• David H Thom, Miguel D Tirado, Tommy L Woon, and Melen R McBride.
• Department of Family and Community Medicine, University of California, San Francisco, San Francisco General Hospital, 1001 Potrero Avenue, Building 80/83, San Francisco, CA 94110, USA. dthom@fcm.ucsf.edu
• BMC Med Educ. 2006 Jan 1; 6: 38.

BackgroundIncreasing the cultural competence of physicians and other health care providers has been suggested as one mechanism for reducing health disparities by improving the quality of care across racial/ethnic groups. While cultural competency training for physicians is increasingly promoted, relatively few studies evaluating the impact of training have been published.MethodsWe recruited 53 primary care physicians at 4 diverse practice sites and enrolled 429 of their patients with diabetes and/or hypertension. Patients completed a baseline survey which included a measure of physician culturally competent behaviors. Cultural competency training was then provided to physicians at 2 of the sites. At all 4 sites, physicians received feedback in the form of their aggregated cultural competency scores compared to the aggregated scores from other physicians in the practice. The primary outcome at 6 months was change in the Patient-Reported Physician Cultural Competence (PRPCC) score; secondary outcomes were changes in patient trust, satisfaction, weight, systolic blood pressure, and glycosylated hemoglobin. Multiple analysis of variance was used to control for differences patient characteristics and baseline levels of the outcome measure between groups.ResultsPatients had a mean of 2.8 + 2.2 visits to the study physician during the study period. Changes in all outcomes were similar in the "Training + Feedback" group compared to the "Feedback Only" group (PRPCC: 3.7 vs.1.8; trust: -0.7 vs. -0.2 ; satisfaction: 1.9 vs. 2.5; weight: -2.5 lbs vs. -0.7 lbs; systolic blood pressure: 1.7 mm Hg vs. 0.1 mm Hg; glycosylated hemoglobin 0.02% vs. 0.07%; p = NS for all).ConclusionThe lack of measurable impact of physician training on patient-reported and disease-specific outcomes in the current has several possible explanations, including the relatively limited nature of the intervention. We hope that the current study will help provide a basis for future studies, using more intensive interventions with different provider groups.

35 keywords...

hide…