Introduction. Injection drug users (IDUs) are aging and tend to have chronic health problems like obesity and diabetes, with a decreased quality of life. However, limited data exist on the longitudinal changes in obesity among current and former IDUs. This dissertation presents an evaluation of two primary exposures (injecting drugs and HIV infection) in relation to longitudinal changes in obesity and to cross-sectional measures of glycemia, using data from the AIDS Link to the IntraVenous Experience (ALIVE) study.

Methods. Since 1988, ALIVE has prospectively followed HIV-infected and at risk but uninfected IDUs through semi-annual study visits including interviews, clinical measures and laboratory testing. Mixed effects models were performed to determine longitudinal changes in body mass index (BMI) and waist circumference (WC) from 2005 to 2009. Random intercept models were employed to determine the effect of antiretroviral therapy (ART) on the difference in mean BMI in the era of effective antiretroviral therapy. Multinomial logistic regression models were examined to determine cross-sectional associations between HIV and measures of glycemia.

Results. Active injection drug use attenuated age-related increases in BMI and WC over 5 years, especially among HIV-infected persons who experienced decreases in these parameters. HIV-uninfected former IDUs however, experienced increases in BMI and WC. Current and former HIV-infected injectors on (HAART) with virologic suppression had higher mean BMI compared to those not on HAART. Among HAART users, BMI increased [0.08 kg/m ²/yr on HAART] with longer duration of HAART use among current injectors. HIV was positively associated with elevated fructosamine levels but negatively associated with elevated levels of HbA1c, after adjusting for potential confounders. Factors like anemia, macrocytosis, alcohol use and liver disease were independently associated with high levels of HbA1c and fructosamine.

Conclusion. In this aging cohort of IDUs, increasing BMI was observed among HIV-uninfected persons and among those HIV-infected persons receiving HAART. Although often considered secondary to other medical concerns among IDUs, programs emphasizing a healthy lifestyle with appropriate diet and exercise could potentially reduce the growing burden of chronic disease morbidity and mortality in this population. However, HbA1c and fructosamine may not be adequate measures of glycemia among HIV-infected persons.