Exploring adolescents' ability to appraise health information: validation of the self-report short-form Critical...

Adolescents' ability to evaluate health information is an increasingly important area in public health and health promotion; however, there is a scarcity of brief, psychometrically sound self-report tools focused on adolescents' information appraisal abilities. In this study, we aim to validate a unidimensional short form of the Critical Health Literacy for Adolescents Questionnaire (CHLA-QA) domain A and to examine its convergent and known-groups validity.

We conducted a cross-sectional survey among Norwegian students (N = 522; response rate 33%) aged 13-15 years old. We employed stepwise confirmatory factor analysis (CFA) to compare three models. Known-groups validity was assessed by examining the distribution of CHLA-QA across variables: sex, grade level, parents' education, books at home, and generic health literacy (HLSAC), with analyses based on the Kruskal-Wallis test, the Wilcoxon rank-sum test, and effect sizes (ε², r). We assessed convergent validity through Spearman correlations with HLSAC, self-rated health, HRQoL, and academic achievement.

Analysis yielded a five-item solution with acceptable fit (CFI = 0.98; RMSEA = 0.097 [0.065-0.132]; SRMR = 0.037, uSRMR/2 = 0.055), strong standardized loadings (0.582-0.839), and satisfactory reliability (α = 0.79). Known-groups analyses showed higher scores among girls than boys (p < 0.001; r = 0.169), a small grade effect (ε² = 0.013; grade 9 > grades 8 and 10), and a socioeconomic gradient was found, with higher scores associated with parents' education (r = 0.111) and number of books at home (ε² = 0.024). HLSAC increased stepwise across low, moderate, and high groups based on the CHLA-QA scale (ε² = 0.132). Convergent validity was supported by moderate correlations with HLSAC (ρ = 0.461), self-rated health (ρ = 0.316), and HRQoL (ρ = 0.346). Correlations with academic achievement were weak but significant for five of six school subjects. Higher scores were correlated with asking parents for help regarding health information (ρ = 0.233).

This study advances the measurement of adolescents' CHL by developing and evaluating a brief, unidimensional instrument for assessment of information appraisal (CHLA-QA). The scale showed strong validity and acceptable reliability. Future studies in independent samples are warranted to further establish validity and measurement invariance across relevant populations.