Abstract
Ninety-four patients with non-specific lumbar pain who were seen by their general practitioners took part in a double-blind controlled trial to compare mobilisation and manipulation with placebo physiotherapy. Immediately after treatment most patients showed improvements in the various features studied, but for several features improvement was more common in the treated group than the controls. At three months the condition of most patients was still improved but the differences between the two groups had largely disappeared. At one year the groups were identical. Prognostic presenting features were sought, but only a shorter history correlated with clinical improvement. This study indicates the high rate of spontaneous resolution of low back pain. A course of mobilisation and manipulation may hasten improvement but does not affect the long-term prognosis.