Abstract
Weight management has become a major focus in worldwide health, and platforms like TikTok and Bilibili are now popular for health information. However, the quality and reliability of weight management content on these platforms remain uncertain. This research systematically evaluated such videos to provide evidence-based guidance for public health communication. We analyzed the top 100 weight management videos from TikTok and Bilibili, recording their sources, content, and characteristics. The DISCERN instrument, JAMA benchmark criteria, and Global Quality Score (GQS) were used to evaluate video quality and reliability. Further analysis was conducted to explore the relationship between video quality and video characteristics. While TikTok videos attracted more likes, saves, comments, and shares, Bilibili videos were longer and exhibited higher quality and reliability (all P < 0.001). Videos from doctors and non-profit organizations had the highest DISCERN and GQS scores, while those from fitness bloggers and individual users were more popular but of lower quality. Video duration was positively associated with quality, whereas engagement indicators (likes, comments, shares, saves) were negatively associated with both GQS and DISCERN. Overall, the quality of weight management videos on TikTok and Bilibili was poor, although Bilibili performed better than TikTok. Doctors and non-profit organizations produced higher-quality content, highlighting the need for stronger platform review and greater professional contributions to improve the dissemination of reliable health information.
Supplementary Information
The online version contains supplementary material available at 10.1038/s41598-026-38404-y.
Keywords: Weight management, TikTok, Bilibili, Social media, Video quality, Video reliability
Subject terms: Health care, Medical research
Introduction
Weight management has become a major focus in worldwide health. According to the latest data from the World Health Organization (WHO), in 2022, there were 2.5 billion overweight adults worldwide, of whom more than 890 million were obese. Over the past three decades, the prevalence of obesity among adults has nearly doubled1. Being overweight and obesity are highly correlated with a variety of comorbidities, such as type 2 diabetes, cardiovascular diseases, sleep apnea, osteoarthritis, and different cancers2, and also impose a substantial socioeconomic burden. The Organization for Economic Co-operation and Development and the World Obesity Federation have projected that obesity-related healthcare costs and productivity losses will reach several trillion U.S. dollars by 2030 and may exceed 18 trillion U.S. dollars by 20603. The development of obesity is multifactorial, driven by genetic, lifestyle, disease- and medication-related, environmental, and social factors4,5. Accordingly, scientific, effective, and sustainable weight management strategies—including dietary modification, physical activity, pharmacological therapy, and surgical interventions—are considered central to the prevention and management of overweight and obesity6,7.
Recently, short video–based social media platforms like TikTok and Bilibili have seen rapid growth. Their entertaining and visually appealing content has made them significant sources for the public to obtain health information8–10. Studies indicate that the spread of health information via short video platforms has been increasing over the past decade11. However, due to the absence of rigorous review mechanisms and peer-review processes, the quality and reliability of video content on these platforms vary greatly, and some videos may even contain misleading or deceptive information2,12. Exposure to inaccurate health information in short videos may misguide the public in their decision-making, potentially leading to adverse health outcomes13–16. Therefore, evaluating the quality and reliability of weight management short videos on TikTok and Bilibili is of considerable importance.
As public interest in weight management rises, exploring the role of emerging media platforms in health education becomes essential. However, there is currently a shortage of systematic studies on the quality and reliability of weight management short videos. This study sought to evaluate the quality and reliability of weight management information on TikTok and Bilibili to fill this gap, thereby providing evidence-based insights to help the public obtain more accurate and reliable information.
Materials and methods
Data collection
TikTok and Bilibili were selected because they are widely used video-based social media platforms in China and represent distinct content ecosystems17. TikTok is characterized by algorithm-recommended short-form videos with broad reach, whereas Bilibili contains a higher proportion of longer-form, information-oriented content, enabling a comparative assessment of weight-management information quality across different media formats and dissemination contexts. Using the keyword “weight management”, we searched the Chinese versions of TikTok and Bilibili on July 2, 2025, as a cross-sectional snapshot and retrieved the top 100 videos returned by each platform (Fig. 1). A pilot search using a closely related term (“losing weight”) yielded largely overlapping results; therefore, we used “weight management” as the single keyword to ensure terminological consistency and to capture a broader range of professional and lifestyle content. Prior to the search, in order to reduce bias from personalized recommendations, all accounts were signed out, and search histories were removed. Search results were shown in the default order without applying any filters. We excluded advertisements, duplicate videos (i.e., identical content uploaded by different users), and videos irrelevant to weight management (e.g., content focusing on other diseases such as breast cancer). Eligible videos were screened until 100 videos per platform were included. The decision to focus on the top 100 videos was based on prior studies suggesting that including videos beyond the top 100 does not materially change analytical conclusions15,18,19. For each video, we recorded basic information including title, upload date, video duration, video content, and the number of likes, saves, comments, and shares, as well as the number of days since uploading. For uploaders, we collected information on username, verification status, and number of followers. Two researchers independently extracted the data, and discrepancies were resolved by discussion before finalizing the dataset. All data were recorded in Microsoft Excel.
