This paper investigates an alternative proxy for individual worker productivity in physical work settings: a direct measure of physical activity using an accelerometer. First, the paper compares worker labor outcomes, such as labor supply and daily productivity obtained from firm personnel data, with physical activity; they are strongly related. Second, the paper investigates the effect of a health intervention on physical activity, using a temporally randomized offer of malaria testing and treatment. Workers who are offered this program reallocate time from lower intensity activities in favor of higher intensity activities when they work.