Tools and tactics for postdischarge medication management interventions

Abstract

Purpose

To identify interventions for organizational pharmacist-leaders and frontline pharmacy staff to optimize peri- and postdischarge medication management.

Summary

An evidence-based toolkit was systematically constructed on the basis of findings of 3 systematic overviews of systematic reviews. The interventions were reviewed by a technical expert panel and categorized as either tools or tactics. The identified tools are instruments such as diagrams, flow charts, lists, tables, and templates used in performing a distinct operation, whereas identified tactics reflect broader methods (eg, reduced dosing frequency). Tools and tactics were chosen on the basis of their potential to improve postdischarge medication management, with a focus on interventions led by pharmacy staff that may reduce hospital readmissions among older, sicker patients. Overall, 23 tools and 2 tactics were identified. The identified tools include items such as education, text messaging, and phone calls. The tactics identified are dose simplification and monetary incentives. Practical information has also been provided to facilitate implementation.

Conclusion

Several tools and tactics are available to optimize peri- and postdischarge medication management. Organizational pharmacist-leaders and frontline pharmacy staff can implement these interventions to improve patient outcomes.

Hospital discharge is an important transition of care that requires optimal management to reduce the risk of rehospitalization and to improve patient outcomes. The literature pertaining to medication management interventions focused on these goals is vast and has been the focus of hundreds of systematic reviews. Many of these interventions fall under the following 3 domains: medication reconciliation, polypharmacy, and medication adherence. In prior issues of this journal, 3 overviews of systematic reviews summarized the evidence base for each of these domains.^1–3 To capitalize on these rigorous overviews in a way that makes their findings most useful to organizational pharmacist-leaders and frontline pharmacy staff, in this article we have summarized the findings in a toolkit. Our aim was to relieve readers of having to repeat rigorous literature reviews in these 3 areas and put forward a toolkit based in evidence rather than expert opinion alone. The text of this article is intentionally concise, and we have used a table to summarize the key points from each primary article on which tools or tactics are based, encouraging the reader to reference the original literature in cases where more detail is needed. Above all, this toolkit was developed to facilitate ease of implementation and includes practical guidance.

Toolkit development

This evidence-based toolkit was constructed on the basis of findings of 3 systematic overviews of systematic reviews. The objective of each systematic overview was to describe and distill the findings from multiple systematic reviews examining pharmacist-led peri- and postdischarge medication management interventions addressing each of 3 domains: medication reconciliation, polypharmacy, and medication adherence. The reader is directed to each systematic overview for details, but the overall process of toolkit development is summarized in Table 1.

Table 1.

Major Steps of Toolkit Development

Step of Development	No. of Articles or Tools in Domain
	Medication Reconciliation	Polypharmacy	Medication Adherence
1. Conduct systematic literature search in 3 domains: medication adherence, medication reconciliation, and polypharmacy (sources: MEDLINE, Cochrane Database of Systematic Reviews, and Database of Abstracts of Reviews of Effects).
2. Identify relevant systematic reviews.a	18	9	70
3. Isolate high-quality systematic reviews.a	11	6	25
4. Extract conclusions of high-quality systematic reviews.a	37	14	50
5. Isolate systematic reviews drawing positive conclusions.	3	3	16
6. Identify primary studies that both drove systematic reviewers’ positive conclusions and used a generalizable tool or tactic.b	3	9	9
7. For each tool and tactic identified, assign a GRADE level of evidence quality based on the systematic review.a	3 tools from low-quality bodies of evidence	9 tools from very low-quality bodies of evidence	2, 4, and 3 tools from moderate-, low-, and very low-quality bodies of evidence, respectively
8. Consider other primary studies recognized by the technical expert panel or during the review process.	2 tools addressing all 3 domains

Abbreviations: GRADE, Grading of Recommendations Assessment, Development and Evaluation⁵ (GRADE ratings available only for 21 tools or tactics identified in systematic overviews).

^aReviews were conducted independently by 2 coauthors prior to reconciling results.

^bGeneralizability was determined by pharmacist coauthors involved in patient care.

