Journal Assignment Week 11

AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 22 | NOVEMBER 15, 2016 1845 Medication reconciliation interventions in ambulatory care: A scoping review Lisa McCarthy, Pharm.D., M.Sc., Women’s College Research Institute, Toronto, Canada.

Xinru (Wendy) Su, B.Sc.Phm., Women’s College Hospital, Toronto, Canada.

Natalie Crown, Pharm.D., Women’s College Hospital, Toronto, Canada.

Jennifer Turple, B.Sc.Pharm., Institute for Safe Medication Practices Canada, Toronto, Canada.

Thomas E. R. Brown, Pharm.D., Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, Canada.

Kate Walsh, B.Sc.Pharm., Toronto Central Community Care Access Centre, Toronto, Canada.

Jessica John (Pharm.D. student), Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, Canada.

Paula Rochon, M.D., M.P.H., Women’s College Research Institute, Toronto, Canada.

DOI 10.2146/ajhp150916 Purpose. The published literature on medication reconciliation (MR) interventions, outcomes, and facilitators in ambulatory care settings is reviewed.

Methods. A scoping review was conducted to characterize ambulatory care–based MR research in terms of study design, elements of interven - tions, and outcomes examined. English-language articles on comparative studies of MR programs targeting adults in ambulatory care settings were identified using data sources including MEDLINE, PreMEDLINE, EMBASE, and International Pharmaceutical Abstracts. For each study, steps under - taken in the MR process were extracted. The Cochrane Effective Practice and Organisation of Care (EPOC) taxonomy was used to classify types of interventions; taxonomies for reported outcomes and factors facilitating implementation of MR initiatives were developed by the authors.

Results. From among 2062 publications screened, 15 were included in the review. In 13 studies, multiple data sources were used to compile a “best possible medication history” (BPMH); however, the BPMH was shared with external healthcare providers in only 4 studies and with patients in only 5 studies. Most reported MR interventions were classified into two EPOC domains: professional (predominantly educational outreach visits and pa - tient reminders) and organizational (predominantly provider-oriented inter - ventions). Process outcomes were reported in 12 studies, with correct per - formance of MR being the most commonly evaluated process outcome, and 9 studies identified factors that facilitated MR implementation.

Conclusion. Few studies have examined clinical outcomes of MR in am - bulatory care settings, with the majority of pertinent reports focusing in - stead on process outcomes. Facilitators of successful MR interventions have been identified at the patient, staff, and clinic setting levels. Am J Health-Syst Pharm. 2016; 73:1845-57 M edication reconciliation (MR) is a formal process in which healthcare providers work together with patients, families, and care pro - viders to ensure that accurate and comprehensive medication informa - tion is communicated consistently across transitions of care. 1 Many ad - verse drug events (ADEs) related to medication errors and discrepancies are preventable. 2-4 Therefore, strate - gies to reduce preventable discrepan - cies and errors, such as MR, can have a significant impact on patient safety and outcomes. Evidence shows that MR reduces the potential for medication discrep - ancies such as omissions, duplica - tions, and dosing errors. 5-8 A study in an acute care setting published in 2011 showed that MR at hospital admission reduced ADEs caused by errors in admission orders by 43%. 9 MR has also been shown to reduce the occurrence of ADEs in long-term care settings. 10-12 In addition, eco - nomic analyses have demonstrated that MR is a cost-effective strategy for preventing medication errors in hospitals. 5-8 PRACTICE RESEARCH REPORTS MEDICATION RECONCILIATION INTERVENTIONS PRACTICE RESEARCH REPORTS 1846 AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 22 | NOVEMBER 15, 2016 Recognizing the importance of MR, the World Health Organization has named it one of five top patient safety strategies. 13 The Institute for Healthcare Improvement’s 100,000 Lives Campaign has championed MR as a key component, 14 and accredita - tion bodies in North America have recognized MR as a critical compo - nent of patient safety initiatives. 15,16 In ambulatory care settings, pa - tients visit health professionals, re - ceive care, and return home the same day, so researching medication safety in ambulatory care poses unique chal - lenges relative to research in institu - tional settings. Ambulatory care en - vironments are generally fragmented (i.e., patients see multiple health pro - fessionals who are not using a cen - tral record for information sharing), visits with providers are periodic, and patients have increased responsibil - ity for safe medication use because they must obtain and manage their own medications. It is known that patients receiving ambulatory care have important risk factors for ADEs, but whether MR reduces ADEs in this setting is not known. In their study of ADEs in four primary care practices, Gandhi and colleagues 17 found an event rate of 27.4 per 100 patients for serious or significant ADEs; 3 of every 100 patients were considered to have a preventable ADE. Given the preva - lence of risk factors for ADEs in am - bulatory care and given that MR has reduced ADEs in other settings within health systems, it can be hypothesized that MR will also reduce ADEs in am - bulatory care. That hypothesis, coupled with current efforts to implement wid - escale MR, creates an imperative to understand the current state of the literature on MR in ambulatory care. This review synthesizes the available literature and identifies its limitations as well as opportunities for future research. Specifically, this scoping review aims to describe and categorize studies of MR interven - tions in ambulatory care in terms of study design, intervention elements, KEY POINTS • The clinical importance of medication reconciliation (MR) in ambulatory care settings is unknown.

