kwint.pdf

Photography: Patricia Nauta, Leiden, and Ruth Veneboer, s-HertogenboschCover design andlay-out inside work: Francis te Nijenhuis, zonnezijn creaties, s-HertogenboschPrinted by: Optima Gra!sche Communicatie, Rotterdam

!e work presented in this thesis was performed at the Division of Pharmaco-epidemiology & Clinical Pharmacology, Utrecht Institute for Pharmaceutical Sciences (UIPS), Faculty of Science, Utrecht University, Utrecht; the Department of Public Health and Primary Care, Leiden University Medical Center, Leiden and the SIR Institute for Pharmacy Practice and Policy, Leiden.

CIP-gegevens Koninklijke Bibliotheek, Den Haag

Kwint, H.F.Improving appropriate medication use for older people in primary care.!esis Utrecht University - with ref. - with summary in DutchISBN/EAN: 978-94-6169-373-0

2013 Henk-Frans Kwint

For articles published or accepted for publication, the copyright has been transferred to the respective publisher. No part of this thesis may be reproduced, stored in a retrieval system, or transmitted in any form or by any means without the permission of the author or, when appropriate, the publisher of the manuscript.

IMPROVING APPROPRIATE MEDICATION USE FOROLDER PEOPLE IN PRIMARY CARE

OPTIMALISEREN VAN JUIST MEDICIJNGEBRUIKD O OR OUDEREN IN DE EERSTELIJNSGEZONDHEIDSZORG(met een samenvatting in het Nederlands)

PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magni"cus, prof.dr. G.J. van der Zwaan, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op woensdag 29 mei des middags te 2.30 uur

door

HENDRIK FRANS KWINT

geboren op 17 januari 1970 te Meppel

PROMOTOREN: Prof.dr. M.L. Bouvy Prof.dr. J. Gussekloo

CO-PROMOTOR: Dr. A. Faber

Financial support for conducting parts of the presented studies was provided by: Apotheek Voorzorg, a provider of multidose drug dispensing systems, and the Royal Dutch Association

for the Advancement of Pharmacy (KNMP) (Chapter 3.2). BENU Apotheken (formerly known as LLOYDS Apotheken); Royal Dutch Association for the

Advancement of Pharmacy (KNMP); Astra-Zeneca and the healthcare insurance companies Achmea and Menzis (Chapter 4.1, 4.2 and 4.3).

Connecting Care cooperation (Chapter 4.4).

Financial support in general and/or for the publication of this thesis was provided by: Apotheek Stevenshof, SIR Institute for Pharmacy Practice and Policy and Nederlands Bijwerkingen

Fonds.

Chapter 1 Introduction 9

Chapter 2 Collaboration in medication review 212 !e relationship between the extent of collaboration

of general practitioners and pharmacists and the implementation of recommendations arising from medication review. A systematic review.

23

Chapter 3 Multidose drug dispensing 473.1 Medication adherence and knowledge of older

patients with and without multidose drug dispensing.49

3.2 E$ects of medication review on drug-related problems in patients using multidose drug dispensing systems. A pragmatic randomised controlled study.

65

Chapter 4 Home Medication Review 934.1 Home medication review in older patients.

A randomised controlled trial in primary care95

4.2 !e contribution of patient interviews to the identi"cation of drug-related problems in home medication review.

115

4.3 Completeness of medication reviews provided by community pharmacists

133

4.4 Evaluation of medication reviews in older patients using STOPP-START criteria

145

Chapter 5 General discussion 159

Chapter 6 From Summary to About the author 1856.1 Summary 1876.2 Samenvatting 1936.3 Dankwoord 1996.4 List of co-authors 2096.5 List of publications 2116.6 About the author 213

11

| Chapter 1I ntroduc t ion

!e proportion of people over 60 years of age is growing faster than any other age group, as a result of longer life expectancy and declining fertility rates.1 While it is important to remember that the increase in life expectancy is the consequence of improvements in living conditions and medical advances, the ageing of the population does create major challenges to health care systems.2 Patients aged 75 years and older o"en present a complex clinical picture involving coexisting conditions and frailty.3 !e implementation of guidelines for the management of these conditions has resulted in an increase in the number of drugs prescribed to older people.2,4 One out of three visitors of 75 years and older of Dutch community pharmacies uses 5 chronic drugs or more, also known as polypharmacy.5While the bene#ts of some individual therapeutic interventions are clear-cut, the co-prescription of multiple drugs has considerable attendant risks.2 Adhering to current clinical practice guidelines in caring for an older person may have undesirable e$ects.6 Clinical practice guidelines especially fail to address the needs of older patients with complex co-morbidity. Older people are at increased risk of adverse drug reactions (ADR) secondary to age-related changes in pharmacokinetics and pharmacodynamics, drug interactions, increased co-morbidity and associated polypharmacy. At least 5% of hospital admissions are directly related to ADRs 7-10 and higher rates have been reported for older people, who are likely to be receiving multiple medications for long-term illnesses.8 !e Hospital Admission Related to Medication (HARM) study in !e Netherlands suggested that almost half of these medication related hospitalizations could be avoided.7 One of the main recommendations of the HARM-study was that the medication use of older patients with polypharmacy should be reviewed regularly for potential drug-related problems (DRPs). !is evaluation should also include identi#cation of barriers for medication regimen adherence and should provide tools to facilitate the proper use of medication.7!is thesis focuses at both the medication management and the quality of pharmacotherapy of older people with polypharmacy in primary care and ignores older people in nursing homes and hospitals. Health policy encourages older people to remain living in their own homes as long as possible. !e proportion of older people living in their own homes is increasing (e.g. from 80% in 2000 till 86% in 2010 for persons of 80 years and older).11 !e ability to remain independent in ones home partly depends on the ability to manage a complicated medication regimen 12 and the use of multiple medicines can give rise to various problems.Clinical pharmacy interventions for community-dwelling older people with polypharmacy can be divided in dispensing services (aimed at support of

12

medication management) and medication reviews (aimed at appropriateness of the pharmacotherapy, see Figure 1).13

Figure 1 Cl inical pharmac y inter ventions for older people with polypharmac y

Patients 65 years and 5 drugs

AssessmentMedication

managementproblems

Drug-relatedproblems

Intervention DispensingserviceMedicationreview

DISPENSING SERVICES

Medication management is a term that is generally used for both problems and solutions related to administration of a medication regimen. In this thesis, medication management is de! ned as the patients ability to self-administrate his or her medication regimen.14 A dispensing service consists of the provision of support (e.g. adherence aids) that enable patients to manage their own medication.13A dispensing service is preceded by an assessment of patients medication management problems (Figure 1). " is consists of di# erent elements. " e ! rst and most important element is to assess patients actual medication use: which prescribed drugs are actually used, which apparently discontinued prescription drugs are still used and which non-prescription drugs (e.g. over-the-counter drugs, complementary and alternative drugs) are used. Other elements may be discussion of patients medication knowledge and assessment of risk factors for medication

13


mismanagement (e.g. adherence and practical issues such as di!culties with opening containers).13,15Dispensing services consist of a variety of strategies to optimize medication management including education, written medication schedules, medication calendars, special packaging to organize medications, disposal of expired/unused medications 13,14 and chronic medication services.16-18"e use of special packagings to organize medications is growing fast in "e Netherlands. "e most frequently used special packaging is multidose drug dispensing (MDD). MDD provides patients with robot-dispensed unit doses. All drugs intended for one dosing moment are gathered in disposable bags and labelled with patient data, drug contents, and the date and time for intake.19-21 In "e Netherlands, patients experiencing di!culties with managing their medications are mostly identi#ed as eligible for MDD by community pharmacists, general practitioners (GPs) or relatives.19 Another development in "e Netherlands is the steep increase of chronic medication services (CMS) by community pharmacies. "is proactive service consists of synchronizing chronic medications and informing patients when their re#lls are available.16,17

MEDICATION REVIEW

"e term medication review is used to describe a plethora of interventions aimed at the appropriateness of pharmacotherapy that might be carried out by prescribers, pharmacists or nurses, both independently as well as in collaboration. Medication review has been de#ned as a structured, critical examination of a patients medicines with the objective of reaching an agreement with the patient about treatment, optimising the impact of medicines, minimizing the number of drug-related problems (DRPs) and reducing waste.22 "e di$erence between a medication review and a dispensing service is that medication review is an in-depth analysis of patients drug-related problems including e$ectiveness and safety issues and potential undertreatment whereas a dispensing service is aimed at patients practical problems with medication use including adherence issues (Figure 1). Ideally, medication reviews also include assessment of these medication management problems.13In 2002, the Medicines Partnership in the United Kingdom described di$erent levels of medication review based on the availability of patient data: Prescription review (medication records), Treatment review (medication and medical records) and Clinical medication review (medication records, medical records and patient

14

interview).22 ! e development of the Medicines Use Review (MUR) was an important reason to introduce a new type of medication review in 2008. ! e MUR was classi" ed as a Concordance or compliance review with the scope on exploring medicine taking including the patients self-reported use and beliefs of medicines.23 On the contrary Prescription review usually takes places without consulting the patient.23 For a Clinical medication review, not only a patient interview is required, but also access to patients medical notes and laboratory test results.23,24 All these types and levels of medication review appear to be de" ned from the perspective of the pharmacist (Figure 2).

