ABSTRACT
Bias is an undesirable feature of study design that tends to produce results which are systematically different from the real values. It can apply to all types of study, and it usually occurs due to faults in the way in which a study is planned and carried out. Errors in data analysis can also produce bias, and should be similarly sought out and dealt with. The main types of bias are selection bias and information bias. Missed responders or non-responders – this is called responder bias. Medical records may contain more information on patients who are ‘cases’ – this is called recording bias. Confounding occurs when a separate factor influences the risk of developing a disease, other than the risk factor being studied. To be a confounder, the factor has to be related to the exposure, and it also has to be an independent risk factor for the disease being studied.
