This review summarizes the current status of and recommendations for prostate cancer screening with prostate-specific antigen in light of recent reductions in prostate cancer incidence and mortality. It describes how the uncertain effectiveness of aggressive treatment for prostate cancer and a reservoir of unsuspected indolent cancers make prostate cancer fit poorly into conventional screening models. The large proportion of men with unsuspected prostate cancers that may not cause morbidity or mortality and are unlikely to benefit from aggressive treatment decrease the effectiveness of a screening program. In addition, indolent, unsuspected prostate cancers in the screening population accentuate the detrimental effects of length bias on studies evaluating the effectiveness of prostate cancer screening. Screening tests for prostate cancer will continue to improve, but chemoprevention or nutritional prevention with vitamins and micronutrients such as tocopherols or selenium may prove to be effective methods of reducing prostate cancer incidence and should be aggressively investigated.