Figure 2 Different t ypes of medicat ion review based on avai labi l i t y of data 22-24

Patientinterview

Medicalrecords

Medicationrecords

Concordance& compliance

reviewClinical

medicationreview

Treatmentreview

Prescriptionreview

Only few studies have evaluated medication reviews by physicians independently 25,26 while time limitations probably impede widespread implementation of these reviews.25 Dependent on the availability of data more types of review appear to be possible, but these are not mentioned in literature (e.g. a patient interview without any pharmacy or prescriber data or a review solely based on prescriber data) (Figure 2).Clinical medication review has been adapted in di# erent countries, like Home Medicines Reviews in Australia,13,27 Medication ! erapy Management services in the United States 28 and Comprehensive Medication Reviews in Finland.29 Home

15


Medicines Reviews (HMRs) in Australia are structured collaborative reviews conducted by an accredited pharmacist in the patients home. !e patient should have at least one risk factor that necessitates a review. Examples of risk factors are: currently taking "ve or more medications; signi"cant changes to the medication regimen in the preceding 3 months, recent hospital discharge; having di#culty with organising medicines; symptoms suggestive of an adverse drug reaction; taking medication with a narrow therapeutic index or requiring therapeutic drug monitoring.13,27 !e HMR process consists of: a formal GP referral to the community pharmacy; HMR by an accredited pharmacist; formal report from the pharmacist, which is discussed with the GP; patient and GP agreement with the pharmaceutical care plan; implementation of the plan with monitoring and follow-up. At the home visit, the patient is asked to present all medicines (prescribed and otherwise) to the pharmacist, who then reviews the patients knowledge, storage, practical problems and perception of the e$ectiveness and adverse e$ects of each medicine. !e pharmacist also arranges for disposal of expired or unused medicines as required.13!e Dutch clinical medication review process is described in the recent Dutch multidisciplinary guideline Polypharmacy in the elderly.30 !is process consists of "ve steps, is continuous and occurs over multiple encounters of pharmacists, patients and physicians (Figure 3).31At the start of the process, data collection has to take place including medication records from the community pharmacy (current medications, past medication use, information on comorbidity and drug intolerances) and medical records from the GP practice (current medical conditions, medical history and laboratory data). !e "ve steps are:(1) Patient interview (pharmaceutical anamnesis)During the patient interview the pharmacist veri"es patients actual medication use and potential use of earlier discontinued prescription drugs, over-the-counter drugs and complementary and alternative drugs. In addition, practical problems and perception of the e$ectiveness and adverse e$ects of each medicine are reviewed. !e patients beliefs about and experiences with pharmacotherapy are discussed in a non-judgmental way.(2) Assessment!e pharmacist identi"es potential drug-related problems (DRPs) and associated recommendations by assessing patient, medical and medication data using both implicit and explicit criteria. A simple implicit approach is to match each of the patients conditions with the medications that the patient is taking. Areas of mismatch can highlight drugs that are used with no indication, underused or

16

misused.25 As explicit criteria for inappropriate prescribing, STOPP (Screening Tool of Older Persons Prescriptions) is recommended which contains 65 explicit rules for avoidance of certain drugs/drug classes. In addition, START (Screening Tool to Alert doctors to Right Treatment) lists 22 common instances of potentially prescribing omissions.30,32,33(3) Pharmaceutical care planDuring a case-conference (preferably face-to-face between pharmacist and physician) potential DRPs and associated recommendations are discussed. In a pharmaceutical care plan the actions are formulated to resolve DRPs. Actions are prioritized based on the patients perspective, focusing on DRPs that causes the most concern. !e pharmaceutical care plan should contain detailed information on who is responsible for implementation and evaluation of actions (e.g. GP, pharmacist or practice nurse).(4) Implementation of action!e responsible care provider explains to the patient why an action is required, for example a change in the medication regimen. In concordance, patient and care provider decide if this adaptation of the regimen can be implemented or should be adjusted. !e number of changes during each patient encounter should be

Figure 3 The medication review process according to Dutch multidisciplinary guideline Polypharmacy in the elderly

1. Patient interview

2. Assessment (identifying DRPs)

3. Pharmaceutical care plan

4. Implementation of actions

5. Follow-up and monitoring

DRP = drug-related problem

17


restricted. It is recommended that the agreed actions are written on an information lea!et as well as the provision of an updated medication list.(5) Follow-up and monitoring"e responsible care provider veri#es if all agreed actions are implemented. Where necessary the patient is consulted once more to examine whether DRPs have been resolved. If needed, new actions will be planned.

OBJECTIVE

Dispensing services and medication reviews are relatively new clinical pharmacy interventions to increase the appropriate use of medicines especially for older patients with polypharmacy. "is o$ers an opportunity for pharmacists to ful#l their professional roles as health care providers in collaboration with physicians and patients. "ere is limited evidence that interventions aimed at appropriateness of medication reduce drug-related problems in older people. However, it is not clear if these result in clinical improvements.34,35 "e heterogeneity in patient populations, settings, interventions and outcomes in earlier studies make it di%cult to draw de#nitive conclusions. "erefore, more research data are needed in order to clarify the e$ects of these new clinical pharmacy services.In this thesis, studies will be presented with the objective to describe the e$ects of clinical pharmacy interventions for older patients with polypharmacy in primary care. Apart from the focus of medication review on appropriateness of the pharmacotherapy, this thesis will provide data on the impact of a dispensing service (i.e. multi dose drug dispensing systems) on the medication management of older patients.

OU TLINE OF THE THESIS

"is thesis consists of three parts. Chapter 2 presents a systematic literature review that investigates how the degree of collaboration between the general practitioner (GP) and the pharmacist impacts on the implementation of recommendations arising from medication review.Chapter 3 focuses on the impact of multi dose drug dispensing systems on the medication management of older patients and the appropriateness of medication in these systems.

18

In Chapter 3.1 we compared the self-reported medication adherence and knowledge of older patients receiving their drugs via multidose drug dispensing (MDD-users) with patients receiving manually-dispensed drugs (non-MDD-users).In Chapter 3.2 the results are presented of a pragmatic randomized controlled study on the e!ect of medication review for older patients using multidose drug dispensing (MDD) on DRPs.Chapter 4 focuses on the impact of medication review on di!erent outcomes and in-depth analyses of the sources for identi"cation of DRPs, the pharmacists conducting medication reviews and the presence of explicit criteria in DRPs.Chapter 4.1 describes the results of a randomised controlled trial on the e!ect of Home Medication Review on DRPs, disease-speci"c outcomes (blood pressure, LDL-cholesterol and HbA1c) and health-related quality of life in older patients.Chapter 4.2 focuses on the contribution of the patient interview to the identi"cation of DRPs within the Home Medication Review.Chapter 4.3 describes the completeness of DRPs identi"ed by community pharmacists.Chapter 4.4 aims to describe the number and types of explicit STOPP-START criteria present in identi"ed potential DRPs and their implementation rate.Finally, the results of these studies are summarized and put into a broader perspective in Chapter 5.

REFERENCES

1. Cerreta F, Eichler HG, Rasi G. Drug policy for an aging population - the European Medicines Agencys geriatric medicines strategy. N Engl J Med 2012;367:1972-4.

2. Hubbard RE, OMahony MS, Woodhouse KW. Medication prescribing in frail older people. Eur J Clin Pharmacol 2012;Sep [epub].

3. Gnjidic D, Hilmer SN, Blyth FM, Naganathan V, Cumming RG, Handelsman DJ, et al. High-risk prescribing and incidence of frailty among older community-dwelling men. Clin Pharmacol #er 2012;91:521-8.

4. Hajjar ER, Ca"ero AC, Hanlon JT. Polypharmacy in elderly patients. Am J Geriatr Pharmacother 2007;5:345-51.

5. Stichting Farmaceutische Kengetallen (SFK). Polyfarmacie bij een op de tien apotheekbezoekers. Available from: http://www.s$.nl/nieuws-publicaties/PW/2012/polyfarmacie-voor-1-op-de-10-apotheekbezoekers (Accessed 3 Jan 2013).

6. Boyd CM, Darer J, Boult C, Fried LP, Boult L, Wu AW. Clinical practice guidelines and quality of care for older patients with multiple comorbid diseases: implications for pay for performance. JAMA 2005;294:716-24.

19


7. Leendertse AJ, Egberts AC, Stoker LJ, van den Bemt PM. Frequency of and risk factors for preventable medication-related hospital admissions in the Netherlands. Arch Intern Med 2008;168:1890-6.

8. Kongkaew C, Noyce PR, Ashcro! DM. Hospital admissions associated with adverse drug reactions: a systematic review of prospective observational studies. Ann Pharmacother 2008;42:1017-25.

9. Beijer HM, Blaey CJd. Hospitalisations caused by adverse drug reactions (ADR): a meta-analysis of observational studies. Pharm World Sci 2002;25:46-54.

10. Howard RL, Avery AJ, Slavenburg S, Royal S, Pipe G, Lucassen P, et al. Which drugs cause preventable admissions to hospital? A systematic review. Br J Clin Pharmacol 2007;63:136-47.

11. Centraal Bureau voor de Statistek (CBS). Demogra"e van de vergrijzing, Bevolkingstrends 2e kwartaal 2011. Available from: http://www.cbs.nl/nl-NL/menu/themas/bevolking/publicaties/artikelen/archief/2011/2011-3434-wm.htm (Accessed 3 Jan 2013).