Fig. 1.
Flow diagram for review of TikTok and Bilibili videos on weight management.
Video classification
Videos were classified by source into four groups: (1) fitness blogger; (2) doctor; (3) individual user (non-medical personnel); and (4) non-profit organization (including hospitals, health departments, research institutions, and universities). Video content was classified into six categories: diet, exercise, sleep, medicine, surgery, and Traditional Chinese Medicine. Traditional Chinese Medicine was retained as a prespecified category to ensure a complete and culturally relevant characterization of weight-management information in China, although it was infrequently observed in this sample. This classification approach allowed grouping of videos with similar content while differentiating distinct thematic areas.
Quality assessment
The quality and reliability of videos were assessed using the Global Quality Score (GQS), the DISCERN instrument, and the Journal of the American Medical Association (JAMA) benchmark criteria. In this study, we applied the first section of the DISCERN instrument to evaluate video reliability, with higher scores indicating better reliability20–23. According to the JAMA benchmark criteria, video sources are rated for reliability on a scale of 0 to 4, based on four different standards24. A five-point Likert scale, known as GQS, is used to measure the overall quality of videos, with higher scores reflecting better quality25,26. Detailed criteria for scoring GQS and DISCERN can be found in Supplementary Tables S1 and S2. The JAMA benchmark score was calculated for each video but was not included in subsequent analyses. This decision was made because the JAMA criteria were originally developed as a brief credibility checklist for text-/website-based health information and may have limited discriminative ability in short videos; therefore, DISCERN and GQS were selected as the primary outcomes for statistical comparisons.
All authors are physicians specializing in weight management. During the screening and scoring process, two reviewers (LH and FL) independently assessed each video using the DISCERN, JAMA benchmark criteria, and GQS. Before scoring, both reviewers reviewed the scoring guidelines and discussed details to minimize cognitive bias. In cases of disagreement, a third reviewer (XL) served as an adjudicator to determine the final score. All authors subsequently reached consensus on the final ratings.
Statistical analysis
Normality of continuous variables was assessed using the Shapiro–Wilk test and visual inspection (histograms). Continuous variables that were not normally distributed were described using medians and interquartile ranges (IQRs), while categorical variables were shown as frequencies and percentages. The Mann–Whitney U test was used for comparing two groups of continuous variables that were not normally distributed, while the Kruskal–Wallis H test was applied for comparisons among more than two groups, followed by pairwise comparisons with the Mann–Whitney U test with Bonferroni adjustment. Categorical variables were compared using the chi-square test, continuity correction, or Fisher’s exact test. We used Cohen’s κ to quantify the agreement between the two reviewers. Spearman correlation analysis was employed to evaluate the associations between variables that do not follow a normal distribution. Given the large number of pairwise correlations, P values were adjusted for multiple comparisons using the Benjamini–Hochberg false discovery rate (FDR) procedure. According to Spearman’s rank correlation coefficient, r > 0 represents a positive correlation and r < 0 represents a negative correlation. The correlation’s strength was classified as: |r|≤ 0.2, no correlation; 0.2 <|r|≤ 0.4, weak correlation; 0.4 <|r|≤ 0.6, moderate correlation; 0.6 <|r|≤ 0.8, strong correlation; and |r|> 0.8, very strong correlation. Statistical analyses were performed using R software (version 4.4.1).
