% CI) 72 (+/− 2.32%) 59 (+/− 3.36%)Median % TTR (IQR) 74 (64.6–81) 58.6 (45.5–73.1)Mean INR tests/patient ± SD

(range)41.7 +/− 6.6 (24–60) 10.7 +/− 5.2 (5–35)

Cost of 6 months of patientmanagement

€226.45 €167.38

Incremental cost of6 months of PST therapyversus AMS

€59.08

PST is marginally more expensive in comparison with central-ised laboratory testing. However, the associated increase inINR control for a modest increase in expenditure providesfurther evidence that optimally managed warfarin therapyremains a successful strategy for anticoagulation management.TTR levels demonstrated in the intervention arm of the studyare associated with significantly improved patient outcomes interms of death, major bleeding and stroke[2].

1. Ryan F, Byrne S, O’Shea S. Randomized controlled trial ofsupervised patient self-testing of warfarin therapy using aninternet-based expert system. J Thromb Haemost 2009; 7:1284–1290.

2. White HD, Gruber M, Feyzi J, Kaatz S, Tse HF, Husted S.Comparison of outcomes among patients randomized towarfarin therapy according to anticoagulant control: resultsfrom SPORTIF III and V. Arch Intern Med 2007; 167: 239–245.

Oral abstracts 27


hsrpp 2015 ijpp pain

Documents