New Diagnostic Tests: Breakthrough Approaches or Expensive Add-ons?

New diagnostic tests are seductive for physicians and patients alike. More information can only be better. Many tests are noninvasive, and it seems that no harm can come from them. With insurance, costs seem secondary. The growth of direct-to-consumer sales of imaging procedures attests to the allure of diagnostic testing in the public mind. Physicians and hospitals often find new diagnostic technologies attractive because of expected financial return.

And yet, no test is perfect. False-positive and false-negative results can lead to additional testing, unnecessary treatments, and avoidable complications. Sometimes, diagnostic test results have little effect on treatment and even less effect on patient survival or quality of life. So when new diagnostic technologies are introduced—such as positron emission tomography (PET) scanning to detect metastases in patients with lung cancer—we should critically ask how they may benefit patients.

Experts have suggested that diagnostic tests be evaluated along a hierarchy of increasingly stringent evaluation (1, 2). First, is the test reproducible, feasible, and safe? Does it produce high-resolution images? Second, does it accurately distinguish individuals with disease from individuals without disease (good sensitivity and specificity)? Third, do the results affect the choice of treatment? Fourth, does the test lead to better patient outcomes than would occur without the test? And finally, what is the impact on society, including efficiency and costs of care?

In this issue, Gould and …