Cost-Effectiveness of Medication Therapy Management Program Across Racial and Ethnic Groups Among Medicare Beneficiaries.

OBJECTIVES: Equity and effectiveness of the medication therapy management (MTM) program in Medicare has been a policy focus since its inception. The objective of this study was to evaluate the cost-effectiveness of the Medicare MTM program in improving medication utilization quality across racial and ethnic groups. METHODS: This study analyzed 2017 Medicare data linked to the Area Health Recourses File. A propensity score was used to match MTM enrollees and nonenrollees, and an incremental cost-effectiveness ratio between the 2 groups was calculated. Effectiveness was measured as the proportion of appropriate medication utilization based on medication utilization measures developed by Pharmacy Quality Alliance. Net monetary benefits were compared across racial and ethnic groups at various societal willingness-to-pay (WTP) thresholds. The 95% confidence intervals were obtained by nonparametric bootstrapping. RESULTS: MTM dominated non-MTM among the total sample (N = 699 992), as MTM enrollees had lower healthcare costs ($31 135.89 vs $32 696.69) and higher proportions of appropriate medication utilization (87.47% vs 85.31%) than nonenrollees. MTM enrollees had both lower medication costs ($10 681.21 vs $11 003.08) and medical costs ($20 454.68 vs $21 693.61) compared with nonenrollees. The cost-effectiveness of MTM was higher among Black patients than White patients across the WTP thresholds. For instance, at a WTP of $3006 per percentage point increase in effectiveness, the net monetary benefit for Black patients was greater than White patients by $2334.57 (95% confidence interval $1606.53-$3028.85). CONCLUSIONS: MTM is cost-effective in improving medication utilization quality among Medicare beneficiaries and can potentially reduce disparities between Black and White patients. Expansion of the current MTM program could maximize these benefits.

Effects of Medicare comprehensive medication review on racial/ethnic disparities in nonadherence to statin medications among patients with Alzheimer's Disease: an observational analysis.

BACKGROUND: Alzheimer's Disease (AD) is the mostcommon cause of dementia, a neurological disorder characterized by memory loss and judgment impairment. Hyperlipidemia, a commonly co-occurring condition, should be treated to prevent associated complications. Medication adherence may be difficult for individuals with AD due to the complexity of AD management. Comprehensive Medication Reviews (CMRs), a required component of Medicare Part D Medication Therapy Management (MTM), have been shown to improve medication adherence. However, many MTM programs do not target AD. Additionally, racial/ethnic disparities in MTM eligibility have been revealed. Thus, this study examined the effects of CMR receipt on reducing racial/ethnic disparities in the likelihood of nonadherence to hyperlipidemia medications (statins) among the AD population. METHODS: This retrospective study used 2015-2017 Medicare data linked to the Area Health Resources Files. The likelihood of nonadherence to statin medications across racial/ethnic groups was compared between propensity-score-matched CMR recipients and non-recipients in a ratio of 1 to 3. A difference-in-differences method was utilized to determine racial/ethnic disparity patterns using a logistic regression by including interaction terms between dummy variables for CMR receipt and each racial/ethnic minority group (non-Hispanic Whites, or Whites, as reference). RESULTS: The study included 623,400 Medicare beneficiaries. Blacks and Hispanics had higher statin nonadherence than Whites: Compared to Whites, Blacks' nonadherence rate was 4.53% higher among CMR recipients and 7.35% higher among non-recipients; Hispanics' nonadherence rate was 2.69% higher among CMR recipients and 7.38% higher among non-recipients. Differences in racial/ethnic disparities between CMR recipients and non-recipients were significant for each minority group (p < 0.05) except Others. The difference between Whites and Hispanics in the odds of statin nonadherence was 11% lower among CMR recipients compared to non-recipients (OR = 0.89; 95% Confidence Interval = 0.85-0.94 for the interaction term between dummy variables for CMR and Hispanics). Interaction terms between dummy variables for CMR and other racial/ethnic minorities were not significant. CONCLUSIONS: Receiving a CMR was associated with a disparity reduction in nonadherence to statin medications between Hispanics and Whites among patients with AD. Strategies need to be explored to increase the number of MTM programs that target AD and promote CMR completion.

Racial and ethnic disparities associated with the measure for drug-drug interactions among Medicare beneficiaries.

