Simulation study of HIV temporal patterns using Bayesian methodology

Abstract

Viral load values and CD4+T cells count are markers currently evaluated in the clinical follow-up of HIV/AIDS patients. In this context, it is relevant to develop methods that provide a more complete temporal description of these markers, e.g. in between clinical appointments. To this end, we combine a mathematical model and a Bayesian methodology to estimate trajectories from a set of observed values. Also, we construct a variation band containing the most central trajectories for one patient, by exploring the range of values in the a posteriori distributions. The methods are illustrated with simulated data.