Counseling by primary care physicians of patients who disclose psychosocial problems

Most descriptive studies of psychosocial counseling by US primary care physicians (PCPs) have relied on the PCPs' recollections rather than researchers' observations of actual visit content. The latter approach should yield more accurate measurement of counseling frequency and duration.

Our sample consisted of 308 patients whose scores signified psychological distress on the 28-item General Health Questionnaire. Their visits to 69 community-based PCPs had been audiotaped for an earlier study of the benefits of communication skills training for PCPs. Using those tapes, we identified the disclosure of psychosocial problems by patients and subsequent counseling and psychotropic drug prescribing by their PCPs. We timed counseling and visit lengths. Effects of patient and PCP variables on visit duration and on counseling occurrence and duration were assessed using generalized estimating equations to accommodate the correlation among patients who shared PCPs.

After adjusting for the effects of the communication skills training, we found that PCPs counseled 60% of patients who disclosed psychosocial problems. Given disclosure, counseling probability was lowest for new patients (P <.001); among patients with previous visits, counseling probability was inversely related to the number of visits (P <.001). When provided, counseling had a mean duration of 5.2 minutes. Counseling was associated with a 28% (95% confidence interval, 9%-49%) increase in visit duration after adjustment for the effects of other significant variables. PCPs prescribed psychotropic medications in 30% of visits with disclosure.

PCPs treated psychosocial problems with brief counseling twice as often as with medication. Brief counseling interventions caused small but significant increases in visit durations.