Abstract
Background/Objective: Inconsistencies exist regarding the exact multidimensional structure underlying the Fear-Avoidance Beliefs Questionnaire (FABQ), with no prior study examining the internal structure of the Arabic FABQ. This study aimed to examine validity evidence of the Arabic FABQ in patients with low back pain (LBP), based on two sources: validity evidence based on the internal structure (dimensionality and reliability) and validity evidence based on relations with other variables (i.e., pain intensity and disability). Methods: Participants (N = 112) with LBP were recruited from physical therapy clinics. Data were collected through the completion of FABQ and other measures of pain and disability. CFA was performed using a diagonally weighted least squares estimation. The fit of the two-factor model recommended by the original scale developer was assessed using multiple fit indices. Reliability of FABQ subscale scores was assessed using McDonald's omega (ω) and Average Variance Extracted (AVE). Results: One hundred and twelve patients with LBP with mostly chronic complaints participated in the study. The CFA supported the two-factor model with modifications to account for residual correlations between items 4-5 and 6-7, yielding improved fit indices (χ(2)(41) = 77.82; p < 0.001; TLI = 0.98, CFI = 0.99, RMSEA = 0.09 (90% CI = 0.06-0.12), and SRMR = 0.08). All factor loadings were salient and significant with values ranging from 0.43 to 0.96. The two underlying factors reflecting physical activity-related and work-related fear avoidance beliefs showed a significant positive correlation of 0.58. These findings confirm the hypothesized dimensionality of the Arabic FABQ. The FABQ work subscale scores demonstrated higher reliability (ω = 0.86; AVE = 0.54) compared to the physical activity subscale scores (ω = 0.63; AVE = 0.44), with both factors measuring related but distinct constructs. The latent scores for the FABQ related to physical activity demonstrated stronger positive correlations with pain intensity (r = 0.37; p < 0.001) and disability (r = 0.43; p < 0.001), compared to the latent scores for work-related FABQ, which showed weaker correlations with pain intensity (r = 0.22; p < 0.001) and disability (r = 0.26; p < 0.001). Conclusions: This study provides evidence to support the two-factor structure of the Arabic FABQ and the common scoring method for the FABQ and facilitates the interpretation of the FABQ subscale scores as reflecting related but distinct domains of fear avoidance beliefs.