Psychometric Properties and Factor Structure of a Long and Shortened Version of the Cognitive and Behavioural Responses Questionnaire.

Abstract

Symptoms of chronic fatigue syndrome (CFS) can be perpetuated by cognitive and behavioral responses to the illness. We aimed to determine the factor structure, reliability, and validity of the 40-item Cognitive and Behavioural Responses Questionnaire (CBRQ) using data gathered from CFS patients. We also propose a short-version CBRQ for greater clinical utility. The psychometric analysis was performed on data sets drawn from two sources: a clinical service for CFS patients (n = 576) and the PACE randomized controlled trial of CFS treatments (n = 640). An exploratory factor analysis was conducted on the clinical data set and a confirmatory factor analysis was performed on the randomized controlled trial data set. Using these results, a short version of the CBRQ was proposed. Reliability, metric invariance across age and sex, and construct validity were assessed. The exploratory factor analysis (relative χ = 2.52, root mean square error of approximation = 0.051, comparative fit index = 0.964, Taylor-Lewis Index = 0.942) and confirmatory factor analysis (relative χ = 4.029, root mean square error of approximation = 0.069, comparative fit index = 0.901, Taylor-Lewis Index = 0.892) revealed that eight-factor models fitted the data well. Satisfactory Cronbach's α values were obtained for the final subscales (≥0.76). The shortened CBRQ was obtained by removing items that cross-loaded onto other factors and/or were the lowest loading items in each factor. The shortened CBRQ contained 18 items that had high factor loadings, good face validity, and reliability (Cronbach's α = 0.67-0.88). The CBRQs, long and short versions, are reliable and valid scales for measuring cognitive and behavioral responses of patients with CFS. Further research is needed to examine the utility of the CBRQ in other long-term conditions.