Critical and interactive digital health literacy and health information adoption among college students: a cross-sectional serial mediation study from a health communication perspective

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Background Digital media have transformed how college students access, evaluate, and use health information. Although digital health literacy has been linked with risk perception, preventive behavior, self-care, and information seeking, less is known about its association with health information adoption as a downstream stage of digital health communication and the statistical pathways that may account for this association. Objective This study examined the cross-sectional association between critical and interactive digital health literacy and self-reported health information adoption among college students and estimated the specific indirect associations through health risk perception and exercise self-efficacy, separately and sequentially. Methods This cross-sectional study used a structured online questionnaire administered in three batches to college students from 12 institutions in Shandong Province, China, between September and November 2025. An independent pilot sample ( n = 331) informed item-level assessment and scale reduction, and 3,148 formal-survey responses were included in the primary analysis. The primary scores were a 10-item health information adoption score, a 9-item critical and interactive digital health literacy score, an 8-item health risk perception score, and a 10-item exercise self-efficacy score. Analyses included descriptive statistics, group comparisons, Pearson correlations, reliability and measurement-model assessment, and ordinary least-squares path models corresponding to PROCESS Model 6. The adjusted model controlled for gender, academic year, major category, residential background, family economic status, and weekly internet-use frequency. Percentile bootstrap confidence intervals were calculated from 5,000 resamples. Analyses were conducted using SPSS 26.0 and Python; additional models assessed survey-batch and alternative-scoring sensitivity. Results Among the 3,148 participants, the mean scores were 30.83 (SD 8.24) for health information adoption, 29.21 (SD 6.25) for critical and interactive digital health literacy, 26.43 (SD 5.86) for health risk perception, and 32.35 (SD 8.17) for exercise self-efficacy. Digital health literacy was positively correlated with health information adoption ( r = 0.416; P < 0.001). In the covariate-adjusted model, the total association between digital health literacy and health information adoption was B = 0.534. The direct association was B = 0.344 (64.3%). The specific indirect associations through health risk perception (effect = 0.090, bootstrap 95% CI 0.058–0.122; 16.9%), exercise self-efficacy (effect = 0.080, bootstrap 95% CI 0.059–0.101; 14.9%), and the serial health-risk-perception-to-exercise-self-efficacy pathway (effect = 0.020, bootstrap 95% CI 0.012–0.029; 3.8%) were statistically detectable. The total indirect association was 0.191 (bootstrap 95% CI 0.150–0.230; 35.7%). Results were materially similar after adjustment for survey batch and under alternative scoring specifications. Conclusions Critical and interactive digital health literacy was positively associated with self-reported health information adoption among college students. Health risk perception and exercise self-efficacy provided separate and serial statistical indirect routes, but the direct association was the largest component and the serial route was comparatively small. Because the data were cross-sectional and self-reported, these estimates should be interpreted as statistical decompositions of associations rather than evidence of temporal or causal mediation.