Authors (including presenting author) :
Sam Shih (1) Ashley Chan (1) Eva Yeung (1) Amily Tsang (1) Rose Chiu (1) Menza Chu (1) Magdalene Poon (1)
Affiliation :
(1)Occupational Therapy Department, Kwai Chung Hospital
Introduction :
People with common mental disorders (CMD) are highly prevalent worldwide and in Hong Kong while functional impairment is significantly associated with people with common mental disorder. Work and social adjustment scale (WSAS) was then translated into Chinese and validated to aid treatment planning and outcome evaluation. WSAS was developed to measure the functional impairment resulting from a health problem. It consists of five domains, comprising ability to work, home management, social leisure, private leisure, and ability to form and maintain close relationship with others. Each item is rated on a 9-point Likert scale from 0 (no impairment) to 8 (very severe impaired) by respondents.
Objectives :
To examine the psychometric properties of CWSAS in outpatient with CMD from classical test theory and Rasch analysis and its clinical implication
Methodology :
Forward and backward translations were done and field test was conducted. Written consent from both participants and their case psychiatrists were sought. Classical testing analysis and Rasch analysis were studied in 252 participants who were recruited from West Kowloon Psychiatric Center and East Kowloon Psychiatric Center from Oct 2018 to March 2020.
Result & Outcome :
Principal component analysis yielded one-factor solution, determining functional impairment, accounting for 68.1% of total variance with factor loading ranged from .74 to .87 and CWSAS demonstrated adequate concurrent validity as it has significantly moderated correlation with other outcome measures including PHQ-9 (r = .54), GAD-7 (r = .46), WHO-5 (r = -.42) & PSS-10 (r = .55). Besides, CWSAS demonstrated adequate internal consistency (Cronbach’s alpha = .88) and acceptable item-total correlation (ranging from .61 to .78) & inter-item correlation (within .44 & .71). Employment status was found to have group difference (F = 4.672, p < .01), showing people without employment had significant higher perceived impairment in item of work while there was no gender difference. The total scores of CWSAS could be predicted by PSS-10 and PHQ-9 through multiple linear regression analysis [F(2, 249) = 69.572, p < .001, R2 = .304; adjusted R2 = 302]. Treatment intervention focusing on these two aspects are indicative. CWSAS was also revealed its distinct factor among other outcome measures i.e. PHQ-9 & GAD-7, WHO-5 by principal component analysis with direct Oblimin rotation. The eigenvalue for PHQ-9 & GAD-7, WHO-5, & WSAS were 13.64, 2.31 & 1.91 which explained 52.45%, 8.89% and 7.35% of the variances respectively. Rasch analysis indicated all items fit the Rasch model and it was a well-targeted scale with no local dependency (all item pairs in largest standardized residual correlations ranging from -.01 to -.47). Rasch analysis also confirmed its unidimensionality as the findings indicated that 65.3% of the variance was explained by the Rasch model and the eigenvalue in the first contrast (1.81) of the residual variance was less than 2. Our result showed that CWSAS had good personal separation index (2.28) and personal reliability (.84), indicating the power of CWSAs was able to distinguish our sample into 2 to 3 levels of impairment and possessed acceptable discriminant ability. No categorical disordering was found while inadequate adjacent threshold distance was reported. Item of work indicated noticeable DIF in employment status (DIF contrast = -.61, p< .001) and source of finance (DIF contrast =-.63, p< .001) according to Linacre’s criterion. Conclusion: Both classical testing theory and Rasch analysis indicated that CWASA was unidimensional and possessed adequate reliability and discriminant ability. It is noteworthy that the treatment interventions consisting of stress management and depression management are highly indicative to promote their daily functioning. To sum up, Chinese version of WSAS is a brief and psychometrically appropriate measure for functional outcome for outpatients with common mental disorders.