Results
Video characteristics
200 videos were analyzed, with 100 each from TikTok and Bilibili. The overall characteristics of these videos are presented in Table 1. TikTok videos showed a significantly higher number of likes, saves, comments, and shares than those from Bilibili (all P < 0.001), whereas Bilibili videos had longer durations (P < 0.001). No significant differences were found between the two platforms in terms of days since uploading (P = 0.11) or number of followers of uploaders (P = 0.05). The distribution of video sources varied greatly across platforms: on TikTok, 86% of uploaders were individual users, while on Bilibili, more than half were individual users, followed by fitness bloggers (Fig. 2). Videos uploaded by physicians accounted for only a small proportion on both platforms. Regarding content categories, Bilibili covered a wider range of topics. TikTok videos primarily focused on diet and exercise, whereas Bilibili emphasized exercise, largely because many Bilibili uploaders shared their personal workouts and training sessions. However, videos related to medicine and surgery therapy were scarce on both platforms. Notably, no TikTok videos mentioned traditional Chinese medicine for weight management, whereas three such videos were identified on Bilibili.
Table 1.
Baseline of weight management-related short videos.
| Variables | TikTok | Bilibili | Statistic | P value |
|---|---|---|---|---|
| Likes, median (IQR)a | 73,044 (26,050, 320,423) | 3514 (167, 43,000) | 7972 | 3.86e-13 |
| Saves, median (IQR)a | 35,556 (8703, 147,380) | 3425 (318, 96,750) | 6845 | 6.58e-06 |
| Comments, median (IQR)a | 6008 (1721, 19,097) | 228 (28, 2103) | 8112 | 2.90e-14 |
| Shares, median (IQR)a | 23,300 (5606, 100,877) | 681 (59, 11,000) | 8067 | 6.74e-14 |
| Duration (s), median (IQR)a | 173 (77, 263) | 510 (270, 828) | 2109 | 1.62e-12 |
| Days since uploading, median (IQR)a | 301 (164, 507) | 603 (28.5, 1193) | 4340 | 0.11 |
| Followers, median (IQR)a | 216,000 (42,750, 1,148,250) | 110,624 (10,750, 672,750) | 5797 | 0.05 |
| Video source, n (%)b | 1.98e-06 | |||
| Fitness Blogger | 5 (5%) | 29 (29%) | ||
| Doctor | 7 (7%) | 7 (7%) | ||
| Individual user | 86 (86%) | 56 (56%) | ||
| Non-profit organization | 2 (2%) | 8 (8%) | ||
| Contents of videos, n (%)b | ||||
| Diet | 68 (68%) | 51 (51%) | 0.02 | |
| Exercise | 64 (64%) | 80 (80%) | 0.02 | |
| Sleep | 11 (11%) | 22 (22%) | 0.06 | |
| Medicine | 2 (2%) | 3 (3%) | 1.00 | |
| Surgery | 1 (1%) | 4 (4%) | 0.37 | |
| TCM* | 0 (0%) | 3 (3%) | 0.25 | |
| GQS score, median (IQR)a | 2.5 (2, 3) | 3 (2, 4) | 3653 | 4.97e-04 |
| DISCERN score, median (IQR)a | 4 (3, 4) | 4 (4, 5) | 3569 | 2.07e-04 |
| JAMA, median (IQR)a | 1 (1, 1) | 1 (1, 1) | 5044 | 0.84 |
Bold text means the P-value < 0.05.
aMann–Whitney U test; bFisher’s exact test.
TCM Traditional Chinese Medicine.
Fig. 2.
Video source of the included videos. (a) TikTok; (b) Bilibili.
Video quality assessment
As shown in Table 1, the median GQS score of TikTok videos was 2.5 (IQR 2–3), and the median DISCERN score was 4 (IQR 3–4), indicating generally poor quality and moderate reliability. In contrast, Bilibili videos had a median GQS score of 3 (IQR 2–4) and a median DISCERN score of 4 (IQR 4–5), demonstrating moderate quality and reliability. Significant differences were observed between the two platforms in both GQS and DISCERN, and Bilibili videos scored significantly higher than TikTok videos in overall quality (Fig. 3A, P < 0.001) and in reliability (Fig. 3B, P < 0.001). These findings suggest that, compared with TikTok, Bilibili provides weight management videos of higher quality and reliability. The Cohen’s κ values of the GQS and DISCERN score were 0.943 and 0.957, indicating good consistency. The JAMA benchmark criteria, however, showed limited discriminatory power in this study. The majority of videos (179/200, 89.5%) scored 1 point, while 18 videos (9.0%) scored 2 points, and only 3 videos (1.5%) scored 3 points. Given its poor differentiation, JAMA scores were excluded from subsequent analyses.