BACKGROUND: Drug-drug interactions (DDIs) cause many preventable hospitalizations and admissions. Efforts have been made to raise DDI awareness and reduce DDI occurrence; for example, Medicare Part D Star Ratings, a health plan quality assessment program, included a DDI measure. Previous research reported racial and ethnic disparities in health services utilization and that racial and ethnic minorities, compared with non-Hispanic whites (whites), may be less likely to be targeted for a similar measure, a Star Ratings adherence measure for diabetes medications. OBJECTIVE: This study aimed to investigate whether any racial and ethnic disparities are associated with the DDI measure in Part D Star Ratings among Medicare populations with diabetes, hypertension, and hyperlipidemia. METHODS: This cross-sectional study analyzed a 2017 Medicare Part D data sample, including 3,960,813 beneficiaries. Because the inclusion in the denominator of the Star Ratings DDI measure was determined by the use of a list of target medications, the likelihood of using a listed target medication was compared between racial and ethnic minorities and whites. Individuals with diabetes, hypertension, and hyperlipidemia were included in the analysis owing to the high prevalence of these conditions. Patient- and community-level characteristics were adjusted by logistic regression. RESULTS: Of the entire study sample, 26.2% used a target medication. Compared with whites, most racial and ethnic minorities were less likely to use a target medication. For example, among individuals with diabetes, blacks, Hispanics, Asians/Pacific Islanders, and others had, respectively, 14% (odds ratio 0.86 [95% CI 0.84-0.88]), 5% (0.95 [0.93-0.98]), 12% (0.88 [0.84-0.92]), and 10% (0.90 [0.87-0.93]) lower odds compared with whites. Findings were similar among hypertension and hyperlipidemia cohorts, except that Hispanics had similar odds of use as whites. CONCLUSION: Most racial and ethnic minorities may have lower likelihood of being targeted for the DDI measure compared with whites. Future studies should examine whether these disparities affect health outcomes and devise new DDI measures for racial and ethnic minorities.

Effects of Medicare Part D medication therapy management on racial/ethnic disparities in adherence to antidementia medications among patients with Alzheimer's disease and related dementias: An observational study.

Background: Evidence is sparse on the effects of Medicare medication therapy management (MTM) on racial/ethnic disparities in medication adherence among patients with Alzheimer's disease and related dementias. Objectives: This study examined the Medicare MTM program's effects on racial/ethnic disparities in the adherence to antidementia medications among patients with Alzheimer's disease and related dementias. Methods: This is a retrospective analysis of 100% of 2010-2017 Medicare Parts A, B, and D data linked to Area Health Resources Files. The study outcome was nonadherence to antidementia medications, and intervention was defined as new MTM enrollment in 2017. Propensity score matching was conducted to create intervention and comparison groups with comparable characteristics. A difference-in-differences model was employed with logistic regression, including interaction terms of dummy variables for the intervention group and racial/ethnic minorities. Results: Unadjusted comparisons revealed that Black, Hispanic, and Asian/Pacific Islander patients were more likely to be nonadherent than non-Hispanic White (White) patients in 2016. Differences in odds of nonadherence between Black and White patients among the intervention group were lower in 2017 than in 2016 by 27% (odds ratios [OR]: 0.73, 95% confidence interval [CI]: 0.65-0.82). A similar lowering was seen between Hispanic and White patients by 26% (OR: 0.74, 95% CI: 0.63-0.87). MTM enrollment was associated with reduced disparities in nonadherence for Black-White patients of 33% (OR: 0.67, 95% CI: 0.57-0.78) and Hispanic-White patients of 19% (OR: 0.81, 95% CI: 0.67-0.99). Discussion: The Medicare MTM program was associated with lower disparities in adherence to antidementia medications between Black and White patients, and between Hispanic and White patients in the population with Alzheimer's disease and related dementias. Conclusions: Expanding the MTM program may particularly benefit racial/ethnic minorities in Alzheimer's disease and related dementia care.

Racial and ethnic disparities in Medicare Part D medication therapy management services utilization.