12. Marek KD, Antle L. Medication Management of the Community-Dwelling Older Adult. In: Hughes RG, editor. Patient Safety and Quality: An Evidence-Based Handbook for Nurses. Rockville (MD); 2008.

13. MacKeigan LD, Nissen LM. Clinical pharmacy services in the home. Disease Management and Health Outcomes 2008;16:227-44.

14. Maddigan SL, Farris KB, Keating N, Wiens CA, Johnson JA. Predictors of older adults capacity for medication management in a self-medication program: a retrospective chart review. J Aging Health 2003;15:332-52.

15. Drenth-van Maanen AC, van Marum RJ, Knol W, van der Linden CM, Jansen PA. Prescribing optimization method for improving prescribing in elderly patients receiving polypharmacy: results of application to case histories by general practitioners. Drugs Aging 2009;26:687-701.

16. van Ge#en K, Vos I, van Horssen N, Winters N, Bouvy ML. Apotheker proactiever. Enquete: rol in herhaalmedicatie groeit. Pharm Weekbl 2011;146(20).

17. de Hertog NJ, Donker AA, Smith R. Patient therapietrouwer met herhaalservice. Signi"cante verbetering bij zeven geneesmiddelgroepen. Pharm Weekbl 2011;146(20).

18. $e Scottish Government. Establishing E#ective $erapeutic Partnerships - A generic framework to underpin the Chronic Medication Service element of the Community Pharmacy Contract. 2009. Available from: http://www.scotland.gov.uk/Publications/2010/01/07144120/0 (Accessed 3 Jan 2013).

19. van Wijck F. Populariteit Baxterrol vraagt om goede leidraad. Pharm Weekbl 2011;146(21).20. Wekre LJ, Spigset O, Sletvold O, Sund JK, Grimsmo A. Multidose drug dispensing and

discrepancies between medication records. Qual Saf Health Care 2010;19:e42.21. Johnell K, Fastbom J. Multi-dose drug dispensing and inappropriate drug use: A nationwide

register-based study of over 700,000 elderly. Scand J Prim Health Care 2008;26:86-91.22. Shaw J, Seal R, Pilling M. Task force on medicines partnership and the national collaborative

medicines management services programme. Room for review: a guide to medication review. 2002. Available from: http://www.npc.nhs.uk/review_medicines/intro/resources/room_for_review.pdf (Accessed 1 Feb 2013).

20

23. Clyne W, Blenkinsopp A, Seal R. National prescribing centre. A guide to medication review. 2008. Available from: http://www.npc.nhs.uk/review_medicines/intro/resources/agtmr_web1.pdf (Accessed 1 Feb 2013).

24. Blenkinsopp A, Bond C, Raynor DK. Medication reviews. Br J Clin Pharmacol 2012;74:573-80.

25. Steinman MA, Hanlon JT. Managing medications in clinically complex elders: !eres got to be a happy medium. JAMA 2010;304:1592-601.

26. Fiss T, Dreier A, Meinke C, van den Berg N, Ritter CA, Ho"mann W. Frequency of inappropriate drugs in primary care: analysis of a sample of immobile patients who received periodic home visits. Age Ageing 2011;40:66-73.

27. Castelino RL, Bajorek BV, Chen TF. Retrospective evaluation of home medicines review by pharmacists in older Australian patients using the medication appropriateness index. Ann Pharmacother 2010;44:1922-9.

28. Bluml BM. De#nition of medication therapy management: development of professionwide consensus. J Am Pharm Assoc 2005;45:566-72.

29. Leikola SN, Virolainen J, Tuomainen L, Tuominen RK, Airaksinen MS. Comprehensive medication reviews for elderly patients: #ndings and recommendations to physicians. J Am Pharm Assoc 2012;52:630-3.

30. Vermeulen Windsant-van den Tweel AM, Verduijn MM, Derijks HJ, van Marum RJ. Detection of inappropriate medication use in the elderly; will the STOPP and START criteria become the new Dutch standards? Ned Tijdschr Geneeskd 2012;156:A5076.

31. Cipolle RJ. Pharmaceutical care practice: the clinicians guide. 2nd ed. New York: McGraw-Hill, Medical Pub. Division; 2004.

32. Gallagher P, Ryan C, Byrne S, Kennedy J, OMahony D. STOPP (Screening Tool of Older Persons Prescriptions) and START (Screening Tool to Alert doctors to Right Treatment). Consensus validation. Int J Clin Pharmacol !er 2008;46:72-83.

33. Gallagher P, OMahony D. STOPP (Screening Tool of Older Persons potentially inappropriate Prescriptions): application to acutely ill elderly patients and comparison with Beers criteria. Age Ageing 2008;37:673-9.

34. Patterson SM, Bradley MC, Kerse N, Cardwell CR, Hughes CM. Interventions to improve the appropriate use of polypharmacy for older people. Cochrane Database Syst Rev. 2012;5.

35. Holland R, Desborough J, Goodyer L, Hall S, Wright D, Loke YK. Does pharmacist-led medication review help to reduce hospital admissions and deaths in older people? A systematic review and meta-analysis. Br J Clin Pharmacol 2008;65:303-16.

24

ABSTRACT

BackgroundMany studies have investigated the e!ect of medication review on a variety of outcomes, but the elements of the interventions have been quite diverse. Moreover, implementation rates of recommendations also vary widely between studies.

Aim"e objective of this study was to investigate how the extent of collaboration between the general practitioner (GP) and the pharmacist impacts on the implementation of recommendations arising from medication review.

MethodsMEDLINE, EMBASE and Web of Science were searched for studies published between January 2000 and April 2012. Keywords included medication review, medication therapy management, pharmaceutical services and drug utilization review. Sixteen articles (describing 14 randomized controlled trials [RCTs]) out of 620 titles met the inclusion criteria. Inclusion criteria for the review were medication review, RCT design, involvement of both pharmacist and GP, and community-dwelling patients (mean age > 70 years) who had not been recently discharged. A#er quality assessment of the article, the presence of the following eight key elements re$ecting collaboration were scored for each intervention: pharmacist with clinical experience, own pharmacist involved, sharing of medical records, patient interview by pharmacist, invitation of patients by GP, case conference between GP and pharmacist, action plan, follow-up. "e primary outcome was the implementation rate of recommendations. Meta-regression analysis was used to assess the association between the implementation rate and the number of key elements present.

ResultsTwelve RCTs were included a#er quality assessment. "e mean number of key elements within the intervention was 5.2 (range 18). "e mean implementation rate of recommendations was 50% (range 1786). "e association between the number of key elements present in the intervention and the implementation rate of recommendations was signi&cant: = 0.085 (95%CI 0.0520.128; P < 0.01).

25

| Chapter 2GP/pharmacist col laborat ion in medicat ion review

Conclusion!is systematic review shows a signi"cant association between the number of key elements of the intervention re#ecting collaborative aspects in medication review and the implementation rate of recommendations.

INTRODUCTION

Polypharmacy and drug-related morbidity is increasingly recognized as a major public health problem among the elderly.1,2 Medication review has been proposed as an important strategy to constrain the negative e$ects of polypharmacy, aiming at safer and more e$ective use of medicines.3,4Medication review has been de"ned as a structured, critical examination of a patients medicines with the objective of reaching an agreement with the patient about treatment, optimising the impact of medicines, minimizing the number of medication-related problems and reducing waste.5 !ree types of medication review have been described, based on the purpose of the review: prescription review technical issues related to prescription(s), concordance or compliance review (issues relating to the patients medicine behaviour) and clinical medication review (issues relating to the patients use of medicines in the context of their condition).6 Concomitantly, e$orts have been made to standardize medication review.5-7However, systematic reviews of pharmacist-led medication review have not shown an e$ect on clinical outcomes such as hospital admissions or mortality.8-10 In some studies, positive e$ects were reported on intermediate outcomes like drug knowledge and adherence.8 !e heterogeneity in patient populations, settings, interventions and outcomes in these studies made it di%cult to draw de"nitive conclusions. !ere may be merit in combining the expertise of the pharmacist and physician with shared decision-making involving the patient in order to improve outcomes.11 Previous systematic reviews did not take into account the variability in collaboration between pharmacists and general practitioners (GPs) in medication reviews.Studies on barriers and facilitators in medication review reveal collaborative aspects that might be essential for conducting successful medication reviews.12-14 !e most commonly cited facilitators were having an established pharmacistphysician relationship 13,14 and a face-to-face meeting (case conference) between pharmacist and physician to discuss the pharmacists recommendations.13,15,16 Using a pharmacist other than the patients regular pharmacist was seen as a barrier,

26

as was inadequate clinical training of the pharmacist.13 Without access to medical records, the pharmacist may make tentative or inappropriate recommendations that are of little help.13 !e GPPC (General PractitionerPharmacist Collaboration) study 17 further suggested that a general practice-based service could be more facilitating than a community pharmacy-based service.17 !is could imply that patients are approached for medication review by the GP practice, which is also common in the Home Medicines Review (HMR) programme in Australia.12,13 !ey further suggested that the pharmacist should meet the patient for interview about their medicines in the physicians o"ce,17-19 while a patient interview at home by an accredited pharmacist is the predominant step of the HMR programme.12,13 Finally, it is important for a collaborative medication review that responsibilities for implementation of the action plan and follow-up are clearly de#ned and divided between physician and pharmacist.13!e aim of this systematic review was to investigate how the degree of collaboration between the GP and the pharmacist impacts on the implementation of recommendations arising from medication review.

