MS Gerrity, JW Williams, AJ Dietrich. Departments of Medicine, Oregon Health Sciences University, Portland, OR and University of Texas at San Antonio, and Department of Community and Family Medicine, Dartmouth Medical School, Hanover, NH.

Purpose: Primary care physicians see many patients with depression; yet, they may miss the diagnosis or under treat the patient. Methods to identify physicians most likely to benefit from educational programs to improve care of depressed patients have not been developed. We used data from a national survey to develop measures to identify these physicians.

Method: A 53-item questionnaire was mailed to a national sample of 2,308 family physicians, general internists, and obstetrician-gynecologists who spent at least 50% of their time seeing continuity patients. We used exploratory factor analysis (EFA) to identify factors underlying physicians responses to 35 attitude items. We named the factors and selected items with loading> 0.40 for the final scales. After collapsing the ratings of each item into dichotomous responses (0 versus 1), we scored the scales by summing physicians' responses to each item. Cronbach's alpha was used to assess reliability. Using a separate physician validation sample, physicians' scores on the scales were compared to actual behavior with unannounced standardized patients and to their scale scores after attending a workshop on depression.

Results: 1,350 (58%) of physicians returned questionnaires. Respondents spent a median of 11 years in practice, 50% were female, 46% family physicians, 35% general internists, and 19% obstetrician-gynecologists. EFA identified 2 factors that we named Confidence in Caring for Depressed Patients (8 items, alpha = .78) and Intention to Change (6 items). Twenty-nine percent of family physicians, 54% of general internists, and 88% of obstetrician-gynecologists reported not being confident (scale score (7) in caring for depressed patients. For all specialties, 64% intended to change some aspect of their care of depressed patients. Of those physicians who were not confident, preliminary data suggest that 25% - 30% are very likely to attend an educational program about depression in the next 6 months. Based on standardized patient observations for the validation sample, physicians who were not confident in caring for depressed patients were less likely to ask about the criteria for depression, discuss depression , prescribe anti-depressants, and schedule a return visit within 2 weeks Physicians scores on the Confidence scale changed significantly after attending a workshop on depression.

Conclusion: The Confidence in Caring for Depressed Patients scale is a reliable and valid measure that can be used to identify physicians most likely to benefit from educational programs about depression. Furthermore, the majority of general internists and obstetricians do not feel confident in their care of depressed patients.
Return to top of page