Question

Please read the following two articles (A and B) before answering the questions noted below.

A. Bruneau J, Lamothe F, et al. High rates of HIV infection among IDU participating in needle exchange programs in Montreal: results from a cohort study. American Journal of Epidemiology. 1997;146:994-1002.

B. Lurie P. Invited commentary: le mystere de Montreal. American Journal of Epidemiology. 1997; 146: 1003-1006.

Question: What confounders were considered in this study? How did the investigators determine if these confounders were associated with the exposure of interest (NEP use)?

Answer 1

The confounders in this study were sociodemographic data's, behavioural changes, drug consumption variables such as number of syringes. The investigators determined by the regression models consisted of covariates. Where some had been left uncounted because of absenteeism. The NEP biasing association towards the effect which was opposite from the expected protective one. Where the study cannot be generalised completely due to net confounding effect by the behavioural characteristics.