Over the past decade, researchers have put a great amount of e#ortinto developing suitable models for the analysis of longitudinal CD4data and other markers of AIDS progression. These models must begeneral enough to allow for di#erent patterns of change in the markerdata. In this paper, we review the existing literature including ourpreferred models whichinvolve mixed e#ects, stochastic terms and independentmeasurement error. Adding stochastic terms to standardmixed e#ects models...