U-shaped association between social media usage frequency and suggestibility by internet health information in Chinese online population with pre-diabetes and diabetes: a cross-sectional study

• Published in: BMC public health Vol. 25; no. 1; pp. 1525 - 11
• Main Authors: Chen, Mutong, Lin, Xiaobing, Zhou, Rui, Fan, Guanhua
• Format: Journal Article
• Language: English
• Published: England BioMed Central Ltd 24.04.2025; BioMed Central; BMC
• Subjects: Adult; Aged; Analysis; Blood; Blood glucose; Blood sugar; Body measurements; China; Chronic illnesses; Consumer health information; Consumer Health Information - statistics & numerical data; Cross-Sectional Studies; Curve fitting; Data analysis; Diabetes; Diabetes Mellitus; Diabetes Mellitus - therapy; Diabetes therapy; Digital media; East Asian People; Evaluation; Female; Glucose; Health Behavior; Health surveys; Humans; Information dissemination; Information management; Internet; Latency; Likert scale; Male; Middle Aged; Online health care information services; Prediabetic state; Prediabetic State - therapy; Psychological aspects; Psychological factors; Questionnaires; Regression analysis; Secondary analysis; Self-care, Health; Self-Management; Social Media; Social Media - statistics & numerical data; Social networks; Subgroups; Surveys and Questionnaires; China; Self-Management; Internet; Diabetes Mellitus; Social Media; Health Behavior
• ISSN: 1471-2458; 1471-2458
• DOI: 10.1186/s12889-025-22724-1

Internet-based self-management of diabetes has been demonstrated to be effective. The frequency of social media is believed to be associated with diabetes management, yet the quality and applicability of online information are still subject to debate. The dynamic nature of online information complicates its study, making it crucial to further assess the online behaviors and psychological aspects of the population engaged in online blood glucose management. The objective of this study was to explore the relationship between the frequency of social media usage and the suggestibility of internet health information. This study is a secondary analysis based on data obtained from a prior cross-sectional survey conducted in multiple online diabetes communities in China, which received a total of 5,504 responses, ultimately including 1,062 individuals with diabetes or prediabetes for analysis. Frequency of social media usage was measured using a 6-point Likert scale across five items, evaluating the frequency of use on platforms such as WeChat, Weibo, QQ, TikTok, and others. Suggestibility by internet health Information was assessed using a 5-point Likert scale across nine items reflecting individuals' trust in, discussion of, and engagement with internet health information and communities. Data analysis was conducted using R language and EmpowerStats software, encompassing Chi-square tests, U tests, multifactor linear regression analysis, smooth curve fitting, and subgroup and interaction effect analysis. After adjusting for other factors, there was a positive correlation between the frequency of social media and the suggestibility of internet health information cues (β = 0.27, 95% CI: 0.24 to 0.30, P < .001). A nonlinear relationship was identified, with a turning point at 3.8. When the social media usage score is below 3.8, each unit increase in the social media score corresponds to a decrease of 0.12 in the internet health trust score (95% CI: -0.19 to -0.05, P < .001). Conversely, when the application usage score is 3.8 or above, each unit increase leads to an increase of 0.46 in the internet health suggestibility score (95% CI: 0.42 to 0.50, P < .001). Interaction analysis revealed significant interactive factors affecting the relationship between social media usage frequency and suggestibility by internet health information, including gender (P = .045), age (P < .001), body measurement index (P = .03), sleep latency (P = .001), self-monitoring of blood glucose frequency (P = .001), and glycemic control status (P < .001). Once a certain threshold of social media usage frequency is reached, suggestibility by internet health information cues increases with increased usage. Monitoring social media usage frequency could thus provide insightful indicators for evaluating the online behavioral and psychological profiles of individuals engaged in online blood glucose management.