To assess the efficacy of exercise therapy for acute low back pain, a randomized, placebo-controlled trial was performed in 40 Dutch general practices. Patients received either exercise instruction with advice for daily life by a physiotherapist; placebo ultrasound therapy by a physiotherapist; or usual care by the general practitioner. All patients received analgesic agents and information on low back pain before randomization. Four hundred seventy-three patients were included. No differences in number of recurrences, functional health status, or medical care usage could be found among the three groups. In the exercise group, duration of recurrences was shorter and patients were less tired during the first 3 months than in the usual care group, but no differences were found between the exercise and placebo groups. It was concluded that exercise therapy for patients with acute low back pain has no advantage over usual care from the general practitioner.