Residential Aged Care Medication Review to Improve the Quality of Medication Use: A Systematic Review

JAMDA xxx (2016) 1.e1e1.e14

JAMDA journal homepage: www.jamda.com

Review Article

Residential Aged Care Medication Review to Improve the Quality of Medication Use: A Systematic Review Kaeshaelya Thiruchelvam BPharm a, *, Syed Shahzad Hasan BPharm, MClinPharm, PhD a, Pei Se Wong MPharm, PhD a, Therese Kairuz DipPharm(Hons), MPharm, PhD, GCHEd b a b

School of Pharmacy, International Medical University, Kuala Lumpur, Malaysia College of Medicine and Dentistry, James Cook University, Townsville, Queensland, Australia

a b s t r a c t Keywords: Aged care medication review medication use quality residential

Background: Aging is often associated with various underlying comorbidities that warrant the use of multiple medications. Various interventions, including medication reviews, to optimize pharmacotherapy in older people residing in aged care facilities have been described and evaluated. Previous systematic reviews support the positive impact of various medication-related interventions but are not conclusive because of several factors. Objectives: The current study aimed to assess the impact of medication reviews in aged care facilities, with additional focus on the types of medication reviews, using randomized controlled trials (RCTs) and observational studies. Methods: A systematic searching of English articles that examined the medication reviews conducted in aged care facilities was performed using the following databases: PubMed, CINAHL, IPA, TRiP, and the Cochrane Library, with the last update in December 2015. Extraction of articles and quality assessment of included articles were performed independently by 2 authors. Data on interventions and outcomes were extracted from the included studies. The SIGN checklist for observational studies and the Cochrane Collaboration’s tool for assessing risk of bias in RCTs were applied. Outcomes assessed were related to medications, reviews, and adverse events. Results: Because of the heterogeneity of the measurements, it was deemed inappropriate to conduct a meta-analysis and thus a narrative approach was employed. Twenty-two studies (10 observational studies and 12 controlled trials) were included from 1141 evaluated references. Of the 12 trials, 8 studies reported findings of pharmacist-led medication reviews and 4 reported findings of multidisciplinary team-based reviews. The medication reviews performed in the included trials were prescription reviews (n ¼ 8) and clinical medication reviews (n ¼ 4). In the case of the observational studies, the majority of the studies (8/12 studies) reported findings of pharmacist-led medication reviews, and only 2 studies reported findings of multidisciplinary team-based reviews. Similarly, 6 studies employed prescription reviews, whereas 4 studies employed clinical medication reviews. The majority of the recommendations put forward by the pharmacist or a multidisciplinary team were accepted by physicians. The number of prescribed medications, inappropriate medications, and adverse outcomes (eg, number of deaths, frequency of hospitalizations) were reduced in the intervention group. Conclusion: Medication reviews conducted by pharmacists, either working independently or with other health care professionals, appear to improve the quality of medication use in aged care settings. However, robust conclusions cannot be drawn because of significant heterogeneity in measurements and potential risk for biases. Ó 2016 AMDA e The Society for Post-Acute and Long-Term Care Medicine.

All authors contributed equally to this work. The authors declare no conflicts of interest. * Address correspondence to Kaeshaelya Thiruchelvam, School of Pharmacy, International Medical University, No. 126, Jalan Jalil Perkasa 19, Bukit Jalil, Kuala Lumpur 57000, Malaysia. E-mail address: [email protected] (K. Thiruchelvam). http://dx.doi.org/10.1016/j.jamda.2016.10.004 1525-8610/Ó 2016 AMDA e The Society for Post-Acute and Long-Term Care Medicine.

Aging is inevitable, and the global proportion of older people is escalating. People aged 60 years and older constituted 11% of the world’s population in 2009, and the figure will double by midcentury.1 Developed countries are expected to experience a steep rise in the number of senior citizens as people are living longer given improved living conditions, medical advances, and implementation of health

1.e2

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

care programs. Although this phenomenon is the outcome of successful health care, the aging society poses a reign of challenges for any health care model.2 Aging is often related to chronic diseases, typically cerebrovascular, cardiovascular, and musculoskeletal conditions. These underlying comorbidities warrant the use of multiple medications, which are often recommended in respective clinical guidelines, leading to polypharmacy. Polypharmacy refers to the use of multiple medications and may affect people of any age3; however, it is more prevalent among older people, especially those residing in aged care facilities where the medications have been increasing over the past decade.4 “Multiple prescribing clinicians” contribute to polypharmacy and may be practicing in primary care as community-based or family physicians, or in secondary care; secondary care can be defined as treatment provided by a specialist after referral of the patient from a primary care family physician.5 Polypharmacy may result in suboptimal management of a disease: multiple medications and the prescribing of medications within similar drug classes. Furthermore, patients may receive subtherapeutic or toxic dosages, or misdiagnoses; and in some cases, suboptimal management may be associated with nonadherence or nontreatment (treatment deprivation).6 Although not unique to long-term care settings, a frequently encountered predicament is the drug prescribing cascade: the prescribed medication causes an adverse drug reaction that is mistakenly diagnosed as a new condition resulting in the prescribing of new medications.7 Various interventions to optimize pharmacotherapy in older people residing in aged care facilities have been described and evaluated. They include medication reviews, pharmacist interventions, multidisciplinary team interventions, geriatric evaluation, computerized decision support systems, management teams, and educational approaches. Medication review is defined as “a structured, critical examination of a patient’s medicines with the objective of reaching an agreement with the patient about treatment, optimizing the impact of medicines, minimizing the number of medication-relation problems and reducing waste.”8 The definition of medication review was simplified in 2008 to 3 levels from its previous 4-level definition of ad hoc, prescription, treatment, and clinical medication review. The latest definition of a medication review comprises prescription review (type 1), concordance and compliance review (type 2), and clinical medication review (type 3).9 Pharmacist-led medication reviews have formed the foundation of numerous intervention studies in Australia, the United States, and Europe.10 They are known as Residential Medication Management Review (RMMR) in Australia, Medication Therapy Management in the United States, Medicines Use Review in the United Kingdom and New Zealand, and MedsCheck in Canada.9,11e14 Medication review services were implemented in the United States as early as 1974 with a quarterly medication regimen review15; the Omnibus Budget Reconciliation Act (OBRA-87) was legislated that necessitated national minimum standards of care for residents of certified nursing facilities, and it became mandatory for clinical pharmacists to conduct monthly medication regimen reviews.16 In Australia, the RMMR is a service offered to government-funded aged care home residents. The RMMR services may be provided by an accredited pharmacist (pharmacist RMRR) or in partnership with a pharmacist and general practitioner (collaborative RMMR).17 Previous systematic reviews support the positive impact of various medication-related interventions; however, they are not conclusive owing to several factors. For instance, although some reviews aimed to reduce inappropriate prescribing, optimize pharmacotherapy, and improve the appropriate use of polypharmacy, the reviews included studies that were conducted in various care settings without specifications about study design; some had mixed interventions that may or may not have included pharmacists (Kaur et al,18 Patterson et al,19

Spinewine et al,20 Spinewine et al,10 and Tjia et al21). Although other reviews have focused on studies conducted in aged care facilities, the interventions were of various types without the compulsory involvement of a pharmacist and study designs were not specified (Alldred et al,4 Nishtala et al,17 Forsetlund et al,22 Kröger et al,23 Loganathan et al,24 and Marcum et al25). Two other reviews focused on studies that did include pharmacists, either working independently or with other health care professionals; however, the studies used various designs and methodologies (Cooper et al26 and Geurts et al27) or included a variety of care settings (Cooper et al,26 Geurts et al,27 Holland et al,28 and Zermansky and Silcock29). Furthermore, the study population in the review by Holland et al included randomized controlled trials (Holland et al28) but may not have restricted the study population to “only” older people as the selection criteria included “mainly older people.” A review by Castelino et al30 included randomized controlled trials and the compulsory involvement of pharmacists, but the studies were conducted in a variety of settings and used mixed interventions. A review by Verrue et al31 included 8 randomized controlled trials, did not focus on medication-review interventions, and did not include longitudinal studies. Da Silva et al32 published findings in the form of a “Letter to the Editor,” which was very brief and included 7 studies without a focus on medication reviews. For the reasons discussed above, a comprehensive critique of current literature is essential to determine the impact of medication reviews in aged care facilities. The gap in current evidence should be addressed to identify the impact of medication reviews conducted by pharmacists, either working independently or with other health care professionals in aged care facilities. The aim of the current study was to assess the impact of medication reviews in aged care facilities, with additional focus on the types of medication reviews (prescription and/ or clinical medication reviews) in a single care setting (aged care homes) using a specific study design (randomized controlled trials and prospective studies). Methods Scope of Review: Eligibility Criteria The systematic review process was conducted in line with the PRISMA guidelines. The primary investigators (KT and SSH) screened abstracts for articles published in English (a) addressing medication reviews, (b) reporting pharmacist-led or multidisciplinary team reviews in aged care facilities, and (c) reporting outcome measures and impact of interventions on medication usage. Studies focusing on cognitive, behavioral, and educational approaches or interventions, and evaluations of cross-sectional or case series data were excluded, whereas randomized controlled trials, longitudinal studies, and those measuring pre and post medication review interventions were included. The publication period was from January 1998 to December 2015 and included studies on people older than 60 years living in aged care facilities, which assessed medication review and its impact on medication use and patient safety. Studies focusing on medication reviews performed by other health care professionals or by a multidisciplinary team without pharmacist involvement were excluded. Information Sources The following databases were searched: MEDLINE, Cumulative index to Nursing and Allied Health Literature (CINAHL), Turning Research into Practice (TRiP), International Pharmaceutical Abstracts (IPA), and Cochrane Database of Systematic Reviews, with the last update in December 31, 2015. Reference lists of articles identified in the search and relevant review articles were included and were subject to the same eligibility evaluation.

