Potentially Inappropriate Medication Use Following Hospital Discharge Among Medicare Beneficiaries

ABSTRACT

Objective: Medicare beneficiaries 65 years and older are susceptible to receiving potentially inappropriate medications (PIMs), for which the risks outweigh the benefits. Unlike stand-alone Part D prescription drug plans serving traditional Medicare (TM) beneficiaries, Medicare Advantage (MA) plans have a financial incentive to improve transitional care following hospitalization and reduce PIM use in order to avoid costly adverse drug events. This study examined how changes in PIM use following hospitalization differed between TM and MA.

Study Design: This observational study analyzed nationwide Medicare Part D Event files, TM fee-for-service claims, and MA encounter data from 2016 to 2019. It included 675,776 hospitalization episodes for Medicare beneficiaries 65 years and older with Part D prescription drug coverage.

Methods: We conducted a difference-in-differences analysis of PIM use before and after hospitalization between TM and MA. PIM use was measured as any Part D dispensing event for a high-risk medication during the 180 days before and after hospitalization.

Results: The prevalence of PIM use in TM was 23.90% preadmission and 23.29% post discharge and in MA was 21.82% preadmission and 21.47% post discharge. After adjusting for patient factors, the decrease in PIM use following hospitalization was 0.25 percentage points (95% CI, 0.02-0.49) less in MA than in TM.

Conclusions: Although MA plans are likely to have stronger financial incentives than TM to reduce PIM use, we found that MA and TM beneficiaries experienced similar changes in PIM use following hospitalization.

Takeaway Points

Medicare Advantage did not provide a clear advantage over traditional Medicare in reducing potentially inappropriate medication (PIM) use following hospitalization.

• PIM use was prevalent among Medicare beneficiaries prior to hospitalization, with minimal decreases after hospital discharge.
• Traditional Medicare beneficiaries and Medicare Advantage enrollees experienced similar changes in PIM use post hospitalization.
• There are untapped opportunities to reduce PIM use through transitional care and medication therapy management programs.

Patients are often discharged from the hospital with new medications or modifications to existing medication treatment prescribed by multiple providers.^1,2 Between one-third and two-thirds of patients 65 years and older are discharged from the hospital with outpatient prescriptions for potentially inappropriate medications (PIMs),^3,4 for which the risks outweigh the benefits. This can lead to preventable adverse drug events, including emergency department visits and hospital readmissions.^3,5 Because patients discharged from the hospital have a heightened risk of adverse drug events,⁵ the postdischarge period provides an opportunity to reconcile medication therapy. Reducing the use of PIMs after hospitalization, therefore, helps prevent adverse drug events and reduce health care expenditures among adults 65 years and older.

Review and reconciliation of medications after discharge are critical in the continuum of care. For Medicare patients, primary care physicians (PCPs) reconcile and manage medications following hospitalization as part of transitional care management, which is a bundle of services that includes a follow-up visit within 7 to 14 days post discharge, medication reconciliation (during or before the follow-up visit), patient education, and coordination of follow-up with other providers or community services.^6,7 Additionally, Medicare Part D prescription drug coverage includes medication therapy management (MTM) programs with postdischarge medication reviews to identify and resolve medication issues, including PIM use.^8,9

Although all Medicare beneficiaries are eligible to receive transitional care management services and all Part D plans include MTM programs, regulatory differences and financial incentives of traditional Medicare (TM) and Medicare Advantage (MA) may result in different medication management outcomes. That is, beneficiaries in TM and MA may experience different patterns of postdischarge medication management, resulting in differences in PIM use. TM provides prescription drug benefits through stand-alone prescription drug plans (PDPs), operated independently from the medical benefits. MA plans typically include Part D prescription drug benefits and medical services under a single plan. This financial alignment in MA creates incentives to avoid unnecessary and costly health services,¹⁰ including adverse drug events resulting from PIMs.^3,5 Consequently, MA enrollees have higher rates of comprehensive medication reviews than TM beneficiaries.^11,12

Moreover, MA plans receive capitated payments for covering medical services, so they encourage PCP use through lower cost sharing for PCP visits than TM,^13,14 emphasizing disease management through PCPs.¹⁵ MA plans must also publicly report the quality of hospital transitions of care experienced by enrollees, including postdischarge medication reconciliation and provider follow-up,¹⁶ which can help reduce PIM use. In contrast, TM does not operate under the same incentives or requirements to enhance medication treatment quality.

