This study was undertaken in order to better understand the detection of depression by primary care physicians. Specifically, we investigated the relationship between information gathered during the course of the medical interview and the subsequent diagnosis of depression. Forty-seven community-based primary care physicians, unaware of the mental health focus of this research, were videotaped in the office setting, as they interviewed two “typical” standardized patients who met DSM-III-R criteria for major depression. One patient presented with headaches and the other presented with palpitations and chest pain. After each interview, physicians were provided with physical findings and results of any diagnostic procedures they ordered, then asked to construct and explicate their differential diagnoses. The two patients were correctly diagnosed as depressed by 53 and 45% of the physicians. Although detection was related to greater amounts of information gathered, inquiry about the DSM-III-R criteria symptoms was generally low, and in no case was sufficient information acquired to make a formal DSM-III-R diagnosis of depression. However, a subset of the DSM-III-R symptoms (those related to disturbances of appetite, sleep, and other neurovegetative functions) were among the reasons cited for inclusion of depression in the differential, as were psychosocial stressors and the patient's appearance. These findings suggest that detection of depression is low by primary care physicians.

OBJECTIVES:To measure primary care physicians' attitudes toward psychosocial issues, determine their relationship to the style of the medical interview, and assess whether attitudes and interview behaviors lead to correct diagnosis in patients with depression.DESIGN:Physicians were videotaped while interviewing four patients standardized with criteria symptoms of major depression. Physicians were unaware of the mental health focus of the study.SETTING:Patient examining rooms.PARTICIPANTS:Physicians were eligible for recruitment if they were board certified or eligible in family practice or internal medicine, practiced primary care medicine, and were listed in regional directories. Standardized patients were recruited from the community.MAIN OUTCOME MEASURES:Attitudes toward psychosocial issues (measured by the Physician Belief Scale), interview content (measured by review of the videotaped encounters), interview behaviors (measured by the Interaction Analysis System for Interview Evaluation), and a listing of depression in the differential diagnosis (determined by physician debriefing interviews).RESULTS:Forty-seven community-based practitioners participated. Forty-eight percent of interviews resulted in a diagnosis of depression. Physician Belief Scale scores were not significantly correlated with patient-centered interviewing, psychosocial questions, inquiry about depression symptoms, or a depression diagnosis. Longer interviews were more likely to result in a depression diagnosis.CONCLUSIONS:High interest in psychosocial issues was not associated with patient-centered interviewing behaviors, questions about psychosocial or depression symptoms, or depression diagnoses. However, certain patient-centered interviewing behaviors, particularly those defined as "affective," did lead to the recognition of depression.