LAUSR

STUDY DESIGN A cohort study with 12 months of follow-up. OBJECTIVE To assess 1) the unidimensionality of the Fear-Avoidance Beliefs Questionnaire (FABQ) and 2) whether single questions in the FABQ predict future sickness absence as well as the whole scale. SUMMARY OF THE BACKGROUND DATA The fear-avoidance model is a leading model in describing the link between musculoskeletal pain and chronic disability. However, reported measurement properties have been inconsistent regarding the FABQ. METHODS Individuals (n = 722) sick listed due to musculoskeletal, unspecified or common mental health disorders undergoing rehabilitation was included. A Rasch analysis was applied to evaluate the measurement properties of FABQ and its two subscales (physical activity and work). Linear regression was used to assess how well single items predicted future sickness absence. RESULTS The Rasch analysis did not support the FABQ or its two subscales representing a unidimensional construct. The 7-point scoring of the items was far too fine meshed and in the present population the data only supported a yes or no or a 3-point response option. The items were invariant to age, whereas two of the items revealed gender differences. The item "I do not think that I will be back to my normal work within 3 months" was the best predictor of future sickness absence. Adding the item "I should not do my regular work with my present pain" improved the prediction model slightly. CONCLUSIONS The FABQ is not a good measure of fear-avoidance beliefs about work or physical activity, and the predictive property of the FABQ questionnaire is most likely related to expectations rather than fear. Based on these results we do not recommend using the FABQ to measure fear-avoidance beliefs. LEVEL OF EVIDENCE 1.