Previous studies examining the effect of transitional care on PIM use have focused on interventions during hospital admission or discharge, with mixed findings on the efficacy of in-hospital interventions on postdischarge PIM use.^17-19 Studies on the impact of postdischarge transitional care on PIM use have been limited to localized pharmacist- or nurse practitioner–led interventions.^20,21 Thus, they are not generalizable to all Medicare beneficiaries because MA enrollees were not included in those studies. Furthermore, to our knowledge, no studies have compared PIM use during hospital transitions of care between TM and MA.

Our study adds to this literature by providing empirical evidence on the association of MA enrollment with changes in PIM use following hospitalization. Given MA plans’ financial incentives and emphasis on primary and transitional care, we expected MA enrollees to experience lower PIM use and a greater decrease in PIM use after hospital discharge than TM beneficiaries. Understanding the potential for MA plans to reduce PIM use during hospital transitions of care is an important step toward assessing the value of the MA program and identifying strategies to prevent unnecessary PIM use and associated adverse events (AEs).

METHODS

Data

This observational study used 2016-2019 TM claims and MA encounter data. We used the Part D Event files to obtain details on drug name, strength, days’ supply, and prescription fill date, which were used to construct measures of PIM utilization.²² We used the Medicare Provider Analysis and Review file for TM beneficiaries and the inpatient encounter files for MA enrollees to identify hospitalization episodes based on admission and discharge dates for hospital stays and the post–acute care destination after discharge.^23,24 We used the Master Beneficiary Summary File for beneficiary demographics, zip code, county of residence, and monthly Medicare enrollment details, including plan type and eligibility for Part D low-
income subsidies.²⁵

We also linked the Medicare data to the Area Health Resources Files to identify county-level health resources,²⁶ to the American Community Survey to identify zip code socioeconomic characteristics,²⁷ and to the Rural-Urban Commuting Area codes to classify beneficiary residence as rural or urban.²⁸ The Ohio State University Institutional Review Board deemed this retrospective study, which used researcher-identifiable secondary data, to be exempt from review and waived the informed consent requirement. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology guidelines for reporting observational studies.²⁹

Study Sample

We identified Medicare beneficiaries 65 years and older with continuous Part A, B, and D coverage through either TM or MA. We excluded beneficiaries younger than 65 years because we examined PIMs that are a clinical concern for adults 65 years and older. We excluded beneficiaries who switched between TM and MA during the observation period (from 180 days before admission to 180 days after discharge). Due to concerns about missing records in MA encounter data, we followed established, validated methods to limit the analysis of MA enrollees to contracts with highly complete data.^30,31 We excluded beneficiaries who lived outside the 50 US states or Washington, DC, or who were enrolled in Program of All-Inclusive Care for the Elderly plans, cost plans, Medicare Medical Savings Account plans, or private fee-for-service plans, which have different structures and financial incentives and may allow concurrent enrollment in a stand-alone PDP and a separate MA plan. eAppendix 1 (eAppendices available at ajmc.com) displays a diagram of the sample selection.

The unit of analysis was a hospitalization episode, defined as an inpatient stay resulting in a live discharge to home with or without home health care services. We excluded hospitalizations that ended in transfer to a skilled nursing facility because a portion of PIM use in this setting may not be observed as outpatient prescription dispensing events.³² Based on the Healthcare Effectiveness Data and Information Set (HEDIS) measure for transitions of care,³³ when patients were transferred to another hospital or readmitted within 30 days of discharge, the hospitalization episode was defined as the time between the first admission date and the last discharge date. Following the existing literature on medication management, the observation period was defined as 180 days before hospital admission through 180 days after discharge.^18,34,35 To accommodate the observation period using 2016-2019 data, we included hospitalization episodes with an admission date on or after June 29, 2016, and a discharge date on or before July 4, 2019. Beneficiaries could have multiple hospitalizations during the study period.

