Psychometric characteristics of the Spanish version of the FAB questionnaire

Study Design. Validation of a translated, culturally adapted questionnaire. Objectives. To translate and culturally adapt the Spanish version of the Fear Avoidance Beliefs Questionnaire (FABQ), and to validate its use in Spanish-speaking patients with low back pain (LBP). Summary of Background Data. The FABQ is a reliable evaluation instrument for fear avoidance beliefs, which includes two subscales (FAB-Work and FAB-Phys). No validated Spanish version was available. Methods. Translation/retranslation of the English version of the FABQ was done blindly and independently by a multidisciplinary team. The study was done in 12 primary care centers and 9 hospital outpatient clinics from seven regions in Spain, with 209 acute, subacute, and chronic patients who visited their physician for LBP: 53 in the pilot phase and 156 in the validation phase. Subjects were given the FABQ, two VAS for LBP and referred pain, and the Roland-Morris and SF-12 Questionnaires on their first visit and 14 days later. In the pilot phase, on day 1 test-retest reliability was estimated by giving a second FABQ in which the name and order of the items had been changed. Results. Time necessary to complete the FABQ was [median, P25, P75] 10 minutes (5,15). FABQ values were not normally distributed. Comprehensibility: No request for aid in interpretation was made during the validation phase and no item was left unanswered by >= 10% of patients. Reliability: Scores [median, P25, P75] of the two FABQ were: 72.00 (47.25, 82.00) and 72.00 (49.50, 83.75), with an intraclass correlation coefficient of 0.9668 (95% confidence interval, 0.9421,0.9823). Mean of kappa values for all items was 0.743. Internal Consistency: Cronbach's alpha was 0.9337. Validity: Values of FAB and FAB-Work were > 37% higher for patients on sick leave. For FAB-Phys differences were below 8%. FABQ, FAB-Work and FAB-Phys strongly correlated with disability on days 1 and 15. Conclusions. The Spanish version of the FABQ has good comprehensibility, internal consistency, and reliability. It cannot be analyzed parametrically, but only nonparametrically. The total FABQ is at least as valid as its subscales and simpler to score, making it more suitable for routine clinical use.