The practice of including many prevalent users, i.e. patients taking a therapy for some time before study follow-up began, in observational studies can cause two types of bias. Firstly, prevalent users are “survivors” of the early period of pharmacotherapy, which can introduce substantial selection bias if risk varies with time. Secondly, covariates for drug users at study entry are often plausibly affected by the drug itself. New user designs help avoid the mistake of adjusting for factors on the causal pathway which may introduce bias towards the null. Evaluating medication effects outside of clinical trials: new-user designs (Am J Epidemiol 2003;158 (9):915–20) reviews designs which avoid these biases by restricting the analysis to persons under observation at the start of the current course of treatment. In addition to defining new-user designs, the article explains how they can be implemented as case-control studies and describes the logistical and sample size limitations involved.
|Annex 1.||Guidance on conducting systematic revies and meta-analyses of completed comparative pharmacoepidemiological studies of safety outcomes|