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

Searching The search strategy identified research on medication review interventions involving pharmacists (alone or in a multidisciplinary team). Search terms were constructed using a Population (P), Intervention (I), Outcome (O) model with the following terms: “(older people OR senior citizen OR elderly OR geriatrics OR aged care residents OR nursing home residents OR long-term care residents) AND (medication review OR medication reconciliation OR pharmacist’s intervention OR pharmaceutical intervention OR pharmacy services) AND (medication appropriateness OR medication use OR drug use OR optimi* medication use/pharmacotherapy/prescribing OR appropriate prescribing/medication use OR polypharmacy OR unnecessary medication use OR suboptimal prescribing OR inappropriate prescribing/medication use OR potential inappropriate use* OR pharmacy practice experience*).” Titles and abstracts were screened to remove studies that were clearly irrelevant to the aim of the review. The full texts of the remaining studies were then examined to determine eligibility. Study Selection Two investigators (KT and SSH) assessed abstracts independently against 5 criteria: (1) medication reviews conducted by pharmacists either independently or in a multidisciplinary team; (2) medication reviews conducted in aged care facilities; (3) randomized controlled trials or observational (longitudinal and pre and post intervention) studies; (4) included older people (mean age of subjects > 60 years); and (5) encompassed patients with a range of diseases. Studies were conducted in aged care facilities with a minimum follow-up period of 1 month, and reported in English. Full papers of potential studies were independently assessed by the 2 investigators for their suitability. Studies involving the transition of residents in aged care facilities to and from hospitals or other care settings were excluded as the nature and impact of the interventions could vary. Data Collection Process The principal investigator extracted data using a data extraction form (table format) that was verified by the second reviewer. The extracted data included the name of the first author; country; publication year; study design; duration of the study; number of subjects, and number of females; mean age of the subjects; types of care setting; any inclusion criteria; types of intervention; description of the intervention; outcomes measured; and impact of the intervention (Table 1). The impact of the interventions as reported by the authors of the included studies was also presented. Data Items/Study Characteristics Classification of interventions We assessed medication reviewebased interventions conducted by pharmacists, either independently or in a multidisciplinary team, aimed at improving medication use and patient safety in aged care settings. Interventions had to be targeted at patients and not at health professionals, and potentially included clinical medication reviews and prescription reviews. “Clinical medication review” is a holistic review that involves a health care professional reviewing the medications and health status of patients, with full access to their medical and care records or laboratory test results, followed by a consultation with the patient and their carer. This comprehensive clinical approach to medication review is often conducted by a prescriber or a specialist practitioner, such as a pharmacist with a special area of

1.e3

interest or accredited to conduct reviews, or a specialist diabetes nurse.9 “Prescription review” considers practical pharmacotherapy issues that aim to enhance the clinical and cost-effectiveness aspects of medications and patient safety. This type of review may be conducted in the absence of the patient, and any medication changes made subsequent to the review should be done with the patient’s consent. A prescription review may be applicable when a patient is transferred from one setting to another, or when changes need to be made during the prescribing or dispensing process.9 Outcome measures Outcome measures included medication-related outcomes (eg, medication-related problems, pharmacotherapy problems, polypharmacy, medication appropriateness, number of medications per patient, use of potentially inappropriate medication [PIMs], number of PIMs, potentially inappropriate prescribing, changes in frequency and type of medication therapy), medication reviewerelated outcomes (eg, rate of acceptance of the recommendations by the pharmacist or team, number of recommendations, type of recommendations, reasons for acceptance of recommendations), and adverse outcomes (eg, potential risks such as falls, sentinel events, mortality, accidents, adverse drug events, disability indices, hospitalization). Studies were included if they reported at least 1 medication-related and/or 1 review-related and/or 1 adverse outcome measure. Aged care settings Various terms are used to describe the homes that provide care for older people. In Australia they are known as “aged care facilities,” in the United States “long-term care facilities,” and in the United Kingdom, “care homes.” Aged care facilities are of 2 types: those that provide continuous nursing care (aged care facilities delivering highlevel care in Australia, skilled-nursing facilities in the United States and nursing homes in the United Kingdom); and those that provide personal care (aged care facilities providing low-level care in Australia, assisted-living in the United States, and residential homes in the United Kingdom). Some homes provide both forms of care.4 This review included studies focusing on any type of aged care facility. Assessment of Quality and Risk of Bias in Included Studies Two investigators (KT and SSH) independently assessed the internal validity of each included study. Each observational study was evaluated against the Scottish Intercollegiate Guidelines Network (SIGN) Methodology Checklist for Observational Studies35 and ranked in terms of deficiencies. For randomized controlled trials and nonrandomized controlled trials, the Cochrane Collaboration’s tool36 and the Effective Practice and Organisation of Care criteria37 were used to assess the risk of bias based on 10 standard criteria: random sequence generation; allocation concealment; blinding; incomplete outcome data; selective reporting; other risk of bias; similar baseline outcome measurements; similar baseline characteristics; reliable primary outcome measures; and adequate protection against contamination. The investigators made judgments as to whether the studies were at low risk, high risk, or unclear risk of bias, and reported all included studies in Figure S1 (see Supplementary File). Results Search Results and Study Characteristics The search yielded 1060 unique abstracts from PubMed and 101 additional abstracts from CINAHL and IPA. After removal of duplicate records, 1141 abstracts were screened. Of these, 912 studies were

1.e4

Table 1 Summary of Randomized and Nonrandomized Controlled Trials Design and Setting

Inclusion Criteria

Intervention

Description of Intervention

Outcome Measures

Impact of Interventions

1. Houghton et al (UK e2014)41

Cluster RCT; 30 care homes for older people (381 participants from 15 intervention homes and 445 participants from 15 control homes). Female: NA Mean age: NA Period: 12 mo RCT; 369 residents (182 in intervention and 187 in the control group). Intervention: Female: 75.8% Mean age: 87 Control: Female: 75.9% Mean age: 87.7 Period: 2 mo

d

Intervention type: Multidisciplinary team medication review Review type: Clinical medication review

Intervention homes received a multiprofessional medication review at study commencement and at 6 mo, with follow-up at 12 mo. Control homes received usual care and a multiprofessional medication review after study completion.

Primary outcomes: Potentially inappropriate prescribing (number of drugs matching STOPP). Secondary outcomes: Number of emergency hospital admissions and mortality

No significant difference between the treatment groups, with the exception of potentially inappropriate prescribing. Significant effects were found for the number of drugs meeting the STOPP criteria at 12 mo. No adverse events were reported during the trial.

Users of the multidose drug-dispensing system aged 75 years or older, living in nursing homes or their own homes with municipally provided home care.

Intervention type: Pharmacist-led medication review Review type: Prescription review

Impact of pharmacist-led medication reviews on the number of PIMs. Intervention model led to reduction in the number of patients using 10 drugs and 3 psychotropics. The study also intended to classify and describe the types of DRPs identified during the intervention period and the medication reviews’ impact on the patients’ medication therapy.

3. Verrue et al (Belgium2012)43

Nonrandomized RCT; 69 residents from 1 intervention home and 79 residents from 1 control home. Intervention: Female: 60% Mean age: 86 Control: Female: 52% Mean age: 79.9 Period: 6 mo

Terminal illness, life expectancy <3 mo, and insufficient clinical data obtained from the GP were excluded.

Intervention type: Pharmacist-led medication review Review type: Prescription review

Pharmacists performed a systematic medication review without personal patient contact. The medication review included assessment of relevant parts of the electronic medical records (EMRs) and collection of data on the clinical variables. To identify DRPs, the clinical pharmacist initiated medication reviews based on the background information [symptom assessment form and the multidose drug dispensing (MDD) cards]. The working process was carried out in a structured way with formularies compiled from the Lund Integrated Medicines Management model. The clinical pharmacist prepared a record with clinical (obtained via GP) and pharmaceutical (taken from medication record in NH) data. The appropriateness of prescribing was determined, and based on this information recommendations to optimize treatment were formulated. The pharmacist consulted these recommendations with a geriatrician, whereafter they were communicated to the GPs to accept or reject.

