Study design: Secondary analysis of a prospective observational cohort study.
Objectives: To evaluate whether depression and somatization subscores of the Symptom Checklist-90-Revised (SCL-90-R), which have been shown to identify chronic disability in individuals with nonspecific low back pain, are applicable to a different population of individuals with low back pain; and to determine if this potential association is confounded by a combination of centralization and subsequent treatment based on centralization.
Background: To help direct management of patients with nonspecific low back pain, recommendations include performing tests designed to identify psychosocial risk factors predictive of poor patient outcomes. SCL-90-R depression and somatization subscores have been shown to predict chronic disability among patients with low back pain.
Methods: SCL-90-R depression and somatization subscores and data on centralization were collected during the initial physical therapy examination of 231 consecutive patients treated for low back pain in 2 clinics. Disability was assessed by the Oswestry Disability Questionnaire at intake and discharge from physical therapy, and work status was determined by patient self-report at 6 and 12 months after discharge. Pain intensity was assessed by the numeric pain rating scale at the initial visit, and at 6- and 12-month follow-ups. Data were analyzed using logistic regression.
Results: Odds ratios for the association between depression and somatization subscores and patient outcomes ranged from 0.76 to 2.93. For analyses in which the data suggested a trend toward an association, the association was less evident following adjustment for centralization and centralization-based treatment.
Conclusions: In our sample, in which all individuals received physical therapy, and those who centralized received interventions based on the direction of centralization, SCL-90-R depression and somatization subscores were moderately associated with chronic pain and disability. This association was reduced when centralization and centralization-based treatment was considered in multivariable analyses.