WHAT IS A SCREENING-TEST - MISCLASSIFICATION BIAS IN OBSERVATIONAL STUDIES OF SCREENING FOR CANCER

OBJECTIVE: To demonstrate the importance of accurately identifying cli nical distinctions of subjects in observational studies of screening. DESIGN: Simulated case-control studies. SETTING: The West Haven Vetera ns Affairs Medical Center. PATIENTS: Fifty-two men diagnosed with pros tate cancer in 1988 or 1989 had 252 digital rectal examinations (DREs) in the preceding 5 years. A classification scheme used patient sympto ms and the results of prior DREs to assign the last DRE before the dia gnosis of cancer to one of the following categories: definite screenin g, likely screening, probable screening, not screening, or other and u nknown. Sixty-five percent of the DREs were classified as definite or likely screening, and another 15% were classified as probable screenin g. MAIN RESULTS: Changing the definition of a screening DRE from one i ncluding to one excluding probable DREs lowered the frequency of scree ning in case subjects more than it did in case controls, and thus lowe red the odds ratio (OR), making screening appear to be more protective . Even when DRE was not protective, the ORs for the effectiveness of s creening with the more restrictive definition ranged from 0.21 to 0.83 in 36 simulated case-control studies that differed according to the f requency of screening, the prevalence of cancer in case controls, and the extent of misclassification error. CONCLUSIONS: If clinical distin ctions in the performance of screening tests are not classified approp riately, observational studies will misrepresent the proportion of sub jects exposed to screening interventions and produce biased results.