Hsu Joe L, Banerjee Dipanjan, Kuschner Ware G
Division of Pulmonary and Critical Care Medicine, Stanford University School of Medicine, Stanford, CA, USA.
South Med J. 2008 Dec;101(12):1240-5. doi: 10.1097/SMJ.0b013e31818860c0.
Bias and confounding are types of error that may be encountered in the collection, analysis, or interpretation of research data. Bias and confounding may result in erroneous research conclusions with adverse consequences for patients and health care providers. In this article, we provide clinician-friendly descriptions and examples of bias (including surveillance, information, selection, lead, length, and publication) and confounding. The purpose of the article is to help clinicians to recognize two important sources of error in research and in turn to help clinicians to assess the validity and generalizability of a research report.