The 28 patients in the Linton et al. (1985) outcome study were contacted for a 20-month follow-up. The follow-up was superimposed on a three-group design consisting of a waiting-list control, a regular treatment group, and a “behavioural” group receiving the regular treatment plus applied relaxation and operant-activities training. Measures similar to the ones used in the original study were employed, in addition to a follow-up questionnaire concerning work, health status, activity level, pain ratings, medicine use, and sleep. Ninety-three per cent of the original patients participated in the follow-up. In general, improvements were maintained by the behavioural group. Specifically, results indicated that the group receiving applied relaxation and operant-activities training had significantly larger improvements with regard to pain and activity at follow-up than the control groups. Moreover, an overall analysis of variance showed that the behavioural group differed significantly from the other two groups. The results definitely suggest that the positive effects of the behavioural programme were maintained at follow-up.