In brief, to begin each systema- tic overview, searches of MEDLINE, the Cochrane Database of Systematic Reviews, and the Database of Abstracts of Reviews of Effects for articles published from January 2004 to February 2017 were performed. Systematic reviews evaluating interventions in each domain that could have substantial potential to prevent hospitalizations among older adult patients in the period following hospital discharge were included. Two of the authors independently screened, selected, appraised, and extracted information from systematic reviews. Reviews with an AMSTAR (A MeaSurement Tool to Assess systematic Reviews)⁴ score below 8 were excluded. After extracting relevant conclusions from each review, authors summarized the body of evidence and compared characteristics and conclusions across reviews. For systematic reviews reporting effective interventions, the supporting primary literature was identified. Input from a technical expert panel (TEP) with training and experience in pharmacy, medicine, and research methods was solicited to provide context to findings and input beyond what was available in the published literature. Primary studies recommended by the TEP were also considered. A team of hospital-based pharmacists with frontline expertise were then responsible for reviewing the primary literature and identifying generalizable interventions for the toolkit.

Interventions were classified into 2 major categories: tools and tactics. The identified tools are instruments such as diagrams, flow charts, lists, tables, and templates used in performing a distinct protocol, whereas identified tactics reflect broader methods or plans designed to achieve an overall aim (eg, reduced dosing frequency). Included tools and tactics were categorized multidimensionally by patient population, setting, disease or condition, and timing.

We applied Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology⁵ to rate the quality of evidence for the body of literature that informed development of each tool or tactic. If such a rating was derived from an adequately conducted systematic review, then the rating was adopted. As per the GRADE framework, evaluation of evidence quality was based on the following GRADE domains: study design, study quality, consistency of effects, applicability of evidence to target population, and other modifying factors, including sample size and magnitude of effect estimates.

Toolkit interventions

Tools.

The systematic overview process yielded 19 tools and 2 tactics. Two other tools were also highlighted for inclusion by the TEP. Nine tools focus on medication adherence, 9 others focus on polypharmacy, 3 focus on medication reconciliation, and 2 tools address all 3 domains. Of the tools directed at medication adherence, 3 prominently feature phone calls, 3 function primarily through patient education, 2 involve use of text messaging, and 2 focus on managing depression. While the medication adherence and medication reconciliation tools tend to be complex multicomponent interventions, the tools addressing polypharmacy tend to be protocols or lists of medications that could be potentially worrisome in an older patient population (eg, Beers criteria, STOPP/START [Screening Tool of Older Persons’ Prescriptions/Screening Tool to Alert to Right Treatment] criteria). Outcome variables assessed included medication prescribing errors, medication adherence as measured by dispensation or patient report, adverse drug events, postdischarge utilization, and mortality. Of the 21 tools identified via the systematic overview process, only 2 tools represented methods that were backed by a moderate-quality body of evidence, whereas low- and very low-quality bodies of evidence backed 7 and 12 tools, respectively. Please note that GRADE ratings were not used as inclusion criteria in toolkit development, and low- and very low-quality ratings may reflect limited study of these interventions. Table 2, Table 3, Table 4, and Table 5 include general characteristics of each tool by domain, including the digital object identifier for the publication that best describes the intervention. Further details, including relevant text and figures from these publications, are available in the eAppendix.

Table 2.

General Characteristics of Tools Addressing All 3 Domains (Medication Adherence, Medication Reconciliation, and Polypharmacy)^a

Tool	Primary Study (Year Published)	Primary Study Characteristics				Study Measurement Methods			Results
		Setting (Country)	Follow-up Duration(s)	Patient Population	Who Used Tool?	No. Subjects	Data Source	Outcome Variable
Multicomponent pharmacist intervention: medication review, patient interview, and postdischarge follow-up to address polypharmacy	Ravn-Nielsen et al (2018)6	Postdischarge (Denmark)	30 and 180 days	New acute admission adults with 5 or more medications and multiple medical conditions	Pharmacist	BI: 494 EI: 476 C: 498	National patient register	Occurrence of readmission, ED visits, and deaths within 30 days and 180 days	Significant effect on readmission within 30 days within the extended intervention group (HR, 0.62; 95% CI, 0.46-0.84) as well as within 180 days (HR, 0.75; 95% CI, 0.62-0.90)
Pharmacist counseling via telephone discharge script: comparison of patients’ self-reported medication list with discharge list as well as medication adherence	Schnipper et al (2006)7	Postdischarge (United States)	30 days	General medicine patients with multiple medical conditions who were being discharged home and could be contacted in 30 days	Pharmacist	I: 92 C: 84	Preventable ADEs measured with a screening questionnaire (developed by Bates and colleagues)	Presence of preventable ADEs in patients within 30 days after discharge	Significant difference in rate of preventable ADEs in control group (11%) vs intervention group (1%) (P = 0.01; unadjusted OR, 0.10; 95% CI, 0.013-0.86)