Outcome Measure

We defined PIM use based on the HEDIS list of high-risk medications for adults 65 years and older.³³ High-risk medications, such as anticholinergics, long-acting sulfonylureas, and benzodiazepines, incur risks that outweigh the benefits for older patients because the AEs become more severe or the therapeutic benefits decrease as patients age.³⁶ We measured PIM use as a binary indicator of any outpatient prescription dispensing for a PIM during the prehospitalization period (180 days prior to admission) and the posthospitalization period (180 days after discharge). eAppendix 2 provides additional measurement details.

Exposure

The main explanatory variable was enrollment in MA, compared with TM as the control group, during the observation period.

Statistical Analysis

Because differences in health care needs and preferences may influence beneficiaries’ selection of TM or MA,^37,38 we used inverse probability of treatment weighting (IPTW) to balance observed characteristics between TM and MA.³⁹ Details are provided in eAppendix 2. We ensured that the sample characteristics were balanced by comparing the standardized mean difference before and after IPTW.⁴⁰ We limited the analysis to hospitalization episodes for which the propensity score distributions of TM and MA overlapped (eAppendix 3).⁴¹

We used linear regression to estimate a 2-group, 2-period difference-in-differences (DID) model for the association between MA enrollment and changes in PIM use following hospitalization. The DID estimate represents the percentage-point change in PIM use from prehospitalization to post hospitalization in MA relative to TM. We used robust SEs clustered by individuals to account for individuals with multiple hospitalization episodes during the study period.

Sensitivity Analyses

First, we repeated the DID analysis while restricting the sample based on PIM use before hospitalization. We analyzed (1) incident PIM use among beneficiaries with no PIM use before hospitalization and (2) PIM cessation among beneficiaries with any PIM use before hospitalization. Second, because the clinical concern with PIM utilization generally increases with age, we repeated the DID analysis within the following age groups: 65 to 74 years, 75 to 84 years, and 85 years and older. Third, we repeated the DID analysis on a sample that included hospitalization episodes resulting in discharge to a skilled nursing facility. Fourth, we repeated the DID analysis for alternate specifications of the outcome. To avoid overidentifying preadmission PIM use, we defined no preadmission PIM use to include episodes with no days covered by a PIM dispensing event 30 days prior to admission. We also considered preadmission and postdischarge observation periods of 30 days and 90 days to identify any PIM use. Finally, we measured the number of days covered by any PIM, based on prescription dispensing event dates and days’ supply. Calendar days with overlapping days’ supply were counted as 1 covered day.

RESULTS

As shown in Table 1, the study sample included 220,855 hospitalization episodes for 190,001 TM beneficiaries and 454,921 hospitalization episodes for 387,617 MA enrollees. The unweighted mean (SD) age was 77.1 (7.7) years in TM and 76.7 (7.3) years in MA. Women comprised 56.9% of TM hospitalization episodes and 55.0% of MA episodes. After IPTW, the TM and MA groups were balanced on all characteristics during the observation period, with standardized mean differences less than 0.1 (Figure).

The unadjusted prevalence of PIM use in TM was 23.90% preadmission and 23.29% post discharge, and it was 21.82% preadmission and 21.47% post discharge in MA (Table 2). The types of PIMs used were highly similar in TM and MA, with the most common being first-generation antihistamines such as meclizine and hydroxyzine, tricyclic antidepressants such as amitriptyline, and long-acting sulfonylureas such as glimepiride (Amaryl) (eAppendix 4). Relative to the prehospitalization period, PIM use declined for both TM and MA post discharge. The decrease in PIM use following hospitalization was 0.25 percentage points (PP) (95% CI, 0.02-0.49) less in MA than in TM.

