Factors predicting clinical outcome 12 and 36 months after an exercise intervention for recurrent low-back pain

Purpose: The aim of this cohort study was to identify early predictive factors for a poor outcome of disability and pain 12- and 36-months after an intervention in patients with recurrent low-back pain, currently at work. Method: Seventy-one patients with recurrent low-back pain, all at work, seeking care in a primary health care setting were included. Predictive indicators including demographic data and health-related variables were derived from questionnaires pre- and post intervention over eight weeks. The dependent outcome variables were perceived disability and present pain at 12- and 36-months. Results: Multivariate regression analyses show that early data on poor self-efficacy for physical activity, greater disability, and higher level of pain-ratings emerged as independent predictors of a poor outcome of disability at 12 and 36 months. Higher ratings of pain and poor self-efficacy appeared again as independent predictors of a poor outcome of pain at the 12-month follow-up. Pain frequency ratings predicted a poor outcome of pain at 36 months. Conclusions: Our results suggest that ratings of poor self-efficacy for physical activity, greater disability, and pain-ratings, are the most consistent independent predictors of long-term poor outcome of disability and pain. This indicates the importance of screening for such factors to optimize the management of low-back pain. However, larger studies in similar patient populations are needed to confirm these results.