Abbreviations: ADE, adverse drug event; BI, basic intervention; C, control; CI, confidence interval; ED, emergency department; EI, extended intervention; HR, hazard ratio; I, intervention; OR, odds ratio.

^aBecause the 2 tools described were identified by the technical expert panel and not included in any systematic review, no evidence ratings were assigned.

Table 3.

General Characteristics of Tools Focusing on Medication Adherence

Tool	Primary Study (Year Published)	Systematic Review and Monthsdified Evidence GRADE	Primary Study Characteristics				Study Measurement Methods			Results
			Setting (Country)	Follow-up Duration	Patient Population	Who Used Tool?	No. Subjects (I/C)	Data Source	Outcome Variable
Interventions Focused on Medication Adherence in General
1. Telephone reminder timeline with 4 parts: medication reconciliation and tailoring + patient education + collaborative care between pharmacist and PCP and/or cardiologist + voice messaging	Lambert-Kerzner et al (2012)8	van Driel et al (2016) Moderate	Postdischarge (United States)	12 months	Adults with acute coronary syndrome	Pharmacist	140/140	Pharmacy refill data on cardioprotective medications	Modified medication possession ratio for each medication (no. days medication supplied/observation time interval)	Significantly better adherence to medication regimens in intervention group vs control group (89.3% vs 73.9%, P = 0.003)
2. Telephone reminder script: automated call from provider organization that prompted patient to get newly prescribed statin 1–2 weeks after visit (if medication not picked up within 1 weeks of call, a letter was sent)	Derose et al (2013)9	van Driel et al (2016) Moderate	Postdischarge (United States)	25 days	Adults with dyslipidemia	Provider organization	2,606/2,610	Medication dispensations from healthcare pharmacies	Dispensation of statin between first phone call and up to 2 weeks after letter was delivered (25 days)	Significantly better postdischarge statin dispensation in intervention group vs control group (42.3% vs 26.0%; absolute difference, 16.3%; P < 0.001)
Interventions Focused on Medication Adherence for Patients with Comonthsrbid Depression
3. Flowchart for managing depression and hypertension: in-person and telephone conversations discussing depression and effect on antihypertensive medication adherence	Bogner et al (2008)10,a	Viswanathan et al (2012) Low	Outpatient (United States)	6 weeks	Adults 50-80 years of age with HTN and depression	Integrated care manager	32/32	MEMS caps	Adherence: 80% or more of pills taken during specified observation period	Significantly better adherence to antidepressant medication (71.9% vs 31.3%, P < 0.01) and antihypertensive medication (78.1% vs 31.3%, P < 0.001) in intervention group vs control group
Interventions Addressing Medication Adherence with Communication or Patient Education
4. Pharmacist-delivered adherence monitoring service: BMQ/MARS/BMU questionnaires + SF12 quality of life questionnaire + 21 food frequency checklist and exercise survey	Aslani et al (2010)12	Rash et al (2016) Very low	Outpatient (Australia)	12 months	Adults with dyslipidemia	Pharmacist	48/49	MARS	MARS	At the second study visit, 3 months after the initial visit, patients were less likely to take a decreased amount of the prescribed dose (F2,178 = 4.3, P < 0.05; contrast F1,89 = 5.7, P < 0.05)
5. Motivational interviewing script: 4 motivational interview sessions every 3 months for 12 months	Ogedegbe et al (2008)13 Primary study	Easthall et al (2013) Very low	Outpatient (United States)	12 months	African American adults with HTN	Research assistant	95/95	MEMS caps	Taking adherence: proportion of days in which the patient took the prescribed medication	Significantly better posttreatment adherence rates in intervention group vs control group (57% vs 43%, P = 0.027)
6. Pharmacist patient education protocol: baseline medication history with patient-centered verbal/written instructions + collaborative care with PCP and/or nurse	Murray et al (2007)14	Demonceau et al (2013) Very low	Outpatient (United States)	12 months	Low-income adults ≥50 years of age with heart failure	Pharmacist	122/192	MEMS caps	Adherence: % of prescribed medication taken	Overall medication adherence was significantly better in intervention group vs control (78.8% vs 67.9%; absolute difference, 10.9%; 95% CI, 5.0-16.7)
