The development and measurement properties of the Dutch version of the fear-avoidance components scale (FACS-D) in persons with chronic musculoskeletal pain

Abstract

Abstract Objectives The Fear-Avoidance Components Scale (FACS) is a recently developed patient-reported instrument assessing different constructs related to the fear-avoidance model of pain. The aim was to translate the original English FACS into Dutch (FACS-D) and assess its measurement properties in persons with chronic musculoskeletal pain. Methods The original English FACS (20 item-scale, range: 0–100) was translated in Dutch through standard forward-backward translation methodology. The FACS-D’s measurement properties were evaluated in 224 persons with chronic musculoskeletal pain. Internal consistency, test-retest reliability and measurement error were assessed with the Cronbach’s alpha coefficient (α), intraclass correlation coefficient (ICC), and standard error of measurement (SEM). Construct validity was assessed through inter-item correlation analyses, exploratory factor analysis, association with other fear-avoidance-related constructs, and hypothesis testing. Results Internal consistency, test-retest reliability and hypotheses testing were good (α=0.92; ICC=0.92, CI 0.80–0.96; 7/8 hypotheses confirmed). Similar to the original FACS and other translated versions, a two-factor model best fit the data. However, the item distribution differed from other versions. One factor represented “pain-related cognitions and emotions” and a second factor represented “avoidance behaviour.” In contrast to the original FACS, low inter-item correlations for item 12 were found. The FACS-D was more strongly associated with fear-avoidance-related constructs of pain severity, perceived disability, feelings of injustice, and depressive/anxiety symptoms than the other fear-avoidance-related scales studied here. Conclusions The FACS-D demonstrated good reliability and construct validity, suggesting that it may be a useful measure for Dutch-speaking healthcare providers. Two clinically relevant factors, with a different item distribution than the original FACS, were identified: one covering items on pain-related cognitions and emotions, and one covering items on avoidance behaviour. The stronger association between FACS-D and fear-avoidance related constructs suggests that the FACS-D may be more effective in evaluating the cognitive, emotional and behavioural constructs of pain-related fear-avoidance than other similar measures.