Psychometric properties of the Knowledge and Attitudes to Mental Health Scales in a Dutch sample (KAMHS-NL): A comprehensive mental health literacy measure in adolescents

Abstract

Background

Mental health literacy (MHL) is crucial for early recognition of and coping with mental health problems, and for the use and acceptance of mental health services, leading to better health outcomes, especially in adolescence. The prevalence of mental health problems among adolescents is seen as a major public health concern and MHL is an important factor in facilitating positive mental health outcomes. However, the availability of valid measurement instruments for assessing the multifaceted nature of MHL is limited, hindering the ability to make meaningful comparisons across studies. The Knowledge and Attitudes to Mental Health Scales (KAMHS) is a promising comprehensive instrument for measuring adolescents’ mental health literacy but its psychometric properties have not been explored in any other contexts than the Welsh. The aim of this study was to translate the KAMHS into Dutch, adapt it in this context, and evaluate its psychometric properties.

Methods

We performed a cross-sectional study with Dutch adolescents between the ages 11–16. We translated the KAHMS and assessed its content validity using cognitive interviewing with n = 16 adolescents. Next, n = 406 adolescents were asked to fill in the translated KAMHS-NL and reference scales, on mental health (SDQ and WHO-5), resilience (BRS), and mental health help-seeking behaviors. We assessed construct validity based on a priori hypotheses regarding convergent and divergent correlations between subscales of KAMHS-NL and the reference scales. Finally, we assessed structural validity via confirmatory factor analysis and exploratory structural equation modeling.

Results

The KAMHS-NL showed good content validity and satisfactory construct validity. In total, 28 of the 48 hypotheses regarding convergent and divergent correlations between the KAMHS and reference scales were confirmed. Contrary to our expectations, weak, but significant associations were found between MHL and resilience. The KAMHS showed an acceptable to good internal consistency (McDonald’s omega ranging from 0.62 to 0.84). Finally, we could generally confirm the postulated structure of the KAMHS-NL in the Dutch sample with a 5-factor solution (RMSEA = 0.033; CFI = 0.96).

Conclusions

The Dutch version of the KAMHS is a valid measure for detecting differences in MHL levels in adolescents. The KAMHS is a promising instrument for assessing MHL in adolescents in a multifaceted manner in other countries which may facilitate rigorous global MHL research. The instrument therefore deserves further validation research in other settings and comparisons across various cultural contexts.