Translation and validation of the traditional Chinese version of the Knowledge and Attitudes to Mental Health Scale in Hong Kong adolescents.

OBJECTIVES: To translate and validate the traditional Chinese version of the Knowledge and Attitudes to Mental Health Scale (KAMHS) among Hong Kong adolescents. METHODS: Workshops for mental health literacy (MHL) were conducted between July and August 2024 for Form 3 to Form 6 students aged 13 to 18 years from 10 invited secondary schools. Participants were asked to complete an online survey for the collection of data regarding age, sex, year of study, family affluence, general well-being, life satisfaction, mental health, and loneliness. MHL was assessed using the KAMHS and the Mental Health Literacy Questionnaire - Short Version for Young Adults (MHLq-SVa). The Reported and Intended Behaviour Scale (RIBS) was used to assess four actual and four intended behavioural outcomes. Structural validity of KAMHS was assessed using confirmatory factor analysis (CFA). Convergent and divergent validity of KAMHS were examined against the MHLq-SVa and the RIBS. Internal consistency of KAMHS subscales was evaluated using McDonald's omega and Cronbach's alpha. Test-retest reliability of KAMHS was assessed. RESULTS: In total, 102 female and 33 male adolescents (mean age, 15.8 years) were included in the analysis. For structural validity, CFA of the refined 25-item, five-factor KAMHS yielded factor loadings ranging from 0.43 to 0.99 and good fit indices. For construct validity, three of the 22 convergent hypotheses were supported. The KAMHS subscale of lack of public stigma was strongly correlated with the RIBS (Spearman's r = 0.65, p < 0.001). The KAMHS subscale of understanding mental disorders and their treatments was moderately correlated with the MHLq-SVa subscale of erroneous beliefs / stereotypes (Spearman's r = 0.39, p < 0.001). The KAMHS subscale of understanding how to optimise and maintain good mental health was moderately correlated with the MHLq-SVa subscale of self-help strategies (Spearman's r = 0.46, p < 0.001). Additionally, all 26 divergent hypotheses were supported. Internal consistency of KAMHS subscales was good (McDonald's omega = 0.76-0.91). Test-retest reliability of KAMHS was high (Pearson's r = 0.75, p < 0.001). CONCLUSION: The traditional Chinese version of the KAMHS shows preliminary psychometric validity for assessing MHL among Hong Kong Chinese adolescents.