IADR Abstract Archives
Jaw Functional Limitation Scale: Revision and Psychometric Properties for TMDs
Objectives: The 20-item Jaw Functional Limitation Scale (JFLS; Ohrbach et al, 2008) is a validated patient-reported outcome measure (PROM) for masticatory system function. This study aimed to increase its utility by maximizing its properties for temporomandibular disorders (TMDs) and for computer adaptive testing (CAT) which provides better precision and efficiency.
Methods: We conducted secondary analysis of data from two cross-sectional studies in which adults with chronic TMDs (n=~1800 , depending on analysis) completed a 21-item JFLS. Exploratory factor analysis (EFA) was used to identify factor structures, and confirmatory factor analysis (CFA) was used to test their psychometric properties. Item response theory was used before EFA and after CFA to evaluate contribution of each item. Bi-factor analysis was used to test for presence of a dominant underlying latent variable.
Results: Initial item analysis identified 6 items with poor properties that were excluded from further analysis. EFA disclosed a 3-factor model: verbal/emotional expression (as reported for the original JFLS) and two novel factors labelled extreme function and non-extreme function. Using CFA, each subscale exhibited excellent psychometric properties with respect to model fit: root mean square error of approximation (RMSEA≤0.1), comparative fit index (CFI≥0.95), Tucker-Lewis Index (TLI 0≥.95) and standardized root mean residuals (SRMR ≤0.08). Subsequent item analysis within each subscale identified two more items for removal, yielding a 13-item instrument comprising three subscales and a 6-item global short form, all of which had excellent reliability (>0.90). Bi-factor analysis isolated one major general factor (omega-H=0.90) with RMSEA=0.099, CFI=0.97, TLI=0.96, and SRMR=0.07; the three secondary factors exhibited small omega-H values of 0.15, 0.33, and 0.13, respectively.
Conclusions: This TMD-specific JFLS has excellent psychometric properties for assessing generalized functional limitation of the masticatory system; its three subscales can be used to answer more specific questions. Properties are excellent for CAT, improving its value as a PROM for TMDs.
Methods: We conducted secondary analysis of data from two cross-sectional studies in which adults with chronic TMDs (n=~1800 , depending on analysis) completed a 21-item JFLS. Exploratory factor analysis (EFA) was used to identify factor structures, and confirmatory factor analysis (CFA) was used to test their psychometric properties. Item response theory was used before EFA and after CFA to evaluate contribution of each item. Bi-factor analysis was used to test for presence of a dominant underlying latent variable.
Results: Initial item analysis identified 6 items with poor properties that were excluded from further analysis. EFA disclosed a 3-factor model: verbal/emotional expression (as reported for the original JFLS) and two novel factors labelled extreme function and non-extreme function. Using CFA, each subscale exhibited excellent psychometric properties with respect to model fit: root mean square error of approximation (RMSEA≤0.1), comparative fit index (CFI≥0.95), Tucker-Lewis Index (TLI 0≥.95) and standardized root mean residuals (SRMR ≤0.08). Subsequent item analysis within each subscale identified two more items for removal, yielding a 13-item instrument comprising three subscales and a 6-item global short form, all of which had excellent reliability (>0.90). Bi-factor analysis isolated one major general factor (omega-H=0.90) with RMSEA=0.099, CFI=0.97, TLI=0.96, and SRMR=0.07; the three secondary factors exhibited small omega-H values of 0.15, 0.33, and 0.13, respectively.
Conclusions: This TMD-specific JFLS has excellent psychometric properties for assessing generalized functional limitation of the masticatory system; its three subscales can be used to answer more specific questions. Properties are excellent for CAT, improving its value as a PROM for TMDs.