Jonathan I. Silverberg, MD, PhD, MPH
Dr. Silverberg is from the Department of Dermatology, St. Luke’s-Roosevelt Hospital Center, New York, New York.
Dr. Silverberg reports no conflicts of interest in relation to this post.
A May 4, 2012, Wall Street Journal (WSJ) article discussed the increased usage of observational studies in clinical research despite being prone to methodological and statistical biases and possibly flawed results. Observational studies are commonly used to study associations between various exposures, such as environmental risk factors or treatment, and disease outcomes. Observational studies differ from prospectively controlled studies in that participants are assigned to an exposure or treatment group that is not controlled in the study. Typically, data are retrospectively collected from sources ranging from small single-site chart reviews to international epidemiologic databases and comprehensive health management organization cohorts. The WSJ article featured 2 studies from 2010: one published in the Journal of the American Medical Association (2010;304:657-663) and the other in the British Medical Journal (BMJ)(2010;341:C4444) on the use of osteoporosis drugs and risk for esophageal cancer. The studies found entirely conflicting results despite being performed on the exact same database.