A controlled trial was performed in a occupational healthcare setting to determine the effectiveness of graded activity as part of a multistage RTW programme. Workers (112) absent from work for more than eight weeks due to low back pain, were randomised to either graded activity (n = 55) or usual care (n = 57). The graded activity, a physical exercise programme aimed at RTW, was based on operant-conditioning behavioural principles. Main outcome measures were: number of days off work until first RTW for more then 28 days; total number of days on sick leave during follow u; functional status; severity of pain. Follow up was 26 weeks. Results showed, among others things, median time until RTW was equal to the total number of days on sick leave. It was 139 (IQR = 69) days in the graded activity group and 111 (IQR = 76) days in the usual care group (hazard ratio = 0.52, 95% CI 0.32 to 0.86). Graded activity did not improve pain or functional status clinically significantly. In is concluded that graded activity was not effective for any of the outcome measures. Different interventions combined can lead to a delay in RTW. Delay in referral to graded activity delays RTW. In implementing graded activity special attention should be paid to the structure and process of care.