4. Lapane et al (USA e2011)39

RCT: Later changed to an observational study (prospective pre and post intervention study design); (part of the Fleetwood Modeldprospective); 12 nursing homes receiving intervention and 13

All residents living in participating nursing homes were eligible for the Fleetwood Model, but those at high risk for medication-related problems were given priority for the prospective DRR

Intervention type: Pharmacist-led medication review Review type: Prescription review

The Fleetwood Model, performed by dispensing and consultant pharmacists, incorporated prospective reviews, direct communication with the prescribers, and formalized pharmaceutical care planning in patients at highest risk for medication-related problems

Use of PIMs Hospitalizations due to potential ADE Mortality

The proportion of patients with at least 1 PIM decreased in the intervention group (by 6%) but not in the control group. The total number of DRPs in the intervention group was 431 [mean 2.5 (1.5) DRPs per patient (range 0-9)]. No significant difference between the number of DRPs in community-dwelling patients [mean 2.55 (1.29)] and nursing home patients [mean 2.53 (1.33)] was seen. The 2 most common recommendations were withdrawal of drug therapy (30%) and reduced dosage (28%). The changes in the actions taken by the physician were significant for “lowered dosage” and there was a clear tendency to withdraw the PIMs, but not significant. Those in the intervention group were more likely to have an improvement in MAI score than in the control group (odds ratio [OR] 3.91; 1.88-8.15) No difference in 1-point reduction in the number of inappropriate ACOVE ratings and in START ratings. Those in the intervention group were significantly more likely to have a reduction of at least 1 Beers drug compared to the control nursing home (OR 6.52, 1.38-30.92). No significant between-group differences in STOPP ratings were observed (OR 2.86, 0.7111.53). Rates of PIM declined in usual care homes during the full-intervention phase. There was favor to the intervention homes, but no statistical significance (HR 0.86, 0.651.12). No statistically significant effect observed for potential ADE-related hospitalizations when adjusted for clustered design (adjusted HR 1.01, 0.84-1.21).

2. Milos et al (Switzerland e2013)42

Appropriateness of prescribing measured at baseline and after 6 mo. eOverall quality of prescribing: MAI eUnderuse: ACOVE and START criteria eOveruse: Beers and STOPP criteria.

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

Authors (CountryeYear)

5. Patterson et al (Northern Irelande 2010)40

7. Zermansky et al (UK e2006)33

Open RCT; 661 residents from 65 nursing, residential, and mixed care homes in UK. Intervention: Female: 25% Mean age: 85.3 Control: Female: 21% Mean age: 84.9 Period: 6-mo

Homes with >30 resident beds included [homes for general nursing category residents and those for elderly mentally infirm (EMI) people]. Homes that cared exclusively for terminally ill people excluded.

Intervention type: Pharmacist-led medication review Review type: Clinical medication review

Specially trained pharmacists visited the intervention homes monthly for 12 mo and reviewed residents’ clinical and prescribing information, applied an algorithm that guided them in assessing the appropriateness of psychoactive medication, and worked with prescribers (GPs) to improve the prescribing of these drugs.

Proportion of residents prescribed 1 or more inappropriate psychoactive medicine at 12 mo after the start of the intervention. Falls evaluation using routinely collected falls data.

Residents taking inappropriate psychoactive medications at 12 mo in the intervention homes (25/128, 19.5%) was much lower than in control homes (62/ 124, 50.0%) (OR 0.26, 0.14-0.49) No difference observed at 12 mo in the falls rate between the intervention and control groups.

Care home residents aged 65 years, and prescribed at least 1 repeat medication, living in nursing, residential and mixed care homes for older people

Intervention type: Pharmacist-led medication review Review type: Clinical medication review

Rate and nature of pharmacist interventions following clinical medication review. Interventions categorized by therapeutic categories Acceptance and implementation of interventions

At baseline: mean of 6.9 medicines per patient (SD 3.3). At least 1 intervention was required by 77% (95% CI 73.1-81.7) of residents. Most common intervention was technical (eg, generic switching, amending quantities) followed by test to monitor therapy (22%), stopping a medicine (13%), and a recommendation to start a new medicine (10%), one-third being for a new indication and two-thirds being for an existing indication. 75.6% of the pharmacist’s recommendations were accepted by the GP, and 77% of accepted interventions were acted upon.

Residents in another clinical trial, terminally ill (life expectancy <1 mo), or already receiving clinical medication reviews (CMRs) by a pharmacist, individuals at the GP’s request excluded.

Intervention type: Pharmacist-led medication review Review type: CMR

The reviews took place within 1 mo of randomization. The method of clinical medication review was based on that described by Lowe et al, but was adapted to the care home setting. The adaptations made were to review the residents’ medicines administration record (MAR), review the residents’ care records held at the home and to consult the resident’s carer. The study pharmacist conducted a thorough review of the prescription and medical notes; this usually took 20-30 min per resident. Written recommendations were then formulated and sent to the GP on a pro forma. A CMR was conducted by the study pharmacist within 28 days of randomization. It comprised a review of the GP clinical record and a consultation with the patient and carer. The pharmacist formulated recommendations and passed them on a written pro forma to the GP for acceptance and implementation. Control patients received usual GP care.

Primary outcome measure was the number of changes in medication per participant. Secondary outcome measures were: Medication outcomes eNumber of repeat medicines per participant eCost of 28 days of repeat medicines per participant. eRecorded medication reviews in the study period Clinical outcomes in 6 mo eFalls, mortality, hospital admissions, number of GP consultations, Barthel Index, SMMSE

Number of drug changes in 6 mo in the case group was 3.1 (2.7) and 2.4 (2.6) in the control group [RR 1.34 (1.21-1.48)]. Mean no of repeat drugs per patient in the case group, 6.7 (3.3) and 8.9 (3.6) in the control group [RR 0.98 (0.92-1.05)]. Mean falls after 6 mo in the case group was 0.8 (1.7) and 1.3 (3.1) in the control group [RR 0.59 (0.49-0.70)]. Number of deaths in the case group was 51 (15.3) and 48 (14.5) [RR 0.89 (0.56-1.41)]. Number of hospitalizations in per patient in the case group was 0.20 (0.48) and 0.26 (0.61) in the control group [RR 0.75 (0.521.07)]. The mean Barthel Index in the case group was 9.8 (6.1) and 9.3 (6.2) in the control group [RR 0.46 (0.02 to 0.94)] and mean SMMSE in the case group was 13.9 (10.0) and 13.8 (10.6) in the control group [RR 0.24 (1.18 to 0.70)].

1.e5

(continued on next page)

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

6. Alldred et al (UKe2007)45

in the comparison group in the US, 2002-2004. Intervention: Female: 72.3% Control: Female: 72.2% Cluster RCT; nursing homes randomized to intervention (n ¼ 11; 173 residents) and to control (n ¼ 11; 161 residents). Intervention: Female: 73.4% Mean age: 82.9 Control: Female: 72.3% Mean age: 82.6 Period: 12 mo RCT; 315 residents from 65 care homes. Female: NA Mean age: NA Period: 1 mo from randomization

1.e6

Table 1 (continued ) Design and Setting

Inclusion Criteria

Intervention

Description of Intervention

Outcome Measures

Impact of Interventions

8. Crotty et al (Australia e2004)44

Cluster RCT; 154 residents; 10 highlevel aged care facilities. Intervention: Female: 79.3% Mean age: 85.3 Control: Female: 79.3% Mean age: 83.6. Period: 3 mo Nonrandomized RCT (retrospective data set); 245 residents from 3 Canberra nursing homes. Intervention: Female: 71% Mean age: 78.9 Control: Female: 71% Mean age: 80.2 Period: 10/4/1996e4/ 12/1996 (8 mo)

Inclusion: Residents with difficult behavior about whom staff would like more advice and information, prescribed >5 medications

Intervention type: Multidisciplinary team medication review. Review type: Prescription review

2 multidisciplinary case conferences conducted 6-12 wk apart. Attendees of the conference were the resident’s GP, a geriatrician, a pharmacist, a residential care staff, and a representative of the Alzheimer’s Association of SA. A medication review was conducted by independent pharmacists, pre and post case conference.

Effect of intervention on the residents’ behavior (NHBPS scores). Monthly drug costs for intervention and control groups.

Mean MAI score, Intervention group: 4.1 (2.1-6.1) Mean MAI score, Control group: 0.4 (0.41.2) Mean MAI score for benzodiazepines intervention group: 0.73 (0.16-1.30) Mean MAI score for benzodiazepines control group: 0.38 (1.02 to 0.27).

Intervention type: Multidisciplinary team medication review Review type: Prescription review. (The residents concerned or their carers only occasionally attended the meetings.)

Medication use and cost Mortality

Review of 69 residents generated 170 recommendations; 102 were initiated by the pharmacist, 39 of which were accepted unaltered by the committee. By follow-up, all 39 of these had been implemented. The committee initiated 68 recommendations and modified 9 of the pharmacist’s recommendations. Only 53 (69%) of these had been implemented by follow up. 1 mo after the reviews, nonsignificant reductions in medication orders, medication cost, and mortality in the reviewed group.

Cluster RCT; 905 residents from 13 intervention nursing homes and 2325 residents from 39 control nursing home in south-east Queensland and north-east New South Wales, Australia. Female: NA Mean age: NA Period: 12 mo

Inclusion (nursing homes having supply of drugs under the Australian government medication subsidy schemed the Pharmaceutical Benefits Scheme, and central record of hospitalizations, adverse events, and death. Nursing homes were matched on resident age, resident classification instrument (RCI) distribution, and bed numbers into groups of 4 homes after baseline data collection.

