The impact of physician factors on health care quality.

摘要

Health care of suboptimal quality has been reported for many medical conditions. Physician factors, such as practice style and expertise, may explain variations in quality. This dissertation empirically evaluates three such factors.; My first paper examines the relationship between length of time in clinical practice and health care quality. Physicians who have been in practice for longer are believed to have accumulated knowledge and skills over time. However, medical advances occur frequently and the explicit knowledge that physicians possess may become out of date. By systematically reviewing the published literature, I demonstrate that more experienced physicians have less factual knowledge and may be at risk for providing lower-quality care. I conclude that this subgroup of physicians may be in particular need for quality improvement interventions.; My second paper evaluates the influence of physician specialty on the quality of care for patients with atrial fibrillation (AF). AF is the most common cardiac arrhythmia and elevates a patient's risk of stroke. This risk is most effectively reduced using warfarin, but warfarin is received by only 30--60% of appropriate patients. In order to assess the influence of physician, patient and hospital characteristics on warfarin use, I used a population-based cohort of 140,185 elderly patients with AF. I demonstrate that after controlling for clinical factors, patients cared for by non-cardiologist physicians with cardiology consultation were more likely to receive warfarin then patients treated in non-consultative environments. I conclude that preferred models of AF care should involve routine collaboration between cardiologists and other physicians.; My final paper assesses the influence of physicians' experiences with adverse events on the quality of care they provide. In the case of AF, warfarin elevates a patient's risk of bleeding; however, for many patients the benefits of therapy outweigh their risks. Nevertheless, physicians' experiences with significant warfarin-associated adverse events may lead to warfarin under-use. Using a matched-pair analysis, I demonstrate that after a physician has had a patient suffer a warfarin-associated hemorrhage, they are less likely to prescribe warfarin to subsequent appropriate patients. This finding highlights the need for strategies to modify physicians' perceptions of the risks associated with anticoagulation.