Predicting treatment outcome of chronic back pain patients in a multidisciplinary pain clinic: Methodological issues and treatment implications

Abstract

Treatment outcome of 72 chronic back pain patients was assessed with 4 standardized measures: the McGill Pain Questionnaire, the Audiovisual Taxonomy of Pain Behavior, the Beck Depression Inventory, and the Profile of Mood States. Patients were also rated by their primary nurse on pain behavior, activity, drug-seeking, and sleep. Variables used to predict scores on outcome measures included patients'' demographics and MMPI scores. Multiple regression analyses indicated that patients receiving worker''s compensation engaged in more pain behavior and rated their pain as more severe, both upon admission and discharge from the pain program. High scores on the MMPI Hy scale were correlated with high self-ratings of pain and several MMPI scales correlated with negative mood. Even though demographic variables predicted admission and discharge scores on a number of treatment outcome measures, there was no relation between demographics and patients'' improvement on treatment outcome measures. Patients with high scores on MMPI Hy and D scales displayed greatest admission to discharge improvement on self-rated pain and mood. Results were integrated with findings from previous studies. It was suggested that by distinguishing between overall scores on treatment outcome measures and improvement on these measures, professionals will be in a better position to devise individualized treatment plans for pain patients.