Background: Patients with chronic illness commonly experience depressive symptoms and physical inactivity, which can further impair their health status. Depressive symptoms are associated with decreased adherence to medical therapies and health-related quality of life, as well as increased disability, symptom burden, functional and role impairment, and use of health care services. Recent studies suggest that antidepressants may not be effective for treating mild to moderate depressive symptoms or in patients with comorbid chronic illnesses. For this reason, interest persists in nonpharmacologic treatments for depression, including exercise. Herring and colleagues conducted a meta-analysis to estimate the effect of exercise on depressive symptoms in patients with chronic illness who have not been diagnosed with depression.

The Study: Included studies enrolled sedentary adults with chronic illness who were assigned randomly to exercise training or a nonexercise treatment. Participants had depressive symptoms assessed at baseline and at the study's conclusion, but did not have a diagnosis of depression. Chronic illnesses represented in the study included cardiovascular disease, fibromyalgia, other chronic pain, obesity, cancer, chronic obstructive pulmonary disease, multiple sclerosis, and other neurologic conditions. Primary outcomes included depressive symptoms and a variety of exercise-related objective criteria and self-reported function-related measures. On average, participants exercised three times per week for 42 minutes per session over 17 weeks. The mean adherence rate was 77 percent of prescribed sessions.

Results: Of the 216 randomized trials identified, 90 were included in the meta-analysis. Effect sizes were calculated for the exercise versus nonexercise treatments, and a larger decrease in depressive symptoms among persons in the exercise group than those in the control group resulted in a positive effect size. In the mixed effects multiple linear regression analysis, the authors included seven primary moderators: physical activity exposure, change in fitness, illness type, change in the trial's primary outcome, blinded allocation, attention-control use, and intention-to-treat analysis. There was significant improvement in baseline depressive symptoms in persons in the exercise group compared with the nonexercise participants. The effect of exercise was greater when patients met moderate or vigorous physical activity recommendations and when the primary trial outcome was significantly improved. The number needed to treat was 6.

Conclusion: Exercise training reduces depressive symptoms in patients with chronic illness. Evidence suggests that improving depression improves outcomes for patients with medical illness.