METHODS

S earchOur search strategy identi#ed research on medication review interventions involving pharmacists and GPs. MEDLINE, EMBASE and Web of Science were searched for articles published between 1 January 2000 and 1 April 2012. !ese dates were chosen because relatively few studies with an elaborate description of the medication review process were published before 2000. Interventions were identi#ed using the following keywords and medical subject headings (MeSH): medication review, medication therapy management, pharmaceutical services and drug utilization review (see Appendix I for detailed search terms). Di$erent publications on the same group of patients were considered as one study.

Study select ionAll titles were reviewed by two investigators (H.K. and L.B.). Studies were excluded if both agreed that the title clearly indicated that the study did not concern medication review and/or focussed on only one drug or drug class. H.K. and L.B. assessed all remaining abstracts independently in this manner. Studies were included if they ful#lled the following criteria: medication review, randomized

27


clinical trial (RCT), pharmacist and GP involved, community-dwelling patients in primary care, mean age 70 years, patients not recently discharged (< 1 month).Only studies in English were included. Finally, full papers from potential studies were assessed independently by the two investigators for their suitability for inclusion. Di"erences were resolved by discussion, or a third investigator (either A.F. or M.B.) was consulted.

Quality assessment of the studiesTrial quality was assessed according to the Delphi list.20 #is list consists of ten criteria: randomization, treatment allocation, similar groups at baseline, eligibility criteria, blinding of outcome assessor, blinding of care provider, blinding of patient, point estimates and measures of variability, intention-to-treat analysis and reporting of withdrawal/drop-out rate. In addition, we added power calculation to this list. Studies with a low score on the quality assessment (5 or fewer items scored yes) were excluded for analysis of outcomes.

Study characterist icsCategor izationStudies were categorized by study author, year of publication, number of pharmacists and GPs, country, number of patients, duration of the study, mean age and sex of patients, mean number of drugs, description of the intervention, setting, number of recommendations in the intervention group, the clinical, intermediate and process outcomes assessed and the quality score.

Outcomes#e primary outcome was the implementation rate of recommendations following drug-related problems (DRPs) identi$ed during medication review. #e implementation rate was de$ned as the percentage of recommendations fully or partly implemented and/or the percentage of DRPs resolved. Partial implementation of recommendations means that an action other than that originally proposed by the pharmacist was implemented. Fully and partly implemented recommendations were counted equally. Data on clinical outcomes (hospital admissions, quality of life), intermediate outcomes (adherence) and other process outcomes (drug changes, number of drugs) were also extracted. #e e"ect on clinical, intermediate and process outcomes was described as a signi$cant e"ect in favour of the intervention group, a signi$cant e"ect in favour of the control group or no signi$cant e"ect.

28

Key elements of the inter vention!e intervention was characterized by the presence or absence of eight key elements re"ecting collaborative aspects between a GP and a pharmacist, based on the aforementioned facilitators and barriers in medication review.12-14 !e choice of the key elements was supported by scienti#c discussion with experienced pharmacist reviewers who regarded these elements as having face validity. !e following key elements were assessed: (1) pharmacist with clinical experience means that the study pharmacist had adequate clinical training and expertise to perform medication reviews; (2) own pharmacist involved means that the study pharmacist is the patients regular pharmacist who has a longer lasting therapeutic relationship with his or her patient; (3) sharing of medical records describes full access for the care provider performing the medication review to GP data on diseases of the patient and clinical values; (4) patient interview by pharmacist means a face-to-face consultation between a pharmacist and a patient this pharmacist must have a relationship with the GP; (5) invitation of the patients by GP means that the patient is invited to the study or referred for medication review by the GP (practice); (6) case conference GP and pharmacist indicates a face-to-face meeting between at least the GP and the pharmacist to discuss the DRPs and recommendations for speci#c patients; (7) action plan means that the study investigators reported that the agreed recommendations were formulated as an action plan and that there were designated persons responsible for implementation of this plan; and (8) follow-up has taken place to assess whether the actions have been implemented, and to assess the patients experience with these actions.

Data synthesis and analysisFor each trial, we extracted data on the primary outcome, implementation rate. When the implementation rate was not present, we derived this rate from the percentage of DRPs resolved or the decrease in the number of potentially inappropriate prescriptions (PIPs). Trial quality and key elements of interventions were assessed independently by the two investigators (H.K. and L.B.) for each included study. Di$erences were resolved by discussion, or a third investigator (either A.F. or M.B.) was consulted.Meta-regression analysis was used to assess the association between the number of key elements and the implementation rate, with the number of recommendations in the di$erent studies as possible e$ect moderator. !is mixed-e$ects analysis was conducted using the metafor statistical package in R (version 2.12.2, R Project for Statistical Computing, Vienna, Austria, 2011, http://www.R-project.org).

29


RESULT S

S earch resultsA total of 620 titles were identi!ed, 16 of which (describing 14 RCTs) met the inclusion criteria and were included in this review (Figure 1).3,15-17,19,21-32

Figure 1 Flow chart describing study selection and excluded studies

Potentially relevant publicationsidenti!ed and titles screened

n = 620Excluded: 467because intervention is notmedication review, and/ortargeting speci!c drugs

Abstracts of potentialpublications screened

n = 153

Excluded: 131no RCT (95)secondary or tertiary care (20)targeting speci!c disease (8)discharge (7)age too low (1)

Potentially appropriatepublications

n = 22describing 20 RCTs

Excluded: 6age too low (3)targeting speci!c drug classes (2)medication review only part of intervention (1)

Publications includedin review

n = 16describing 14 RCTs

RCTs = randomized controlled trials

Quality assessment of studies"e methodological quality of 12 of the 14 studies was assessed as adequate (i.e. 6 or more of 11 items scored yes)3,15-17,19,21,23-27,29-32 (see Table 1). "e quality of the trial in the two remaining studies was scored as low 22,28 and they were therefore

30

Tab

le 1

Qua

lity

asse

ssm

ent o

f stu

dies

Stud

y au

thor

Met

hod

of

rand

omis

atio

n pe

rfor

med

Trea

tmen

t al

loca

tion

conc

eale

d

Sim

ilar

grou

ps a

t ba

selin

e

Elig

ibili

ty

crite

ria

spec

i!ed

Out

com

e as

sess

or

blin

ded

Care

pr

ovid

er

blin

ded

Patie

nt

blin

ded

Poin

t of

estim

ates

an

d m

ea-

sure

s of

varia

bilit

y

Incl

udin

g in

tent

ion-

to-t

reat

With

draw

al

rate

(%)

unlik

ely

to

caus

e bi

as?

Pow

er

calc

ulat

ion

repo

rted

Qua

lity

Scor

e

Alla

rd (2

001)

Yes

Not

cle

arYe

sYe

sYe

sN

oN

oYe

sYe

sYe

s (9%

)Ye

s8

Bern

sten

(200

3)Ye

sYe

sYe

sYe

sN

oN

oN

oYe

sN

oN

o (3

4%)

No

5

Brya

nt (2

011)

Yes

Yes

Yes

Yes

Yes

No

No

Yes

No

No

(48%

)Ye

s7

Den

nebo

om (2

007)

Yes

Yes

Yes

Yes

Not

cle

arN

oYe

sYe

sN

oYe

s (7%

)Ye

s7

Gry

mon

pre

(200

1)Ye

sYe

sYe

sYe

sN

oN

oYe

sYe

sYe

sYe

s (16

%)

Yes

9

Krsk

a (2

001)

Yes

Yes

Yes

Yes

Not

cle

arN

oN

oYe

sN

oYe

s (13

%)

No

6

Kwin

t (20

11)

Yes

Yes

Yes

Yes

Not

cle

arN

oN

ot c

lear

Yes

No

Yes (

9%)

Yes

7

Lena

ghan

(200

7)Ye

sYe

sYe

sYe

sN

ot c

lear

No

No

Yes

Yes

Yes (

2%)

Yes

7

Sello

rs (2

003)

Yes

Yes

Yes

Yes

Not

cle

arN

oN

oYe

sN

ot c

lear

Yes (

11%

)Ye

s7

Sore

nsen

(200

4)Ye

sN

ot c

lear

Yes

Yes

Not

cle

arN

oN

oYe

sYe

sYe

s (25

%)

Yes

7

Stur

gess

(200

3)Ye

sYe

sN

oYe

sN

oN

oN

oYe

sN

oN

o (4

2%)

No

4

Volu

me

(200

1)Ye

sYe

sN

oYe

sN

ot c

lear

No

No

Yes

No

Yes (

20%

)Ye

s6

Will

iam

s (20

04)

Yes

Yes

Yes

Yes

Not

cle

arN

oN

oYe

sN

oYe

s (5%

)Ye

s7

Zerm

ansk

y (2

001)

