Original Contribution Individual Variation in CD4 Cell Count Trajectory among Human Immunodeficiency Virus-infected Men and Women on Long-term Highly Active Antiretroviral Therapy: An Application using a Bayesian Random Change-Point Model

The authors evaluated population- and individual-level CD4-positive T-lymphocyte (CD4 cell) count trajectories over a 7-year period (July 1995–March 2004) following initiation of highly active antiretroviral therapy (HAART) in the Multicenter AIDS Cohort Study and the Women’s Interagency HIV Study. The study population included 404 human immunodeficiency virus (HIV)-infected men and 609 HIV-infected women who 1) had a CD4 cell count measurement available from their last pre-HAART study visit, 2) provided at least four post-HAART CD4 cell count measurements, and 3) reported HAART usage for at least 80 % of the post-HAART visits. The CD4 cell count trajectory was analyzed by means of a Bayesian random change-point model. The results indicated that CD4 cell count trajectories for long-term frequentHAARTusers canbewellmodeledwith changepoints at both the population and individual levels. At the population level, regardless of CD4 cell count before HAART initiation, the gains in CD4 cell count ended approximately 2 years after HAART initiation in bothmenandwomen. At the individual level, 35%of men in the Multicenter AIDS Cohort Study versus 25 % of women in the Women’s Interagency HIV Study had a statistically significant change in CD4 cell count trajectory within 7 years after HAART initiation. antiretroviral therapy, highly active; Bayes theorem; CD4 lymphocyte count; change point; HIV infections; models