HIV infection and development of diabetes in a cohort of men receiving care from a large urban community-based health center

Background:Antiretroviral therapy has resulted in fewer AIDS-related deaths among persons living with HIV. However, this reduction has been accompanied by increases in morbidity and mortality from non-AIDS related diseases, including diabetes. To assess the incidence of diabetes in a contemporary clinical care environment, the present study examined diabetes incidence among HIV-infected and uninfected men engaged in primary care at a large urban community-based health center.

Methods: From 2001 to 2002, participants completed a brief survey assessing sociodemographic characteristics and health behaviors. Responses were linked to electronic medical records, and outcome information was extracted from the survey date until December 31, 2012. Proportional hazards models were used to estimate the association between HIV infection and development of diabetes before and after controlling for covariates.

Results: Of the 1058 study participants, 189 (17.9%) were HIV-infected at baseline. Significant (p<0.05) differences were observed on a number of sociodemographic characteristics, health behaviors and health outcomes by HIV status. Proportionally more HIV-infected participants developed diabetes during the follow-up period than HIV-uninfected participants (5.03% vs. 1.15% respectively, p<0.001); however, HIV infection was not associated with time to diabetes diagnosis in either the unadjusted (hazard ratio [HR]=1.23, 95% confidence interval [CI]: 0.50, 3.00), or adjusted model (HR=1.89, 95% CI: 0.70, 5.10). 

Conclusions: We did not find an association between HIV infection and development of diabetes in our cohort, despite observing proportionally more diabetes diagnoses among our HIV-infected participants. These contradictory findings suggest future research is needed in this area with a larger and more diverse sample.