• The majority of reported MR interventions in ambulatory care settings were aimed solely at healthcare providers, with very few published studies examin - ing clinical outcomes.

• Studies of MR featuring robust designs, measuring clinical outcomes, and including inter - ventions based on identified MR implementation facilitators at the patient, provider, and clinic setting levels are needed. outcomes examined, and implemen - tation facilitators and to identify gaps in research in order to inform future studies.

Methods The scoping review followed the framework originally proposed by Arksey and O’Malley 18 and enhanced by Levac and colleagues 19 while in - corporating relevant, generally ac - cepted systematic review techniques for search strategies, study selec - tion, data abstraction, and narrative synthesis. 20,21 Search strategy. The following databases were searched for English- language publications (from date of database inception to April 2014, with an update performed in Octo - ber 2015): MEDLINE, PreMEDLINE, CINAHL, EMBASE, International Phar - maceutical Abstracts, and Ovid EBM Reviews (includes ACP Journal Club, Cochrane Central Register of Con - trolled Trials, Cochrane Database of Systematic Reviews, and Database of Abstracts of Reviews of Effects). Two librarians with expertise in ambulatory care research helped to develop unique search strategies for each of the data - bases. A third librarian peer reviewed the search strategies. The search strat - egy for MEDLINE and PreMEDLINE can be found in the appendix. The reference lists of included published articles were manually searched for additional studies that may not have been previously identi - fied. When only an abstract was lo - cated, first and senior authors were contacted by e-mail to determine if a full-text publication was available. Gray literature sources, includ - ing the Agency for Healthcare Re - search and Quality website (www.

ahrq.gov/), the OpenGrey database (www.opengrey.eu/), and the Institute for Safe Medication Practices Canada “Cross-Country MedRec Check-Up” tool (www.ismp-canada.org/medrec/ map/), were searched in January 2015 (the search was updated in August 2015 after a criterion specifying com - parative study design was added to the search protocol). Study selection. The initial study inclusion criteria limited the search to publications on research focused on community-dwelling adults (18 years of age or older) receiving am - bulatory care in clinic-based set - tings and with MR reported as the intervention of interest. Articles on research in primary care and day sur - gery clinics were included; however, studies examining interventions in an emergency room setting or those exploring discharge from hospitals were excluded (the rationale was that at the time of protocol preparation, the literature on MR at transitions of care had recently been systematically reviewed 22). The search was not initially re - stricted with regard to the type of research outcome or study design.

Therefore, studies examining both clinical outcomes (e.g., numbers and types of potentially significant errors and ADEs) and process out - comes (e.g., discrepancies between patient-reported and clinic medica - tion lists, reasons for discrepancies, accuracy of clinic medication lists, PRACTICE RESEARCH REPORTS MEDICATION RECONCILIATION INTERVENTIONS AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 22 | NOVEMBER 15, 2016 1847 time spent on MR activities, MR costs, patient knowledge of medications) were initially included in the search results. Consistent with the approach suggested by Levac et al., 19 we later added a search criterion restricting the results to articles on compara - tive studies, as they were thought to be more informative of the actual im - pact of MR. Two reviewers independently as - sessed the eligibility of potentially rel - evant citations retrieved via the search strategy. The initial article screening was based on title and abstract only. If the title and abstract provided insuf - ficient information, full-text articles were retrieved and reviewed. Dis - agreements between reviewers were resolved by discussion; if consensus could not be reached, a third reviewer made the final decision. Data extraction. The fields in - cluded in the data extraction form were adapted from those recom - mended by the Centre for Reviews and Dissemination and the Cochrane Col - laboration. 20,21 The process suggested by the Cochrane Collaboration 21 was used for development of the data ex - traction form, and the form was regu - larly updated by the research team as suggested by Levac et al. 19 The data ex - traction form addressed study design, clinic setting, patient characteristics, components of MR undertaken, type of healthcare professional conduct - ing MR and obtaining a “best possible medication history” (BPMH), MR ini - tiative characteristics, MR-associated outcomes, and MR implementation facilitators. Two reviewers independently ex - tracted data. The completed data ex - traction forms were then reviewed and compared. Differences between re - viewers were highlighted and summa - rized in table format. The two review - ers met over three days to discuss and resolve discrepancies. A third investi - gator reviewed areas where agreement could not be reached and made final decisions. Data synthesis. The scoping re - view aimed to synthesize data on MR interventions in four ways: by MR components, by type of intervention, by outcomes reported, and by facilita - tors of implementation. MR components. While the terms medication reconciliation and medi - cation history-taking are sometimes used interchangeably, MR involves more than taking a medication his - tory (i.e., gathering and recording a patient’s medications). The first step is to obtain a BPMH, which involves interviewing patients about their medication use and comparing the patient-provided data with infor - mation from at least one alternative source (e.g., electronic drug informa - tion system, community pharmacy– provided medication list, medication vials). Next, the BPMH is compared with prescriber orders, unintended discrepancies are identified and re - solved, and changes are documented.