Reviews were held weekly, rotating between the 3 nursing homes, and were arranged so that 3 residents could be reviewed in 1 session. The pharmacist collected and recorded the data on the proceedings and outcomes. These data included the problem, recommendations, and management plan that was implemented. These multidisciplinary conferences were attended by the 3 presenting GPs, a clinical pharmacist, senior nursing staff, and other health professionals The GP chose the resident he or she reviewed and was responsible for obtaining the resident’s consent. An intervention home was matched to 3 controlled homes. There were 3 phases during the intervention; introducing a new professional role to stakeholders with relationship building; nurse education; and medication review by pharmacists who had a postgraduate diploma in clinical pharmacy. There were focus groups facilitated by a research nurse, written and telephone communication, and face-to-face professional contact between nursing home staff and clinical pharmacists on issues such as drug policy and specific resident problems, together with education and medication review. Written, referenced drug regimen reviews were prepared by the clinical pharmacists for 500 individual residents selected by the nursing home staff.

Continuous drug use data from government prescription subsidy claims, crosssectional drug use data on prescribed and administered medications, deaths and morbidity indices (hospitalization rates, adverse events, and disability indices).

The percentage change (14.8%) observed from the prescription claims data was similar to the reductions of 11.4% and 11.7% for prescribed and administered drugs in the cross-sectional sample of 8 nursing home clusters (1328 residents). No differences between the intervention (mean 6.60, 95% CI 5.60-7.60) and control (mean 7.45, 95% CI 6.98-7.92) nursing homes for the total number of drugs prescribed or between the intervention (mean 4.62, 95% CI 4.09-5.15) and control (mean 5.23, 95% CI 4.92-5.54) nursing homes for the total number of drugs administered per person with clustering. Cumulative survival in the intervention group was better than controls with HR of 0.85 (95% CI 0.76-0.95 unclustered; 95% CI 0.69-1.05 adjusted for clustering). No significant changes were observed in annual mortality rates, frequency of hospitalizations, number of residents with adverse events, or changes in the RCI disability index between intervention and control groups.

9. King and Roberts et al (Australia e2001)48

10. Roberts et al (Australia e2001)46

Intervention type: Pharmacist-led medication review Review type: Prescription review

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

Authors (CountryeYear)

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

excluded. Of the remaining 42 articles, 20 were excluded as they measured educational or behavioral interventions, or were conducted in other settings. The search yielded 11 transitional studies that were excluded because they involved residents transitioning between aged care facilities and hospitals, academic health centers, ambulatory care practices, and other community-based settings. The search also resulted in 2 studies of the same population (Claesson and Schmidt34 and Schmidt et al38); hence, one was excluded (Schmidt et al38). Two studies of the same national-level intervention were included because the study populations were different (Lapane et al39 and Patterson et al40). A total of 22 studies were ultimately included in the review, 10 observational studies and 12 controlled trials (Figure 1). The 22 studies evaluated the effectiveness of medication review interventions to improve the medication use among residents of aged care facilities. The controlled trials (n ¼ 12) included 10 randomized and 2 nonrandomized trials (Table 1). Of the 12 controlled trials, 3 were conducted in Australia, 4 in the United Kingdom, and 1 each in Switzerland, Belgium, North Ireland, Sweden, and the United States. The observational studies (n ¼ 10) used the following study designs: 5 prospective and 1 retrospective pre and post interventions, and 4 longitudinal studies (Table 2). Of the 10 studies, 3 were conducted in the United States, 2 in Netherlands, and 1 each in Spain, Switzerland, Belgium, Singapore, and the United Kingdom. Risk of Bias Within Studies In the case of controlled trials, 6 studies used randomized methods to generate the sequence, and 7 studies used some form of data checking. However, only 2 studies clearly described a form of concealed allocation, and none of the studies definitely or probably used blinding. Hence none of the studies satisfied all 6 key criteria together (Higgins et al36). Regarding the observational studies, quality assessment was considered against the SIGN Methodology Checklist, and the majority (6/10) met at least 50% of the criteria. Quality issues often lacking were blinding of the outcome assessment, identification of potential confounders, assessment of the subjects likelihood of the outcome upon enrolment, and validity and reliability of the outcome assessment tools. Study Outcomes

SMMSE, standardized mini-mental state examination.

Proportion of residents receiving any particular drug Number of drugs classified by guideline as inappropriate before and after the intervention List of each resident’s prescriptions were collected 1 mo before and after the 12-mo intervention. The aim was to comment on the impact of regular multidisciplinary team interventions on the quantity of inappropriate medications RCT; 1854 residents from 33 nursing homes. Female: 70% Mean age: 83 Period: 12 mo 12. Claesson and Schmidt (Sweden e1998)34

d

Intervention type: Multidisciplinary Team medication review. Review type: Prescription review

Mean difference in MMSE scores at the end of 8 mo was 1.6 (0.1 to 3.3). No statistical difference for mean GDS scores, 0.75 (2.03 to 0.52) and mean BASDEC scores, 0.14 (1.45 to 1.09). The CRBRS mean score became significant at the end of 8 mo, 2.2 (4.1 to 0.3). The mean number of drugs prescribed to all residents at beginning: 4.91 (range: 0-17). In 54% of cases, the prescribing of neuroleptics was inappropriate. Reduction in the homes where medication was reviewed was greater than in control homes (P ¼ .070) Of 261 recommendations, 239 (91.6%) were accepted by the GP and resulted in 144 actual changes. The mean number of recommendations per resident: 2.46 (range: 0-7). Residents were prescribed an average of 7.7 (range: 6-11) medications. Most frequently prescribed drugs: laxatives (70%), psychotropic drugs (77%, range: 50%-95%), and cardiovascular drugs (63%). Administration of Mini Mental State Examination (MMSE), Geriatric Depression Scale (GDS), Brief Assessment Schedule Depression Cards (BASDEC), and Crichton-Royal Behaviour Rating Scale (CRBRS) at 0, 4, and 8 mo. Types and numbers of drugs each resident was taking, the reason for the use of any neuroleptic drugs. No of accidents and deaths. A medication review by the study pharmacist for all consenting residents in the intervention homes. The pharmacist compiled lists of the current medications together with a brief medical history. Three weeks on the review, the homes were revisited to check if the changes in medications have been implemented and if there were any problems. The initial assessments were made by the study pharmacist and psychiatrist. The second and third assessments were carried out by 6 Registered Mental Nurses trained in the administration of the instruments to ensure consistency and reliability. Intervention type: Pharmacist-led medication review. Review type: Prescription review The homes were matched into 7-pair equivalent in number of beds, registration status, and resident mix. RCT; 330 residents in 14 nursing homes, followed for 8- to 4mo observation phase and 4-mo intervention phase Female: 73% Mean age: Intervention: 83.5 Control: 78.9 11. Furniss et al (UKe2000)47

1.e7

The outcome measures were categorized into 3 main groups, namely medication-related, review or intervention-related, and adverse outcomes. The medication reviews were either performed by pharmacists (pharmacist-led or with physician consultation) or in a multidisciplinary-team approach involving pharmacists. Of the 12 controlled trials, 8 studies reported findings of pharmacist-led medication reviews and 4 reported findings of multidisciplinary team-based reviews. The medication reviews performed in the included controlled trials were prescription reviews (performed in 8 studies) and clinical medication reviews (performed in 4 studies). In the case of observational studies, the majority of the studies (8 of 12 studies) reported findings of pharmacist-led medication reviews, and only 2 studies reported findings of multidisciplinary teambased reviews. Similarly, 6 studies employed prescription reviews, whereas 4 studies employed clinical medication reviews. Overall, the outcomes were diverse with differing definitions, methods of data collection, varying time points, and different reporting methods. Outcomes in controlled trials Almost all studies (11 of 12 studies) reported some form of medication-related outcome. The outcome measures were potentially inappropriate prescribing (Houghton et al41), potentially inappropriate

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

IdenƟficaƟon

1.e8

Records idenƟfied through PubMed searching (n = 1060)

AddiƟonal records idenƟfied through CINAHL and IPA (n = 101)

Screening

Records aŌer duplicates removed (n =1141)

1141 screened by Ɵtles

912 excluded

Eligibility

229 screened by abstracts 187 excluded 153 irrelevant arƟcles 34 review arƟcles 42 full-text arƟcles assessed for eligibility

Included

20 excluded Studies with the same dataset Studies without pharmacist involvement Studies conducted in other care seƫngs TransiƟonal studies Studies included in qualitaƟve synthesis (n = 22) 12 = Controlled Trials 10 = ObservaƟonal studies Fig. 1. Study selection process (PRISMA flowchart).

medications (Milos et al42 and Lapane et al39), medication-related problems (Milos et al42), medication appropriateness (Verrue et al43 and Crotty et al44), inappropriate psychoactive medication use (Patterson et al40), mean number of medications per patient (Alldred et al45), number of medication changes (Zermansky et al33), mean number of repeat medications per patient (Zermansky et al33), and total number of medications prescribed or administered (Claesson and Schmidt,34 Roberts et al,46 and Furniss et al47). Of the 12 studies, only 4 reported medication reviewerelated outcomes. The outcomes investigated were number of recommendations (Milos et al,42 Alldred et al,45 and King and Roberts48), type of recommendations (Milos et al42 and Alldred et al45), action taken by the physician against recommendations (Milos et al,42 Alldred et al,45 and King and Roberts48), and acceptance of recommendations (Alldred et al,45 Furniss et al,47 and King and Roberts48). Six studies reported adverse outcomes, which included rate of falls (Patterson et al40 and Zermansky et al33), number of deaths/mortality (Zermansky et al,33 Roberts et al,46 and King and Roberts48), number of

hospitalizations (Lapane et al,39 Zermansky et al,33 Roberts et al,46 and Furniss et al47), disability assessment (Roberts et al46), and mental status (Furniss et al47). Outcomes in observational studies The majority of the observational studies (9 of 10 studies) reported medication-related outcomes. The outcome measures were pharmacotherapy-related or medication-related problems (Chia et al,49 Mestres et al,50 Stuijt et al,51 Brulhart and Wermeille,52 and Finkers et al53), medication appropriateness (Mestres et al,50 Stuijt et al,51 and Briesacher et al54), number of medications prescribed (Mestres et al,50 Brulhart and Wermeille,52 Finkers et al,53 and Baqir et al55), and change in medications (Dilles et al56). Eight studies reported medication reviewerelated outcomes that included acceptance of the recommendations (Chia et al,49 Mestres et al,50 Stuijt et al,51 and Christensen et al57), type of recommendations (Mestres et al,50 Finkers et al,53 and Baqir et al55), and reasons for acceptance (Baqir et al55).

