Patients with medically unexplained symptoms are frequent users of health services. Smith and colleagues sought to evaluate the success of a primary care–based treatment approach for these patients.

The study randomized 206 patients with medically unexplained symptoms to treatment or usual care for 12 months. Patients were selected on the basis of having eight or more physician visits within two years and a chart review indicating medically unexplained symptoms (i.e., patients having no identified organic disease to account for their symptoms). Over the study period, treatment was provided by four nurse practitioners who saw patients during 12 sessions lasting 20 minutes with additional scheduled telephone contacts and visits as needed. Treatment included antidepressants, reduction in controlled substance use, exercise, relaxation, physical therapy, and treatment of identified organic disease. Outcomes were scores on the Mental Component Summary, physical and mental subscales, and depression and anxiety scales, as well as antidepressant and controlled substance use.

Completion was high, with 200 of 206 enrolled participants finishing the study. Improvement was higher in the treatment group compared with the control group (48 [49.0 percent] and 34 [33.3 percent], respectively). The relative benefit of treatment was 1.47. It was determined that six or seven patients would have to undergo treatment for one patient to improve (number needed to treat = 6.4). The authors indicated that the four-point increase in the outcome measure scale was comparable to mental health improvement in patients following mitral or aortic valve replacement. Those with more severe mental health dysfunction at baseline were more likely to improve, whereas those with poorer physical function at baseline were less likely to improve.

Disability score improvement was statistically significant in the treatment group but not in the control group. The treatment group had a greater number of patients using antidepressant medication at full doses compared with the control group (68.4 and 19.8 percent, respectively), and more persons in the treatment group were able to decrease their use of controlled substances. Antidepressant use, but not patient satisfaction, was strongly associated with improved Mental Component Summary scores, and controlling for antidepressant use eliminated the association between treatment and higher scores.

This study incorporated a method of treating patients with medically unexplained symptoms in the primary care setting and found that treated patients had improved mental health function at one year compared with those receiving usual care. The authors were not able to identify the mediating component that led to improvement, and although it was not patient satisfaction, specific subcategories of patient satisfaction that could have played a role were not measured.

Antidepressant use with high adherence rates to full doses appeared to have an effect, but this variable did not explain the improvement in the one fifth of patients who did not take full doses. It would be useful to identify the factors that led to the high adherence rates. The authors conclude that their findings underscore the opportunity to treat patients with medically unexplained symptoms in a primary care setting with a focus on psychological distress rather than pursuing a search for underlying organic disease.