The ultimate goal of medical care, including diagnostic testing, is to improve patient outcome. Accordingly, it has been advocated widely that when establishing a test's diagnostic accuracy, the impact of the test on patient outcome subsequently must be quantified. When studying patient outcome in medical research, the use of randomized comparisons comes into perspective. In our view, randomized studies often are not necessary to validly estimate the effect of the diagnostic test on patient outcome. Results of cross-sectional diagnostic studies, combined with results from therapeutic studies, often will suffice.