Lastly, the changes and reconciled medication list are shared with pa - tients, caregivers, and other healthcare providers. 1,22 Given the evolution of MR as a distinct activity widely known to health professionals over the last decade, it was hypothesized that the way in which MR was conducted may vary among studies; therefore, data on the components of the MR proc - ess undertaken, the source or sources of medication information used, and the occupation or profession of the in - dividuals taking medication histories and conducting MR were extracted. Types of interventions. The Cochrane Effective Practice and Organisation of Care (EPOC) taxonomy was used to further classify types of interventions into four major categories: (1) profes - sional interventions (e.g., distribu - tion of education materials, audit and feedback, reminders), (2) organiza - tional interventions (e.g., revision of professional roles, including patient involvement in organizational gov - ernance), (3) financial interventions (e.g., provider or patient incentives), and (4) regulatory interventions (de - fined as changes to healthcare deliv - ery or associated costs brought about by regulations or laws). Further exam - ples of interventions falling into each of these categories are available at the EPOC website. 23 The potential for cat - egory overlap is a noted challenge of applying the EPOC framework. Study outcomes. We were unable to identify an established framework to guide our collection of outcomes.

At the study’s outset, we planned to categorize them as process outcomes (e.g., medication discrepancies), clini - cal outcomes (e.g., health-related quality of life), or economic outcomes (e.g., costs associated with use of healthcare services). This framework was independently tested by two au - thors, with further subdivision of the process outcomes category based on feedback. Process outcomes were sub - divided according to accuracy (e.g., completeness, correctness), 24 engage - ment (patient and provider), and im - pact on patient medication practices (e.g., knowledge, adherence). Implementation facilitators. We were also unable to identify an exist - ing framework on which to base data extraction for MR implementation facilitators. Therefore, we stratified facilitators of MR initiatives into three levels: patient, staff, and clinic setting.

Patient-level interventions included those targeted at recipients of care, staff-level interventions included those directed at individual members of the healthcare team, and clinic setting–based interventions included those involving multiple providers or systems (e.g., workflow adjustments, use of communication tools).

Results Search results. The initial search yielded 3376 reports (Figure 1). After duplicate removal, 2062 potentially relevant articles remained; on the basis of abstract and title review, 158 were identified for full-text review. Af - ter reviewing full-text articles for eligi - bility, 88 citations remained. After the addition of a search criterion stipulat - ing the inclusion of comparative stud - ies only, the final review set consisted of 15 articles. 24-37,38 The gray literature search did not yield additional studies.

MEDICATION RECONCILIATION INTERVENTIONS PRACTICE RESEARCH REPORTS 1848 AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 22 | NOVEMBER 15, 2016 Characteristics of included stud - ies. Study design, study setting, and MR components. Table 1 summarizes the characteristics of the studies in - cluded in the review. Of the 15 includ - ed studies, 9 were before-and-after studies, 24,25,28-31,34,36,38 3 were cohort studies, 27,32,33 2 were randomized con - trolled trials (one publication reported only the protocol for a study complet - ed in April 2016; the study findings have not been published), 35,37 and 1 was a quality-improvement initiative with a quasi-experimental compo - nent. 26 Seven studies were conducted in primary care settings 25,29,31,32,35-37 , 3 Identification Eligibility Screening Records identified through database searching (n = 3376) Records after duplicates removed (n = 2062) Records screened (n = 2062) Full-text articles assessed for eligibility (n = 158) Studies included in qualitative synthesis (n = 88) Additional eligibility criterion (comparative studies only) imposed(n = 15) Records excluded (n = 1904) Full-text articles excluded, with reasons (n = 70) Not pertaining to ambulatory care (n = 18) Not pertaining to MR (n = 32) Other (n = 20) Figure 1. Flow diagram of article selection. MR = medication reconciliation. in internal medicine clinics, 24,28,30 and 3 in specialty ambulatory care clin - ics (oncology, pain management, and geriatrics). 26,34,38 The remaining 2 stud - ies were conducted in a pharmacy call center and a medication therapy man - agement clinic, respectively. 27,33 In terms of MR components, medi - cation history-taking was reported to be part of the MR process in all 15 re - viewed studies; however, 3 reports did not explicitly mention identification and resolution of discrepancies. 31,34,38 Consistent with MR best practices, patients were provided a copy of the BPMH in 5 studies 28,29,31,33,36 ; in 4 stud - ies, MR documentation was shared with healthcare providers external to the participating organization. 26,32,33,35 Healthcare professionals, primarily nurses and pharmacists, took medi - cation histories in all but 2 of the studies; in those 2 studies, medical assistants performed medication history-taking. 29,37 In all reviewed studies, healthcare professionals were noted to perform the reconciliation step of the MR process. Multiple sources of medication information were used in all but 2 studies. 26,37 Types of interventions. Table 2 pro - vides a summary of the types of MR interventions reported. No included studies used financial or regula - tory interventions as outlined by the Cochrane EPOC taxonomy. 23 Among the reported professional interven - tions, the most commonly used in - cluded educational outreach visits (e.g., a visit in a provider practice set - ting from a trained individual seek - ing to gather information meant to change practice) 24-26,28,30,34,35 and audit and feedback. 24,25,28,30 Of the reported organizational interventions, provider- oriented interventions were the most commonly reported. 26,27,32,33,35,36 Reported outcomes. The framework created by our team for the purposes of this study is outlined in Table 3. Process outcomes were classi - fied into three categories: accuracy, engagement, and patient medica - tion practices. In terms of accuracy, outcomes related to the correctness of MR were reported in 8 of the 15 included studies, 24-26,28-30,37,38 with 3 studies discussing MR complete - ness. 24,28,37 Patient engagement was examined as an outcome in 3 stud - ies, 25,28,31 all of which measured the proportion of patients bringing medi - cation vials or an updated medica - tion list to clinic appointments, and 3 studies examined healthcare pro - fessional engagement by measuring compliance with MR policies. 34,35,37 Two studies reported patient medica - tion adherence as an outcome, 33,35 and 1 study assessed changes in patient medication knowledge. 35 Three of the PRACTICE RESEARCH REPORTS MEDICATION RECONCILIATION INTERVENTIONS AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 22 | NOVEMBER 15, 2016 1849 Table 1.

