Family socioeconomic resources, family differentiation, and short-term change in health literacy among Chinese university students: a three-wave longitudinal study

Abstract

Family resources and relationships may shape the opportunities through which students acquire and use health information. Longitudinal evidence is needed to distinguish their associations with initial health literacy from those with subsequent change. This three-wave study followed students at one Chinese public university over 160 days, from October 2025 to March 2026. Of 912 invited students, 877 provided usable baseline data and 801 formed the complete-case primary sample; all 877 were included in a full information maximum likelihood sensitivity analysis. Health literacy was assessed with a 50-item comprehensive score (0 to 66), family socioeconomic resources with parental education and a 14-item household-resource count, and perceived relational climate with the Differentiation in the Family System Scale. Parallel-process latent growth models distinguished initial levels from change and provided an exploratory statistical path decomposition. Mean health literacy increased from 44.32 to 44.99, a small change of 0.67 points (dz = 0.076). Family socioeconomic resources were associated with initial health literacy (b = 3.637, 95% CI 1.956 to 5.318) and short-term change (b = 2.349, 95% CI 0.797 to 3.901). Initial family differentiation was also associated with initial health literacy (b = 7.098, 95% CI 5.211 to 8.984) and change (b = 2.993, 95% CI 1.023 to 4.963). The indirect association with change through initial family differentiation was 1.310 (95% CI 0.467 to 2.152). Focal associations were similar across sensitivity analyses. With alternative scoring of unanswered items, the model-estimated mean change was 0.54 points (95% CI -0.07 to 1.15). In this university sample, family resources and perceived relational climate were associated with both initial health literacy and short-term change. The findings support considering family relationships alongside educational and material resources when examining differences in students’ health literacy in comparable university settings. Not applicable.