Twenty-six patients attending the pain clinic of a large metropolitan general hospital were randomly assigned to receive either twelve sessions of psychodynamic psychotherapy or six sessions of cognitively orientated supportive psychotherapy. Questionnaires measuring aspects of illness behaviour, depression and anxiety were administered before and after treatment, but did not reveal any significant differences between the treatment groups at any time. Global subjective estimates of outcome reported to an independent observer were made after completion of therapy and showed no significant differences. All measures were repeated six and twelve months after completion of therapy and revealed significant improvement in the dynamic therapy group only with respect to levels of activity. These findings suggest that brief psychodynamically oriented psychotherapy may have a role in the management of chronic pain but further evaluation of this approach is clearly necessary, involving larger numbers of patients, before a more definitive conclusion can be reached.