Characteristics of Studies Included in Scoping Review a Investigators and Year Published Study Design and Sample Sizes ( n)b Setting c or Care Focus Steps of MR Process Performed Responsible Person or Professional Medication Information Source(s) Medication History-Taking MR Discrepancies Resolved Communication/Documentation Medication History-Taking MR Nassaralla et al., 2009 28 Before-and-after (108/217 for evaluation of MR completeness; 61/121 for evaluation of MR correctness) Internal medicine Ye s Ye s EMR updated, list given to patient Nurse Physician Patient/caregiver interview, patient/caregiver medication list, medication vials Stock et al., 2009 29 Before-and-after (15–30 patients at each of 11 included clinics) Primary care/family medicine Ye s Ye s EMR updated, list given to patient Nurse, medical office assistant Physician Patient/caregiver interview, patient/caregiver medication list, medication vials Peyton et al., 2010 30 Before-and-after (90/90) Internal medicine Ye s Ye s Unclear Nurse Nurse Patient/caregiver interview, patient/caregiver medication list, medication vials Moczygemba et al., 2012 31 Before-and-after (379/379) Primary care/family medicine Ye s No EMR updated, list given to patient, patient encouraged to share updated medication list with all healthcare providers Pharmacist Pharmacist Patient/caregiver interview, medication vials Kilcup et al., 2013 32 Cohort study (251/243) Primary care/family medicine Ye s Ye s Direct communication with healthcare team, EMR updated, shared with primary care provider or, if appropriate, specialty provider Pharmacist Pharmacist Patient/caregiver interview, external physician’s record (discharge summary and, in some cases, hospital EMR) Continued from previous page Continued on next page PRACTICE RESEARCH REPORTS MEDICATION RECONCILIATION INTERVENTIONS AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 22 | NOVEMBER 15, 2016 1851 Table 1.

Characteristics of Studies Included in Scoping Review a Investigators and Year Published Study Design and Sample Sizes ( n)b Setting c or Care Focus Steps of MR Process Performed Responsible Person or Professional Medication Information Source(s) Medication History-Taking MR Discrepancies Resolved Communication/Documentation Medication History-Taking MR Moore et al., 2013 33 Cohort study (2250/2250) Medication therapy management program Ye s Ye s Direction communication with healthcare team, care plan faxed to primary care provider, patient encouraged to share care plan with healthcare providers Pharmacist Pharmacist Patient/caregiver interview, patient faxed lab data to pharmacy prior to appointment Neufeld et al., 2013 34 Before-and-after (1733 patient charts) Pain management Unclear Unclear EMR updated Physician Physician Patient/caregiver interview, patient/caregiver medication list Persell et al., 2013 35 RCT (study in progress) Primary care/family medicine Ye s Ye s Direct communication with healthcare team, documentation shared with healthcare practitioners outside clinic (nurse-identified medication problems conveyed to patient’s primary physician by e-mail/phone/page) Nurse Nurse Patient/caregiver interview, medication vials Roth et al., 2013 36 Before-and-after (64/64) Primary care/family medicine Ye s Ye s Direction communication with healthcare team, EMR updated, list for patient Pharmacist Pharmacist Patient/caregiver interview, medication vials Continued from previous page Continued on next page MEDICATION RECONCILIATION INTERVENTIONS PRACTICE RESEARCH REPORTS 1852 AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 22 | NOVEMBER 15, 2016 15 studies did not report any process outcomes. 27,32,36 Only 6 of the 15 studies exam - ined clinical outcomes, meaning the majority did not. 25,27,32,33,35,36 Use of healthcare services was the most commonly explored clinical out - come, reported in 4 studies. 27,32,33,36 Medication-related problems, the next most commonly explored clini - cal outcome, was evaluated in 2 stud - ies. 25,36 The protocol for the study of Persell and colleagues 35 was novel in that it called for patient follow-up to evaluate blood pressure change from baseline as a surrogate outcome and exploration of the impact of MR on health-related quality of life. Delate et al. 27 examined the impact of an MR initiative on mortality. Facilitators. The categories our team developed to guide data extrac - tion are illustrated in Figure 2. Facili - tators were classified as occurring at three levels: patient, staff, and clinic.

