Incidence of diabetes mellitus in a population-based cohort of HIV-infected and non-HIV-infected persons: the impact of clinical and therapeutic factors over time

Aims To examine incidence density rate and correlates of incident diabetes mellitus in a cohort of HIV‐infected individuals compared with matched non‐HIV‐infected persons. Methods Data were obtained from the South Carolina Medicaid system and the enhanced HIV/AIDS Reporting System surveillance database for persons ≥ 18 years of age who had been attended to during the period 1994 to 2011. Time‐dependent proportional hazards analysis and marginal structural models were used to analyse the data. Results A total of 13 632 individuals (6816, 1:1 matched HIV‐infected and non‐HIV‐infected persons; median age 39 years; 57% male) contributed 88 359 person‐years of follow‐up. Incidence rate of diabetes was higher in the non‐HIV‐infected group compared with the HIV‐infected group (13.60 vs. 11.35 per 1000 person‐years). Multivariable hazards analysis suggested a significantly lower risk of incident diabetes among HIV‐infected persons treated with combination antiretroviral therapy compared with the matched non‐HIV‐infected persons (adjusted hazards ratio 0.55; 95% CI 0.46–0.65). Among HIV‐infected persons, marginal structural modelling suggested a significantly higher risk of diabetes with cumulative exposure to protease inhibitors over the observation period (adjusted relative risk 1.35; 95% CI 1.03–1.78), but this association was not significant for exposure to non‐nucleoside reverse transcriptase inhibitors. Overall, female gender, older age, non‐white race/ethnicity, and pre‐existing hypertension, dyslipidaemia, obesity and hepatitis C infection were associated with higher risk of diabetes incidence. Conclusions HIV infection may not be independently associated with increased risk of diabetes. Among HIV‐infected persons, exposure to protease inhibitor‐based regimens may increase the risk of diabetes. Healthcare providers should make every effort to use combination antiretroviral therapy regimens with a better cardiometabolic profile. What's new? This is the first study to examine the incidence density rate of diabetes among HIV‐infected persons compared with a non‐HIV‐infected control group in the Southern USA, which has disproportionally both HIV and cardiometabolic disorders, especially obesity. Time‐dependent statistical methodology was used to examine the retrospective observation data over an 18‐year study period for the pertinent risk factors associated with new‐onset diabetes. Results empirically reiterate the importance of monitoring and managing risk fac