Background: The Medicare Part D medication therapy management (MTM) program has positive effects on medication and health service utilization. However, little is known about its utilization, much less so about the use among racial and ethnic minorities. Objective: To examine MTM service utilization among older Medicare beneficiaries and to identify any racial and ethnic disparity patterns. Methods: A retrospective cross-sectional analysis of 2017 Medicare administrative data, linked to the Area Health Resources Files. Fourteen outcomes related to MTM service nature, initiation, quantity, and delivery were examined using logistic, negative binomial, and Cox proportional hazards regression models. Results: Racial and ethnic disparities were found with varying patterns across outcomes. For example, compared with White patients, the odds of opting out of MTM were 8% higher for Black patients (odds ratio [OR] = 1.08, 95% confidence interval [CI] = 1.03-1.14), 57% higher for Hispanic patients (OR = 1.57, 95% CI = 1.42-1.72), and 57% higher for Asian patients (OR = 1.57, 95% CI = 1.33-1.85). The odds of continuing MTM from the previous years were 12% lower for Black patients (OR = 0.88, 95% CI = 0.86-0.90) and 3% lower for other patients (OR = 0.97, 95% CI = 0.95-0.99). In addition, the probability of being offered a comprehensive medication review (CMR) after MTM enrollment was 9% lower for Hispanic patients (hazard ratio [HR] = 0.91, 95% CI = 0.85-0.97), 9% lower for Asian patients (HR = 0.91, 95% CI = 0.87-0.94), and 3% lower for other patients (HR = 0.97, 95% CI = 0.95-0.99). Hispanic and Asian patients were more likely to have someone other than themselves receive a CMR. Conclusions: Racial and ethnic disparities in MTM service utilization were identified. Although the disparities in specific utilization outcomes vary across racial/ethnic groups, it is evident that these disparities exist and may result in vulnerable communities not fully benefiting from the MTM services. Causes of the disparities should be explored to inform future reform of the Medicare Part D MTM program.

Racial and ethnic disparities in the enrolment of medicare medication therapy management programs.

Objectives: Racial/ethnic disparities have been found in prior literature examining enrolment in Medicare medication therapy management programs. However, those studies were based on various eligibility scenarios because enrolment data were unavailable. This study tested for potential disparities in enrolment using actual MTM enrolment data. Methods: Medicare Parts A&B claims, Medication Therapy Management Data Files, and the Area Health Resources File from 2013 to 2014 and 2016 to 2017 were analysed in this retrospective analysis. An adjusted logistic regression compared odds of enrolment between racial/ethnic minorities and non-Hispanic Whites (Whites) in the total sample and subpopulations with diabetes, hypertension, or hyperlipidaemia. Trends in disparities were analysed by including interaction terms in regressions between dummy variables for race/ethnic minority groups and period 2016-2017. Key Findings: Disparities in MTM enrolment were detected between Blacks and Whites with diabetes in 2013-2014 (Odds Ratio = 0.78, 95% Confidence Interval = 0.75-0.81). This disparity improved from 2013-2014 to 2016-2017 for Blacks (Odds Ratio=1.08, 95% Confidence Interval = 1.04-1.11) but persisted in 2016-2017 (Odds Ratio = 0.84, 95% Confidence Interval = 0.81-0.87). A disparity was identified between Blacks and Whites with hypertension in 2013-2014 (Odds Ratio = 0.92, 95% Confidence Interval = 0.89-0.95) but not in 2016-2017. Enrolment for all groups, however, declined between periods. For example, in the total sample, the odds of enrolment declined from 2013-2014 to 2016-2017 by 22% (Odds Ratio=0.78, 95% Confidence Interval=0.75-0.81). Conclusions: Racial disparities in MTM enrolment were found between Blacks and Whites among Medicare beneficiaries with diabetes in both periods and among individuals with hypertension in 2013-2014. As overall enrolment fell between periods, concerns about program enrolment remain.

Solving racial/ethnic disparities associated with Medicare Part D Star Ratings.

OBJECTIVE: Medicare Part D Star Ratings are instrumental in shaping healthcare quality improvement efforts. However, the calculation metrics for medication performance measures for this program have been associated with racial/ethnic disparities. In this study, we aimed to explore whether an alternative program, named Star Plus by us that included all medication performance measures developed by Pharmacy Quality Alliance and applicable to our study population, would reduce such disparities among Medicare beneficiaries with diabetes, hypertension, and/or hyperlipidemia. METHOD: We conducted an analysis of a 10% random sample of Medicare A/B/D claims linked to the Area Health Resources File. Multivariate logistic regressions with minority dummy variables were used to examine racial/ethnic disparities in measure calculations of Star Ratings and Star Plus, respectively. RESULTS: Adjusted results indicated that relative to non-Hispanic Whites (Whites), racial/ethnic minorities had significantly lower odds of being included in the Star Ratings measure calculations: the odds ratios (ORs) for Blacks, Hispanics, Asians, and Others were 0.68 (95% confidence interval [CI] = 0.66-0.71), 0.73 (CI = 0.69-0.78), 0.88 (CI = 0.82-0.93), and 0.92 (CI = 0.88-0.97), respectively. In contrast, every beneficiary in the sample was included in Star Plus. Further, racial/ethnic minorities had significantly higher increase in the odds of being included in measure calculation in Star Plus than Star Ratings. The ORs for Blacks, Hispanics, Asians, and Others were 1.47 (CI = 1.41-1.52), 1.37 (CI = 1.29-1.45), 1.14 (CI = 1.07-1.22), and 1.09 (CI = 1.03-1.14), respectively. CONCLUSIONS: Our study demonstrated that racial/ethnic disparities may be eliminated by including additional medication performance measures to Star Ratings.