In total, nine studies discussed factors perceived to improve the success of MR initiatives. 24,25,27-29,31,34,36,37 At the patient level, engaging pa - tients in the MR process (e.g., asking them to bring their current medica - tion vials or most up-to-date medica - tion lists to clinic visits) was reported to facilitate MR initiatives and was a reported outcome in three stud - ies. 25,28,31 Helping patients understand the importance of MR and having the process conducted in a familiar setting also seemed to help with the process. For example, in the study by Delate et al., 27 the investigators thought that when patients received home-based counseling they were more likely to update information provided by pharmacists. At the staff level, providing the staff with individualized feedback was reported to be helpful in imple - menting successful MR initiatives. 28 Engaging all clinic personnel—from front desk assistants to healthcare providers such as nurses, pharma - cists, and physicians—was another reported facilitator. 29,31,36 A third re - ported staff-level facilitator was pro - Table 1.

Discussion Our scoping review of MR inter - ventions in ambulatory care identi - fied 15 relevant comparative studies.

The studies identified were mainly before-and-after studies and gener - ally had small sample sizes. In most of the studies, nurses, physicians, or pharmacists were responsible for medication history-taking and resolu - tion of discrepancies; however, stud - ies based in hospitals have suggested that other personnel, including phar - macy technicians, can also perform these tasks. 39,40 In relatively few stud - ies were BPMH documents shared with patients or external healthcare providers. Given the resource inten - sity of well-done MR, maximizing the sharing of the BPMH with everyone in a patient’s circle of care has several potential patient and health-system benefits. Few studies focused on clini - cal outcomes, as was expected by our team at study outset. With regard to the development of future studies of MR initiatives, a focus on more robust study designs, ideally with comparators and possi - bly more participants, is encouraged.

Additional research questions about the health professional best suited to conduct MR, strategies for sharing the BPMH across a healthcare system, and the impact of MR on clinical out - comes abound. Given the paucity of data about clinical outcomes, our scoping review cannot speak to the clinical impor - tance of MR in ambulatory care set - tings. However, its findings are impor - tant because they provide a synthesis of the types of MR initiatives that have been reported and factors that facilitate their implementation while providing insights that can guide future quality-improvement and re - search initiatives. To our knowledge, there is only one published systematic review that tackled a similar research question.

In 2008, Bayoumi and colleagues 41 conducted a systematic review of MR interventions in primary care. The eli - gibility criteria were designed to limit the review to randomized controlled trials or before-and-after studies of in - terventions for community-dwelling adults receiving MR in ambulatory care settings or at hospital discharge; only four studies met the inclusion criteria. Further, the results of the four studies were conflicting. The lim - ited body of literature in the review of Bayoumi et al. influenced our choice of the scoping review methodology. It was decided that such methodology allowed broader and more flexible in - clusion criteria, which we believe en - abled us to capture more potentially relevant studies from the literature. In our scoping review of studies evaluating MR interventions in am - bulatory care, we identified two types of interventions (using the Cochrane EPOC taxonomy): professional and organizational interventions. By defi - nition, professional interventions are interventions that target healthcare providers with the goal of changing practice behavior. 23 The most com - monly reported professional interven - tions included educational outreach visits, reminders, and audit and feed - back. Ivers et al. 42 conducted a sys - tematic review to assess the effects of audit and feedback on professional Figure 2. Schematic of three levels of implementation facilitators. MR = medication reconciliation.

Patient-Level Facilitators Engagement in MR process Understanding of MR importance Familiar setting for MR Staff-Level Facilitators Individualized feedback Engagement of all personnel Education about MR process Clinic-Level Facilitators Collaboration with outside providers Ease of integration into workflow Low cost Implementation of MR MEDICATION RECONCILIATION INTERVENTIONS PRACTICE RESEARCH REPORTS 1854 AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 22 | NOVEMBER 15, 2016 Table 2.

Medication Reconciliation Interventions Performed in Reviewed Studies Professional Interventions Organizational Interventions Investigators and Year Published Distribution of Educational Materials Local Consensus Process Audit and Feedback Educational Outreach Visits Patient-Mediated Interventions Provider Reminders Patient Reminders Mass Media Campaign Other Provider-Oriented Interventions Structural Interventions Workflow Interventions Nassaralla et al., 2007 24      Varkey et al., 2007 25     Weingart et al., 2007 26     Delate et al., 2008 27  Nassaralla et al., 2009 28     Stock et al., 2009 29     Peyton et al., 2010 30      Moczygemba et al., 2012 31   Kilcup et al., 2013 32  Moore et al., 2013 33  Neufeld et al., 2013 34   a Persell et al., 2013 35     Roth et al., 2013 36  Vouri and Marcum, 2013 38  Wolff et al., 2014 37     aPositive reinforcement.

