The WHODAS-2 and Community Psychiatry

Dear Editor,

The third and fourth editions of the Diagnostic and Statistical Manual of Mental Disorders (DSM) included a multi-axial diagnostic system that included an assessment of functioning, the Global Assessment of Functioning (GAF) scale. While DSM-5 does not use the multi-axial system, it recognizes that functioning is an important component of mental health and is relevant psychiatric practice, and identifies the World Health Organization Disability Assessment Schedule 2.0 (WHODAS-2) as a possible measurement strategy. ¹ The WHODAS-2 was developed through extensive international collaboration and with close attention to cross-cultural validity. ² A brief 12-item self-report version is available. Despite these advantages, the instrument has not generally been embraced by psychiatrists in Canada.

Hoenhe et al. recently evaluated the WHODAS-2 in a psychiatric emergency population, confirming its strong performance in that setting. ³ These authors conducted a confirmatory factor analysis, whereas a previous analysis of WHODAS-2 using data from the Canadian Community Health Survey (CCHS-MH) did not confirm the instrument’s factor structure. ⁴ This leaves some questions about its factor structure in community populations with mood and anxiety disorders.

Motivated by Hoenhe et al.’s work, we conducted a confirmatory factors analysis of CCHS-MH respondents who reported that they had been diagnosed with mood or anxiety disorders, and those who fulfilled diagnostic criteria for these disorders according to a Canadian adaptation of the Composite International Diagnostic Interview (CIDI), ⁵ a fully structured diagnostic interview administered to CCHS-MH respondents. We linked items associated with each of the instrument’s six domains of functioning assessed by the WHODAS-2, allowing each of these domains to correlate with each other. We used the root mean square error of approximation (RMSEA), the Tucker-Lewis Fit Index (TLI) and Comparative Fit Index (CFI) to assess the fit of this model in each of the sub-populations of interest.

The performance of the WHODAS-2 was consistently good. In respondents reporting that they had been diagnosed with a mood disorder the RMSEA was 0.048 (95% CI, 0.046 to 0.049), with values <0.08 generally being regarded as representing good fit. The CFI was 0.974 and the TLI was 0.962, with values > 0.95 on each being generally regarded as evidence of good fit. Among those reporting anxiety disorders, the RMSEA was 0.046 (95% CI, 0.045 to 0.048), the CFI was 0.977, and the TLI was 0.966. Similarly, in a population with any disorder diagnosed by the CIDI (including substance use disorders), the RMSEA was 0.048 (95% CI, 0.046 to 0.050), the CFI was 0.974, and the TLI was 0.961.

The WHODAS-2 appears to provide a practical and valid means of assessing functioning both in acute settings and in community populations with common disorders. Psychiatrists should consider incorporating this measure into their practices.