Table 2 Summary of Observational Studies (Including Longitudinal and Pre and Post Intervention Studies) Authors (Country eYear)

Design and Setting

Pre-post intervention studies 1. Chia et al Observational (retrospective (Singapore pre and post intervention 49 e2015) study); 480 residents from 3 nursing homes, over 6 mo. Female: 48.7%

Intervention

Description of Intervention

Outcome Measures

Impact of Interventions

Nursing homes serviced by NHG Pharmacy

Intervention type: Pharmacist-led medication review Review type: Clinical medication review

A 1-mo pre-setup period (during which all the enrolled patients were reviewed; a 1time pharmacist review was conducted before the weekly pharmacist visits) followed by a 6-mo post-setup period (during which all the patients were reviewed again). Pharmacotherapy problems were classified according to a clinical pharmacist recommendation. Medication reviews by pharmacists using an algorithm developed according to criteria of efficacy, safety, efficiency, and standard in geriatrics Drug-related problems (DRPs) detected by the pharmacists in the pharmacologic profiles were communicated using encrypted email to physicians. Answers to pharmacists about the acceptance were obtained through email, telephone, or through meetings. The time the pharmacists devoted to the review ranged between 15 and 40 min. The intervention comprises 3 main components: A detailed review of medicines by clinical pharmacists, a multidisciplinary team discussion and resident (and/ or family) involvement. The multidisciplinary team comprised a clinical pharmacist, care home nurse, GP, psychiatry of old age consultant, and psychiatry of old age service staff (case-tocase basis) The care home staff monitored the residents for adverse events following the intervention and this was followed up by the pharmacist for safe management.

Analysis of pharmacotherapy problems highlighted by pharmacists Rate of acceptance of the recommendations by the medical team Possible risks associated with polypharmacy and the inappropriate use of medication

392 pharmacotherapy problems, of which 236 (60.2%) were accepted. Each resident had 0.429 pharmacotherapy problems during pre-setup and 0.388 problems in post-setup. There were 375 potential risks found: 177 from pre- and 198 from postsetup periods. Most common risks were falls (16%) and constipation (13.1%).

Medication appropriateness index (MAI) DRPs

Of 424 pharmacists’ interventions at prescription level, mean degree of acceptance by physicians was 80%. DRPs totally solved in 73% of the interventions, partially solved in 3%, no possibility to solve problem in 16%, problem not solved due to lack of cooperation of prescriber in 6%. Before the interventions: mean number of drugs prescribed was 10.5 (3.5); after the interventions, dropped to 9.6 (3.5). Recommendations of stopping a drug: 106/132 accepted. Recommendations of starting a drug, 34/60 accepted. Mean summed MAI score before intervention: 5.36; after the interventions: 1.51 for 127 patients.

The most common interventions The number of medicines stopped Reasons for stopping medicines

1346 interventions were made in 91% of residents with 15 types on interventions. The most common intervention (52.3%) was to stop medicines; 704 medicines stopped in 298 residents (70.6%) 1.7 medicines were stopped for every resident reviewed, with a 17.4% reduction in medicines prescribed (3602 medicines prescribed before and 2975 after review). Main reasons for stopping medicines were no current indication (401 medicines, 57%), resident not wanting medicine after risks and benefits were explained (120 medicines; 17%), and safety concerns (42 medicines; 6%).

2. Mestres et al (Spaine2015)50

Observational (prospective pre and post intervention study); 3354 patients from 70 nursing homes in Barcelona, August 2013 eDecember 2013). Female: 61% Mean age: 83.6

Included patients were classified as Complex Chronic Patient (PCCdpatients with multimorbidity or 1 disease that implies difficult clinical management) in the Health Plan, with the defined criteria.

Intervention type: Pharmacist-led medication review Review type: Prescription review

3. Baqir et al (UKe2014)55

Observational (prospective pre and post intervention study); 422 residents from 20 care homes in North Tyneside, 12 mo

d

Intervention type: Multidisciplinary medication review of a pharmacist, doctor, and care home nurse Review type: Clinical medication review

1.e9

(continued on next page)

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

Inclusion Criteria

Authors (Country eYear)

Inclusion Criteria

Intervention

Description of Intervention

Outcome Measures

Impact of Interventions

4. Dilles et al (Belgium e2013)56

Observational (prospective pre and post intervention study); 418 residents from 8 nursing homes % Female: 76.6 Mean age: 84.9 Period: 12 mo

Included residents needed to reside at least 1 mo in the nursing home and to take 4 or more different drugs.

Interdisciplinary medication review prepared by nurse observations of potential ADRs using personalized screening lists generated by the Pharmanurse software. It was validated by an expert team, a clinical pharmacologist, pharmacist, GP, geriatrician, and 3 nurses.

Number of ADRs detected. ADRs confirmed by GPs Medication changes

Nurses observed 1527 ADRs in 81% of the residents Physicians confirmed 821 ADRs in 60% of the residents; corresponds to a confirmation rate of 53.8% of the nurse observations. Alterations were planned for 13.2% of the confirmed ADRs in 21% of the residents. Medication was changed in 275 (67%) of 409 residents.

5. Stuijt et al (Netherlands e2008)51

Observational (prospective pre and post intervention study); 1 residential nursing home with 54 residents, April 2003 to April 2004. Mean age: 85.8 Female: 90%

Included residents who were receiving ongoing medical care from 2 GPs and 1 dispensing community pharmacist.

Intervention type: Multidisciplinary medication review (Mainly physicians and nurses, supported by pharmacists in some sessions) Review type: Clinical medication review Intervention type: Pharmacist-led medication review. Review type: Clinical medication review

Compilation of the patients’ medication profile Identification of drug-related problems Discussion of recommendations by the health care team and implementation by the GP, if deemed necessary.

Acceptance rate of the recommendations made by the pharmacist.

6. Christensen (USAe2004)57

Observational (prospective pre and post intervention study); 253 nursing homes served by a consultant pharmacy organization. Female: 75% Mean age: 76.8 Medication review period: October 2002e March 2003.

Residents who had 18 prescription fills in the previous 90-d period were enrolled.

Intervention type: Pharmacist-led medication review plus consultations with prescribing physicians Review type: Prescription review

The intervention consisted of a systematic pharmacist drug regimen review and consultation with prescribing physicians, supplemented with the existing requisite Omnibus Budget Reconciliation Act of 1987 consultation programs in nursing homes.

Frequency of recommendations by pharmacists in response to targeted profile alerts aimed at high-risk patients Changes in frequency and type of drug therapy. The impact on drug-related quality and costs.

Mean summed MAI score before intervention 23.7 [CI 17.0, 30.3 (078)] and after intervention 16.0 [CI 9.48, 22.6 (0-81)]. Mean per medication MAI score before intervention 3.79 [CI 2.89, 4.68 (018)] and after intervention 2.43 [CI 1.75, 3.11 (0-18)]. 115 DRPs were identified, and 78 recommendations (67.8%) were accepted. A total of 6520 flagged drug therapy recommendations for 4136 patients (mean: 1.58 per patient). Physicians concurred with 58% of the recommendations to change drug therapy. The mean drug costs were $2.49 and $65.04 per patient with those alerts, respectively.

Observational (longitudinaldconcurrent cohort analysis); residents in long-term care facility (LTCF) who received methotrexate from January 1, 2007 to December 31, 2009 Female: N/A Mean age: N/A

Included if residents in LTCF who received methotrexate.

Intervention type: Pharmacist-led quality improvement process to improve methotrexate (MTX) safety in nursing home residents. Review type: Prescription review

A 3-pronged approach involving modification to dispensing systems and practices, mandatory staff training, and measurement was implemented in June 2008 and monitored through December 2009. Software programming forcing a mandatory second clinical review of all MTX orders before dispensing.

Number of prescriptions filled, number of patients treated, number of sentinel events, number of safety variances identified

Observational (longitudinal study); 251 residents in 10 nursing homes (January 2007eDecember 2009). Female: 76% Mean age: 83

Included randomly selected palliative care and terminally ill patients.

