Validity evidence for the back beliefs questionnaire in the general population

Abstract

Background: Despite common usage of the back beliefs questionnaire (BBQ) in a variety of studies, important validity evidence is missing. The objective of this study was to examine the validity of the BBQ in the general population.

Methods: A population-based, cross-sectional study design was used. Adult residents in two Canadian provinces were randomly sampled. To examine structural validity, items from the BBQ were subjected to factor analysis. Construct validity was tested by examining two hypotheses: BBQ scores would be most pessimistic in those with a recent history of back pain and in those who utilized passive treatments for back pain. Multiple linear regression was used to analyse the two hypotheses.

Results: Complete data were available for 6171 subjects. The best structure for the BBQ was to use eight or nine of the 14 items for scoring, which is consistent with the structure reported by the BBQ developers. BBQ scores varied based on participants' history of low back pain (LBP) and depended on pain severity. Those with severe pain and a recent history of LBP had the most pessimistic BBQ scores. In addition, participants who utilized passive management behaviours such as bed rest and activity avoidance had more pessimistic BBQ scores compared with those that did not.

Conclusions: This study provides strong validity evidence supporting the current structure and scoring of the BBQ. In addition, construct validity was evidenced by the behaviour of BBQ scores in a manner congruent with our hypotheses, further supporting use of the BBQ in the general population.