Fig. 3.
GQS scores and DISCERN scores for videos of different platforms. (a) GQS scores; (b) DISCERN scores; “***” means P < 0.001.
Source of videos
As shown in Table 2, videos uploaded by fitness bloggers received significantly more likes, saves, shares, longer durations, more days since uploading, and larger follower counts compared with those uploaded by doctors, individual users, and non-profit organizations (all P < 0.05). This indicates that videos from fitness bloggers were generally more popular among viewers. However, the quality and reliability of these videos were relatively poor, with median GQS and DISCERN scores of 2 (IQR 2–3) and 4 (IQR 4–4), respectively. Videos uploaded by doctors achieved the highest scores, with a median GQS of 4 (IQR 3–5) and a median DISCERN of 5 (IQR 5–5.8). Non-profit organizations ranked second, with a mean GQS of 3.8 ± 0.9 and a median DISCERN of 5 (IQR 5–6). Differences in video quality and reliability were significant across various sources (P < 0.001). Overall, videos uploaded by doctors and non-profit organizations were of higher quality, whereas those from fitness bloggers and individual users were of lower quality (Fig. 4).
Table 2.
Comparison of different video sources.
| Variables | Individual user (N = 142) | Doctor (N = 14) | Fitness blogger (N = 34) | Non-profit organization (N = 10) | P value |
|---|---|---|---|---|---|
| Likes | 29,300 (3531, 183,732) | 1719 (62 27,523) | 68,218 (10,040, 200,750) | 2796 (205, 52,402) | 0.0033 |
| Saves | 14,372 (2322, 103,955) | 422 (50, 15,326) | 107,500 (6612, 623,405) | 2702(528, 26,649 | < 0.001 |
| Comments | 19,340 (216, 8097) | 229 (15, 2084) | 3543 (521, 11,685) | 200 (39, 1183) | 0.0065 |
| Shares | 6784 (714, 53,932) | 621 (12, 16,986) | 17,232 (1208, 86,500) | 565 (87, 13,103) | 0.0226 |
| Duration (s) | 231 (121, 513) | 105 (61, 189) | 618 ± 384 | 609 ± 383 | 0.0001 |
| Days since uploading | 347 (139, 735) | 138 (41, 419) | 997 ± 635 | 79 (28, 201) | < 0.001 |
| Followers | 82,000 (16,000, 457,000) | 65,500 (3895, 1,224,500) | 938,500 (327,000, 4,232,000) | 257,500 (16,750, 1,189,000) | < 0.001 |
| GQS score | 3.0 (2.0, 3.0) | 4.0 (3.0, 5.0) | 2.0 (2.0, 3.0) | 3.8 ± 0.9 | < 0.001 |
| DISCERN score | 4.0 (3.0, 4.0) | 5.0 (5.0, 5.8) | 4.0 (4.0, 4.0) | 5.0 (5.0, 6.0) | < 0.001 |
All the P-values were obtained from the Kruskal–Wallis H-test.
Fig. 4.
GQS scores and DISCERN scores for videos of different sources. (a) GQS scores; (b) DISCERN scores; “**” means P < 0.01; “***” means P < 0.001; “****” means P < 0.0001.
Correlation analysis
As presented in Table 3, Spearman correlation analysis showed a strong positive correlation between video quality (GQS) and information reliability (DISCERN) (r = 0.79, P < 0.001), indicating high consistency between the two scoring tools. Among video engagement indicators, likes, shares, comments, and saves were all highly intercorrelated (r > 0.9, P < 0.001). However, each of these metrics was negatively correlated with GQS (likes: r = –0.28; shares: r = –0.30; comments: r = –0.35; saves: r = –0.26; all P < 0.001). DISCERN scores showed the same negative trend but with weaker correlations (r = –0.12 to –0.25). These findings suggest that videos with higher engagement tended to have lower quality and reliability.
Table 3.