Racial/ethnic disparities in the enrollment of Medication Therapy Management programs among Medicare beneficiaries with Alzheimer's disease and related dementias.

OBJECTIVE: Previous analysis of policy scenarios reported potential disparities in eligibility in the Medicare Medication Therapy Management (MTM) program. With recently released MTM data, this study aimed to determine if racial/ethnic disparities exist in MTM enrollment among Medicare beneficiaries with Alzheimer's disease and related dementias (ADRD). METHODS: Medicare claims/records (from 2013-2014 and 2016-2017) linked to the Area Health Resources File were examined. Included individuals were patients with ADRD and diabetes, hypertension or hyperlipidemia. The proportions of MTM enrollment were compared between non-Hispanic White (White) patients and racial/ethnic minority groups in descriptive analysis. Racial/ethnic disparities were then examined using a logistic regression adjusting for patient and community characteristics. Disparities across study periods were compared by estimating a logistic regression model with interaction terms between dummy variables for each racial/ethnic minority group and 2016-2017. RESULTS: In unadjusted analyses, minorities had higher enrollment proportions than Whites. In 2016-2017, for example, enrollment percentages for Whites, Blacks, Hispanics, Asian/Pacific Islanders (Asians) and Others were respectively 14.44%, 16.71%, 19.83%, 16.66%, and 17.78%. In adjusted analyses, Blacks had lower enrollment odds than Whites within all cohorts. In the entire study sample in 2016-2017, for example, Blacks with ADRD had 9% lower odds of MTM enrollment (odds ratio 0.91, 95% confidence interval [CI] = 0.86-0.97) than Whites. These disparities decreased over time among the ADRD sample and all sub-groups. The interaction term between Blacks and 2016-2017, for instance, indicated that disparities were lowered by 11% (odds ratio 1.11, 95% CI = 1.05-1.16) across study periods among those with ADRD. CONCLUSIONS: Blacks with ADRD, and diabetes, hypertension or hyperlipidemia have lower likelihood of MTM enrollment than Whites. Racial disparities were reduced over time but not eliminated.

Using machine learning to classify patients on opioid use.

OBJECTIVES: High-frequent opioid use tends to increase an individual's risk of opioid use disorder, overdose and death. Thus, it is important to predict an individuals' opioid use frequency to improve opioid prescription utilization outcomes. METHODS: Individuals receiving at least one opioid prescription from 2016 to 2018 in the national representative data, Medical Expenditure Panel Survey, were included. This study applied five machine learning (ML) techniques, including support vector machine, random forest, neural network, gradient boosting and XGBoost (extreme gradient boosting), to predict opioid use frequency. This study compared the performance of these ML models with penalized logistic regression. The study outcome was whether an individual lied in the upper 10% of the opioid prescription distribution. Predictors were selected based on Gelberg-Andersen's Behavioral Model of Health Services Utilization. The prediction performance was assessed using the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC) in the test data. Patient characteristics as predictors for high-frequency use of opioids were ranked by the relative importance in prediction in the test data. KEY FINDINGS: Random forest and gradient boosting achieved the top values of both AUROC and AUPRC, outperforming logistic regression and three other ML methods. In the best performing model, the random forest, the following characteristics had high predictive power in the frequency of opioid use: age, number of chronic conditions, public insurance and self-perceived health status. CONCLUSIONS: The results of this study demonstrate that ML techniques can be a promising and powerful technique in predicting the frequency of opioid use and health outcomes.

Effects of the Medicare Part D comprehensive medication review on medication adherence among patients with Alzheimer's disease.