Interventions Addressing Medication Adherence with SMS Text Messages
7. SMS text message scripts for asthma: SMS text messages that were tailored to each patient based on BIPQ responses at baseline	Petrie et al (2012)15	Tao et al (2015) Low	Outpatient (New Zealand)	9 mo	Adults and adolescents (16-45 years of age) with asthma	A bank of 166 text messages tailored for asthmatic patients	73/74	Self-report	BIPQ Adherence: 80% or greater of taking medication	Significant improvement in SMS group in perception of medication necessity, nature of patient’s asthma, and patient’s control of asthma (P < 0.05); significantly higher percentage of SMS group vs control group took 80% of their inhaler doses (37.7% vs 23.9%, P < 0.05)
8. SMS text message scripts for type 1 diabetes: SweetTalk Software System (motivational SMS text messages)	Franklin et al (2006)16	Tao et al (2015) Low	Outpatient (United Kingdom)	12 months	Pediatric patients with type 1 diabetes	Automated system with over 400 text messages	33/28	Self-report	Visual analogue adherence score	Significantly better adherence in the intervention group vs control group (mean [SD], 77.2 [16.1] vs 70.4 [20.0], P = 0.042)
Interventions Addressing Medication Adherence with Communication or Patient Education
9. Pharmacist-delivered adherence monitoring service: BMQ/MARS/BMU questionnaires + SF12 quality of life questionnaire + 21 food frequency checklist and exercise survey	Aslani et al (2010)12	Rash et al (2016) Very low	Outpatient (Australia)	12 months	Adults with dyslipidemia	Pharmacist	48/49	MARS	MARS	At second study visit, which occurred 3 mo after the initial visit, patients were less likely to take a decreased amount of the prescribed dose (F2,178 = 4.3, P < 0.05; contrast F1,89 = 5.7, P < 0.05)
10. Motivational interviewing script: 4 motivational interview sessions every 3 mo for 12 mo	Ogedegbe et al (2008)13 Primary study	Easthall et al (2013) Very low	Outpatient (United States)	12 months	African-American adults with HTN	Research assistant	95/95	MEMS caps	Taking adherence (proportion of days on which the patient took the prescribed medication)	Significantly better posttreatment adherence rates in intervention group vs control group (57% vs 43%, P = 0.027)
11. Pharmacist patient education protocol: baseline medication history with patient-centered verbal/written instructions + collaborative care with PCP/nurse	Murray et al (2007)14	Demonceau et al (2013) Very low	Outpatient (United States)	12 months	Low-income adults age ≥50 years with heart failure	Pharmacist	122/192	MEMS caps	Adherence (percentage of prescribed medication taken)	Overall medication adherence was significantly better in intervention group vs control group (78.8% vs 67.9%; absolute difference, 10.9%; 95% CI, 5.0-16.7)
Interventions Addressing Medication Adherence with SMS Text Messages
12. SMS text message scripts for asthma: SMS text messages that were tailored to each patient based on BIPQ responses at baseline	Petrie et al (2012)15	Tao et al (2015) Low	Outpatient (New Zealand)	9 months	Adults and adolescents (16-45 years of age) with asthma	A bank of 166 text messages tailored for asthmatic patients	73/74	Self-report	BIPQ Adherence: 80% or greater of taking medication	Significant improvement in SMS group in perception of medication necessity, nature of patient’s asthma, and patient’s control of asthma (P < 0.05); significantly higher percentage of SMS group vs control group took 80% of their inhaler doses (37.7% vs 23.9%, P < 0.05)
13. SMS text message scripts for type 1 diabetes: SweetTalk Software System (motivational SMS text messages)	Franklin et al (2006)16	Tao et al (2015) Low	Outpatient (United Kingdom)	12 months	Pediatric patients with type 1 diabetes	Automated system with over 400 text messages	33/28	Self-report	Visual analogue adherence score	Significantly better adherence in the intervention group vs control group (mean [SD], 77.2 [16.1] vs 70.4 [20.0], P = 0.042)

