Health Literacy Online May Cut Both Ways for Stressed Chinese Students

A large survey of Chinese university students has uncovered a surprising split in how digital health skills relate to mental wellbeing, and the difference depends on what students study. Researchers at Shandong University found that among medical students, stronger electronic health literacy was linked to worse negative mental health states, while among their non-medical peers the same relationship was not statistically significant. The study, published in BMC Public Health, also found that social capital, meaning the networks and trust that connect people, was consistently associated with better mental health across both groups.

The research team, led by Zhengna Sun and corresponding author Fanlei Kong of the Department of Social Medicine and Health Management at Shandong University’s Cheeloo College of Medicine, recruited 1,472 university students through the online survey platform Wenjuan.com using a snowball sampling method. Of these, 665 students, or 45.18 percent, were studying medicine, while 807 students, or 54.82 percent, came from non-medical disciplines. This deliberate split allowed the investigators to test whether the pathways connecting online health information skills, social relationships and psychological distress operate differently depending on a student’s academic background.

To measure the key variables, the team used three validated instruments. Negative mental health states were assessed with the Depression Anxiety Stress Scale, known as the DASS-21, a widely used questionnaire that screens for symptoms of depression, anxiety and stress. Electronic health literacy, the ability to seek, find, understand and appraise health information from electronic sources, was measured with the eHealth Literacy Scale, or eHEALS. Social capital was evaluated using the Chinese Shortened Social Capital Scale, abbreviated as CSSCS, which captures the cognitive and structural dimensions of social ties within a community.

The headline numbers reveal a substantial burden of psychological distress across the sample. The total score for negative mental health states averaged 30.87 with a standard deviation of 26.72. When the researchers broke this down by discipline, medical students scored 25.26 on average, with a standard deviation of 24.86, while non-medical students scored markedly higher at 35.49, with a standard deviation of 27.33. In other words, students outside medicine reported consistently higher levels of depression, anxiety and stress symptoms than their medically trained counterparts, a finding that challenges assumptions about which student populations carry the heaviest psychological load.

The most striking result, however, concerned electronic health literacy. Among medical students, higher eHealth literacy was significantly and positively associated with worse negative mental health states. Among non-medical students, this association was not statistically significant. The authors of the study suggest that this divergence may reflect the particular position of medical students, who are trained to understand health information in depth and may therefore be more acutely aware of health risks, including their own psychological symptoms, when they encounter health content online. Greater digital health literacy may sharpen awareness of problems rather than buffer against them, at least in this population.

By contrast, social capital behaved as many public health researchers would expect. Across both medical and non-medical students, higher social capital was negatively associated with negative mental health states, meaning that students with stronger social networks and greater interpersonal trust reported fewer symptoms of depression, anxiety and stress. This protective pattern held regardless of academic discipline, underscoring the role of social connection as a consistent buffer against psychological distress in university settings.

A third relationship tied the two predictors together. Electronic health literacy showed a significant positive association with social capital among both medical and non-medical students. Students who were better at navigating digital health resources also tended to report richer social capital. This suggests that the skills involved in finding and evaluating online health information may travel alongside, or even support, the social connections that protect mental health, even as those same skills correlate with heightened distress among medical students.

Methodologically, the team combined classical statistical analysis with structural equation modeling. Descriptive statistics, chi-square tests and t-tests were performed in SPSS to characterize the sample and compare groups, while AMOS software was used to build and test structural equation models linking the three constructs. The study reports standard goodness-of-fit indicators for these models, including the chi-square statistic, the root mean square error of approximation, the goodness-of-fit index, the adjusted goodness-of-fit index, the comparative fit index, the normed fit index and the incremental fit index. This dual approach allowed the researchers to test whether a single model could describe the whole sample or whether separate pathways were needed for medical and non-medical students, which is precisely the major difference highlighted in the paper’s title.

The study was approved by the Ethics Committee of Shandong University under task number LL2022025’s companion approval LL20220425, in compliance with the 1964 Helsinki Declaration and its later amendments. Every participant provided written informed consent in electronic form by clicking an agreement button on the online survey platform before proceeding to the questionnaire, and consent was required before anyone could take part. The work was funded by the National Natural Science Foundation of China under grants 72574130 and 71804094, the Shandong Provincial Natural Science Foundation under grant ZR2025MS1149, and the Fundamental Research Funds for the Central Universities under grants 2022KJGL01 and 2018JC055.

For universities designing mental health support programs, the findings carry a practical message. Interventions that simply boost digital health literacy may not help all students equally, and could even coincide with greater distress among those trained to interpret health information critically. Meanwhile, efforts to strengthen social capital, such as building peer networks and community trust, appear to offer benefits that cut across disciplines. The authors note that the data were collected through snowball sampling, a method that can limit how representative the sample is of all Chinese university students, and that the cross-sectional design captures associations rather than proving cause and effect. Even so, the study, published open access on 13 October 2026 with the DOI 10.1186/s12889-026-29817-5, adds a nuanced piece to the growing evidence that the digital and social environments shaping student mental health do not operate the same way for everyone.

Subject of Research: The relationship between electronic health literacy, social capital and negative mental health states among Chinese university students

Article Title: Major differences in the relationship between electronic health literacy, social capital and negative mental health states among the university students in China

Article References: Major differences in the relationship between electronic health literacy, social capital and negative mental health states among the university students in China. (n.d.). https://doi.org/10.1186/s12889-026-29817-5

Image Credits: AI Generated

DOI: 10.1186/s12889-026-29817-5

Keywords: electronic health literacy, social capital, mental health, university students, medical students, DASS-21, eHEALS, structural equation modeling, China, public health, depression, anxiety