|
D. R. Bangsberg*, S. Perry, E. D. Charlebois, R. Clark, A. R. Zolopa, and A. R. Moss.
San Francisco Gen. Hosp./Univ. California, San Francisco.
Background:While adherence (ADH) predicts of viral suppression, its role in disease progression is not established.
Methods:ADH was prospectively measured in a population-based cohort using electronic pill cap monitoring (MEMS) and unannounced home pill counts in 76 AIDS-free individuals on HAART. These ADH measures explain 40-60% of the variation in VL between individuals (Bangsberg, AIDS 2000). Progression was compared by level of ADH (<50% Low, 50-89%, Middle,>90% High) using the Kaplan-Meier method. Cox proportional
hazards analysis was used to examine the relation between ADH and time-to-AIDS with adjustment for baseline characteristics.
Results:With 92% retention over 988 person-months, AIDS rates were 38%, 8%, and 0% for Low, Middle, and High adherers respectively. A 10% difference in average adherence was associated with a 21% reduction in risk adjusted for baseline CD4 count (adherence: RR=0.79 [95% CI: .64-.99], p=0.04; CD4: RR=0.82 [95% CI .67-1] p=0.03 ). Injection drug use (p=0.08), and months of HAART at baseline (p=0.1) were marginally predictive in a bivariate analysis, but not in the Cox model. Housing status, ETOH use, and previous psychiatric admission were not predictive. Consistent with previous results, adherence was also closely associated with the proportion of months with a VL <400 (Spearman's Rho=0.55, p=0.001). Median proportion of months with viral load <400 were 0%, 49%, and 100%, respectively.
Conclusion:ADH to HAART is associated with progression to AIDS. Efforts to improve HAART adherence are necessary to maximize AIDS-free survival.
© 8th Conference on Retroviruses and Opportunistic Infections