Confirmatory Factor Analysis of the 12-Item Whodas 2.0 within the Clubhouse Model of Psychosocial Rehabilitation for Serious Mental Illness

Background and Purpose: The Clubhouse Model (CM) of psychosocial rehabilitation for serious mental illness (SMI) originates within the field of social work and is gaining attention as a preferred strengths-focused approach for mitigating functional disability. The CM is recovery-oriented and member-driven, with aims to reduce social isolation and promote functional improvement through replication of the “real world” with a work-ordered day. Despite a promising evidence-base, efficacy and mechanism evaluation of the CM is limited by the lack of uniform functional disability assessment, that can be feasibly implemented in this unique setting. The World Health Organization (WHO) Disability Assessment Schedule 2.0 (WHODAS-2.0) is a widely accepted assessment of functional disability that represents a viable option for standardized CM outcomes evaluation. However, the psychometric properties of the measure, including the conceptual factor structure, require establishment within the CM.

Methods: The primary aim of this research was to confirm the conceptual factor structure of the 12-item WHODAS-2.0 within the CM using retrospective administrative data from 339 adults with SMI from an accredited and established Clubhouse in Cleveland, OH. The 12-item WHODAS-2.0 was interview-administered by trained staff at the time of member treatment planning. Only the members’ initial WHODAS-2.0 was used for analysis. This measure is completed via self-report and generates a general disability score (GDS) from a multidimensional factor structure across six subdomains: mobility, life activities, cognition, participation in community, self-care, and getting along with others. Items are scored on a 5-point Likert scale, with 1 indicating no impairment to 5 indicating extreme impairment. Second-order confirmatory factor analysis (CFA) was conducted in R using conventional fit statistics (i.e., CFI ≥ 0.90). Secondary known-groups analyses were also conducted to confirm whether the WHODAS-2.0 adequately differentiates between subgroups known to report varying degrees of disability.

Results: The 12-item WHODAS-2.0 demonstrated good overall reliability (GDS: Cronbach’s α = 0.81) among CM participants. Despite excellent fit, the generic six-factor structure produced non-significant loadings within the ‘participation in community’ subdomain. Further investigation revealed high covariance (0.92) between this subdomain and the ‘getting along with others’ subdomain. These subdomains were combined and the final five-factor model presented excellent fit (χ² = 52.01, df = 49, p = .357; CFI = 0.998; RMSEA = 0.013). All paths were significant (all p’s <0.038) with adequate-to-strong loadings (0.29-0.94) and no correlation among error terms. The WHODAS-2.0 significantly differentiated members by length of membership, receipt of public assistance, and number of medical comorbidities.

Conclusion and Implications: These results provide initial support for the use of the 12-item WHODAS-2.0 as a CM-related outcome measure. However, adjustments to the factor structure will likely be needed for use in the CM. The intentional community approach of the CM, which supports the development of a sense of purpose, people, and place, is unique in community-based mental health care. Therefore, it is likely that the items constructing ‘participation in community’ and ‘getting along with others’ are measuring equivalent constructs in this model. These findings support continued investigation of the psychometric properties of the 12-item WHODAS-2.0 within the CM.