Clinical Question

Do psychological therapies reduce the severity of medically unexplained physical symptoms?

Evidence-Based Answer

Psychological therapy, specifically cognitive behavior therapy (CBT), has been shown to reduce the severity of medically unexplained physical symptoms in patients with somatoform disorders. The effect is small to moderate in magnitude, but in these studies CBT was as acceptable as usual care to patients. (Strength of Recommendation: B, based on inconsistent or limited-quality patient-oriented evidence.)

Practice Pointers

Up to one-third of patients who present to their physician with physical symptoms will receive no medical explanation for those symptoms.¹ Persistent medically unexplained physical symptoms are part of the diagnostic criteria for many somatoform disorders. They place a heavy burden on the patient and can strain the patient-physician relationship.² An earlier systematic review concluded that new generation antidepressants are slightly better than placebo for short-term treatment of these symptoms.³

The authors of this Cochrane review sought to determine whether there were any effective nonpharmacologic treatments for medically unexplained physical symptoms. This meta-analysis included 21 studies with 2,658 participants. All studies were randomized and examined some form of psychological therapy, with most examining CBT. All participants were required to meet the criteria for a somatoform disorder as well as to have medically unexplained physical symptoms described as their primary medical problem. Primary outcomes examined were changes in the severity of medically unexplained physical symptoms and acceptability of treatment. Secondary outcomes included depression and anxiety, adverse effects, behavioral or emotional dysfunction, overall treatment response, functional disability, and health care use.

Fifteen of the studies evaluated patients receiving psychological therapy vs. usual care or wait list control patients; 10 of the studies examined CBT. Psychological therapy as a whole group (standardized mean difference [SMD] = −0.34; 95% confidence interval [CI], −0.53 to −0.16) and CBT as a subgroup (SMD = −0.37; 95% CI, −0.69 to −0.05) were both more effective than usual care. These effects persisted for both groups at one year of follow-up. Usual care was considered slightly more acceptable by patients (relative risk [RR] = 0.93; 95% CI, 0.88 to 0.99) than psychological therapy overall. However, CBT as a subgroup was judged by patients in these studies to be as acceptable as usual care. Clinician-rated symptoms of anxiety (SMD = −0.40; 95% CI, −0.63 to −0.17) and depression (SMD = −0.25; 95% CI, −0.48 to −0.02) favored psychological therapies over usual care. However, participant ratings of the anxiety and depression symptoms did not show a significant difference. Outcomes of clinician-rated treatment response (RR = 3.30; 95% CI, 2.08 to 5.21), functional disability (SMD = 0.17; 95% CI, 0.03 to 0.32), and health care usage (SMD = −0.68; 95% CI, −1.06 to −0.30) favored psychological therapy. There were no significant differences in adverse effects or behavioral/emotional dysfunction between groups.

Five studies examined psychological therapies vs. enhanced care, defined by the review as usual care with added enhancements of various types that could include participant education, a psychiatric interview, or reattribution training for the primary care physician. Reduction in the severity of medically unexplained physical symptoms at one year of follow-up favored the psychological therapies (SMD = −0.21; 95% CI, −0.40 to −0.02). Acceptability of treatment favored enhanced care (RR = 0.93; 95% CI, 0.87 to 1.00). Behavioral/emotional dysfunction (SMD = −0.24; 95% CI, −0.49 to 0.00) and functional disability (SMD = 0.20; 95% CI, 0.02 to 0.38) favored psychological therapies at one year of follow-up.

One study compared CBT with progressive muscle relaxation. No significant differences were noted in any primary or secondary outcomes in this comparison.

All studies in this review included participants who were willing to receive psychological treatment. There was a high risk of bias caused by a lack of blinding, which was not possible with the treatments studied. In most cases, there were too few studies to draw strong conclusions about secondary outcomes, or even about primary outcomes for therapies other than CBT. Finally, there were no studies examining therapies that were both nonpharmacologic and nonbehavioral, such as physical therapy.

Medically unexplained physical symptoms are common and often persistent in those with somatoform disorders. Although psychological therapies such as CBT may have some benefit over usual care, there are currently no practice guidelines regarding the best treatment for these symptoms. More high-quality studies are needed to determine the effectiveness and acceptability of nonpharmacologic interventions for medically unexplained physical symptoms.