Abbreviations: BIPQ, Brief Illness Perception Questionnaire; BMQ, Brief Medication Questionnaire; BMU, Barriers to Medication Use Questionnaire; CI, confidence interval; GRADE, Grading of Recommendations Assessment, Development and Evaluation⁵; MARS, Medication Adherence Report Scale; MEMS, Medication Event Monitoring System; PCP = primary care provider; SD, standard deviation; SF12, 12-item Short Form Survey; SMS, short message service.

^aIntervention methods also discussed in reference 11.

Table 4.

General Characteristics of Each Tool Focusing on Medication Reconciliation

Tool	Primary Study (Year Published)	Systematic Review and Modified Evidence GRADE	Primary Study Characteristics				Study Measurement Methods			Results
			Setting (Country)	Follow-up Duration	Patient Population	Who Used Tool?	No. Subjects (I/C)	Data Source	Outcome Variable(s)
14. Care Transitions Innovation (C-TraIn): transitional care nurse + pharmacy consultation + postdischarge primary care medical home linkage + monthly improvement meetings	Englander et al (2014)17	Mekonnen et al (2016) Low	Postdischarge (United States)	30 days	Uninsured low-income adults with any medical condition	Pharmacist and nurse level	209/173	Electronic medical record	Primary: hospital readmissions (per Oregon Hospital Discharge Dataset), ED visits Secondary: Mortality per chart review and state death records	No significant difference in proportions of patients readmitted at 30 days post discharge (OR, 0.88; 95% CI, 0.50-1.54) or 30-day ED visits (OR, 1.38; 95% CI, 0.83–2.31); intervention patients had a lower mortalitythan control group (0 vs 5 patients, unadjusted P = 0.02)
15. Multifaceted clinical pharmacist discharge service intervention	Eggink et al (2010)18	Mekonnen et al (2016) Low	Postdischarge (Netherlands)	At discharge	Adults with heart failure taking ≥5 medications	Pharmacist	41/44	Medication list	Total percentage sum of prescription errors and discrepancies post discharge, BMQ	Significantly fewer patients with at least one prescription error or medication discrepancy in the intervention vs control (39% vs 68%; RR, 0.57; CI 95%, 0.37-0.88)
16. Discharge summary with medication report: quality improvement of medication list in the discharge summary before patient discharge	Bergkvist et al (2009)19	Mekonnen et al (2016) Low	Admission through discharge (Sweden)	At discharge	Adults age ≥65 years with any medical condition	Pharmacist and physician level	52/63	ApoDos (Medication Dispensing System) at pharmacy, or manual medication list at community health center	Medication error classified as one of the following when patient discharged to community health care setting: (1) missing medication in list, (2) medication added to list, or (3) total dosage changed over a 24-hour period	Decrease of 45% in medication errors per patient in intervention group (P = 0.012); proportion of patients without medication errors was 63.5% in the control group and 73.1% in the intervention group (P = 0.319)

Abbreviations: BMQ, Brief Medication Questionnaire; CI, confidence interval; C, control; ED, emergency department; GRADE, Grading of Recommendations Assessment, Development and Evaluation⁵; I, intervention; OR, odds ratio; RR, relative risk.

Table 5.