OBJECTIVE: Older patients with Alzheimer's disease (AD) are challenged with adhering to complex medication regimens. We examined effects of Comprehensive Medication Review (CMR), a required Medicare Part D Medication Therapy Management (MTM) program component, on medication adherence among AD patients. METHODS: This retrospective study analyzed 100% of 2016-2017 Medicare claims covering the entire United States, linked to Area Health Resources Files. Medicare beneficiaries aged ≥65 years were included. Propensity score matching identified comparable intervention and comparison groups with the intervention defined as receiving a CMR in 2017. A difference-in-differences analysis included in multivariate logistic regressions an interaction term between CMR receipt and year 2017. The outcome measured was nonadherence to diabetes, hypertension and hyperlipidemia medications, with nonadherence defined as proportion of days covered <80% for study medications. RESULTS: Unadjusted comparisons indicated the proportion of nonadherence for intervention group members decreased from 2016 to 2017 but increased for the comparison group. In adjusted analyses, reduction in medication nonadherence among the intervention group remained higher: odds ratios for the interaction term were 0.62 (95% confidence interval [CI] = 0.54-0.71), 0.54 (95% CI = 0.50-0.58) and 0.50 (95% CI = 0.47-0.53) respectively for diabetes, hypertension and hyperlipidemia medications. This suggests that the likelihood of nonadherence in the intervention group was respectively reduced by 38%, 46% and 50% more than the comparison group. CONCLUSIONS: CMR was found to reduce nonadherence to diabetes, hypertension and hyperlipidemia medications among older Medicare beneficiaries with AD. This provides evidence that the MTM program is effective for a population with unique medication compliance challenges.

Factors associated with medication nonadherence among Medicare low-income subsidy beneficiaries with diabetes, hypertension, and/or heart failure.

BACKGROUND: Previous studies have documented factors influencing medication nonadherence among the Medicare population, but few studies have examined medication nonadherence among the Medicare low-income subsidy (LIS) population. Furthermore, little is known about the factors associated with nonadherence among this population, especially those with prevalent chronic conditions such as type 2 diabetes, hypertension, or heart failure. OBJECTIVE: To examine factors associated with the likelihood of medication nonadherence among Medicare LIS recipients with type 2 diabetes, hypertension, or heart failure. METHODS: This was a retrospective analysis of 2012-2013 Medicare Parts A, B, and D claims (most recent available for this research) linked to the Area Health Resources Files. Beneficiaries aged 65 years or older with continuous Medicare coverage and receiving any LIS were included. Individuals were categorized into full LIS or partial LIS groups. Nonadherence was determined by the proportion of days covered less than 80% for specified oral type 2 diabetes, hypertension, and heart failure medications, as defined by the Pharmacy Quality Alliance. A multivariate logistic regression was used to determine and compare individual-level and community-level characteristics associated with nonadherence among the entire study sample, the full LIS group, and the partial LIS group. RESULTS: The study sample included 505,771 Medicare beneficiaries, with 448,509 (88.7%) receiving full LIS and 57,262 (11.3%) receiving partial LIS. The proportion of individuals nonadherent was higher among the full LIS population (33.2%) than that of the partial LIS population (30.8%). Among the entire population, younger age was associated with nonadherence (OR = 0.98; 95% CI = 0.98-0.99). Men were more likely to be nonadherent than women (OR = 1.12; 95% CI = 1.11-1.14). Compared with non-Hispanic Whites, racial/ethnic minorities had higher nonadherence. Compared with beneficiaries who were non-Hispanic White, the ORs for those who were Black, Hispanic, Asian, and other were 1.41 (95% CI = 1.38-1.43), 1.58 (95% CI = 1.55-1.61), 1.08 (95% CI = 1.05-1.11), and 1.63 (95% CI = 1.56-1.70), respectively. There were higher nonadherence rates among patients living in communities with lower socioeconomic characteristics, such as a metropolitan statistical area (MSA vs non-MSA; OR = 1.05, 95% CI = 1.04-1.07). A higher risk adjustment summary score, indicating worse health status, was associated with an increased likelihood of medication nonadherence (OR = 1.21; 95% CI = 1.20-1.22). These patterns were similar among the full and partial LIS groups. CONCLUSIONS: Individual- and community-level characteristics were associated with the likelihood of medication nonadherence among Medicare LIS recipients with type 2 diabetes, hypertension, or heart failure. These characteristics included younger age, male sex, racial/ethnic minorities, living in lower socioeconomic communities, and a higher risk adjustment summary score. This study provided insight into medication nonadherence within the Medicare LIS population and identified the need to consider these factors when developing future policies to improve medication adherence. DISCLOSURES: This study was funded by the Pharmaceutical Research & Manufacturers of America (PhRMA), which was involved in the preparation and revision of the manuscript. Dougherty is employed by PhRMA. Todor was a PQA-CVS Health Foundation Scholar who was funded to work on this study. Hines is employed by Pharmacy Quality Alliance. Wang reports grants from AbbVie, Curo, Bristol Myers Squibb, and Pfizer, during the time of this study, and fees from the PhRMA Foundation for work on its Heath Outcomes Research Advisor Committee. The other authors have nothing to disclose. This study was presented as a poster at the online 2020 PQA Annual Meeting, May 7, 2020.