Management of infection among Medicare beneficiaries with HIV/AIDS: risk of diabetes with protease inhibitors and associated racial disparities using big data approach

Association between protease inhibitors (PI) and Type II diabetes mellitus (T2DM) in human immunodeficiency virus/acquired immune deficiency syndrome (HIV/AIDS) patients is largely debated. This study examined the odds of developing T2DM among HIV/AIDS Medicare beneficiaries treated with PI and possible racial disparities in the odds. We performed a nested casecontrol study of Medicare database 2013-2017. We included HIV/AIDS positive beneficiaries who were enrolled continuously in Medicare Part A/B with no previous history of T2DM. PI-users were matched to non-PI users and non-anti-retroviral therapies (ART) users using a1:1 greedy propensity score (PS) matching . Multivariablee logistic regressions were performed to assess the odds of developing T2DM. The analysis included 2,353 HIV/AIDS beneficiaries. Matched samples were generated for PI vs. non-PI groups (n = 484) and PI vs. non-ART groups (n = 490). Compared to the non-PI group, the odds of developing T2DM were higher in PI-users (AOR: 1.76; 95% CI: 1.17-2.64), in Caucasian PI-users (AOR: 1.81; 95% CI: 1.02-3.22) and in African-American PI-users (AOR: 1.86; 95% CI: 1.03-3.36). Compared to the non-ART group, the odds of developing T2DM were higher in PI-users (AOR: 1.87; 95% CI: 1.25-2.81), in Caucasian PI-users (AOR: 1.96; 95% CI: 1.14-3.39) and in African-American PI-users (AOR: 2.05; 95% CI: 1.03-4.09). The use of PI is associated with higher odds of T2DM; odds were higher among African-Americans than Caucasians.