Data in the preceding editorial show that, across most of the adult lifespan, energy expenditure attributable to sport and recreation is much lower than that attributable to occupation. The editorial makes the point that most studies on the relationship between physical activity and health have focused largely on leisure-time activity, and may therefore be vulnerable to ‘missing important exposure information’.1

While this is true, we would do well to recall that the earliest studies of physical activity epidemiology relied primarily on measures of occupational physical activity, and that this field of research has now turned full circle, as outlined in our earlier review paper (from early roots in occupational sitting, through aerobic fitness training, then moderate-intensity physical activity, to a contemporary perspective on the balance between activity and inactivity in different domains of everyday life).2 It is also important to recall that all these studies showed relationships between physical activity …