General Characteristics of Each Tool Focusing on Polypharmacy

Tool(s)	Primary Study (Year Published)	Systematic Review and Modified Evidence GRADE	Primary Study Characteristics			Who Used Tool?	Study Measurement Methods			Results
			Setting (Country)	Follow-up Duration	Patient Population
							No. Subjects (I/C)	Data Source	Outcome Variable
17. Clinician Rated Anticholinergic Score (CR-ACHS)	Yeh et al (2013)20	Kroger et al (2015) Very low	Veterans home (Taiwan)	12 weeks	Adults with dementia	Physician	40/27	CR-ACHS	CR-ACHS (scale: 0 = no effect 3 = strong effect)	Significant reduction in anticholinergic scores in intervention group vs control group at 12 weeks (mean [SD], 0.5 [1.1] vs 1.1 [1.3], P = 0.021)
18. Medication Appropriateness Index (MAI): structured medication review	Verrue (2012)21	Kroger et al (2015) Very low	Nursing home (Belgium)	6 months	Older adults with any medical condition	Pharmacist	230/154	Medication list review	Quality of prescribing assessed with MAI scale	Significant reduction in MAI in intervention group vs control group (OR, 3.91; 95% CI, 1.88-8.15)
Assessing Care of Vulnerable Adults (ACOVE): structured medication review for older adults	Verrue (2012)21	Kroger et al (2015) Very low	Nursing home (Belgium)	6 months	Older adults with any medical condition	Pharmacist	230/154	Medication list review	Drug underuse assessed with ACOVE scale	Significant reduction in inappropriate ACOVE ratings in intervention group vs control group (OR, 4.35; 95% CI, 0.87-21.67)
Beers criteria: structured medication review of potentially inappropriate medications for older adults based on patient age, comorbidities, medications, and kidney function	Verrue (2012)21	Kroger et al (2015) Very low	Nursing home (Belgium)	6 months	Older adults with any medical condition	Pharmacist	230/154	Medication list review	Drug overuse assessed with Beers criteria	Significant reduction in the number of Beers criteria drugs in intervention groups vs control group (OR, 6.52; 95% CI, 1.38-30.92)
STOPP/START: structured medication review for older adults	Verrue (2012)21	Kroger et al (2015) Very low	Nursing home (Belgium)	6 months	Older adults with any medical condition	Pharmacist	230/154	Medication list review	Drug overuse assessed with STOPP/START scales	No statistically significant between-group difference in inappropriate START ratings per patient (OR, 10.92; 95% CI, 0.58-206.64); no statistically significant between-group difference in number of STOPP drugs per patient (OR, 2.86; 95% CI, 0.71-11.53)
19. Deprescribing algorithm for psychoactive medications	Patterson et al (2010)22	Kroger et al (2015) Very low	Nursing home (Northern Ireland)	NA	Older adults with any medical condition	Nursing home	173/161	Manual medication list review	Proportion of patients prescribed 1 or more inappropriate psychoactive medications	Proportion of patients taking inappropriate psychoactive medications was significantly lower in intervention nursing homes compared to control nursing homes (OR, 0.26; 95% CI, 0.14-0.49)
20. Deprescribing tool for antipsychotic and benzodiazepine use	Westbury (2010)23	Kroger et al (2015) Very low	Nursing home (Australia)	26 weeks	Older adults with dementia	Nursing home	898/693	Drug Use Evaluation Audit Program	Prescribing rates (mean proportion) of benzodiazepines and antipsychotics in nursing homes	Significant reductions in mean use of benzodiazepines in the intervention group from baseline to 26 weeks (mean [SD], 31.8 [8.6] vs 26.9 [8.6], P < 0.005); significant reductions in mean use of antipsychotics within the intervention group from baseline to 26 weeks (mean [SD], 20.3 [8.7] vs 18.6 [8.4], P < 0.05)
21. Algorithm for improving drug therapy in disabled/frail elderly patients in nursing facilities	Garfinkel et al (2007)24	Kroger et al (2015) Very low	Geriatric medical center (Israel)	12 months	Adults with any medical condition	Physician and nurse	119/71	Medication list review	Annual incidence of death and referral to acute care hospital	Significant difference in 1-year mortality rate in intervention vs control (21% vs 45%, P < 0.001); significant decrease in yearly referrals to acute care hospital in intervention group vs control group (11.8% vs 30.0%, P < 0.002)
22. Deprescribing algorithm for NSAIDs	Stein et al (2001)25	Kroger et al (2015) Very low	Nursing home (United States)	3 months	Adults 65 years of age or older with muscle or joint pain	Nursing home	1,065/1,067	Medication list review	Post-intervention change: 3-months follow-up value minus baseline value	Significant decrease in mean number of days of NSAID use from baseline to 3-months follow-up in intervention group (from 7.0 days to 1.9 days) vs control group (from 7.0 days to 6.2 days); P = 0.0001

