Evidence Based Medicine Part 5. Observational evidence
Observational evidence offers a huge volume of data to support health care, and can detect harms that randomised controlled trials may not be designed or powered to measure. Many research questions are outside the remit of clinical trials. Observational study designs include population-based longitudinal surveys, cross-sectional surveys, cohort studies, non-randomised controlled studies, case-control studies, case reports and case series. The fifth in a series looking at evidence-based medicine, this article explores observational evidence using the recent controversy surrounding the safety of pioglitazone as an example.