Rasch analysis of the Italian version of fear avoidance beliefs questionnaire (FABQ-I)

To perform a Rasch analysis (RA) of the Fear Avoidance Beliefs questionnaire (FABQ-I) in a sample of chronic non-specific low back pain (CLBP) patients. The RA is expected to provide information about the internal construct validity and all the single items consistency of the FABQ-I and then contributing to the knowledge about the questionnaire's clinimetric properties. Methods: One hundred and fifty subjects with CLBP were enrolled into the study. The FABQ-I have been studied by means of RA. The sample size needed to obtain stable person, item and rating-scale calibrations for use with the Rasch model was based on the level of error expected in the measure. Results: RA did not support the total scale as an unidimensional measure of Fear Avodiance Beliefs. Also the analysis of the Physical Activity (PA) and Word (WO) subscales as well as the attempts of manipulation failed in achieving an acceptable fit to the Rasch model. Conclusion: FABQ-I may be considered to reflect a multiple psychological constructs describing scale. The raw score of the FABQ-I and changes in scores must be interpreted with caution because as a general measure of fear avoid beliefs was not supported.Implication for RehabilitationFear avoidance beliefs questionnaire (FABQ-I) as a whole, as well as the PA and WO subscales does not fit Rasch model thus indicating a "Fear Avoidance" represent a multidimensional construct.Practitioners should be aware that FABQ-I need to be used with caution in clinical settings. The score of different items should not be summed in order to obtain an "overall score of Fear Avoidance Beliefs". Every item of the FABQ-I potentially defines its own dimension.Changes in scores must be interpreted with caution because FABQ-I as a general measure of fear avoid beliefs was not supported, also the ability to distinguish between distinct levels of the underlying trait is poor.Caution: This study was conducted on a sample of Chronic Low Back Pain patients; thus, the results might not be generalized on patients with other pathologies/conditions.