Development and Assessment of a Brief Tool to Measure Melanoma-Related Health Literacy and Attitude Among Adolescents.

The aim of this study is to develop a tool to measure health literacy and attitude towards melanoma and to assess the tool in a group of adolescents through a multicenter cross-sectional survey. The concept, dimensionality, and item pool of the tool were developed by a focus group. The Delphi method was applied to determine the content validity. Newly enrolled students in five universities were invited to an online questionnaire survey. Items were selected according to correlation, factor loading, and item response parameters. Psychometric properties (reliability, construct validity, and measurement invariance) were assessed using McDonald's ω and confirmatory factor analysis (CFA), respectively. A total of 21,086 valid questionnaires were collected. The focus group drafted two subscales and 13 items. Content validity was good for all items (Kappa > 0.7). One item was removed from the tool owing to low factor loading and discrimination parameter. McDonald's ω of the subscales were 0.84 (health literacy) and 0.86 (attitude). Local dependencies were identified in CFA; after modification, the goodness-of-fit was satisfactory (comparative fit index, CFI > 0.98). The tool showed measurement invariance across subgroups of gender, ethnicity, and university (CFI change < 0.01 across models). The brief tool to measure health literacy and attitude towards nevus and melanoma shows good psychometric properties and measurement invariance. It can be used in further investigation.