OBJECTIVE: To assess the benefit and harm of exercise training in adults with clinical depression. METHOD: The DEMO trial is a randomized pragmatic trial for patients with unipolar depression conducted from January 2005 through July 2007. Patients were referred from general practitioners or psychiatrists and were eligible if they fulfilled the International Classification of Diseases, Tenth Revision, criteria for unipolar depression and were aged between 18 and 55 years. Patients (N = 165) were allocated to supervised strength, aerobic, or relaxation training during a 4-month period. The primary outcome measure was the 17-item Hamilton Rating Scale for Depression (HAM-D(17)), the secondary outcome measure was the percentage of days absent from work during the last 10 working days, and the tertiary outcome measure was effect on cognitive abilities. RESULTS: At 4 months, the strength measured by 1 repetition maximum for chest press increased by a mean (95% CI) of 4.0 kg (0.8 to 7.2; p = .014) in the strength training group versus the relaxation group, and maximal oxygen uptake increased by 2.7 mL/kg/min (1.2 to 4.3; p = .001) in the aerobic group versus the relaxation group. At 4 months, the mean change in HAM-D(17) score was -1.3 (-3.7 to 1.2; p = .3) and 0.4 (-2.0 to 2.9; p = .3) for the strength and aerobic groups versus the relaxation group. At 12 months, the mean differences in HAM-D(17) score were -0.2 (-2.7 to 2.3; p = .8) and 0.6 (-1.9 to 3.1; p = .6) for the strength and aerobic groups versus the relaxation group. At 12 months, the mean differences in absence from work were -12.1% (-21.1% to -3.1%; p = .009) and -2.7% (-11.7% to 6.2%; p = .5) for the strength and aerobic groups versus the relaxation group. No statistically significant effect on cognitive abilities was found. CONCLUSION: Our findings do not support a biologically mediated effect of exercise on symptom severity in depressed patients, but they do support a beneficial effect of strength training on work capacity. TRIAL REGISTRATION: (ClinicalTrials.gov) Identifier: NCT00103415.
Journal of Clinical Psychiatry, 2009, Vol 70, Issue 6, p. 790-800