Spearman correlation analysis among different video variables, GQS and DISCERN score.
| r, q-value | GQS | DISCERN | Likes | Shares | Comments | Saves | Duration (s) | Days since uploading | Followers |
|---|---|---|---|---|---|---|---|---|---|
| GQS | 1 | ||||||||
| DISCERN | 0.79, < 0.001 | 1 | |||||||
| Likes | − 0.28, < 0.001 | − 0.20, < 0.05 | 1 | ||||||
| Shares | − 0.30, < 0.001 | − 0.18, < 0.05 | 0.97, < 0.001 | 1 | |||||
| Comments | − 0.35, < 0.001 | − 0.25, < 0.001 | 0.94, < 0.001 | 0.94, < 0.001 | 1 | ||||
| Saves | − 0.26, < 0.001 | − 0.12, 0.10 | 0.93, < 0.001 | 0.92, < 0.001 | 0.85, < 0.001 | 1 | |||
| Duration (s) | 0.21, < 0.05 | 0.231, < 0.05 | − 0.10, 0.17 | − 0.12, 0.11 | − 0.16, < 0.05 | 0.06, 0.43 | 1 | ||
| Days since uploading | − 0.17, < 0.05 | − 0.09, 0.25 | 0.25, < 0.001 | 0.22, < 0.05 | 0.22, < 0.05 | 0.32, < 0.001 | 0.24, < 0.05 | 1 | |
| Followers | − 0.09, 0.21 | 0.01, 0.85 | 0.61, < 0.001 | 0.58, < 0.001 | 0.51, < 0.001 | 0.61, < 0.001 | 0.15, < 0.05 | 0.40, < 0.001 | 1 |
Bold text indicates FDR-adjusted P values (Benjamini–Hochberg q values) < 0.05; |r|≤ 0.2 no relationship; 0.2 <|r|≤ 0.4 weak relationship; 0.4 <|r|≤ 0.6 moderate relationship; 0.6 <|r|≤ 0.8 strong relationship; |r|> 0.8 very strong relationship.
Video duration was weakly but positively correlated with both GQS (r = 0.21, P < 0.01) and DISCERN (r = 0.23, P < 0.01), indicating that longer videos were generally of higher quality and reliability. The number of days since uploading was weakly and negatively correlated with GQS (r = –0.17, P < 0.05), but not significantly associated with DISCERN (r = –0.09, P = 0.23). The number of followers was not significantly correlated with GQS (r = –0.09, P = 0.19) or DISCERN (r = 0.01, P = 0.85), but showed moderate-to-strong positive correlations with likes, shares, comments, and saves (r = 0.51–0.61, all P < 0.001). These relationships are illustrated in the correlation heatmap (Fig. 5), which visually confirms the inverse association between video engagement and video quality.
Fig. 5.
Spearman correlation analysis among different video characters, GQS, and DISCERN score concerning weight management.
Discussion
Major findings
The research analyzed weight management videos on TikTok and Bilibili at one specific time, evaluating their quality with the DISCERN and the GQS. We found that Bilibili videos had significantly lower engagement metrics than TikTok but achieved higher quality and reliability scores. On both platforms, most weight management videos were uploaded by individual users, whose content generally demonstrated poor quality and reliability. In contrast, videos uploaded by doctors and non-profit organizations were fewer in number but of markedly better quality and reliability. These findings align with previous studies reporting that doctors and non-profit organizations tend to offer superior health information27. Furthermore, engagement indicators such as likes, shares, comments, and saves were negatively related to quality and reliability scores, suggesting that popularity does not necessarily reflect the scientific accuracy or reliability of content.
Factors influencing video dissemination and impact
Likes, shares, and comments are commonly used indicators of video appeal and dissemination effectiveness, and thus are widely regarded as measures of video influence28. Our study found that video influence, as reflected by engagement indicators (likes, shares, comments, and saves), was associated with both platform and source. TikTok videos showed significantly higher overall engagement than those on Bilibili, suggesting that platform characteristics may play an important role in shaping dissemination. In terms of video source, weight management videos uploaded by fitness bloggers and individual users consistently outperformed those from doctors and non-profit organizations in engagement metrics, indicating greater audience appeal. This advantage is likely attributable to their strengths in content packaging, visual presentation, and audience interaction. From a health communication perspective, however, the influence of doctors and non-profit organizations remains limited. Therefore, while maintaining the professional accuracy of their content, these sources should also focus on innovation in presentation style and audience friendliness to avoid reduced dissemination due to content being perceived as monotonous or less engaging.
