Diagnostic errors are increasingly recognized as an important source of preventable harm in many health care settings.1 Missed, wrong, and delayed diagnoses have been underappreciated by internal peer review, autopsy reports, and examination of malpractice claims. All of these methodological approaches have limitations. Internal peer review is often challenging because of local hospital politics, physician-vested interest, and sampling error. Autopsy studies may overestimate diagnostic performance when necropsy rates are low,2 and they often miss nonlethal diagnostic errors. Malpractice claims may capture nonlethal errors; however, they are most often associated with permanent disability or death.3 Only about 1% of adverse events due to medical negligence result in a claim.4 Thus, malpractice-based rates of diagnostic errors substantially underrepresent the true impact of these events and are biased toward cases with a clear paper trail (eg, missed cancers evident on radiographic images), in which the burden of legal proof can be met more easily. None of these approaches is well suited to real-time surveillance for errors that might be rectified before harm occurs.