Expected-utility perspectives on defensive testing - Torts, tradeoffs, and thresholds - Is defensive medicine defensible?

Medical decision making must address tradeoffs: between uncertainty and the risk of obtaining information, between quality of life and quantity of life, between the short term and the long run, between cost and effectiveness, between the individual and society, and perhaps between physicians and their patients. This journal focuses on making such tradeoffs explicit, helping clinicians and patients better understand these conflicts and their implications. DeKay and Asch(1) extend the simple logic of the threshold approach(2) to examine the implications of one bedside conflict of interest-the clinician's desire to avoid liability and criticism while trying to optimize the patient's health outcomes. Using a global utility scale that incorporates risk, benefit, and liability, these authors demonstrate that even the potential of liability either for withholding treatment from a patient ultimately shown to have disease or for administering treatment to a patient ultimately shown not to have disease will change the range of probabilities of disease for which diagnostic testing would seem to be appropriate. When physicians attempt to avoid liability by practicing defensive medicine, some patients will experience decrements in expected health, but no patients experience increments in expected health.