Correlation between video quality and video characteristics
As shown in Table 3, video duration was positively associated with video quality, with longer videos generally scoring higher. This finding aligns with previous research15. However, typical video duration differs between TikTok and Bilibili; therefore, platform-specific duration distributions may partially influence the observed association between video duration and quality. We also observed negative correlations between engagement metrics (likes, shares, comments, and saves) and both GQS and DISCERN, indicating that videos with greater engagement tended to be of lower quality and reliability. Similar trends have been reported in studies of YouTube content15. A possible explanation is that high-quality videos, while more professional and comprehensive, often lack entertainment value, may appear monotonous, and are less visually appealing, thereby attracting limited audience engagement29,30. In contrast, low-quality videos are often presented in a more vivid and eye-catching manner, making them more likely to gain popularity. This trend is particularly evident in the context of weight management, a widely popular topic, where many individual users share personal experiences or even highlight unconventional approaches to capture attention31–33. However, such videos often lack scientific validity and may be subject to survivor bias, potentially misleading audiences. Furthermore, recommendation algorithms on TikTok and Bilibili prioritize highly engaged content, further amplifying the dissemination of low-quality popular videos.
Given the complexity and professional nature of health information, introducing expert review mechanisms could help improve the overall quality and reliability of health-related videos. By optimizing recommendation algorithms, platforms can prioritize professionally verified, high-quality content in search results, thus supporting the spread of accurate health information. In parallel, government agencies and medical institutions should actively contribute more evidence-based weight management content through social media to enhance public health literacy.
Evaluation of quantitative scoring tools
We found that the JAMA benchmark criteria were insufficient to accurately assess the informational quality of videos, as the tool contains only four items and lacks granularity. This limitation is consistent with previous reports18. Accordingly, JAMA scores were excluded from subsequent analyses. In contrast, the consistency between DISCERN and GQS was acceptable, supporting their use as complementary tools for evaluating video quality and reliability.
Strengths and limitations
Several strengths should be highlighted. First, we examined two of the most representative Chinese short video platforms: TikTok, which reaches users across all age groups and educational backgrounds, and Bilibili, which primarily serves younger audiences. The simultaneous evaluation of both platforms makes our findings more reflective of the real-world context and avoids the limitations of single-platform studies. Second, this is the initial study to use multiple evaluation tools (DISCERN, GQS, and JAMA) to determine the quality and reliability of weight management videos on TikTok and Bilibili. Third, beyond comparing video quality across platforms and sources, we further analyzed correlations between video characteristics and quality, finding that video duration was positively correlated with quality and reliability, whereas likes, comments, shares, and saves were negatively correlated. These findings provide new insights into the association between social media video dissemination and informational quality. Overall, this is the initial systematic analysis of the quality and reliability of weight management videos on TikTok and Bilibili.
Nevertheless, several limitations should be acknowledged. First, the evaluation instruments used were originally developed for text-based materials, and thus may not fully capture aspects unique to videos, such as production quality and audiovisual design. Second, comment counts may not exclusively represent positive feedback, as some may reflect dissatisfaction. However, in this study, the correlation between comments and likes and shares was positive, implying it could still be a proxy for influence. Future research might separate positive from negative comments to investigate their link to video quality. Third, most content on TikTok and Bilibili remains entertainment-oriented, with health-related videos representing only a small proportion, warranting ongoing observation of their evolution. Fourth, the video search was conducted on a single day as a cross-sectional snapshot. Given the dynamic nature of platform algorithms and ranking mechanisms, the retrieved results may vary over time; future studies could consider repeated or multi-time-point sampling to assess temporal stability. In addition, this study included only Chinese-language videos, and the findings may not generalize to other languages, such as English. Future research should extend to multilingual contexts to provide a more comprehensive understanding of video quality across cultures. Finally, the relatively small sample size of certain uploader categories (e.g., physicians) may have limited statistical power. This limitation could be addressed in future studies by expanding sample sizes or broadening search strategies.
Conclusions
This study found that the overall quality of weight management videos on TikTok and Bilibili was low, although Bilibili videos demonstrated higher quality and reliability than those on TikTok. Videos uploaded by doctors and non-profit organizations were of higher quality and reliability, whereas those from fitness bloggers and individual users achieved greater reach but were of lower scientific value. Strengthening platform-level content review and increasing the supply of professional, evidence-based content are needed. Future studies should use larger multi-platform, longitudinal samples to track changes over time and evaluate platform/policy interventions that enhance transparency and accountability.