As shown in Table 3, 10.25% of TM episodes and 9.35% of MA episodes had incident PIM use after hospital discharge. Compared with TM, MA was 0.90 PP (95% CI, –1.11 to –0.69) less likely to start a new PIM. Among hospitalization episodes for beneficiaries with prehospitalization PIM use, 64.84% of TM episodes and 64.92% of MA episodes continued to have at least 1 PIM after hospitalization, although the difference was not significant (0.08 PP; 95% CI, –0.52 to 0.67).

In the stratified analysis by age group, the association between MA enrollment and posthospitalization PIM use was similar in magnitude to that in the main analysis, although the TM-MA differences within each age group were not significant. The DID estimate was 0.16 PP (95% CI, –0.21 to 0.53) for those aged 65 to 74 years, 0.34 PP (95% CI, –0.02 to 0.71) for those aged 75 to 84 years, and 0.32 PP (95% CI, –0.17 to 0.81) for those 85 years and older (Table 4).

The results of the sensitivity analysis for the sample, including hospitalization episodes resulting in discharge to a skilled nursing facility, were similar to the main analysis (eAppendices 5-6). TM had a mean of 21.33 days covered by any PIM preadmission and 20.43 days post discharge vs 19.21 days preadmission and 18.80 days post discharge in MA. The DID estimates for the alternate outcome specifications were similar to the main analysis (eAppendices 7-9).

DISCUSSION

Transitions of care surrounding hospitalization present opportunities to evaluate appropriate medication use and discontinue PIMs, but they also present challenges that may lead to new PIM use. In this study, we found that among Medicare beneficiaries, PIM use is prevalent before hospital admission and decreases only modestly following hospital discharge. Although MA plans typically have greater incentives than TM to reduce PIM use post hospitalization, we observed that MA and TM experienced similar changes in PIM use.

Our findings suggest that hospitalization and transitional care following discharge may not provide adequate opportunities for MA plans to reduce PIM use. Although PIMs, by definition, should be avoided in all adults 65 years and older, prescribers may determine that PIM use is appropriate in certain clinical situations on a case-by-case basis. Because PIM use was lower in MA than in TM before hospitalization, there may be less room for improvement following discharge. That is, because MA enrollees are more likely than TM beneficiaries to have a PCP,¹⁵ deprescribing PIMs may be more likely to occur during routine outpatient visits, such as annual wellness visits,⁴² than during postdischarge PCP follow-up. It is also possible that the higher prehospital PIM use in TM may have led to a greater proportion of admissions due to an AE related to PIM use.⁴³ Hospitalization resulting from PIM use would be more likely to prompt the discontinuation of the offending PIM than a routine medication review and reconciliation following a hospitalization for an unrelated reason, which may partially explain the greater decrease in PIM use post hospitalization in TM.

The small differences in PIM use surrounding hospitalization between TM and MA were reflected in separate analyses by age group. PIM use may result in more severe AEs with increasing age.⁴⁴ Therefore, any differences between TM and MA would be most impactful on PIM use around hospitalization for older beneficiaries who would be at higher risk of PIM-related AEs. However, even the oldest age group (≥ 85 years) experienced negligible differences between TM and MA in the change in PIM use surrounding hospitalization.

We also found that MA enrollment was associated with lower incident PIM use but no difference in stopping PIM use vs TM. This suggests that PIM use is a complex clinical issue that may be inconsistently influenced by insurance type. It is possible that MA plans’ focus on primary care and transitional care services is better suited to prevent the start of new PIMs than to discontinue PIMs that patients were taking prior to admission. Overprescribing is a particular concern for older adults who, compared with younger individuals, tend to have more disease conditions for which medications are prescribed. Providers are more likely to prescribe PIMs when challenged with balancing multiple disease-specific clinical guidelines, symptom management, and time constraints.^45,46 Meanwhile, providers may hesitate to deprescribe PIMs started by another prescriber, particularly given incomplete information about past clinical rationales.⁴⁷