PRACTICE RESEARCH REPORTS MEDICATION RECONCILIATION INTERVENTIONS AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 22 | NOVEMBER 15, 2016 1855 practices and patient outcomes and found that audit and feedback gener - ally led to small but potentially im - portant practice improvements; the effectiveness of such interventions seemed to depend on baseline per - formance and how the feedback was provided. Similarly, O’Brien et al. 43 conducted a systematic review to as - sess the effects of educational out - reach visits on healthcare practice or patient outcomes and concluded that educational outreach visits alone or in combination with other interven - tions have relatively small effects on prescribing behavior and that the ef - fects on other types of professional performance vary from small to mod - est. Therefore, given that educational outreach visits and audit and feedback Table 3. Framework for Categorizing Medication Reconciliation Process and Clinical Outcomes a Outcome Category Subcategory Definition Process Outcomes MR accuracy Correctness Outcomes concerned with numbers of, types of, or reasons for discrepancies in medication name, dose, and frequency between clinic documentation and what patient reports to be taking at home Completeness Outcomes concerned with the complete documentation of medication name, dose, frequency, and route of administration in clinic record and reasons for incomplete documentation Program engagement Patient engagement Patient behaviors that facilitate part of the MR process Provider engagement Healthcare provider compliance with process for performing MR Patient medication practices Patient medication adherence Degree to which patient is taking medication(s) as prescribed Patient medication knowledge Patient ability to comprehend information about his or her medication(s), including indication and directions for use Clinical Outcomes Surrogate outcome Blood pressure Outcomes that may be associated with an important clinical endpoint Medication-related problems Drug therapy problems, adverse drugs events, medication-use errors (and associated severity) Adverse events or potential problems that are related to medication use Healthcare utilization Hospitalization/rehospitalization, emergency department visits, ambulatory care visits Use of healthcare services Health-related quality of life Quality of life General well-being of individuals Mortality Death Death aMR = medication reconciliation. are the two most frequently used in - terventions and that their effects vary from small to modest, future studies could consider other types of profes - sional interventions (e.g., reminders or involving local opinion leaders) or a more complex intervention involv - ing various such components. In ad - dition, the impact of financial and regulatory interventions is unknown, as they were not discussed in any of the articles included in our review.

Future research efforts in this field could also be directed toward looking at the impact of these novel catego - ries of interventions. In our review, the majority of study outcomes were process-focused out - comes, with the accuracy of MR being the most frequently reported. While it is important to reduce discrepancies between patient-reported medica - tion information and clinic records (the most commonly reported out - come in MR initiatives), the clini - cal impact of MR initiatives remains largely unexplored. Examining clini - cal outcomes such as healthcare utili - zation and medication-related prob - lems such as ADEs in future studies is critical for determining the clinical value of performing MR in ambula - tory care settings. Researchers con - ducting work in this area are encour - aged to direct their attention toward determining the clinical impact of MR initiatives. As stated previously, factors that facilitate MR implementation at ambulatory care clinics occur at the MEDICATION RECONCILIATION INTERVENTIONS PRACTICE RESEARCH REPORTS 1856 AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 22 | NOVEMBER 15, 2016 patient, staff, and clinic levels. This suggests that future research involv - ing multiple-component interven - tions (i.e., interventions that target all three levels) may be more successful than narrowly focused interventions. A possible limitation of our review is that the gray literature search was con - ducted after the requirement of a com - parative study design was added to the study inclusion criteria; no applicable search results were identified. Another limitation is that some included stud - ies did not meet all of the scoping re - view’s objectives. As an example, only 9 of the 15 studies discussed factors that facilitated the initiatives; therefore, the corresponding conclusions may be bi - ased, as they were based on fewer stud - ies. A final potential concern with our review, a concern inherent in the scop - ing review methodology, is that the risk of bias for each study was not as - sessed. Ultimately, our team was more interested in learning about existing MR models and evaluation methods (in order to inform future research) than about potential bias in published studies.

Conclusion Few studies have examined clini - cal outcomes of MR in ambulatory care settings, with the majority of pertinent reports focusing on process outcomes. Facilitators of successful MR interventions have been identified at the patient, staff, and clinic setting levels.

Acknowledgments Undergraduate pharmacy students Rebecca Babaei-Rad, Melissa Baranski, Anna Bowes, Thomas Boyang Huang, Michelle Liu, and Victor Tsang are acknowledged for their as - sistance with data extraction and updating of literature searches. Gail Nichols, M.I.St., and Joanna Bielecki, M.I.St., designed the literature search strategies. Chaim Bell, M.D., Ph.D., and Janice Kwan, M.D., pro - vided content knowledge and expertise as the study protocol was designed.

Disclosures This work was supported by the Canadian Institutes of Health Research (KRS 124597). Dr. Rochon holds the Retired Teachers of Ontario Chair in Geriatric Medicine at the University of Toronto. The authors have declared no potential conflicts of interest.

