760
Comparison of Diabetes Mellitus Prevalence and Risk Factors among HIV-infected ART-naïve Individuals vs Individuals in the NHANES III Cohort
Indira Brar*1, J Shuter2, D Wentworth3, A Thomas3, E Daniels4, J Absalon5, and Minority and Women's Task Force, Terry Beirn Community Prgms for Clin Res on AIDS
1Henry Ford Hosp, Detroit, MI, US; 2Montefiore Med Ctr, New York, NY, US; 3Univ of Minnesota, Minneapolis, US; 4Social & Sci Systems, Silver Spring, MD, US; and 5Columbia Univ Mailman Sch of Publ Hlth, New York, NY, US
Background: Racial
and ethnic minorities are at risk for diabetes mellitus (DM). Women may be
at risk due to higher body mass index (BMI). Among HIV-infected persons, ART use
and hepatitis C (HCV) infection have been associated with DM. To determine whether
“traditional” risk factors (age, BMI, race) are risk factors for DM in
HIV-infected individuals we compared factors associated with DM among HIV-infected
adults and the general population.
Methods: Cross-sectional analysis of ART-naïve
HIV-infected adults enrolled in 3 CPCRA clinical trials vs
adults enrolled in the National Health and Nutritional Examination Survey (NHANES)
III. We used logistic regression to identify factors independently associated
with DM.
Results: In CPCRA (n
= 2535) 50% were African American (AA), 15% Latino, 22% female (F), 15% were injection
drug users (IDU), and 53% men who had sex with men (MSM). In NHANES (n = 6585), 20% were AA, 31% Latino, 54% F,
2% IDU, and 3% MSM. Mean BMI was lower in CPCRA vs
NHANES (25 kg/m2 vs 28 kg/m2, p <0.01). Prevalence of DM was lower
in CPCRA vs NHANES (3.3% vs
4.8%, p <0.01). On bivariate analysis, race, older age, and higher BMI were
associated with DM in both cohorts; HCV, female gender, and non-MSM status were
associated with DM in the CPCRA. After adjustment, AA, Latino race, age, and
BMI were associated with DM in both cohorts. HCV was associated with DM in the
CPCRA cohort (see the table). We also explored associations by gender and race.
Among AA, older age and higher BMI were associated with DM for men and women in
both cohorts. HCV infection was associated with DM among AA men in CPCRA (AOR =
2.94, p = 0.03). The sample did not
permit separate analyses among whites and Latinos.
Conclusions: While our analyses confirmed the association between
hepatitis C and DM, traditional risk factors for DM were the dominant ones
among ART-naive HIV-infected adults and mirrored those of the general population.
Prospective studies are needed to determine the effect of ART on the development
of DM and the relationship between gender and HCV/HIV co-infection in this
population.
|