Intervention type: Pharmacist-led medication review. Review type: Prescription review

2 pharmacists, 11 physicians, and 23 nurses reviewed in the form of multidisciplinary monthly meetings of 1 h until reached an annual average of 20% of patient medication records per nursing home.

Medication review, acceptance by physicians, DRPs, interventions and treatment changes implemented. Decrease of medicines per patient before and after team meeting, and

Longitudinal studies 7. Zarowitz et al (USAe2012)58

8. Brulhart and Wermeille (Switzerland e2011)52

Number of patients treated (10,399 pre intervention, 10,762 post intervention) No of sentinel events (3 in 2007, 4 in 2008, and 0 in 2009). Of 43,652 prescriptions dispensed in 2008, 497 variances (1.1%); of 43,890 prescriptions dispensed in 2009, 693 variances (1.6%) noteddcould have resulted in sentinel events if not detected and corrected. A total of 1225 DRPs was detected by the pharmacist, with a mean of 3.7 per patient. 35% DRPs related to dose and duration of treatment, 32% to drug-drug interactions, and 26% to drug choice in 26%. The DRPs detected led to 1228 interventions. A total of 93% of the

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

Design and Setting

1.e10

Table 2 (continued )

number of medication changes implemented.

Observational (longitudinal study); 742 beds from 5 nursing homes in the northern Netherlands, reviews between October 2005 and May 2006. Female: 68% Mean age: 80

Included if using >9 medications. Excluded if without active drug, antiseptic preparations, phytopharmaceuticals, and food supplements.

Intervention type: Pharmacist and physician-led medication review Review type: Prescription review

10. Briesacher et al (USA e2005)54

Observational (quasiexperimental, longitudinal); MCB Survey. Period: 1997-2000. Female: 68% Mean age: 1997 Nursing home (NH): 80.3; assisted living facility (ALF): 73.6 2000 NH: 80.5; ALF: 74.6 NH: 8 million ALF: 2 million

Persons excluded if they lived in both NH and ALF during the year as the medication use is incomplete for transitional people.

Intervention type: Surveyor and consultant pharmacisteled drug use review for nursing home certification. Review type: Prescription review

A medication review by 1 pharmacist and nursing home physician. The pharmacist identified all polypharmacy patients and sent the list of patients to the physician. At initial meeting, the appropriateness of each prescribed drug was discussed and prepared a plan for optimizing the medication profile. Six weeks later, pharmacist and physician met again to identify any alteration in the medication profile. In total, 3 hospital pharmacists and 12 nursing home physicians participated in the intervention. State surveyors þ consultant pharmacists evaluate the medical records of NH and ALF residents’ exposure to Beers criteria drugs and determine whether adverse reactions have resulted, and whether the medical notes justify exempted use. Note: The Centers for Medicare and Medicaid Services (CMS) expanded the drug review protocol to NH to include a list of inappropriate medications (CMS criteria and 1997 Beers list).

The number and nature of DRPs of nursing home patients receiving >9 drugs.

Prevalence of PIM and related adverse drug reactions

Exposure for “medication to always avoid,” by institutional residence use in NH (before CPE: 3.7%; after CPE: 3.4%), in ALF (before CPE: 3.0%; after CPE: 2.8%). Exposure for “medications sometimes appropriate” by institutional residence use in NH (before CPE: 17.4%; after CPE: 17.6%), in ALF (before CPE: 10.3%; after CPE: 12.7%). Exposure for “inappropriate medication use restricted with specific diagnosis, by institutional residence” in NH (before CPE: 8.1%; after CPE: 14.7%), in ALF (before CPE: 3.0%; after CPE: 7.2%). In 2000, 16.4% Medicare beneficiaries in NH (95% CI: 13.9%-18.9%) and 16.6% in ALF (95% CI: 12.3%-20.9%) received new prescriptions for PIM. Prevalence exposure for the same period were 25.6% of Medicare beneficiaries in NH (95% CI: 22.6%-28.5%) and 19.0% of those in ALFs (95% CI: 14.5%-23.5%).

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

9. Finkers et al (Netherlands e2007)53

interventions led to a rapid change in medication within a few days. Mean number of drugs per patient for continuous and on-demand use decreased from 12.8 (range 2-27) to 11.8 (range 1-27), respectively. At initial meeting, 323 DRPs in 91 polypharmacy patients (3.5 per patient). About 96% of the polypharmacy patients had 1 or more DRP. Of 323 DRPs, 304 were directly coupled to a prescribed drug and 19 problems were classified as “need for additional drug” or “need for review.” At 6 mo, of 323 DRPs, 80 drugs were stopped, 79 were altered, and 164 DRPs remained unchanged (eg, patient’s discharge or death). Between initial meetings and followups, 27 new drugs were started, whereas 103 drugs were stopped. A significant decrease of 0.7 drugs per patient post intervention.

1.e11

1.e12

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

Three studies reported adverse outcomes that included risk of falls (Chia et al49), adverse drug reactions (ADR; Dilles et al56), and sentinel events (Zarowitz et al58). Study Interventions All interventions were delivered to older residents in aged care facilities. Because of the heterogeneity of the interventions and outcomes, it was deemed inappropriate to conduct a meta-analysis, and therefore the effectiveness of the interventions is described below. Interventions in controlled trials Houghton et al41 demonstrated significant difference in inappropriate prescribing between the control and intervention groups. Milos et al42 tested pharmacist-led medication reviews in the intervention group and found that the proportion of patients with at least 1 PIM decreased in the intervention group. Verrue et al43 and Crotty et al44 identified significant improvement in medication appropriateness in the intervention group. Patterson et al40 ascertained pharmacist-led medication review interventions as effective in reducing inappropriate use of psychoactive medications (intervention: 19.5% vs control: 50%). Zermansky et al33 investigated the number of changes in medication per patient and number of repeat medications per patient, and identified that the number of medication changes were higher in the intervention group compared with the control group (3.1 vs 2.4). The mean number of medications per patient was also lower in the intervention group (6.7) compared with the control group (8.9). Lapane et al39 found a decline in PIM rates, in favor of the intervention homes. However, regarding the total number of medications prescribed or administered, Roberts et al46 found no evidence of improvement or difference between intervention and control groups. Milos et al42 reported the withdrawal of medication therapy (30%) and reduced dosage (28%) as the 2 most common recommendations; however, the acceptance rate for reduced dosage was significantly higher than that of withdrawal of medication therapy. Alldred et al45 reported that about 76% of pharmacist recommendations were accepted by the general practitioner. The 2 most common recommendations reported were technical (eg, generic switching) and tests to monitor the medication therapy (22%). Furniss et al47 identified the mean number of recommendations per resident as 2.46, and of these recommendations 92% were accepted by the general practitioner. Crotty et al44 reported a 38% acceptance rate of pharmacist recommendations, all of which were implemented. Patterson et al40 found no difference at 12 months in the rate of falls between the intervention and control groups. However, Zermansky et al33 detected a significant difference in the rate of falls between the intervention and control groups [0.8 vs 1.3; relative risk (RR) 0.59, 95% confidence interval (CI): 0.49-0.70]. They also reported a lower number of deaths (51 vs 48; RR 0.89, 95% CI: 0.56-1.41) and hospitalizations per patient (0.20 vs 0.26; RR 0.75, 95% CI: 0.52-1.07) in the intervention group as opposed to the control group. Roberts et al46 found evidence of better cumulative survival in the intervention group compared to the control group (hazard ratio [HR] 0.85 95% CI: 0.76  0.95). However, they found no significant changes in annual mortality rates, frequency of hospitalizations, number of residents with adverse events, and changes in disability index between the intervention and control nursing home groups. Interventions in observational studies Chia et al49 found that medication review interventions were effective in reducing drug-related problems (DRPs), from 0.429 preintervention to 0.388 postintervention. Mestres et al50 reported a reduction in the number of medications prescribed, from 10.5 [standard deviation (SD): 3.5] to 9.6 (SD: 3.5) after medication review interventions. They also found evidence of improvement in the