Yes

No

Yes

Yes

Not

cle

arN

oN

oYe

sYe

sYe

s (5%

)Ye

s7

31


excluded from further analysis. !ese studies were related; one trial formed part of a larger co-ordinated project with more countries, which was described in the other paper.22,28In all included studies, a method of randomization was performed and eligibility criteria were speci"ed. !e majority of studies reported a method of treatment allocation,3,15-17,22-26,29-31 while in three studies this was either not clearly described or not conducted.19,21,27,32 All except two studies reported similar groups at baseline.28-30 An independent outcome assessor who was blinded to the intervention allocation was clearly described in only two studies.17,21 Because of the nature of the studied intervention, the care provider was never blinded to the intervention allocation. !e patient was blinded for the intervention allocation in three studies.3,15,16,23 In two of these studies, patient interviews were conducted for both the intervention and the control group, but a pharmaceutical care plan was implemented only for the intervention group.3,23 In the third study, no patient interview was conducted and there was no description of patient involvement with the study.15,16 Point estimates and measures of variability were described in all studies. Intention-to-treat analysis was conducted in "ve studies.19,21,23,25,27,32 !e withdrawal rate was likely to have caused bias in 3 of 14 studies.3,15,16,19,21,23-27,29-32

Study characterist icsIn Table 2, the study characteristics and outcomes are presented for the 12 included studies. !e number of participants in these studies ranged from 118 to 1,188. !e mean age of the participants was 76.6 years (range 71.884.3) and 66% were females (range 5690). !e mean number of prescribed drugs was 7.2 (range 4.512). !ree of these studies were performed in the US,23,29-31 three in the UK,3,19,25,32 and two in !e Netherlands.15,16,24

OutcomesSeven of 12 studies provided data on clinical outcomes.3,17,19,25-27,29,32 Six of these studies reported on quality of life measured using the 36-item Short Form Health Survey (SF-36)3,17,19,26,27,29,32 or the EuroQOL-5D/visual analogue scale (VAS).25 No e$ects were found on total scores for quality of life, and one study reported only negative e$ects on some domains in one study.17 Data on hospital admissions were provided by four studies,19,25-27,32 and no signi"cant e$ects were reported. Twelve studies provided data on intermediate outcomes. Two studies reported on adherence, either self-reported 29,30 or measured by re"ll rate,23 with no e$ect. Two studies reported on DRPs in both the intervention and the control group, with positive e$ects on DRPs resolved.3,24 Two studies 17,21 reported on potentially inappropriate

32

Tab

le 2

Des

crip

tion

of in

clud

ed st

udie

s: e!e

ct o

n cl

inic

al, i

nter

med

iate

and

pro

cess

out

com

es

Stud

y(y

ear o

f pu

blic

atio

n)

No.

of

phar

mac

ists

an

d G

Ps

(cou

ntry

)

Patie

nts

n (%

F);

mea

n ag

e in

yea

rs

Dur

atio

nN

o. o

fdr

ugs

Inte

rven

tion

(set

ting)

No.

of

reco

mm

en-

datio

ns (m

ean

per p

t)

Clin

ical

out

com

es

(sig

ni"c

ant e!e

ct)

Inte

rmed

iate

and

pr

oces

s out

com

es

(sig

ni"c

ant e!e

ct)

Qua

lity

scor

e

Alla

rd e

t al.2

1

(200

1)Te

am o

f 2

phys

icia

ns,

phar

mac

ist

and

nurs

e;52

GPs

(Can

)

266

(90%

);80

.512

mo

6.3

Nur

se m

et p

ts a

t hom

e fo

r inv

ento

ry o

f dru

gs.

MR

was

per

form

ed b

y a

mul

tidisc

iplin

ary

team

and

m

aile

d to

GPs

(Fac

ulty

of

Med

icin

e, U

nive

rsity

)

147

(1.1

)-

No.

of P

IMs (C

); no

. of p

ts w

ith

1

PIM

(C);

no.

of p

ts w

ith d

rug

impr

ovem

ents

(C);

no. o

f pre

scrib

ed

drug

s (C

)

8

Brya

nt e

t al.1

7

(201

1)26

CPs

;57

GPs

(NZ)

498

(59.

0%);

75.4

12 m

o5.

0H

ome

MR

follo

wed

by

CC w

ith G

P (c

omm

unity

ph

arm

acie

s)

462

(1.7

)Q

oL (S

F-36

): em

otio

nal r

ole

(?);

soci

al fu

nctio

ning

(?

); ot

her d

omai

ns

(C)

MAI

(B);

no. o

f PIM

s (B

); no

. of d

rug

chan

ges (B

)

7

Den

nebo

om e

t al

.15(2

007)

29 C

Ps;

84 G

Ps(N

L)

738

(62.

1%);

81.0

9 m

o7.

2M

R w

ithou

t pt i

nter

view

(t

reat

men

t rev

iew

) fo

llow

ed b

y CC

with

GP

(stu

dy a

rm A

) or w

ritte

n fe

edba

ck (s

tudy

arm

B)

(com

mun

ity p

harm

acie

s)

(A) 1

41 (0

.5)

(B) 1

28 (0

.3)

N

o. o

f dru

g ch

ange

s: 6

mon

ths

(B);

9 m

onth

s (C

)

7

Gry

mon

pre

et

al.23

(200

1)

Phar

mac

ists

NS;

35 G

Ps(U

S)

135

(79.

3%);

77.0

12 m

o6.

2H

ome

med

icat

ion

hist

ory

take

n by

lay

pers

on a

nd re

view

ed b

y ph

arm

acy

cons

ulta

nt (o

ne

inte

rdisc

iplin

ary

heal

th

clin

ic)

794

(11.

5)-

Adhe

renc

e (re"l

l ra

te) (C

); dr

ug

know

ledg

e (C

); no

. of p

resc

ribed

dr

ugs (C

)

9

Krsk

a et

al.3

(200

1)Ph

arm

acist

s N

S, 6

gen

eral

pr

actic

es

(UK)

332

(60.

5%);

75.1

3 m

o7.

5H

ome-

base

d M

R. C

are

plan

sent

to G

P, ag

reed

ac

tions

impl

emen

ted

by

phar

mac

ist a

ssist

ed b

y pr

actic

e st

a# (g

ener

al

prac

tices

)

1,20

6 (6

.3)

QoL

(SF-

36) (C

)Pr

opor

tion

of P

CIs

reso

lved

(B);

use

of

heal

th se

rvic

es (C

)

6

33


Tab

le 2

Des

crip

tion

of in

clud

ed st

udie

s: e!e

ct o

n cl

inic

al, i

nter

med

iate

and

pro

cess

out

com

es

Stud

y(y

ear o

f pu

blic

atio

n)

No.

of

phar

mac

ists

an

d G

Ps

(cou

ntry

)

Patie

nts

n (%

F);

mea

n ag

e in

yea

rs

Dur

atio

nN

o. o

fdr

ugs

Inte

rven

tion

(set

ting)

No.

of

reco

mm

en-

datio

ns (m

ean

per p

t)

Clin

ical

out

com

es

(sig

ni"c

ant e!e

ct)

Inte

rmed

iate

and

pr

oces

s out

com

es

(sig

ni"c

ant e!e

ct)

Qua

lity

scor

e

Kwin

t et a

l.24

(201

1)6

CPs;

GPs

NS

(NL)

118

(68.

5%);

79.3

6 m

o10

.1M

Rs w

ithou

t pt i

nter

view

w

ere

perf

orm

ed b

y ex

tern

al

phar

mac

ist re

view

ers a

t di

stan

ce, b

ut th

e fo

llow

ing

CC w

ith G

P w

as c

ondu

cted

by

pts

ow

n CP

s (co

mm

unity

ph

arm

acie

s)

249

(4.0

)-

No.

of D

RPs

lead

ing

to

reco

mm

enda

tion

for d

rug

chan

ge

(B);

no. o

f dru

g ch

ange

s rel

ated

to

a re

com

men

datio

n (B

); no

. of d

rug

chan

ges (B

)

7

Lena

ghan

et a

l.25

(200

7)1

CP;

9 G

Ps(U

K)

136

(65.

7%);

84.3

6

mo

7.4

Hom

e-ba

sed

MR

(one

di

spen

sing

gene

ral

prac

tice)

71 (1

.0)

No.

of h

ospi

tal

adm

issio

ns (C

); de

aths

(C);

QoL

(E

Q-5

D/V

AS) (C

)

No.

of p

resc

ribed

dr

ugs (C

)7

Sello

rs e

t al.2

6

(200

3)24

CPs

;48

GPs

(US)

889

(62.

8%);

74.0

5 m

o7.

9Cl

inic

-bas

ed M

R, th

en

tele

phon

e in

terv

iew

s (fa

mily

pra

ctic

es in

24

sites

)

1,09

3 (2

.5)

No.

of h

ospi

tal

adm

issio

ns (C

); no

. of d

rug-

rela

ted

hosp

ital

adm

issio

ns (C

); Q

oL (S

F-36

) (C

)

No.

of d

aily

uni

ts o

f dr

ugs t

aken

(C);

no. o

f dru

gs ta

ken

(C

7

Sore

nsen

et a

l.27

(200

4)32

CPs

;84

GPs

(Aus

)

400

(63.

8%);

71.8

6 m

o8.

