Background: Wide variations in disability duration and magnitude have been noted among recipients of workers' compensation for low back pain. Findings from recent studies have indicated that inclusion of a broad array of variables (i.e., physical, occupation, social, economic) is needed to understand differences in workers' responses to occupational low back pain.
Methods: Workers' compensation and questionnaire data from 340 Oregon workers with low back claims were merged to develop multivariate models predicting: (1) absenteeism days, (2) residual symptoms, (3) functional impairment, and (4) medical costs.
Results: Forty-two percent of the variation in low back symptoms was explained by: discontinuing physical fitness activities post-injury (beta = -.419), self-reported low energy/high fatigue (beta = -.227), poorer general health (beta = .137), and attorney involvement in claim (beta = .117), (adjusted R2 = .418, p < 0.001). Survival curves revealed significantly longer claim durations among workers who discontinued physical fitness activities post-injury, compared with workers who did not; these differences remained significant even after controlling for severity of the initial injury.
Conclusion: Continuation of physical fitness activities during the recovery process was found to be a significant predictor in three of four regression models, providing evidence on behalf of a relationship between fitness and positive health outcomes. However, it was not possible to clearly differentiate pre-morbid from post-injury fitness, nor to determine if this relationship was due to a therapeutic effect on the back, the general restorative benefits of remaining active, or represents a proxy variable for workers' self-care efforts during recovery.