medication appropriateness score after the intervention (difference 3.85). Brulhart and Wermeille52 identified a mean DRP of 3.7 per patient, where the majority of the DRPs were related to dose and duration of the treatment. The mean number of medications per patient decreased from 12.8 to 11.8. Briesacher et al54 found evidence for improvement in medication appropriateness on interventions. Mestres et al50 identified an 80% mean acceptance of recommendations, 73% of which were implemented. The 2 most common recommendations were ceasing a medication and starting a new medication. Baqir et al55 recorded 15 different types of recommendations, whereby the most common recommendation was to cease medications (1.7 medicines stopped for every resident reviewed). Chia et al49 discovered an increase in the potential risks, from 177 preintervention to 198 postintervention. The risk of falls (16%) was the most common potential risk identified in their study. Dilles et al56 reviewed residents for adverse drug reactions risk and found 1527 adverse drug reactions in 81% of the residents. Zarowitz et al58 reviewed the use of methotrexate among nursing home residents and found a reduction in the number of sentinel events, from 3 in 2007 to zero in 2009. Discussion Main Findings This systematic review has identified a large and rapidly growing body of evidence investigating the effectiveness of medication review interventions in aged care facilities. Medication review interventions may slightly reduce the number of prescribed medications and inappropriate medications, may improve medication appropriateness, and would appear to have positive effects on reducing the number of deaths, frequency of hospitalizations, and disability index. Medication review promotes appropriate polypharmacy, decreases the use of inappropriate medications, and aids in identifying potential and actual medication-related adverse events; it also offers an opportunity to encourage medication adherence. Medication reviews may also assist to identify and resolve challenging medication issues for nursing home staff and caregivers; for instance, simplifying medication regimes and reducing the number of doses.59 Pharmacists have been in the forefront of many of these developments.20 Indeed an exemplary approach to a pharmacist’s involvement in optimizing pharmacotherapy in aged care facilities should be one that is patient-centered. Patient-centered care is defined as an approach to health care that is responsive to, and respectful of, the preferences and values of the patients.60 Patientcentered care encourages collaboration with other health care professionals and families of the patient to promote quality and safety of health care.60 Medication review is an essential component of patientcentered care. Impact on Patient Safety Patient-centeredness is an essential concept for achieving patient safety. There is increasing discussion about implementing patientcentered, systems-based health care to improve safety in health care. Older people are particularly vulnerable and are disproportionately affected by safety incidents and adverse events, which could result in trauma or even death.60,61 Adverse drug events are frequently caused by medication errors and adverse drug reactions. Residents in aged care facilities are susceptible to complications arising from their pharmacokinetic and pharmacodynamic characteristics; normal doses may be suboptimal for them.62 Although not every resident in a nursing home is frail, the frail elderly residing in nursing homes consume 4 or more medications compared with their nonfrail counterparts living in the community,63 and have a higher possibility of

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

experiencing adverse drug events (ADEs). More than 4.3 million visits to health care facilities were attributed to an ADE,64 and rates of ADEs are double among residents taking 9 or more medications as opposed to those taking fewer.65 Impact on Cost The conundrum of polypharmacy and a lack of pharmacotherapy interventions results in increased health care costs to both the individual and the health care system. A study in England that investigated 686 million NHS prescriptions dispensed with an estimated cost of £8 billion66 found that polypharmacy increased the risk of outpatient clinic visits and hospitalizations that led to a 30% increase in medical expenses.67 Increased health care cost may adversely affect a nation’s economic status. Pharmacist interventions in aged care facilities have been reported to confer financial benefits. The cost of prescribing extends beyond the cost of the medication supply, as cost is a factor when bodily biochemical repercussions occur on starting or ceasing therapy, or switching brands of medications even though these may not always cause adverse drug reactions.29 The American Society of Consultant Pharmacists explored an alternative model for clinical pharmacy services in long-term care facilities, the Fleetwood model. During the first phase, when medication-related problems were prevented by consultant pharmacist-conducted medication regimen reviews, the authors reported a saving of US$3.6 billion.68 Similarly, the North Carolina Long-Term Care Polypharmacy Initiative identified a medication cost saving of US$21.63 per resident per month, a reduction in medication alerts, and a reduction in the relative risk of hospitalizations among residents who received a retrospective medication review.69 Possible Weaknesses To minimize risk of biases, the current study involved an extensive literature search and screening of relevant studies. However, robust conclusions cannot be drawn because of significant heterogeneity in outcome measures, interventions, and publication bias. Overall, this review had several limitations that may affect its generalizability, including language bias (only English-language databases and journals were searched), selection bias (allocation concealment), and detection bias or performance bias (blinding related). Blinding of all study participants, personnel, and outcome assessors was not possible across all included studies because of the nature of the intervention itself; participants may have been unaware of whether reviews had been conducted, although health professionals conducting the medication reviews in intervention groups may have been aware. Moreover, reporting bias cannot be ruled out. Conclusions The current study indicates that evidence of medication reviews conducted by pharmacists, either working independently or with other health care professionals, in aged care facilities under certain circumstances may improve the quality of medication use among residents of aged care facilities. Although the evidence for the results from randomized controlled trials is of low quality, it is supported by findings from observational studies. Medication misadventure defies the maintenance of good health among residents in aged care facilities and should be curbed to ensure the safety of the residents. Optimizing pharmacotherapy using a patient-centered approach is crucial in this population.

1.e13

Supplementary Data Supplementary data related to this article can be found online at http://dx.doi.org/10.1016/j.jamda.2016.10.004. References 1. United Nations (Department of Economic and Social Affairs). World population ageing. Available at: http://www.un.org/esa/population/publications/ WPA2009/WPA2009-report.pdf; 2009. Accessed June 10, 2016. 2. Hajjar ER, Cafiero AC, Hanlon JT. Polypharmacy in elderly patients. Am J Geriatr Pharmacother 2007;5:345e351. 3. Stewart RB. Polypharmacy in the elderly: A fait accompli? Ann Pharmacother 1990;24:321e323. 4. Alldred DP, Kennedy Claire M, Hughes C, et al. Interventions to optimise prescribing for older people in care homes. Cochrane Database Syst Rev; 2016:CD009095. 5. Merriam-Webster. Definition of “secondary care.” Available at: http://www. merriam-webster.com/dictionary/secondaryþcare; 2016. Accessed September 30, 2016. 6. Planton J, Edlund BJ. Strategies for reducing polypharmacy in older adults. J Gerontol Nurs 2010;36:8e12. 7. Rochon PA, Gurwitz JH. Optimising drug treatment for elderly people: The prescribing cascade. Br J Med (Clini Res Ed) 1997;315:1096e1099. 8. Task Force on Medicines Partnership, The National Collaborative Medicines Management Services Programme. Room for review: A guide to medication review: The agenda for patients, practitioners and managers. Available at: http://myweb.tiscali.co.uk/bedpgme/CG/Room%20for%20Review%20-%20Medication %20review.pdf; 2002. Accessed June 13, 2016. 9. Clyne W, Blenkinsopp W, Seal R. A guide to medication review. Available at: http://www.cff.org.br/userfiles/52%20%20CLYNE%20W%20A%20guide%20to% 20medication%20review%202008.pdf; 2008. Accessed June 13, 2016. 10. Spinewine A, Fialová D, Byrne S. The role of the pharmacist in optimizing pharmacotherapy in older people. Drugs Aging 2012;29:495e510. 11. Australian Government Department of Health and Ageing. Medication management reviews. Available at: http://www.health.gov.au/internet/ main/publishing.nsf/Content/medication_management_reviews.htm; 2014. Accessed June 14, 2016. 12. American Pharmacists Association and National Association of Chain Drug Stores Foundation. Medication therapy management in pharmacy practice: Core elements of an MTM service model. Available at: http://www.pharmacist. com/sites/default/files/files/core_elements_of_an_mtm_practice.pdf; 2008. Accessed June 14, 2016. 13. Government of Ontario. MedsCheck: Public Information; Ministry of Health and Long-term Care. 2014. Available at: http://www.health.gov.on.ca/en/ public/programs/drugs/medscheck/. Accessed June 14, 2016. 14. Pharmacy Council of New Zealand. Medicines management: Definition, levels, competence framework. 2006. Available at: http://www.pharmacycouncil.org. nz/cms_show_download.php?id¼124. Accessed June 14, 2016. 15. Levenson SA, Saffel DA. The consultant pharmacist and the physician in the nursing home: Roles, relationships, and a recipe for success. J Am Med Dir Assoc 2007;8:55e64. 16. Furniss L, Craig SK, Burns A. Medication use in nursing homes for elderly people. Int J Geriatr Psychiatry 1998;13:433e439. 17. Nishtala PS, McLachlan AJ, Bell JS, Chen TF. Psychotropic prescribing in longterm care facilities: Impact of medication reviews and educational interventions. Am J Geriatr Psychiatry 2008;16:621e632. 18. Kaur S, Mitchell G, Vitetta L, Roberts MS. Interventions that can reduce inappropriate prescribing in the elderly: A systematic review. Drugs Aging 2009; 26:1013e1028. 19. Patterson SM, Hughes C, Kerse N, et al. Interventions to improve the appropriate use of polypharmacy for older people. Cochrane Database Syst Rev; 2012:CD008165. 20. Spinewine A, Schmader KE, Barber N, et al. Appropriate prescribing in elderly people: How well can it be measured and optimised? Lancet 2007;370: 173e184. 21. Tjia J, Velten SJ, Parsons C, et al. Studies to reduce unnecessary medication use in frail older adults: A systematic review. Drugs Aging 2013;30:285e307. 22. Forsetlund L, Eike MC, Gjerberg E, Vist GE. Effect of interventions to reduce potentially inappropriate use of drugs in nursing homes: A systematic review of randomised controlled trials. BMC Geriatr 2011;11:16. 23. Kröger E, Wilchesky M, Marcotte M, et al. Medication use among nursing home residents with severe dementia: Identifying categories of appropriateness and elements of a successful intervention. J Am Med Dir Assoc 2015;16:629. e1e629.e17. 24. Loganathan M, Singh S, Franklin BD, et al. Interventions to optimise prescribing in care homes: Systematic review. Age Ageing 2011;40:150e162. 25. Marcum ZA, Handler SM, Wright R, Hanlon JT. Interventions to improve suboptimal prescribing in nursing homes: A narrative review. Am J Geriatr Pharmacother 2010;8:183e200. 26. Cooper JA, Cadogan CA, Patterson SM, et al. Interventions to improve the appropriate use of polypharmacy in older people: A Cochrane systematic review. BMJ Open 2015;5:e009235.

