PURPOSE: To examine the underlying structure of a new work disability assessment tool, the Work Disability Functional Assessment Battery (WD–FAB)

BACKGROUND: The WD–FAB was developed as a systematic, comprehensive assessment of work-related function to supplement the current Social Security Administration (SSA) disability determination process. Development of the WD–FAB was grounded in notions of assessing whole-person work relation function and has potential to enhance occupational therapy practice in the area of work rehabilitation and work capacity assessment.

DESIGN: Cross-sectional study of working-age adults in the United States to assess construct validity of the WD–FAB scales

PARTICIPANTS: A sample of adults (ages 21–66 yr) representative of working-age adults with regard to gender, race, ethnicity, education, and geographic region, with oversampling of racial and ethnic minorities: Black, Asian, and Hispanic.

METHOD: Data were obtained through a survey research organization that maintains an opt-in Internet survey panel of approximately 1.5 million U.S. adults. Respondents completed the WD–FAB survey questions for calibration across all WD–FAB content domains; the item sets were divided into two administrations to reduce respondent burden.

ANALYSIS: Construct validity was examined using exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). The factor analyses were performed in two primary phases, and each model was tested using multiple goodness-of-fit indices. In addition to statistical criteria, factor interpretability was considered in determining the final WD–FAB factor structure.

RESULTS: An analytic sample of 2,023 adults with an average age of 42 yr (range = 21–66); 45% male; and 53% White was obtained. EFA and CFA yielded the following dimensions for the WD–FAB: Basic Mobility (56 items; root mean square error of approximation [RMSEA] = 0.048, CFI = 0.986, Tucker–Lewis Index [TLI] = 0.985), Upper Body Function (54 items; RMSEA = 0.04, CFI = 0.985, TLI = 0.984), Applied Cognition (63 items; RMSEA = 0.064, CFI = 0.911, TLI = 0.908), Interpersonal Interactions (32 items; RMSEA = 0.079, CFI = 0.906, TLI = 0.9), Self-Efficacy (33 items; RMSEA = 0.074, CFI = 0.906, TLI = 0.9), and Mood and Emotions (34 items; RMSEA = 0.09, CFI = 0.908, TLI = 0.902).

DISCUSSION: CFA model testing confirmed the unidimensionality of six factors identified by the EFA and conceptual considerations. The CFA fit statistics support further psychometric testing using item response theory to develop scales for implementation of the WD–FAB using computer adaptive testing. This will allow systematic, efficient profiling across six important domains of work-related function. The WD–FAB is unique in its multidimensional, whole-person approach to characterizing a person’s physical, cognitive, mental, and communication function related to work.

IMPACT STATEMENT: This research represents a significant advance both conceptually and psychometrically in assessment methodologies for work-related function. The measurement of a person’s potential ability to work requires the assessment of multiple dimensions. The WD–FAB assesses key domains of work-related health functioning and may prove useful for application in the context of the SSA disability determination process as well as among occupational therapists working in the area of work rehabilitation and work capacity assessment.