OBJECTIVE: To critically evaluate the evidence for recommending the screening of asymptomatic men for prostate cancer with a blood test to detect a prostate-specific antigen (PSA). DATA SOURCES: Relevant articles on screening for prostate cancer were identified from MEDLINE searches, from the authors' files, and from the bibliographies of identified articles. STUDY SELECTION: In the absence of controlled prospective trials, the studies are primarily retrospective and contain information about the sensitivity, specificity, and predictive values of tests used to screen for prostate cancer; the natural history of untreated prostate cancer; the morbidity, mortality, and costs of definitive treatment; and reviews of screening study biases. DATA EXTRACTION: Potential treatment-related mortality and costs that could be incurred by screening were estimated using defined assumptions. RESULTS: Although screening for prostate cancer has the potential to save lives, because of possible overdiagnosis, screening and subsequent therapy could actually have a net unfavorable effect on mortality or quality of life or both. Given the performance characteristics of the test, widespread screening efforts would probably cost billions of dollars. CONCLUSIONS: The net benefit from widespread screening is unclear. A randomized prospective study of the effect of screening on prostate cancer mortality has therefore been initiated by the National Cancer Institute.