Functional Index Questionnaire: structural validity study in Brazilian patients with anterior knee pain

To assess the Functional Index Questionnaire (FIQ) structure using confirmatory factor analysis (CFA) in Brazilian patients with anterior knee pain.

Brazilian patients of both sexes (n = 100), aged ≥ 18 years, with anterior knee pain for at least 3 months were included. Eligible participants completed an online form that collected personal and clinical data as well as responses to the assessment tools. We used CFA and the following fit indices: chi-square/degrees of freedom (DF), comparative fit index (CFI), Tucker-Lewis index (TLI), root mean square error of approximation (RMSEA), and standardized root mean square residuals (SRMR).

The majority of the respondents were women, young adults, overweight, with incomplete higher education, were physically active, and had pain in sitting or squatting positions. The mean duration of pain was 38.24 months, and the mean pain intensity was 4.54 points. The model fit indices were as follows: χ2/DF = 2.08, TLI = 0.978, CFI = 0.969, RMSEA = 0.104, and SRMR = 0.077. Therefore, the one-dimensional structure with eight items yielded an RMSEA value above the 0.08 cutoff point, suggesting a poorer fit and more residual error than is acceptable for a well-fitting model. Using the modification indices within the CFA, we observed a correlation between Items 2 (climbing up two flights of stairs [16 steps]) and 6 (climbing up four flights of stairs [32 steps]) and Items 3 (squatting) and 4 (kneeling), indicating the similarity in the response pattern for these items. After adding these correlations to the model, we obtained improved fit indices (χ2/DF = 1.51, TLI = 0.990, CFI = 0.985, RMSEA = 0.072, and SRMR = 0.061).

This version of the FIQ should be used with caution, as the unidimensional model demonstrates substantial residuals, mainly because of item redundancy.