Measuring the variability of prescription use in patients with HIV infection or AIDS. The contribution of a French hospital longitudinal database

Abstract

This study is based on a sample of 937 patients with HIV infection or AIDS who were treated in 5 French hospitals. It sets out to describe the prescription of antiretroviral and prophylactic drugs used in treating such patients and, aims to interpret the various treatment strategies used. The study showed the contribution of longitudinal data in the descriptive analysis of patient follow-up, given the evolving nature of the illness. The Principal Components Analysis method allowed the temporal and quantitative aspects of the data, as well as their combination with qualitative variables, to be taken into account. The results revealed the stability over time of the choice of prescriptions, and allowed joint prescription/substitution phenomena between drugs to be evaluated. At the same time, the study provided evidence for a "site' effect, which showed that the heterogeneity of professional practice is not solely the result of differences in patients or in severity of the illness.