Ratings of physicians relying on experts versus physicians relying on decision aids

Abstract

Errors in the field of medicine are costly, both in terms of monetary expenses and the cost of lives. Computerbased decision aids have been shown to decrease physician error. These decision aids decreased many types of errors, reduced hospital admissions and costs and fostered the implementation of evidence-based medicine. However, patients give lower ratings to physicians who use such aids on many variables (diagnostic ability, professionalism, overall satisfaction with exam). The current research attempts to ascertain whether patients are uncomfortable with the use of a computer in a humancentered field or the consultation of an outside source by a physician. Participants were asked to evaluate physicians in one of three scenarios: the physician makes an unaided diagnosis, the physician consults a human expert for a second opinion, or the physician uses a computer-based diagnostic aid to diagnose a patient. Two planned Man-Witney U tests showed that physicians who used the computer-based aid were given lower ratings on diagnostic ability, professionalism, and overall satisfaction compared to physicians making unaided diagnoses. Ratings of physician making unaided diagnosis were not significantly different from that of the physician who consulted human experts. We conclude that it is not the consultation of an outside source that patients find objectionable. However, the use of a computer in a human-centered field such as medicine is responsible for the derogated ratings of physicians.