References 1. Institute for Safe Medication Prac - tices Canada. Medication reconcili - ation in acute care: getting started kit, version 3.0 (September 2011). www.ismp-canada.org/download/MedRec/Medrec_AC_English_GSK_V3.pdf (accessed 2016 Aug 8).2. Zed PJ, Abu-Laban RB, Balen RM et al. Incidence, severity and prevent - ability of medication-related visits to the emergency department: a prospective study. CMAJ . 2008; 178:1563-9.3. Samoy LJ, Zed PJ, Wilbur K et al. Drug-related hospitalizations in a ter - tiary care internal medicine service of a Canadian hospital: a prospec - tive study. Pharmacotherapy . 2006; 26:1578-86.4. Thomsen LA, Winterstein AG, Søndergaard B et al. Systematic review of the incidence and char - acteristics of preventable adverse drug events in ambulatory care. Ann Pharmacother . 2007; 41:1411- 26.5. Kwan Y, Fernandes OA, Nagge JJ et al. Pharmacist medication assessments in a surgical preadmission clinic. Arch Intern Med . 2007; 167:1034-40. 6. Vira T, Colquhoun M, Etchells E. Reconcilable differences: correcting medication errors at hospital admis - sion and discharge. Qual Saf Health Care . 2006; 15:122-6. 7. Cornish PL, Knowles SR, Marchesano R et al. Unintended medication discrepancies at the time of hospital admission. Arch Intern Med . 2005; 165:424-9.8. Karnon J, Campbell F, Czoski-Murray C. Model-based cost-effectiveness analysis of interventions aimed at preventing medication error at hospital admission (medicines rec - onciliation). J Eval Clin Pract . 2009; 15:299-306.9. Boockvar KS, Blum S, Kugler A et al. Effect of admission medication reconciliation on adverse drug events from admission medication changes. Arch Intern Med . 2011; 171:860-1. 10. Whittington J, Cohen H. OSF health - care’s journey in patient safety. Qual Manag Health Care . 2004; 13:53-9. 11. Boockvar K, Fishman E, Kyriacou CK et al. Adverse events due to discontin - uations in drug use and dose changes in patients transferred between acute and long-term care facilities. Arch Intern Med . 2004; 164:545-50. 12. Boockvar KS, Carlson LaCorte H, Giambanco V et al. Medication reconciliation for reducing drug- discrepancy adverse events. Am J Geriatr Pharmacother . 2006; 4:236- 43. 13. World Health Organization. High 5s fact sheet: assuring medica - tion accuracy at transitions in care standard operating protocol fact sheet. www.who.int/patientsafety/ implementation/solutions/high5s/ h5s-fact-sheet.pdf?ua=1 (accessed 2016 Aug 8). 14. Institute for Healthcare Improve - ment. 5 million lives campaign:

overview. www.ihi.org/offerings/ Initiatives/PastStrategicInitiatives/5M illionLivesCampaign/Pages/default.

aspx (accessed 2016 Aug 8). 15. Joint Commission. 2016 National Patient Safety Goals. www.jointcom - mission.org/standards_information/ npsgs.aspx (accessed 2016 Aug 8). 16. Accreditation Canada. Required organizational practices handbook 2016, version 2. https://accreditation.

ca/sites/default/files/rop-handbook- 2016-en.pdf (accessed 2016 Aug 8). 17. Gandhi TK, Weingart SN, Borus J et al.

Adverse drug events in ambulatory care. N Engl J Med . 2003; 348:1556-64. 18. Arksey H, O’Malley L. Scoping studies: towards a methodological framework. Int J Soc Res Methodol . 2005; 8:19-32. 19. Levac D, Colquhoun H, O’Brien KK. Scoping studies: advancing the methodology. Implement Sci . 2010; 5:69. 20. Centre for Reviews and Dissemina - tion, University of York. Systematic reviews: CRD’s guidance for under - taking reviews in health care ( January 2009). www.york.ac.uk/crd/guid - ance/ (accessed 2016 Aug 8). 21. Higgins JP, Green S, eds. The Cochrane handbook for systematic reviews of interventions, version 5.1.0 (updated March 2011). http://handbook.cochrane.org/ (accessed 2016 Aug 8). 22. Kwan JL, Lo L, Sampson M, Shojania KG. Medication reconciliation during transitions of care as a patient safety strategy: a systematic review. Ann Intern Med . 2013; 158(5, Pt 2):397- 403. 23. The Cochrane Collaboration. Effec - tive Practice and Organisation of Care (EPOC) taxonomy. https://epoc.co - chrane.org/epoc-taxonomy (accessed 2016 Aug 8).

PRACTICE RESEARCH REPORTS MEDICATION RECONCILIATION INTERVENTIONS AM J HEALTH-SYST PHARM | VOLUME 73 | NUMBER 22 | NOVEMBER 15, 2016 1857 24. Nassaralla CL, Naessens JM, Chaudhry R et al. Implementation of a medication reconciliation process in an ambulatory internal medicine clinic. Qual Saf Health Care . 2007; 16:90-4. 25. Varkey P, Cunningham J, Bisping DS.

Improving medication reconciliation in the outpatient setting. Jt Comm J Qual Patient Saf . 2007; 33:286-92. 26. Weingart SN, Cleary A, Seger A et al.

