Date of Award

Spring 2020

Document Type

Open Access Dissertation

Department

Pharmacology, Physiology and Neuroscience

First Advisor

Kevin Lu

Abstract

Background: The possibility of an association between the use of protease inhibitors (PI) by HIV/AIDS patients and the occurrence of T2DM mellitus (T2DM) is largely debated. Medicare recipients are disproportionally affected by T2DM. Unfortunately, evidence is unavailable from that particular segment of the population. Clinical management of HIV/AIDS is progressively expanding to include chronic/metabolic complications, which may pose a significant economic burden to both the patients and the Medicare system, which are disproportionally impacted. Objectives: The aims of this project were to (1) examine the association between the use of PIs and the odds of developing T2DM among Medicare beneficiaries with HIV/AIDS, (2) assess any racial/ethnic disparity in odds of developing T2DM among Medicare beneficiaries with HIV/AIDS and (3) to determine the economic burden of comorbid T2DM among Medicare beneficiaries with HIV/AIDS.

Methods: This study used a nationwide Medicare claims data from 2013 to 2017 to analyze a sample of Medicare beneficiaries diagnosed with HIV/AIDS based on the International Classification of Diseases, Ninth & Tenth Revision, Clinical Modification (ICD-9/10-CM) codes. In study aim 1 and 2, a nested case-control study design was used to analyze the odds of developing T2DM among beneficiaries with HIV/AIDS. HIV/AIDS positive beneficiaries enrolled continuously in Medicare Part A and Part B were included as well as those who never enrolled in a Health Maintenance Organization (HMO) and those without a previous history of T2DM. A T2DM diagnosis was determined using T2DM specific ICD-9/10-CM codes. Two matched therapy group pairs – (PI versus non-PIs, PI versus no-ART) were generated using a 1:1 greedy Propensity Score (PS) matching procedure. Multivariate logistic regressions were performed to assess the odds of developing T2DM in both groups and for each race sub-group.

In study aim 3, a pooled cross-sectional study design was used to determine the economic burden of comorbidity T2DM in beneficiaries with HIV/AIDS. The analytical sample consists of HIV/AIDS positive beneficiaries enrolled continuously in Part A/B and never enrolled in an HMO. We assessed records of T2DM diagnosis using T2DM specific ICD-9/10-CM codes. Total medical costs, total prescription costs, total inpatient costs, total outpatient costs, total out of pocket (OOP) and total Medicare costs were assessed from Medicare claims and prescription drug files. Generalized linear models with a log-link and a gamma distribution were used to examine the impact of comorbid T2DM on different costs. All costs were adjusted to the 2017-dollar value using the medical component of a consumer price index based on Medical Expenditure Panel Survey (MEPS) guidelines.

Results: In study aims 1 and 2, the analytical sample consists of 2,353 beneficiaries with HIV/AIDS which includes 342 beneficiaries with T2DM, 2011 beneficiaries without T2DM, 1005 beneficiaries treated with PIs, 766 beneficiaries treated with non-PIs and 582 beneficiaries who had no-ART. Exactly 484 matched beneficiaries per therapy group was generated for PI versus non-PI pair and 490 beneficiaries per therapy group for the PI versus no-ART pair. Matched beneficiaries in the PI versus non-PI therapy group are mostly older 55 years and above per group, mostly male beneficiaries – 77.1% (n=373) vii and consists mainly of Caucasians – 49% (n=237) and African Americans -45% (n=218) per group. Matched beneficiaries in the PI versus no-ART therapy group are mostly older than 55 years and above per group, mostly male beneficiaries – 75.9% (n=372) per group and consists mainly of Caucasians – 42.7 % (n=209) and African Americans -50% (n=245) per group. After adjusting for covariates: (1) in the PI versus non-PI pair: the odds of a T2DM diagnosis was higher among PI-users: AOR= AOR=1.76 [95% CI: 1.17-2.64], Caucasian PI-users: AOR=1.81 [95% CI: 1.02-3.22] and African-American PI-users: AOR= 1.86 [95% CI: 1.03-3.36] compared to non-PI users on average, and (2) in the PI versus no-ART pair: the odds of a T2DM diagnosis was higher among PI users AOR=1.87 [95% CI: 1.25-2.81], Caucasian PI-users: AOR=1.96 [95% CI: 1.14- 3.39] and African-American PI-users: AOR=2.05 [95% CI: 1.03-4.09] compared to ART naïve beneficiaries on average.

In study aim 3, a total of 2,509 eligible beneficiaries were identified of which 19.9 % (n=498) had T2DM and 80.2% (n=2,011) are non-T2DM beneficiaries. Beneficiaries with comorbid T2DM had a higher total prescription cost than non-T2DM beneficiaries across all costs: (mean total medical: T2DM beneficiaries ($189,543) versus non-T2DM beneficiaries ($124,052), P= <.0001). After adjusting for covariates, compared to beneficiaries without comorbid T2DM, beneficiaries with comorbid T2DM had higher: total hospitalization cost: 63.34% (95% CI: 42.73% -86.94 %), total outpatient cost: 50.26% (95% CI: 30.70%-72.75%), total OOP cost: 59.15% (95% CI: 40.02%-80.92%), total Medicare cost: 27.95% (95% CI: 13.81%-43.84%) and total medical cost: 27.83% (95% CI: 14.27%-43.00%), compare to non-T2DM beneficiaries, on average.

Conclusion: Use of PIs is associated with a higher odd of T2DM diagnosis. Results are consistent within African Americans and Caucasian race-sub-groups; however, odds were higher among African Americans beneficiaries than Caucasians. Comorbid T2DM may impose a significant economic burden on Medicare beneficiaries with HIV/AIDS. The findings of this study suggest evidence-based risk management approach in the clinical use of PIs to avoid HIV treatment-related T2DM among Medicare population, who are already enormously predisposed as well as personalized risk management approach in the context of racial variation in treatment-related T2DM. The findings of this study could be helpful to the Medicare they seek to address concerns about its future financial solvency amidst a growing aging population and increasing per capita costs. Evidence of total OOP costs benefit the Medicare as they seek to reduce drug costs to benefit HIV positive beneficiaries who face high OOP cost.

Rights

© 2020, Eric Ejike Chinaeke

Download

Share

COinS