The relationship between activity level and depression in primary care outpatients was examined. Forty-eight patients with depression and 25 general medical controls wore a heart and activity level monitor for 24 h. Minute-averaged horizontal movement was collected. Patients with depression were sorted into two equal groups based on a median split of their Beck Depression Inventory-II scores. ANOVAs revealed significant main effects for average movement over the 24-h interval, average movement between 12 and 6 P.M., and a measure of “high activity” level. Patients with more severe depression had lower activity levels than the other two patient groups. The nondepressed and less severely depressed groups were combined and compared to the more severely depressed group using a standard discriminant function analysis. The overall correct classification rate was 74.0%, with 86% of the combined group and 50% of the more severely depressed group correctly classified. Patients with low activity levels were 6.0 times more likely to fall in the severely depressed group than patients with more normal activity levels.