ArticleThe impact of psychologically different patient groups on outcome after a vocational rehabilitation program for long-term spinal pain patientsBergström, Gunnara,*; Jensen, Irene B.a; Bodin, Lennartb; Linton, Steven J.c; Nygren, Åke L.aAuthor Information aSection for Personal Injury Prevention, Karolinska Institute, Box 127 18, S-112 94 Stockholm, Sweden bÖrebro University, Department of Statistics; 701 82 Örebro, Sweden cDepartment of Occupational and Environmental Medicine, Örebro Medical Center, Örebro, Sweden *Corresponding author. Tel.: +46-8-692-2257; fax: +46-8-653-9413 E-mail: [email protected] Submitted June 28, 2000; revised March 9, 2001; accepted March 30, 2001. Pain: September 2001 - Volume 93 - Issue 3 - p 229-237 doi: 10.1016/S0304-3959(01)00320-7 Buy Metrics Abstract A better knowledge of differential treatment outcomes for subgroups of chronic spinal pain patients may, for instance, help clinicians in treatment planning or pain researchers in treatment outcome research. The purpose of this prospective study was to evaluate the predictive validity of a subgroup classification based on the Swedish version of the (West Haven Yale) Multidimensional Pain Inventory, the MPI-S. Patients referred to a vocational rehabilitation program were classified into one of three groups, labeled ‘adaptive copers’, ‘dysfunctional’ patients, and ‘interpersonally distressed’ patients, and followed over an 18-month follow-up period. The outcome variables were absence from work (defined as sick listing plus early retirement), general health status, and utilization of health care resources. To our knowledge, the predictive validity of the MPI subgroups has not been evaluated regarding sick listing and early retirement after rehabilitation. As hypothesized, the results showed that the ‘dysfunctional’ patient group had significantly more registered absences from work and reported higher utilization of health care, over the follow-up period compared to the ‘adaptive copers’. Furthermore, as hypothesized, the ‘interpersonally distressed’ and ‘dysfunctional’ patient groups report a poorer general health status than the ‘adaptive copers’ over the whole follow-up period. However, contrary to our hypothesis, the proportion of improved patients did not differ significantly between the subgroups. Altogether, the predictive validity of the MPI-S subgroup classification was mainly confirmed. The clinical implications of this study suggest that the matching of treatment to patient needs may enhance treatment outcome, reduce pain and suffering among chronic spinal pain patients and facilitate a better health economic allocation of treatment resources. © 2001 Lippincott Williams & Wilkins, Inc.