0H

ome

visit

and

MR

by

phar

mac

ist a

nd te

am

conf

eren

ces w

ith G

P an

d im

plem

enta

tion

at

next

pt/

GP

visit

(gen

eral

pr

actic

es a

nd c

omm

unity

ph

arm

acie

s)

564

(3.2

)N

o. o

f hos

pita

l ad

miss

ions

(C);

no. o

f hos

pita

l se

rvic

es (C

); Q

oL (S

F-36

) (C

); se

verit

y of

illn

ess

(DU

SOI-A

) (C

); AD

Es (C

)

7

Tabl

e 2

cont

inue

d

34

Tab

le 2

Des

crip

tion

of in

clud

ed st

udie

s: e!e

ct o

n cl

inic

al, i

nter

med

iate

and

pro

cess

out

com

es

Stud

y(y

ear o

f pu

blic

atio

n)

No.

of

phar

mac

ists

an

d G

Ps

(cou

ntry

)

Patie

nts

n (%

F);

mea

n ag

e in

yea

rs

Dur

atio

nN

o. o

fdr

ugs

Inte

rven

tion

(set

ting)

No.

of

reco

mm

en-

datio

ns (m

ean

per p

t)

Clin

ical

out

com

es

(sig

ni"c

ant e!e

ct)

Inte

rmed

iate

and

pr

oces

s out

com

es

(sig

ni"c

ant e!e

ct)

Qua

lity

scor

e

Volu

me

et a

l.29

(200

1)5

CPs;

GPs

NS

(US)

363

(66.

9%);

74.0

15 m

o4.

5Ph

arm

aceu

tical

car

e se

rvic

es in

clud

ing

clin

ical

MR

(com

mun

ity

phar

mac

ies)

559

(3.5

)Q

oL (S

F-36

) (C

)Ad

here

nce

(sel

f-re

port

ed) (C

)6

Will

iam

s et a

l.31

(200

4)1

CP,

1 G

P(U

S)

140

(57.

1%);

73.7

6 w

k12

.0Cl

inic

al M

R (o

ne h

ealth

ce

ntre

am

bula

tory

clin

ic)

257

(4.1

)

No.

of p

resc

ribed

dr

ugs (?

)7

Zerm

ansk

y et

al.3

2

(200

1)1

clin

ical

ph

arm

acist

;4

GPs

(UK)

1188

(56%

);73

.512

mo

4.7

One

clin

ical

pha

rmac

ist

held

pt c

onsu

ltatio

ns in

GP

o!ce

follo

wed

by

clin

ical

M

R (4

gen

eral

pra

ctic

es)

502

(0.8

)N

o. o

f hos

pita

l ad

miss

ions

(C)

No.

of d

rug

chan

ges (B

); fre

quen

cy o

f do

se (C

); no

. of

pre

scrib

ed

drug

s (?

); no

. of

hosp

ital o

utpa

tient

at

tend

ance

s (C

); no

. of G

P co

nsul

tatio

ns (C

)

7

GP

= ge

nera

l pra

ctiti

oner

; F =

fem

ale;

pt(s

) = p

atie

nt(s

); Ca

n =

Cana

da; M

R =

med

icat

ion

revi

ew; P

IMs =

pot

entia

lly in

appr

opria

te m

edic

atio

ns; C

P =

com

mun

ity p

harm

acist

; N

Z =

New

Zea

land

; CC

= ca

se c

onfe

renc

e; Q

oL =

qua

lity

of li

fe; M

AI =

Med

icat

ion

Appr

opria

tene

ss In

dex;

NL

= N

ethe

rland

s; N

S =

not s

peci"e

d; U

S =

Uni

ted

Stat

es; U

K =

Uni

ted

King

dom

; PCI

s = p

harm

aceu

tical

car

e iss

ues;

DRP

= d

rug-

rela

ted

prob

lem

; VAS

= v

isual

ana

logu

e sc

ale;

Aus

= A

ustr

alia

; SF-

36 =

36-

item

Sho

rt F

orm

Hea

lth S

urve

y;

DU

SOI-A

= D

ukes

Sev

erity

of I

llnes

s Visu

al A

nalo

gue

Scal

e; A

DEs

= a

dver

se d

rug

even

tsB

sign

i"ca

nt e#e

ct in

favo

ur o

f int

erve

ntio

n gr

oup;

? si

gni"

cant

e#e

ct in

favo

ur o

f con

trol

gro

up; C

no

signi"c

ant d

i#er

ence

Tabl

e 2

cont

inue

d

35


medications (PIMs), with positive e!ects for one study.17 Process outcomes were reported in all studies. Two studies reported a reduction in the number of (prescribed) drugs,19,31,32 while in four studies no e!ect was reported.21,23,25,26 Five studies reported an increase in the number of drug changes.15-17,19,24,31,32Implementation rates of recommendations in the intervention group are shown in Table 2. "e percentage of implemented recommendations was reported in seven studies,15-17,19,23,25-27,32 while a percentage of resolved DRPs was mentioned in three studies.3,24,29,30 In two studies, the implementation rate was derived from the decrease in the number of PIPs compared with the total number of PIPs.21,31

Key elements of the inter ventionKey elements of 13 interventions from the 12 studies are shown in Table 3. One study compared outcomes between two intervention groups (case conference and written feedback)15,16 and therefore both study arms (A and B) are shown. Pharmacists had clinical experience in 10 of the 13 interventions.3,17,19,23,25-27,29-31 Pharmacists were accredited pharmacists,17,27 consultant pharmacists 23,31 or clinical pharmacists.19,32 "ey followed a university accredited externship programme,26 were clinically trained,3 experienced in medication reviews 24,25 or had a post-graduate quali#cation in pharmacy practice.25 In 8 of 13 interventions, the patients own GP was involved.15-17,25-27,29,30 In the other interventions, the study pharmacist had no existing therapeutic relationship with the patient or this was not described 3,19,21,23,31,32 (see also Table 1). Pharmacists had full access to GPs medical records of the patient in 8 of the 13 interventions.3,17,19,25-27,29,32 Patient interviews were conducted in 11 of 13 interventions, at home 3,17,21,23,25,27 or in the GPs o$ce or clinic.19,26,31,32 In 3 of 13 interventions, eligible patients were invited by the GP to participate in the study.17,19,27,32 Case conferences between GPs and pharmacists were conducted in 7 of 13 interventions.15-17,19,24-27,32 In three interventions, letters with recommendations or care plans were sent to the GPs (written feedback).3,15,16,23 As mentioned earlier, one study compared the process outcomes of case conferences with written feedback.15,16 As part of two interventions, case conferences were held by external multidisciplinary teams without the patients own GP 21,31 and recommendations were mailed to the GP 21 or implemented with endorsement of the GP.31 Action plans were used for implementation of agreed recommendations in 9 of 13 interventions.3,15-17,19,24-27,29,32 A follow-up of the implementation of actions was described in 11 of 13 interventions,3,15-17,19,21,23,25-27,29,31,32 most o%en conducted by a pharmacist.15,17,19,23,25,26,31,32

36

Tab

le 3

Key

elem

ents

of i

nter

vent

ion

vers

us im

plem

enta

tion

rate

of r

ecom

men

datio

ns

Stud

y(y

ear o

f pub

licat

ion)

Phar

mac

ist

with

cl

inic

al

expe

rienc

e

Ow

n ph

arm

acis

t in

volv

ed

Acce

ss to

m

edic

al

reco

rds

Patie

nt

inte

rvie

w

(loca

tion/

care

pr

ovid

er)

Invi

tatio

n by

GP

Case

co

nfer

ence

G

P an

d ph

arm

acis

t

Actio

n pl

an (c

are

prov

ider

)

Follo

w

up (c

are

prov

ider

)

Impl

emen

-ta

tion

rate

Out

com

e m

easu

re-

men

t (m

onth

s)

Den

nebo

om e

t al.1

5 stu

dy a

rm B

(200

7)

917

%6

Alla

rd e

t al.2

1 (20

01)

99

h/O

9

O25

%12

Gry

mon

pre

et a

l.23

(200

1)9

9 h

/P+O

9 P

29%

NS

Den

nebo

om e

t al.1

5 stu

dy a

rm A

(200

7)9

99

NS

9 C

P30

%6

Kwin

t et a

l.24

(201

1)9

99

9 N

S30

%6

Will

iam

s et a

l.31

(200

4)9

99

NS/

CP9

33%

1.5

Sore

nsen

et a

l.27

(200

4)9

99

9 h

/CP

99

9 G

P+CP

9 G

P54

%N

S

Sello

rs e

t al.2

6 (2

003)

9

99

9 G

PO/C

P9

9 G

P9

CP

56 %

5

Volu

me

et a

l.29

(200

1)9

99

9 N

S/CP

99

58%

NS

Brya

nt e

t al.1

7 (2

011)

99

99

h/C

P9

99

NS

9 C

P62

%12

Krsk

a et

al.3

(200

1)9

99

h/P

99

P+O

9 P

83%

3

Lena

ghan

et a

l.25

(200

7)9

99

9 h

/CP

99

9 G

P+O

9 C

P85

%N

S

Zerm

ansk

y et

al.3

2 (2

001)

99

9 G

PO/P

99

9 P

9 P

86%

12

GP

= ge

nera

l pra

ctiti

oner

; h =

pat

ient

s ho

me;

O =

oth

er c

are

prov

ider

; P =

pha

rmac

ist; N

S =

not s

peci!e

d; C

P =

com

mun

ity p

harm

acist

; GPO

= G

Ps o"

ce

37


Associat ion between number of ke y elements and outcomesKey elements of the intervention and the implementation rate are shown in Table 2. !e mean number of key elements present in the interventions was 5.2 (range 18). !e mean implementation rate was 50% (range 1786).!e association between the number of key elements present in the intervention and the implementation rate of recommendations was positive: an increase in number of key elements was related to an increase in implementation rate meta-regression, = 0.085 (95%CI 0.0520.128; P < 0.01) (Figure 2). Figure 3 is a forest plot showing the observed and expected implementation rates, estimated on the meta-analysis association between number of key elements and the implementation rate. In all but three interventions, the observed implementation rate was within the 95% con$dence interval of the expected value.3,19,27,32 No meta-regression analyses were possible for the association between the number of key elements and the number of hospital admissions (n = 4 studies), the number of drug changes (n = 5 studies) and the number of prescribed drugs (n = 5 studies), because of the low number of studies and participants with these outcomes.