Medication reconciliation in ambula - tory oncology. Jt Comm J Qual Patient Saf . 2007; 33:750-7. 27. Delate T, Chester EA, Stubbings TW, Barnes CA. Clinical outcomes of a home-based medication reconcili - ation program after discharge from a skilled nursing facility. Pharmaco - therapy . 2008; 28:444-52. 28. Nassaralla CL, Naessens JM, Hunt VL et al. Medication reconciliation in ambulatory care: attempts at improvement. Qual Saf Health Care. 2009; 18:402-7. 29. Stock R, Scott J, Gurtel S. Using an electronic prescribing system to ensure accurate medication lists in a large multidisciplinary medical group. Jt Comm J Qual Patient Saf . 2009; 35:271-7. 30. Peyton L, Ramser K, Hamann G et al. Evaluation of medication reconcilia - tion in an ambulatory setting before and after pharmacist intervention. J Am Pharm Assoc . 2010; 50:490-5. 31. Moczygemba LR, Gatewood SB, Kennedy AK et al. Medication rec - onciliation campaign in a clinic for homeless patients. Am J Health-Syst Pharm . 2012; 69:558,560-2. 32. Kilcup M, Schultz D, Carlson J, Wilson B. Postdischarge pharmacist medica - tion reconciliation: impact on read - mission rates and financial savings.

J Am Pharm Assoc . 2013; 53:78-84. 33. Moore JM, Shartle D, Faudskar L et al. Impact of a patient-centered pharmacy program and intervention in a high-risk group. J Manage Care Pharm . 2013; 19:228-36. 34. Neufeld NJ, Gonzalez Fernandez M, Christo PJ, Williams KA. Positive rec - ognition program increases compli - ance with medication reconciliation by resident physicians in an outpa - tient clinic. Am J Med Qual . 2013; 28:40-5. 35. Persell SD, Eder M, Friesema E et al. EHR-based medication support and nurse-led medication therapy management: rationale and design for a three-arm clinic random - ized trial. J Am Heart Assoc . 2013; 2(5):e000311. 36. Roth MT, Ivey JL, Esserman DA et al. Individualized medication assess - ment and planning: optimizing medication use in older adults in the primary care setting. Pharmacothera - py. 2013; 33:787-97. 37. Wolff CM, Nowacki AS, Yeh JY, Hickner JM. A randomized controlled trial of two interventions to improve medication reconciliation. J Am Board Fam Med . 2014; 27:347-55. 38. Vouri SM, Marcum ZA. Use of a medication reconciliation tool in an outpatient geriatric clinic. J Am Pharm Assoc . 2013; 53:652-8. 39. Chan C, Woo R, Seto W et al. Medi - cation reconciliation in pediatric cardiology performed by a pharmacy technician: a prospective cohort comparison study. Can J Hosp Pharm . 2015; 68:8-15. 40. Sen S, Siemianowski L, Murphy M, McAllister SC. Implementation of a pharmacy technician–centered medi - cation reconciliation program at an urban teaching medical center. Am J Health-Syst Pharm . 2014; 71:51-6. 41. Bayoumi I, Howard M, Holbrook AM, Schabort I. Interventions to improve medication reconciliation in pri - mary care. Ann Pharmacother . 2009; 43:1667-75. 42. Ivers N, Jamtvedt G, Flottorp S et al. Audit and feedback: effects on professional practice and healthcare outcomes. Cochrane Database Syst Rev . 2012; 6:CD000259. 43. O’Brien MA, Rogers S, Jamtvedt G et al. Educational outreach visits: effects on professional practice and health care outcomes. Cochrane Database Syst Rev . 2007; 4:CD000409. Appendix—MEDLINE and PreMEDLINE search strategy a 1. exp medication therapy management/2. (medication$ adj3 reconcili$).mp.3. (medicine$ adj3 reconcili$).mp.4. exp medication error/5. (medication$ adj3 error$).mp.6. (medicine$ adj3 error$).mp.7. (discrepanc$ and (drug$ or medication$ or medicine$)).mp.8. 1 or 2 or 3 or 4 or 5 or 6 or 79. exp primary health care/10. (primary adj2 care).mp.11. primary healthcare.mp.12. exp general practitioner/13. (family adj physician$).mp.14. exp general practice/15. (family adj practice$).mp.16. exp community medicine/17. exp group practice/18. (physician$ adj office$).mp.19. medical practice$ management.mp.20. exp ambulatory care/21. exp outpatient department/22. exp private practice/23. exp outpatient/24. exp outpatient care/25. (free-standing adj2 clinic$).mp.26. (free-standing adj2 facilit$).mp.27. ambulatory care facilit$.mp.28. general practice$.mp.29. general practitioner$.mp.30. 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29 31. 8 and 3032. 3133. limit 32 to (english language and (adult <18 to 64 years> or aged <65+ years>)) aSearch strategies for EMBASE, CINAHL, and International Pharmaceutical Abstracts available by request.

MEDICATION RECONCILIATION INTERVENTIONS PRACTICE RESEARCH REPORTS Copyright ofAmerican JournalofHealth- System Pharmacy isthe property ofAmerican Society ofHealth System Pharmacists anditscontent maynotbecopied oremailed to multiple sitesorposted toalistserv without thecopyright holder'sexpresswrittenpermission.

However, usersmayprint, download, oremail articles forindividual use.