Abbreviations: CI, confidence interval; GRADE, Grading of Recommendations Assessment, Development and Evaluation⁵; NA, not applicable; NSAID, nonsteroidal anti-inflammatory drug; OR, odds ratio; SD, standard deviation; STOPP, Screening Tool for Older Persons of Potentially inappropriate Prescriptions; START, Screening Tool to Alert doctors to Right Treatment.

Tactics.

Dose simplification. Medication adherence literature consistently demonstrates the positive effects of dose simplification. Our systematic overview found that 5 of 6 reviews examining dose simplification reported adherence improvements of moderate effect size in several populations, including patients with cardiovascular disease and patients with osteoporosis as well as patients taking antihyperglycemic drugs, statins, and blood pressure medications.³ Examples of dose simplification include reduction of dosing frequency (eg, from twice daily to once daily) and incorporation of fixed-dose combinations instead of coadministered individual therapies.

Incentives and reductions in out-of-pocket spending.

Incentives and reductions in out-of-pocket spending can be effective and scalable interventions to improve medication adherence.³ From the standpoint of pharmacists caring for patients, this effect may be most commonly achieved by changing prescriptions to generic or preferred-brand medications with lower copayments or, in some cases, by helping patients to access pharmaceutical manufacturers’ drug assistance programs. This finding comes from a 2016 systematic review focusing on patients taking statins that reported a small positive effect of prescription cost coverage on adherence rates (effect size, 0.15; 95% confidence interval, 0.11–0.21)²⁶; that finding was based exclusively on a 2011 publication by Choudhry et al,²⁷ wherein those authors assessed the impact of full prescription drug coverage (vs usual coverage) for all statins, β-blockers, angiotensin-converting enzyme inhibitors, and angiotensin receptor blockers prescribed after myocardial infarction. Rates of adherence were 5.6 percentage points higher in the intervention group than in a control group (38.9% vs 43.9%). Additionally, a 2012 systematic review found moderate-quality evidence that policy interventions that reduced out-of-pocket expenses, including reduction of patient copayments and improved prescription drug coverage, improved medication adherence to both cardiac and diabetes medications.²⁸

Summary

To aid organizational pharmacist-leaders and frontline pharmacy staff in optimizing peri- and postdischarge medication management, this journal has published 3 systematic overviews of systematic reviews focused on medication reconciliation,² medication adherence,³ and polypharmacy,¹ respectively. Whereas those articles focus more on rigorous methodology and summarizing findings, this article aims to offer a usable list of tools felt by practicing pharmacists to be both readily generalizable and promising for their potential to improve postdischarge medication management. These tools might be considered by frontline pharmacy staff in day-to-day patient care, as well as by pharmacist-leaders as they design and improve systems of care. In the latter case, tools might be considered not only directly but also in terms of what support and contextual interventions might be necessary to allow frontline pharmacists to use them. For example, pharmacy technicians might be deployed for certain tasks to free up pharmacist time for more complex postdischarge medication management activities. Table 2, Table 3, Table 4, and Table 5 have been designed to offer pharmacist-leaders and frontline pharmacy staff a survey of the identified tools, with the idea that interventions can be matched to provider organizations’ priorities and patient populations based on the listed summary characteristics, which include the setting, patient population, and results achieved in the primary study. The authors and collaborators are hopeful that successful use of these tools will allow pharmacists to support better care transitions, ultimately resulting in improved outcomes and reduced unnecessary healthcare resource utilization by older patients on complex medication regimens.

Disclosures

This research was supported by the American Society of Health-System Pharmacists Research and Education Foundation and the National Institute on Aging (NIA) of the National Institutes of Health under awards L30AG048588, K23AG049181, and R01AG058911 (all NIA awards were to Dr. Pevnick). The authors have declared no potential conflicts of interest.