DISCUSSION

!is systematic review shows a signi$cant association between the number of key elements of the intervention re%ecting collaborative aspects in medication review and the implementation rate. !is suggests that more intensive collaboration between GP and pharmacist in medication review leads to higher recommendation implementation rates.!e expected implementation rate could be predicted from the number of key elements estimated from this association (Figure 3). !is model gives a good prediction of the implementation rate for the majority of the studies. For three studies, the expected implementation rate was di&erent from the observed value.3,19,27,32 A higher implementation rate than expected was observed in the studies by Krska et al.3 and Zermansky et al.19,32 In the study by Zermansky et al.,19,32 one pharmacist collaborated with only a few GPs, similar to the study by Lenaghan et al.,25 resulting in comparably high implementation rates. !e major di&erence with the Lenaghan study was that the patients own pharmacist was not involved. However, it is conceivable that this (clinical) pharmacist established a good relationship with this small number of GPs and their patients whilst he was consulting patients in the GPs o'ce. Studies with pharmacists working at a GP practice yielded high rates of acceptance of recommendations.18,19,26,32 Conversely,

38

Fig

ure

2Bu

bble

plo

t of n

umbe

r of k

ey e

lem

ents

of i

nter

vent

ion

vs. i

mpl

emen

tatio

n ra

te o

f rec

omm

enda

tions

24

68

20406080

No.

of e

lem

ents

Implementation rate (%)

Gry

mon

pre

et a

l. (2

001)

Alla

rd e

t al.

(200

1)

Den

nebo

om e

t al.

stud

y ar

m B

(200

7)

Kwin

t et a

l.(2

011)

Will

iam

s et a

l.(2

004)

Den

nebo

om e

t al.

stud

y ar

m A

(200

7)

Volu

me

et a

l.(2

001)

Brya

nt e

t al.

(201

1)

Lena

ghan

et a

l. (2

007)

Sello

rs e

t al.

(200

3)

Sore

nsen

et a

l. (2

004)

Krsk

a et

al.

(200

1)

Zerm

ansk

yet

al.

(200

1)

y =

0.05

38 +

0.0

850x

The

size

of th

e ci

rcle

s re!

ects

the

num

ber o

f rec

omm

enda

tions

in th

e in

terv

entio

n gr

oup

of th

e di"e

rent

stud

ies.

39


Fig

ure

3Fo

rest

plo

t of o

bser

ved

and

expe

cted

impl

emen

tatio

n ra

te o

f rec

omm

enda

tions

025

5075

100

Impl

emen

tatio

n ra

te (%

)

Zerm

ansk

y et

al.

(200

1)

Lena

ghan

et a

l. (2

007)

Krsk

a et

al.

(200

1)

Brya

nt e

t al.

(201

1)

Volu

me

et a

l. (2

001)

Sello

rs e

t al.

(200

3)

Sore

nsen

et a

l. (2

004)

Will

iam

s et a

l. (2

004)

Den

nebo

om e

t al.

stud

y ar

m A

(200

7)

Kwin

t et a

l. (2

011)

Gry

mon

pre

et a

l. (2

001)

Alla

rd e

t al. (

2001

)

Den

nebo

om e

t al.

stud

y ar

m B

(200

7)

86 (8

389

)

85 (7

693

)

83 (8

185

)

62 (5

766

)

58 (5

462

)

56 (5

359

)

54 (5

059

)

33 (2

839

)

30 (2

237

)

30 (2

435

)

29 (2

632

)

25 (1

832

)

17 (1

124

)

Stud

y (y

ear o

f pub

licat

ion)

Mea

n ob

serv

ed im

plem

enta

tion

rate

with

con

!den

ce in

terv

als (

95%

CIs)

. The

gre

y di

amon

ds re

pres

ent t

he 9

5%CI

s of t

he e

xpec

ted

impl

emen

tatio

n ra

te e

stim

ated

on

the

met

a-re

gres

sion

anal

ysis

asso

ciat

ion

betw

een

the

num

ber o

f key

ele

men

ts a

nd th

e im

plem

enta

tion

rate

.

40

Krska et al.3 was the only study with a high implementation rate without a case conference. !is face-to-face meeting between GPs and pharmacists to discuss the pharmacists recommendations is o"en considered one of the most important and key elements of the collaborative approach in medication review.13,15,16,33 In the study by Krska et al.,3 pharmacists were assisted by practice sta# in the implementation of accepted actions. Possibly, this partly explains the high implementation rate. Furthermore, the nature and number of pharmacists and their relationship with patients was not speci$ed.3 On the other hand, a lower than expected implementation rate was observed in the intervention by Sorensen et al.27 In this study, a large number of pharmacists collaborated with an even larger number of GPs, which could have made it di%cult to achieve high implementation rates. !e implementation rate and numbers of GPs and pharmacists in Sorensen et al.27 were similar to those in the study by Sellors et al.26 !e major di#erence was that, in Sorensen et al.27, patients were invited by the GP.!ere have been no earlier systematic reviews investigating the implementation rate of recommendations. We found a signi$cant association between the number of key elements re&ecting collaborative aspects and the implementation rate. !is $nding is in agreement with other medication review studies in secondary and tertiary care where direct communication between healthcare providers revealed higher acceptance rates of recommendations.34,35 For clinical and intermediate outcomes, no association could be assessed, because the number of studies reporting these outcomes was too low. Earlier systematic reviews reported no e#ect on hospital admissions and quality of life.8,10 Compared with these reviews, our scope was more focused, as we included only RCTs, community-dwelling patients in primary care, a mean age of 70 years and no recent discharge, yielding only 12 trials a"er quality assessment. However, due to our inclusion criteria, patients in our review were relatively healthy and not directly at risk for hospital admission. In contrast, studies on medication review in hospitals were more successful in preventing hospital (re)admissions because they generally reviewed patients who were admitted to hospital and at high risk for readmission.36,37!ere may be additional reasons for low implementation rates. Sellors et al.26 showed that these reasons might include patient reluctance, previous failed attempts at the same strategy and a relatively short period for implementation combined with the occurrence of more urgent issues. In particular, the periods over which implementation rates were measured varied between the di#erent studies in this review. Nor do we know if the recommendations in the di#erent studies were clinically appropriate.17 GPs perceptions of pharmacists recommendations in the GPPC study revealed that they generally found the recommendations

41


useful although at times theoretical.38 Pharmacists recommendations may be less appropriate if a high proportion of patients are already receiving the recommended treatment, for example, in the MEDMAN study (ceiling e!ect).39"ere were several strengths to this study. First of all, like in other systematic reviews on medication review, trials reported very heterogeneous outcomes that could not be pooled. In this systematic review, we could compare di!erent trials using the implementation rate as the common (process) outcome. Implementation rates are also reported in many (observational) studies in community-dwelling patients 40,41 as well as for patients in nursing homes.34,42 Implementation rate has a greater signi#cance than acceptance rate of pharmacists recommendations by physicians because it includes enactment of the recommendation by a care provider and the level of acceptance of recommendations by the patient. Secondly, we described eight di!erent elements of the intervention that re$ect collaborative aspects between GPs and pharmacists. "ese key elements were based on described facilitators and barriers in medication review.12-14 "irdly, the importance of patient involvement in medication review was also re$ected by the key elements own pharmacist involved, patient interview and follow-up.Our decision to consider all eight key elements of the intervention as equivalent weighted determinants for the implementation rate could be seen as a limitation. For example, the face-to-face discussion between pharmacists and GPs seemed a key element that could have more weight. "e small number of studies precluded us from studying the association of the individual key elements in a multivariate design. Also, there could have been other key elements re$ecting collaborative aspects that we may have missed. Furthermore, it was not possible to discriminate between the clinical relevance of the implemented recommendations in the di!erent studies. "is clinical relevance was only described in the study by Denneboom et al.15,16 Finally, it cannot be ruled out that some relevant RCTs may have been missed or excluded.

CONCLUSION

"is systematic review showed that the number of key elements re$ecting collaborative aspects of medication review was signi#cantly associated with the recommendation implementation rate. Further clinical trials could demonstrate whether an increase in collaborative aspects leads to higher implementation rates.Based on this model, future studies for elderly in primary care could consider these key elements of intervention to design a standardized medication review process.

42

More research is needed to assess which key elements of this collaborative approach are the most important and if there are additional elements that may in!uence implementation rates. Next to the physician and the pharmacist, the patient is the third main player in the medication review process. Future studies could focus on the in!uence of the patient on the implementation rate. Large multicentre trials in primary care are needed to draw de"nitive conclusions on whether a standardized collaborative approach in medication review could a#ect clinical outcomes. Such trials may be expensive, di$cult to organize in practice settings and it may be questioned how many and which elderly community-dwelling patients in primary care are at greatest risk for negative clinical outcomes.

