Behavioral approaches to treating patients following lumbar disc surgery are becoming increasingly popular. The treatment method is based on the assumption that pain and pain disability are not only influenced by somatic pathology, if found, but also by psychological and social factors. A recent study highlighted the effectiveness of cognitive-behavioral interventions, as compared to no treatment, for chronic low back patients. However, to the authors' knowledge, there is no randomized controlled trial that evaluates a behavioral program for patients following lumbar disc surgery. The purpose of this study was to assess the effectiveness of a behavioral graded activity (BGA) program compared to usual care (UC) in physiotherapy following first-time lumbar disc surgery. The BGA program was a patient-tailored intervention based upon operant therapy. The essence of the BGA is to teach patients that it is safe to increase activity levels. The study was designed as a randomized controlled trial. Assessments were carried out before and after treatment by an observer blinded to treatment allocation. Patients suffering residual symptoms restricting their activities of daily living and/or work at the 6 weeks post-surgery consultation by the neurosurgeon were included. The exclusion criteria were: complications during surgery, any relevant underlying pathology, and any contraindication to physiotherapy or the BGA program. Primary outcome measures were the patient's Global Perceived Effect and the functional status. Secondary measures were: fear of movement, viewing pain as extremely threatening, pain, severity of the main complaint, range of motion, and relapses. Physiotherapists in the BGA group received proper training. Between November 1997 and December 1999, 105 patients were randomized; 53 into the UC group and 52 into the BGA group. The unadjusted analysis shows a 19.3% (95% CI: 0.1 to 38.5) statistically significant difference to the advantage of the UC group on Global Perceived Effect. This result, however, is not robust, as the adjusted analyses reveal a difference of 15.7% (95% CI: -3.9 to 35.2), which is not statistically significant. For all other outcome measures there were no statistically significant or clinically relevant differences between the two intervention groups. In general, the physiotherapists' compliance with the BGA program was satisfactory, although not all treatments, either in the BGA or the UC group, were delivered exactly as planned, resulting in less contrast between the two interventions than had been planned for. There was one re-operation in each group. The BGA program was not more effective than UC in patients following first-time lumbar disc surgery. For Global Perceived Effect there was a borderline statistically significant difference to the advantage of the UC group. On functional status and all other outcome measures there were no relevant differences between interventions. The number of re-operations was negligible, indicating that it is safe to exercise after first-time disc surgery.