To Test or Not to Test, That Is the Question
This article requires a subscription to view the full text. If you have a subscription you may use the login form below to view the article. Access to this article can also be purchased.
Physicians’ primary responsibility has always been to provide the best care for their individual patients, but the proliferation of modern tests and treatments has complicated their task. With so many options, it is often hard during an office visit to know whether to order a test and what to do with the results. And the ever-rising cost of health care makes patient management even more complicated because physicians must also choose wisely and not waste societal resources.1 Although the plight of the physician is not as fraught as Hamlet’s, modern physicians do have difficult decisions about optimal selection of tests and treatments.
Article see p 276
Whether or not to order a test is a question that cuts across diseases and specialties. Tests provide information that might be important or might be irrelevant, depending on the clinical context. In caring for an individual patient, the question is whether the information from the test might change a clinical decision. In their classic analysis of whether to test or not in choosing a treatment, Pauker and Kassirer2 showed that when patients have a high chance of benefiting from treatment, the optimal strategy is simply to treat without doing a test, whereas for patients with a low chance of benefiting from treatment, the optimal strategy is to neither test nor treat. However, for patients in whom the chance of benefitting from treatment is intermediate, the optimal strategy is to obtain a test and use the results to choose to treat or not. The precise range of this intermediate probability sweet spot for testing depends on the benefits, harms, and costs of the disease, the treatment, and the test. This conceptual framework …