Acknowledgements - %e authors thank J.C. Riemens-Louisse, student, for her contribution to the data extraction forms and quality assessment forms and S.V. Belitser for her contribution to the statistical analysis.

REFERENCES

1. Rollason V, Vogt N. Reduction of polypharmacy in the elderly: a systematic review of the role of the pharmacist. Drugs Aging 2003;20:817-32.

2. Castelino RL, Bajorek BV, Chen TF. Targeting suboptimal prescribing in the elderly: a review of the impact of pharmacy services. Ann Pharmacother 2009;43:1096-1106.

3. Krska J, Cromarty JA, Arris F, Jamieson D, Hansford D, Du#us PRS, et al. Pharmacist-led medication review in patients over 65: a randomized, controlled trial in primary care. Age Ageing 2001;30:205-11.

4. Hanlon JT, Schmader KE, Ruby CM, Weinberger M. Suboptimal prescribing in older inpatients and outpatients. J Am Geriatr Soc 2001;49:200-9.

5. Shaw J, Seal R, Pilling M. Task Force on Medicines Partnership and the National Collaborative Medicines Management Services Programme. Room for review: a guide to medication review. 2002. Available from: http://www.npc.nhs.uk/review_medicines/intro/resources/room_for_review.pdf (Accessed 1 Feb 2013).

6. Clyne W, Blenkinsopp A, Seal R. National Prescribing Centre. A guide to medication review. 2008. Available from: http://www.npc.nhs.uk/review_medicines/intro/resources/agtmr_web1.pdf (Accessed 1 Feb 2013).

7. Lowe CJ, Petty DR, Zermansky AG, Raynor DK. Development of a method for clinical medication review by a pharmacist in general practice. Pharm World Sci 2000;22:121-6.

8. Holland R, Desborough J, Goodyer L, Hall S, Wright D, Loke YK. Does pharmacist-led medication review help to reduce hospital admissions and deaths in older people? A systematic review and meta-analysis. Br J Clin Pharmacol 2008; 65:303-16.

43


9. Royal S, Smeaton L, Avery AJ, Hurwitz B, Sheikh A. Interventions in primary care to reduce medication related adverse events and hospital admissions: systematic review and meta-analysis. Qual Saf Health Care 2006;15:23-31.

10. Nkansah N, Mostovetsky O, Yu C, Chheng T, Beney J, Bond CM, et al. E!ect of outpatient pharmacists non-dispensing roles on patient outcomes and prescribing patterns. Cochrane Database Syst Rev 2010;(7):CD000336.

11. Holland R, Smith R, Harvey I. Where now for pharmacist led medication review? J Epidemiol Community Health 2006;60:92-93.

12. Urbis Keys Young. Evaluation of the home medicines review program (pharmacy component): "nal report. 2005. Available from: http://beta.guild.org.au/uploaded"les/Medication_Management_Reviews/Overview/Urbis%20Keys%20Young%20evaluation.pdf (Accessed 3 Feb 2013).

13. MacKeigan LD, Nissen LM. Clinical pharmacy services in the home. Dis Manage Health Outcomes 2008;16:227-44.

14. Niquille A, Lattman C, Bugnon O. Medication reviews led by community pharmacists in Switzerland: a qualitative survey to evaluate barriers and facilitators. Pharm Pract 2010;8:35-42.

15. Denneboom W, Dautzenberg MG, Grol R, De Smet PA. Treatment reviews of older people on polypharmacy in primary care: cluster controlled trial comparing two approaches. Br J Gen Pract 2007;57:723-31.

16. Denneboom W, Dautzenberg MGH, Grol R, De Smet PAGM. Comparison of two methods for performing treatment reviews by pharmacists and general practitioners for home-dwelling elderly people. J Eval Clin Pract 2008;14:446-52.

17. Bryant LJ, Coster G, Gamble GD, McCormick RN. #e General Practitioner-Pharmacist Collaboration (GPPC) study: a randomised controlled trial of clinical medication reviews in community pharmacy. Int J Pharm Pract 2011;19:94-105.

18. Isetts BJ, Brown LM, Schondelmeyer SW, Lenarz LA. Quality assessment of a collaborative approach for decreasing drug-related morbidity and achieving therapeutic goals. Arch Intern Med 2003;163:1813-20.

19. Zermansky AG, Petty DR, Raynor DK, Lowe CJ, Freemantle N, Vail A. Clinical medication review by a pharmacist of patients on repeat prescriptions in general practice: a randomised controlled trial. Health Technol Assess 2002;6:1-86.

20. Verhagen AP, de Vet HC, de Bie RA, Kessels AG, Boers M, Bouter LM, et al. #e Delphi list: a criteria list for quality assessment of randomized clinical trials for conducting systematic reviews developed by Delphi consensus. J Clin Epidemiol 1998; 51: 1235-41.

21. Allard J, Hebert R, Rioux M, Asselin J, Voyer L. E$cacy of a clinical medication review on the number of potentially inappropriate prescriptions prescribed for community-dwelling elderly people. CMAJ 2001;164: 1291-96.

22. Bernsten C, Bjorkman I, Caramona M, Crealey G, Frokjaer B, Grundberger E, et al. Improving the well-being of elderly patients via community pharmacy-based provision of pharmaceutical care: a multicentre study in seven European countries. Drugs Aging 2001;18:63-77.

23. Grymonpre RE, Williamson DA, Montgomery PR. Impact of a pharmaceutical care model for non-institutionalised elderly: results of a randomised, controlled trial. Int J Pharm Pract 2001;9:235-41.

44

24. Kwint HF, Faber A, Gussekloo J, Bouvy ML. E!ects of medication review on drug-related problems in patients using automated drug-dispensing systems: a pragmatic randomized controlled study. Drugs Aging 2011;28:305-14.

25. Lenaghan E, Holland R, Brooks A. Home-based medication review in a high risk elderly population in primary care - "e POLYMED randomised controlled trial. Age Ageing 2007;36:292-7.

26. Sellors J, Kaczorowski J, Sellors C, Dolovich L, Woodward C, Willan A, et al. A randomized controlled trial of a pharmacist consultation program for family physicians and their elderly patients. CMAJ 2003;169:17-22.

27. Sorensen L, Stokes JA, Purdie DM, Woodward M, Elliott R, Roberts MS. Medication reviews in the community: results of a randomized, controlled e!ectiveness trial. Br J Clin Pharmacol 2004;58: 648-64.

28. Sturgess IK, McElnay JC, Hughes CM, Crealey G. Community pharmacy based provision of pharmaceutical care to older patients. Pharm World Sci 2003;25:218-26.

29. Volume CI, Farris KB, Kassam R, Cox CE, Cave A. Pharmaceutical care research and education project: patient outcomes. J Am Pharm Assoc 2001;41:411-20.

30. Kassam R, Farris KB, Burback L, Volume CI, Cox CE, Cave A. Pharmaceutical care research and education project: pharmacists interventions. J Am Pharm Assoc 2001 ;41:401-10.

31. Williams ME, Pulliam CC, Hunter R, Johnson TM, Owens JE, Kincaid J, et al. "e short-term e!ect of interdisciplinary medication review on function and cost in ambulatory elderly people. J Am Geriatr Soc 2004;52:93-8.

32. Zermansky AG, Petty DR, Raynor DK, Freemantle N, Vail A, Lowe CJ. Randomised controlled trial of clinical medication review by a pharmacist of elderly patients receiving repeat prescriptions in general practice. BMJ 2001;323:1340-3.

33. Almeida Neto AC, Chen TF. Exploring elements of interprofessional collaboration between pharmacists and physicians in medication review. Pharm World Sci 2007;29:574-6.

34. Nishtala PS, McLachlan AJ, Bell JS, Chen TF. A retrospective study of drug-related problems in Australian aged care homes: medication reviews involving pharmacists and general practitioners. J Eval Clin Pract 2011;17:97-103.

35. Spinewine A, Swine C, Dhillon S, Lambert P, Nachega JB, Wilmotte L, et al. E!ect of a collaborative approach on the quality of prescribing for geriatric inpatients: a randomized, controlled trial. J Am Geriatr Soc 2007;55:658-65.

36. Gillespie U, Alassaad A, Henrohn D, Garmo H, Hammarlund-Udenaes M, Toss H, et al. A comprehensive pharmacist intervention to reduce morbidity in patients 80 years or older: a randomized controlled trial. Arch Intern Med 2009;169:894-900.

37. Makowsky MJ, Koshman SL, Midodzi WK, Tsuyuki RT. Capturing outcomes of clinical activities performed by a rounding pharmacist practicing in a team environment: the COLLABORATE study. Med Care 2009;47:642-50.

38. Bryant L, Coster G, McCormick R. General practitioner perceptions of clinical medication reviews undertaken by community pharmacists. J Prim Health Care 2010;2:225-33.

39. Bond C. "e MEDMAN study: a randomized controlled trial of community pharmacy-led medicines management for patients with coronary heart disease. Fam Pract 2007;24:189-200.

45


40. Farris KB, Cote I, Feeny D, Johnson JA, Tsuyuki RT, Brilliant S, et al. Enhancing primary care for complex patients: demonstration project using multidisciplinary teams. Can Fam Physician 2004;50:998-1003.

41. Kra

kwint.pdf

Documents

home medication review

medication adherence

utrecht institute

menzis chapter

systematic review

care cooperation chapter

identication of drug

drugrelated problems