Neuropsychopharmacology Reports (Jun 2024)

Comparison of the 12‐item and 36‐item versions of the World Health Organization Disability Assessment Schedule (WHODAS) 2.0 using longitudinal data from community mental health outreach service users

  • Mai Iwanaga,
  • Sosei Yamaguchi,
  • Sayaka Sato,
  • Kaori Usui,
  • Kiyoaki Nakanishi,
  • Erisa Nishiuchi,
  • Michiyo Shimodaira,
  • Yugan So,
  • Chiyo Fujii

DOI
https://doi.org/10.1002/npr2.12426
Journal volume & issue
Vol. 44, no. 2
pp. 457 – 463

Abstract

Read online

Abstract Aim This study aimed to compare the 12‐item and 36‐item versions of the World Health Organization Disability Assessment Schedule (WHODAS) 2.0 using longitudinal data from community mental health outreach service users. Methods Using data from Tokorozawa City mental health outreach service users in Japan, total and domain WHODAS‐12 and WHODAS‐36 scores were compared. First, we examined score‐change differences by domain at the start of outreach services (T1) and 1 year later (T2) for each version. Next, we compared differences between the two versions using Pearson's correlation, Wilcoxon signed‐rank test, and Bland–Altman analysis. Results Among 20 participants, total scores and scores of some domains (i.e., cognition, getting along, life activities, and participation) were significantly lower at T2 than at T1 on both versions (p < 0.010). WHODAS‐36 scores were significantly lower at T2 than at T1 for the self‐care domain (p = 0.018). Except for self‐care, strong correlations were found between scores from the two versions (p < 0.001). In the Wilcoxon signed‐rank test and Bland–Altman analysis, we found significant differences between the scores of the two versions in the mobility, self‐care, and participation domains. There were no significant differences in the distribution or systematic errors between the two versions in scores for the other domains or total score. Conclusion We found strong positive correlations between WHODAS‐12 and WHODAS‐36 total scores with no statistical differences between them. For some domains, differences in distribution and systematic errors were found.

Keywords