This study has several important implications for policy and clinical practice. PIM use is common among all Medicare beneficiaries, and there are small decreases in PIM use following hospitalization. Accordingly, there are untapped opportunities to reduce PIM use through transitional care. Despite the fact that MA has greater financial incentives for PIM reduction following hospitalization through PCP follow-up, transitional care services, and robust MTM programs, MA enrollees have minimal changes in PIM use following hospitalization, even for enrollees 85 years and older. However, we did not observe utilization of these postdischarge medication management services or their direct impact on PIM use, which were beyond the scope of this study. PDPs and MA plans with prescription drug coverage have substantial flexibility in operating their MTM programs and may provide medication reviews through third-party MTM providers.⁴⁸ As such, financial incentives for MA plans to reduce the medical costs of AEs by reducing PIM use post discharge may not translate to the operations of a contracted MTM provider. Future research should examine how heterogeneity among MTM programs’ operations may impact PIM use.

Furthermore, some high-risk medication use may be clinically appropriate in specific situations. Although the types of PIMs used in TM and MA were similar, it is possible that the clinical scenarios for prescribing PIMs differed between the two. However, prescription drug claims do not capture the clinical rationale for PIM use, which may explain some portion of PIM use. Future research should examine specific drug-disease interactions and prescribers’ exceptions to PIM guideline recommendations. It is also possible that the full impact of differences in insurance coverage and transitional care may not be observed immediately after discharge but may require continued follow-up to reduce chronic PIM use,⁴⁹ especially if patients or providers are unwilling or unable to make multiple simultaneous changes.⁵⁰ Future studies should assess how both short- and long-term PIM use changes following transitions of care.

Limitations

This study has several limitations. First, we did not have data on whether hospitalizations were due to PIM use, which may affect the likelihood of changing PIM use after discharge. Second, prescription drug claims may overestimate PIM exposure for patients who fill prescriptions but do not take them. Third, PIM use in this study was limited to outpatient prescription drugs, and the findings may not be generalizable to institutional settings, such as skilled nursing facilities. Fourth, the PIM use outcomes in this study do not capture dose reduction, which is another beneficial step in deprescribing PIMs.⁴⁷ Fifth, although we used IPTW to make the TM and MA groups as comparable as possible, it is plausible that unobservable confounders, such as patient preferences for prescription medications, affect both MA enrollment and posthospitalization PIM use. Finally, we were unable to account for the clinical context of PIM use because such information is not available in prescription drug claims.

CONCLUSIONS

In this study of Medicare beneficiaries 65 years and older, we found that MA had lower PIM use than TM before and after hospitalization. However, TM experienced greater decreases in PIM use following hospitalization. Despite stronger financial incentives and an emphasis on primary and transitional care, MA does not provide a clear advantage over TM in reducing PIM use post hospitalization.

Author Affiliations: Department of Health Policy and Management, College of Public Health, University of Arkansas for Medical Sciences (EWR), Little Rock, AR; Department of Health Administration and Policy, College of Public Health, George Mason University (JJ), Fairfax, VA; Division of Pharmacy Practice and Science, College of Pharmacy, The Ohio State University (MD), Columbus, OH; Department of Internal Medicine, College of Medicine, The Ohio State University (TRG), Columbus, OH; Division of Epidemiology (JW) and Division of Health Services Management and Policy (WYX), College of Public Health, The Ohio State University, Columbus, OH.

Source of Funding: National Institute on Aging (R01AG069352).

Author Disclosures: Dr Jung reports receipt of National Institute on Aging grant R01AG069352. Dr Gure has been an unpaid advisory board member of the Geriatric Medicine Advisory Panel for the National Committee for Quality Assurance since 2021. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (EWR, WYX); acquisition of data (JJ); analysis and interpretation of data (EWR, MD, TRG, JW, WYX); drafting of the manuscript (EWR, TRG, WYX); critical revision of the manuscript for important intellectual content (EWR, JJ, MD, TRG, JW, WYX); statistical analysis (EWR, MD, JW); administrative, technical, or logistic support (EWR, WYX); and supervision (JJ, WYX).

