Predictors of poor outcome in patients with neck pain treated by physical therapy

Objectives: This study evaluated predictors of poor outcome in patients with neck pain treated by physical therapy, and sought to compare the findings of empirical data with physical therapy practitioners' subjective perceptions about predictors of outcome. Methods: This study was a secondary analysis of data from a randomized controlled trial assessing physical therapy treatments for neck pain (n = 346). A baseline questionnaire provided data on a number of potential risk factors of poor outcome at follow-up. Follow-up was conducted at 6 weeks and 6 months by postal questionnaire with outcome defined separately by perceived (global) change and minimal clinically important differences in the Northwick Park Neck Pain Questionnaire (MCID-NPQ). Therapists' perceptions of predictors for treatment outcome were captured using a separate questionnaire, and the ratings compared with ranks derived from the 6-month trial data. Results: Baseline characteristics accounted for a much greater proportion of explained variance for global change compared with MCID-NPQ at 6 months. Independent biopsychosocial characteristics included manual social class, catastrophizing, anxiety and depression, low treatment expectations, severity of baseline neck pain/disability, presence of comorbid back pain, and older age. Physical therapist ranks correlated highly with those derived from the trial data. Conclusions: Significant predictors of outcome were identified, particularly at 6 months, including psychosocial, functional, and demographic indicators. Our findings suggest that physical therapists are collectively aware of the relative importance of physical and psychosocial factors in predicting clinical outcome. However, a significant amount of variability in outcomes in our prognostic models remained largely unexplained, indicating that we need to explore further underlying factors to inform clinical decision-making.