Supplementary Information
Acknowledgements
The authors thank the Department of Clinical Nutrition, Weifang People’s Hospital, for technical support. We also acknowledge the assistance of medical writers, proofreaders, and editors in improving the clarity and language of the manuscript.
Author contributions
All authors contributed to the study conception and design. Han L: Conceptualization, Methodology, Software, Validation, Investigation, Data Curation, Writing—Original Draft, Visualization. Liu F: Software, Validation, Investigation. Wang Z and Liu X: Software, Validation, Investigation. Tang N and Wang Y: Software. Liu X: Methodology, Visualization. Liu H: Conceptualization, Writing—Review & Editing, Visualization, Supervision, Project administration. All authors read and approved the final manuscript.
Funding
This project was supported by the research project on the high-quality development of clinical nutrition work of the National Institute of Hospital Administration, NHC (2025–1-Y-05).
Data availability
The investigation did not involve clinical data, human subjects, or animal testing. All data were obtained from publicly available videos on TikTok and Bilibili, with no collection of personal or identifiable information, so ethical approval was not needed.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Yárnoz-Esquiroz, P. et al. Obesities: Position statement on a complex disease entity with multifaceted drivers. Eur J Clin Invest.52, e13811 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lauby-Secretan, B. et al. Body fatness and cancer-viewpoint of the IARC working group. N. Engl. J. Med.375, 794–798 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Okunogbe, A., Nugent, R., Spencer, G., Powis, J., Ralston, J., Wilding, J. Economic impacts of overweight and obesity: Current and future estimates for 161 countries. BMJ Glob Health.7, (2022). [DOI] [PMC free article] [PubMed]
- 4.Zhang, M. et al. How do lifestyle factors modify the association between genetic predisposition and obesity-related phenotypes? A 4-way decomposition analysis using UK Biobank. BMC Med.22, 230 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Saeed, S., Bonnefond, A. & Froguel, P. Obesity: exploring its connection to brain function through genetic and genomic perspectives. Mol Psychiatry.30, 651–658 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bray, G. A., Frühbeck, G., Ryan, D. H. & Wilding, J. P. H. Management of obesity. Lancet387, 1947–1956 (2016). [DOI] [PubMed] [Google Scholar]
- 7.Koliaki, C., Spinos, T., Spinou, Μ., Brinia Μ, E., Mitsopoulou, D., Katsilambros, N. Defining the optimal dietary approach for safe, effective and sustainable weight loss in overweight and obese adults. Healthcare (Basel).6, (2018). [DOI] [PMC free article] [PubMed]
- 8.Rus, H. M. & Cameron, L. D. Health communication in social media: message features predicting user engagement on diabetes-related facebook pages. Ann. Behav. Med.50, 678–689 (2016). [DOI] [PubMed] [Google Scholar]
- 9.Song, S., Zhao, Y. C., Yao, X., Ba, Z. & Zhu, Q. Short video apps as a health information source: an investigation of affordances, user experience and users’ intention to continue the use of TikTok. Int. Res.31(6), 2120–2142 (2021). [Google Scholar]
- 10.Zhang, R. et al. Analyzing dissemination, quality, and reliability of Chinese brain tumor-related short videos on TikTok and Bilibili: A cross-sectional study. Front Neurol.15, 1404038 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.McMullan, R. D., Berle, D., Arnáez, S. & Starcevic, V. The relationships between health anxiety, online health information seeking, and cyberchondria: Systematic review and meta-analysis. J Affect Disord.245, 270–278 (2019). [DOI] [PubMed] [Google Scholar]
- 12.Fortinsky, K. J., Fournier, M. R. & Benchimol, E. I. Internet and electronic resources for inflammatory bowel disease: A primer for providers and patients. Inflamm Bowel Dis.18, 1156–1163 (2012). [DOI] [PubMed] [Google Scholar]
- 13.Scanfeld, D., Scanfeld, V. & Larson, E. L. Dissemination of health information through social networks: Twitter and antibiotics. Am. J. Infect. Control.38, 182–188 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Song, S., Zhang, Y. & Yu, B. Interventions to support consumer evaluation of online health information credibility: A scoping review. Int J Med Inform.145, 104321 (2021). [DOI] [PubMed] [Google Scholar]