Address Correspondence to: Eli Wei Raver, PhD, PharmD, University of Arkansas for Medical Sciences, 4301 W Markham St #820, Little Rock, AR 72205. Email: eraver@uams.edu.

REFERENCES

1. Kripalani S, Jackson AT, Schnipper JL, Coleman EA. Promoting effective transitions of care at hospital discharge: a review of key issues for hospitalists. J Hosp Med. 2007;2(5):314-323. doi:10.1002/jhm.228

2. Kahn JM, Angus DC. Going home on the right medications: prescription errors and transitions of care. JAMA. 2011;306(8):878-879. doi:10.1001/jama.2011.1209

3. Weir DL, Lee TC, McDonald EG, et al. Both new and chronic potentially inappropriate medications continued at hospital discharge are associated with increased risk of adverse events. J Am Geriatr Soc. 2020;68(6):1184-1192. doi:10.1111/jgs.16413

4. Tian F, Chen Z, Zeng Y, Feng Q, Chen X. Prevalence of use of potentially inappropriate medications among older adults worldwide: a systematic review and meta-analysis. JAMA Netw Open. 2023;6(8):e2326910. doi:10.1001/jamanetworkopen.2023.26910

5. Forster AJ, Murff HJ, Peterson JF, Gandhi TK, Bates DW. Adverse drug events occurring following hospital discharge. J Gen Intern Med. 2005;20(4):317-323. doi:10.1111/j.1525-1497.2005.30390.x

6. Transitional care management (TCM): Medicare codes and toolkit. American Academy of Family Physicians. Accessed October 16, 2024. https://www.aafp.org/family-physician/practice-and-career/getting-paid/coding/transitional-care-management.html

7. Beyond the discharge: principles of effective care transitions between settings. American College of Physicians. 2023. Accessed November 20, 2024. https://www.acponline.org/sites/default/files/acp-policy-library/policies/beyond_the_discharge_principles_of_effective_care_transitions_between_settings_2023.pdf

8. Haag JD, Davis AZ, Hoel RW, et al. Impact of pharmacist-provided medication therapy management on healthcare quality and utilization in recently discharged elderly patients. Am Health Drug Benefits. 2016;9(5):259-268.

9. Shcherbakova N, Tereso G. Clinical pharmacist home visits and 30-day readmissions in Medicare Advantage beneficiaries. J Eval Clin Pract. 2016;22(3):363-368. doi:10.1111/jep.12495

10. Jung J, Carlin C, Feldman R. Medicare Advantage has lower resource use and better quality of care than traditional Medicare. Am J Health Econ. 2025;11(4):628-667. doi:10.1086/730436

11. Almodovar AS, Axon DR, Coleman AM, Warholak T, Nahata MC. The effect of plan type and
comprehensive medication reviews on high-risk medication use. J Manag Care Spec Pharm. 2018;24(5):416-422. doi:10.18553/jmcp.2018.24.5.416

12. Stuart B, Hendrick FB, Shen X, et al. Eligibility for and enrollment in Medicare Part D medication therapy management programs varies by plan sponsor. Health Aff (Millwood). 2016;35(9):1572-1580. doi:10.1377/hlthaff.2016.0258

13. Report to the congress: benefit design and cost sharing in Medicare Advantage plans. Medicare Payment Advisory Committee. December 2004. Accessed November 8, 2024. https://www.medpac.gov/wp-content/uploads/import_data/scrape_files/docs/default-source/reports/Dec04_CostSharing.pdf

14. Ippolito B, Trish E, Vabson B. Expected out-of-pocket costs: comparing Medicare Advantage with fee-for-service Medicare. Health Aff (Millwood). 2024;43(11):1502-1507. doi:10.1377/hlthaff.2024.00295