1.e14

K. Thiruchelvam et al. / JAMDA xxx (2016) 1.e1e1.e14

27. Geurts MM, Talsma J, Brouwers JR, de Gier JJ. Medication review and reconciliation with cooperation between pharmacist and general practitioner and the benefit for the patient: A systematic review. Br J Clin Pharmacol 2012;74: 16e33. 28. Holland R, Desborough J, Goodyer L, et al. 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 2007;65: 303e316. 29. Zermansky AG, Silcock J. Is medication review by primary-care pharmacists for older people cost effective? A narrative review of the literature, focusing on costs and benefits. Pharmacoeconomics 2009;27:11e24. 30. Castelino RL, Bajorek BV, Chen TF. Targeting suboptimal prescribing in the elderly: A review of the impact of pharmacy services. Ann Pharmacother 2009; 43:1096e1106. 31. Verrue CLR, Petrovic M, Mehuy E, et al. Pharmacists’ interventions for optimization of medication use in nursing homes: A systematic review. Drugs Aging 2009;26:37e49. 32. Da Silva DT, Santos APAL, Aguiar PM, et al. Analysis of research quality regarding pharmaceutical intervention in elderly residents of long-term care facilities: A systematic review. J Am Geriatr Soc 2010;58:1404e1406. 33. Zermansky AG, Alldred DP, Petty DR, et al. Clinical medication review by a pharmacist of elderly people living in care homesdRandomised controlled trial. Age Ageing 2006;35:586e591. 34. Claesson CB, Schmidt IK. Drug use in Swedish nursing homes. Clin Drug Investig 1998;16:441e452. 35. Scottish Intercollegiate Guidelines Network. Methodology Checklist 3: Cohort studies. Scotland, Edinburgh: Scottish Intercollegiate Guidelines Network; 2012. 36. Higgins JPT, Altman DG, Sterne JAC. Chapter 8: Assessing risk of bias in included studies. Cochrane Statistical Methods Group and the Cochrane Bias Methods Group. Available at: http://handbook.cochrane.org/chapter_8/8_ assessing_risk_of_bias_in_included_studies.htm; 2011. Accessed May 12, 2016. 37. Effective Practice and Organisation of Care (EPOC). Norwegian Knowledge Centre for the Health Services. Suggested risk of bias criteria for EPOC reviews. Available at: https://epoc.cochrane.org/sites/epoc.cochrane.org/files/ uploads/Suggested%20risk%20of%20bias%20criteria%20for%20EPOC%20reviews. pdf; 2015. Accessed May 12, 2016. 38. Schmidt I, Claesson CB, Westerholm B, et al. The impact of regular multidisciplinary team interventions on psychotropic prescribing in Swedish nursing homes. J Am Geriatr Soc 1998;46:77e82. 39. Lapane KL, Hughes CM, Christian JB, et al. Evaluation of the Fleetwood model of long-term care pharmacy. J Am Med Dir Assoc 2011;12:355e363. 40. Patterson SM, Hughes CM, Crealey G, et al. An evaluation of an adapted U.S. model of pharmaceutical care to improve psychoactive prescribing for nursing home residents in Northern Ireland (Fleetwood northern Ireland study). J Am Geriatr Soc 2010;58:44e53. 41. Houghton J, Kirthisinga V, Desborough JA. Multi-professional medication reviews in care homes for older people: Results from the Care Med Randomised Controlled Trial. Age Ageing 2014;43:ii19eii20. 42. Milos V, Rekman E, Bondesson Å, et al. Improving the quality of pharmacotherapy in elderly primary care patients through medication reviews: A randomised controlled study. Drugs Aging 2013;30:235e246. 43. Verrue C, Mehuys E, Boussery K, Adriaens E, Remon JP, Petrovic M. A pharmacist-conducted medication review in nursing home residents: Impact on the appropriateness of prescribing. Acta Clinica Belgica 2012;67:423e429. 44. Crotty M, Halbert J, Rowett D, et al. An outreach geriatric medication advisory service in residential aged care: A randomised controlled trial of case conferencing. Age Ageing 2004;33:612e617. 45. Alldred DP, Zermansky AG, Petty DR, et al. Clinical medication review by a pharmacist of elderly people living in care homes: Pharmacist interventions. Int J Pharm Pract 2007;15:93e99. 46. Roberts MS, Stokes JA, King MA, et al. Outcomes of a randomized controlled trial of a clinical pharmacy intervention in 52 nursing homes. Br J Clin Pharmacol 2001;51:257e265. 47. Furniss L, Burns A, Craig SK, et al. Effects of a pharmacist’s medication review in nursing homes. Randomised controlled trial. Br J Psychiatry 2000;176: 563e567.

48. King MA, Roberts MS. Multidisciplinary case conference reviews: Improving outcomes for nursing home residents, carers and health professionals. Pharm World Sci 2001;23:41e45. 49. Chia HS, Ho JA, Lim BD. Pharmacist review and its impact on Singapore nursing homes. Singapore Med J 2015;56:493e501. 50. Mestres C, Hernandez M, Llagostera B, et al. Improvement of pharmacological treatments in nursing homes: Medication review by consultant pharmacists. Eur J Hosp Pharm 2015;22:207e211. 51. Stuijt CC, Franssen EJ, Egberts AC, Hudson SA. Appropriateness of prescribing among elderly patients in a Dutch residential home: Observational study of outcomes after a pharmacist-led medication review. Drugs Aging 2008;25: 947e954. 52. Brulhart MI, Wermeille JP. Multidisciplinary medication review: Evaluation of a pharmaceutical care model for nursing homes. Int J Clin Pharm 2011;33: 549e557. 53. Finkers F, Maring JG, Boersma F, Taxis K. A study of medication reviews to identify drug-related problems of polypharmacy patients in the Dutch nursing home setting. J Clin Pharm Ther 2007;32:469e476. 54. Briesacher BA, Limcangco MR, Simoni-Wastila L, et al. The quality of antipsychotic drug prescribing in nursing homes. Arch Intern Med 2005;165: 1280e1285. 55. Baqir W, Barrett S, Desai N, et al. A clinico-ethical framework for multidisciplinary review of medication in nursing homes. BMJ Qual Improv Rep, http:// doi.org/10.1136/bmjquality.u203261.w2538; 2014. 56. Dilles T, Vander Stichele RH, Van Bortel LM, Elseviers MM. The development and test of an intervention to improve ADR screening in nursing homes. J Am Med Dir Assoc 2013;14:379.e1e379.e6. 57. Christensen D, Trygstad T, Sullivan R, et al. A pharmacy management intervention for optimizing drug therapy for nursing home patients. Am J Geriatr Pharmacother 2004;2:248e256. 58. Zarowitz BJ, Erwin WG, Ferris M, et al. Methotrexate safety improvement in nursing home residents. J Am Med Dir Assoc 2012;13:69e74. 59. Parsons C, Alldred D, Daiell L, Hughe C. Prescribing for older people in nursing homes: Strategies to improve prescribing and medicines use in nursing homes. Int J Older People Nurs 2011;6:55e62. 60. Australian Commission on Safety and Quality in Healthcare. Patient-centred care: Improving quality and safety through partnerships with patients and consumers. Available at: http://www.safetyandquality.gov.au/wpcontent/ uploads/2012/03/PCC_Paper_August.pdf; 2011. Accessed June 14, 2016. 61. World Health Organization. Patients for Patient Safety e Statement of Case. Available at: http://www.who.int/patientsafety/patients_for_patient/ statement/en/; 2016. Accessed June 30, 2016. 62. Gurwitz JH, Sanchez-Cross MT, Eckler MA, Matulis J. The epidemiology of adverse and unexpected events in the long-term care setting. J Am Geriatr Soc 1994;42:33e38. 63. Walley T, Scott AK. Prescribing in the elderly. Postgrad Med J 1995;71: 466e471. 64. Bourgeois FT, Shannon MW, Valim C, Mandl KD. Adverse drug events in the outpatient setting: An 11-year national analysis. Pharmacoepidemiol Drug Saf 2010;19:901e910. 65. Nguyen JK, Fouts MM, Kotabe SE, Lo E. Polypharmacy as a risk factor for adverse drug reactions in geriatric nursing home residents. Am J Geriatr Pharmacother 2006;4:36e41. 66. Barber N, Elliott R, Morgan M, Horne R. Concordance, adherence and compliance in medicine taking (Report for the National Coordinating Centre for NHS Service delivery and Organisation R&D). Brighton, United Kingdom: NHS National Coordinating Centre for Service Delivery and Organisation; 2005. Available from: http://www.nets.nihr.ac.uk/__data/assets/pdf_file/0009/64494/ FR-08-1412-076.pdf. Accessed June 30, 2016. 67. Akazawa M, Imai H, Igarashi A, Tsutani K. Potentially inappropriate medication use in elderly Japanese patients. Am J Geriatr Pharmacother 2010; 8:146e160. 68. Bootman JL, Harrison DL, Cox E. The health care cost of drug-related morbidity and mortality in nursing facilities. Arch Intern Med 1997;157:2089e2096. 69. Trygstad TK, Christensen DB, Wegner SE, et al. Analysis of the North Carolina long-term care polypharmacy initiative: A multiple-cohort approach using propensity-score matching for both evaluation and targeting. Clin Ther 2009; 31:2018e2037.