- 15.Mueller, S. M. et al. Fiction, falsehoods, and few facts: cross-sectional study on the content-related quality of atopic eczema-related videos on Youtube. J Med Internet Res.22, e15599 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sampige, R., Rodgers, E. G., Huang, A. & Zhu, D. Education and misinformation: exploring ophthalmology content on TikTok. Ophthalmol Ther.13, 97–112 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wang, M. et al. Bilibili, TikTok, and YouTube as sources of information on gastric cancer: assessment and analysis of the content and quality. BMC Public Health24, 57 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sun, F., Zheng, S. & Wu, J. Quality of information in gallstone disease videos on TikTok: cross-sectional study. J. Med. Internet. Res.25, e39162 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ferhatoglu, M. F., Kartal, A., Ekici, U. & Gurkan, A. Evaluation of the reliability, utility, and quality of the information in sleeve gastrectomy videos shared on open access video sharing platform YouTube. Obes. Surg.29, 1477–1484 (2019). [DOI] [PubMed] [Google Scholar]
- 20.Sun, F., Yang, F. & Zheng, S. Evaluation of the liver disease information in Baidu Encyclopedia and wikipedia: Longitudinal study. J. Med. Internet. Res.23, e17680 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Charnock, D., Shepperd, S., Needham, G. & Gann, R. DISCERN: an instrument for judging the quality of written consumer health information on treatment choices. J Epidemiol Community Health.53, 105–111 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Moens, M. et al. Examining the type, quality, and content of web-based information for people with chronic pain interested in spinal cord stimulation: social listening study. J. Med. Internet. Res.26, e48599 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lai, Y. et al. The status quo of short videos as a health information source of Helicobacter pylori: A cross-sectional study. Front Public Health.11, 1344212 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Silberg, W. M., Lundberg, G. D. & Musacchio, R. A. Assessing, controlling, and assuring the quality of medical information on the Internet: Caveant lector et viewor–Let the reader and viewer beware. JAMA277, 1244–1245 (1997). [PubMed] [Google Scholar]
- 25.Bernard, A. et al. A systematic review of patient inflammatory bowel disease information resources on the World Wide Web. Am. J. Gastroenterol.102, 2070–2077 (2007). [DOI] [PubMed] [Google Scholar]
- 26.Kyarunts, M., Mansukhani, M. P., Loukianova, L. L. & Kolla, B. P. Assessing the quality of publicly available videos on MDMA-assisted psychotherapy for PTSD. Am. J. Addict.31, 502–507 (2022). [DOI] [PubMed] [Google Scholar]
- 27.Kong, W., Song, S., Zhao, Y. C., Zhu, Q. & Sha, L. TikTok as a health information source: Assessment of the quality of information in diabetes-related videos. J. Med. Internet. Res.23, e30409 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Luarn, P., Lin, Y. F. & Chiu, Y. P. Influence of Facebook brand-page posts on online engagement. Online Inform. Rev.39(4), 1–16 (2015). [Google Scholar]
- 29.FerranSabatea, J. M., Cañabatec, A. & Lebherzd, P. R. Factors influencing popularity of branded content in Facebook fan pages. Europ. Manag. J.32, 1001–1011 (2014). [Google Scholar]
- 30.Li, H.O., Bailey, A., Huynh, D., Chan, J. YouTube as a source of information on COVID-19: a pandemic of misinformation? BMJ Glob Health.5, (2020). [DOI] [PMC free article] [PubMed]
- 31.Lenczowski, E. & Dahiya, M. Psoriasis and the digital landscape: YouTube as an information source for patients and medical professionals. J. Clin. Aesthet Dermatol.11, 36–38 (2018). [PMC free article] [PubMed] [Google Scholar]
- 32.Berland, G. K. et al. Health information on the Internet: Accessibility, quality, and readability in English and Spanish. JAMA285, 2612–2621 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Walji, M. et al. Efficacy of quality criteria to identify potentially harmful information: a cross-sectional survey of complementary and alternative medicine web sites. J. Med. Internet. Res.6, e21 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The investigation did not involve clinical data, human subjects, or animal testing. All data were obtained from publicly available videos on TikTok and Bilibili, with no collection of personal or identifiable information, so ethical approval was not needed.