15. Jacobson G, Leonard F, Sciupac E, Rapoport R. What do Medicare beneficiaries value about their coverage? The Commonwealth Fund. February 22, 2024. Accessed July 15, 2024. https://www.commonwealthfund.org/publications/surveys/2024/feb/what-do-medicare-beneficiaries-value-about-their-coverage

16. Medicare 2024 Part C & D Display Measure Technical Notes. CMS. Updated December 14, 2023. Accessed April 22, 2024. https://www.cms.gov/medicare/health-drug-plans/part-c-d-performance-data

17. Atey TM, Peterson GM, Salahudeen MS, Wimmer BC. The impact of partnered pharmacist medication charting in the emergency department on the use of potentially inappropriate medications in older people. Front Pharmacol. 2023;14:1273655. doi:10.3389/fphar.2023.1273655

18. Debacq C, Bourgueil J, Aidoud A, et al. Persistence of effect of medication review on potentially inappropriate prescriptions in older patients following hospital discharge. Drugs Aging. 2021;38(3):243-252. doi:10.1007/s40266-020-00830-6

19. Morandi A, Vasilevskis E, Pandharipande PP, et al. Inappropriate medication prescriptions in elderly adults surviving an intensive care unit hospitalization. J Am Geriatr Soc. 2013;61(7):1128-1134. doi:10.1111/jgs.12329

20. Gurwitz JH, Kapoor A, Garber L, et al. Effect of a multifaceted clinical pharmacist intervention on medication safety after hospitalization in persons prescribed high-risk medications: a randomized clinical trial. JAMA Intern Med. 2021;181(5):610-618. doi:10.1001/jamainternmed.2020.9285

21. Vasilevskis EE, Shah AS, Hollingsworth EK, et al. Deprescribing medications among older adults from end of hospitalization through postacute care: a Shed-MEDS randomized clinical trial. JAMA Intern Med. 2023;183(3):223-231. doi:10.1001/jamainternmed.2022.6545

22. Part D Event. Research Data Assistance Center. Accessed November 20, 2024. https://resdac.org/cms-data/files/pde

23. Medicare Provider Analysis and Review. Research Data Assistance Center. Accessed November 20, 2024. https://resdac.org/cms-data/files/medpar

24. Encounter Inpatient. Research Data Assistance Center. Accessed November 20, 2024. https://resdac.org/cms-data/files/ip-encounter

25. Master Beneficiary Summary File Base A/B/C/D. Research Data Assistance Center. Accessed November 20, 2024. https://resdac.org/cms-data/files/mbsf-base

26. Area Health Resources Files. Health Resources and Services Administration. Updated July 31, 2023. Accessed November 20, 2024. https://data.hrsa.gov/topics/health-workforce/ahrf

27. American Community Survey (ACS). US Census Bureau. Updated October 22, 2024. Accessed November 20, 2024. https://www.census.gov/programs-surveys/acs

28. RUCA data. Rural Health Research Center. Accessed November 10, 2022. https://depts.washington.edu/uwruca/ruca-uses.php

29. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007;147(8):573-577. doi:10.7326/0003-4819-147-8-200710160-00010

30. Jung J, Carlin C, Feldman R. Measuring resource use in Medicare Advantage using Encounter data. Health Serv Res. 2022;57(1):172-181. doi:10.1111/1475-6773.13879

31. Jung J, Carlin C, Feldman R, Tran L. Implementation of resource use measures in Medicare Advantage. Health Serv Res. 2022;57(4):957-962. doi:10.1111/1475-6773.13970

32. Medicare Drug Coverage Under Part A, Part B, and Part D. CMS. Updated March 2018. Accessed July 7, 2025. https://www.cms.gov/files/document/medicare-drug-coverage-under-parts-b-and-d-508-final-v2pdf

33. HEDIS 2020 Volume 2: Technical Specifications for Health Plans. National Committee for Quality Assurance; 2020. Accessed April 4, 2024.

34. Growdon ME, Gan S, Yaffe K, et al. New psychotropic medication use among Medicare beneficiaries with dementia after hospital discharge. J Am Geriatr Soc. 2023;71(4):1134-1144. doi:10.1111/jgs.18161

35. Bongiovanni T, Gan S, Finlayson E, et al. Prolonged use of newly prescribed gabapentin after surgery. J Am Geriatr Soc. 2022;70(12):3560-3569. doi:10.1111/jgs.18005

36. 2019 American Geriatrics Society Beers Criteria Update Expert Panel. American Geriatrics Society 2019 Updated AGS Beers Criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc. 2019;67(4):674-694. doi:10.1111/jgs.15767

37. Jacobson G, Cicchiello A, Sutton JP, Shah A. Medicare Advantage vs. traditional Medicare: how do beneficiaries’ characteristics and experiences differ? The Commonwealth Fund. October 14, 2021. Accessed November 10, 2022. https://www.commonwealthfund.org/publications/issue-briefs/2021/oct/medicare-advantage-vs-traditional-medicare-beneficiaries-differ

38. Agarwal R, Connolly J, Gupta S, Navathe AS. Comparing Medicare Advantage and traditional Medicare: a systematic review. Health Aff (Millwood). 2021;40(6):937-944. doi:10.1377/hlthaff.2020.02149

39. Austin PC, Mamdani MM. A comparison of propensity score methods: a case-study estimating the effectiveness of post-AMI statin use. Stat Med. 2006;25(12):2084-2106. doi:10.1002/sim.2328

40. Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med. 2009;28(25):3083-3107. doi:10.1002/sim.3697

41. Garrido MM, Kelley AS, Paris J, et al. Methods for constructing and assessing propensity scores. Health Serv Res. 2014;49(5):1701-1720. doi:10.1111/1475-6773.12182

42. Park S, Nguyen AM. Determinants and effectiveness of annual wellness visits among Medicare beneficiaries in 2020. Fam Pract. 2024;41(2):203-206. doi:10.1093/fampra/cmad108

43. Clark CM, Shaver AL, Aurelio LA, et al. Potentially inappropriate medications are associated with increased healthcare utilization and costs. J Am Geriatr Soc. 2020;68(11):2542-2550. doi:10.1111/jgs.16743

44. Hilmer SN, McLachlan AJ, Le Couteur DG. Clinical pharmacology in the geriatric patient. Fundam Clin Pharmacol. 2007;21(3):217-230. doi:10.1111/j.1472-8206.2007.00473.x

45. Bennett F, Ferner R, Sofat R. Overprescribing and rational therapeutics: barriers to change and opportunities to improve. Br J Clin Pharmacol. 2021;87(1):34-38. doi:10.1111/bcp.14291

46. Boyd CM, Darer J, Boult C, Fried LP, Boult L, Wu AW. Clinical practice guidelines and quality of care for older patients with multiple comorbid diseases: implications for pay for performance. JAMA. 2005;294(6):716-724. doi:10.1001/jama.294.6.716

47. Scott IA, Hilmer SN, Reeve E, et al. Reducing inappropriate polypharmacy: the process of deprescribing. JAMA Intern Med. 2015;175(5):827-834. doi:10.1001/jamainternmed.2015.0324

48. Thomas DJ, Tran J, eds. The Medication Therapy Management Pharmacist Reference Book. National Board of Medication Therapy Management; 2020. Accessed June 14, 2024. https://www.nbmtm.org/mtm-reference/

49. Jungo KT, Choudhry NK, Chaitoff A, Lauffenburger JC. Associations between sex, race/ethnicity, and age and the initiation of chronic high-risk medication in US older adults. J Am Geriatr Soc. 2024;72(12):3705-3718. doi:10.1111/jgs.19173

50. Ibrahim K, Cox NJ, Stevenson JM, Lim S, Fraser SDS, Roberts HC. A systematic review of the evidence for deprescribing interventions among older people living with frailty. BMC Geriatr. 2021;21(1):258. doi:10.1186/s12877-021-02208-8