Abstract
Introduction
Social media platforms have become major sources of health information, including content related to metabolic bariatric surgery (MBS). While these platforms offer wide accessibility and high user engagement, concerns remain regarding the quality, reliability, and completeness of the information presented. Comparative evaluations across platforms integrating both quantitative quality assessment and qualitative insights are limited.
Methods
A cross-sectional mixed-methods content analysis was conducted on 45 publicly available videos (15 per platform) from TikTok, YouTube, and Facebook, identified using standardized search terms. Data extraction included video metadata, engagement metrics, uploader characteristics, and content features. Video quality and reliability were assessed using the Video Information and Quality Index (VIQI), Global Quality Score (GQS), and DISCERN instrument. Inter-rater reliability was evaluated using intraclass correlation coefficients. Additionally, thematic analysis of rater comments was performed to explore qualitative content patterns.
Results
Significant differences were observed across platforms in video duration, content characteristics, and engagement (p < 0.001). TikTok videos were shortest and demonstrated the highest median views and engagement rates, while YouTube videos were longest and Facebook videos showed moderate duration and engagement. Quality assessment revealed significantly lower VIQI, GQS, and DISCERN scores for YouTube compared with TikTok and Facebook (p < 0.001), with no significant difference between TikTok and Facebook. High-quality content (GQS ≥ 4) was absent on YouTube but present in 60–66.7% of TikTok and Facebook videos. Inter-rater reliability was good to excellent across all tools (ICC range: 0.81–0.88). Qualitative analysis demonstrated recurring themes of incomplete clinical information, limited discussion of risks, and a predominance of positively framed narratives, especially in short-form content.
Conclusion
MBS-related content on social media varies significantly across platforms in terms of quality, reliability, and engagement. While TikTok and Facebook demonstrated higher overall quality scores than YouTube, all platforms exhibited important deficiencies in critical patient-centered information. High engagement does not equate to high informational quality. These findings highlight the need for improved content regulation, increased involvement of healthcare professionals, and development of standardized guidelines to enhance the educational value of bariatric surgery information on social media.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1007/s11695-026-08922-9.
Keywords: Metabolic bariatric surgery, Social media, Video quality assessment, DISCERN, Global quality score, Video information and quality index
Introduction
Metabolic bariatric surgery (MBS) remains the most effective long-term treatment for severe obesity, producing substantial and sustained weight loss while improving obesity-related comorbidities, including type 2 diabetes mellitus, hypertension, and cardiovascular disease [1, 2]. As the number of bariatric procedures increases worldwide, patients are increasingly seeking information regarding surgical options, expected outcomes, risks, and postoperative lifestyle modifications [3]. Social media has emerged as one of the most influential sources of health information, enabling patients to access educational content, share personal experiences, and engage with healthcare communities [4, 5]. Platforms such as YouTube, Facebook, and TikTok are among the most frequently used social media platforms for health-related information seeking and have become important channels through which MBS information is disseminated [4, 5]. The accessibility and wide reach of these platforms provide opportunities for patient education but also raise concerns regarding the accuracy, completeness, and reliability of the information presented [3, 5].
YouTube has long been used as a source of medical and surgical information through longer-form educational videos produced by healthcare professionals, institutions, and patients. However, the absence of formal peer review and content regulation has generated concerns regarding the quality and reliability of information available to viewers [4]. More recently, TikTok has experienced rapid growth as a platform for short-form video content, offering high levels of user engagement and information dissemination. Nevertheless, the brief and highly visual nature of TikTok content may encourage oversimplification of complex medical topics, while much of the available information is produced by non-specialists. Facebook continues to serve as a major platform for health communication and patient support, but limited research has examined the quality and reliability of MBS information available on the platform [6]. These concerns have also been highlighted by the evolving approach of major social-media companies to content moderation. In January 2025, Meta CEO Mark Zuckerberg announced changes to the company’s content-moderation policies, emphasizing greater reliance on user-generated community notes and a reduction in certain forms of content moderation. Such developments underscore the ongoing tension between reducing censorship and ensuring the accuracy and reliability of information disseminated through social-media platforms [7].
The quality of online MBS information has important clinical implications. Previous research has demonstrated that educational videos can significantly improve patient understanding during the informed consent process and enhance shared decision-making [3]. Therefore, social media content may influence patient expectations, perceptions of surgical risks and benefits, and decisions regarding treatment. However, high levels of engagement do not necessarily indicate high-quality information. Studies evaluating bariatric surgery–related content on YouTube and TikTok have reported substantial variability in quality and reliability, and have shown that engagement metrics such as views, likes, and shares are often poorly correlated with validated measures of informational quality [3, 6].
Although previous studies have assessed MBS content on individual social media platforms, comparative evaluations across multiple platforms remain limited. Furthermore, most investigations have relied exclusively on quantitative quality assessment tools, providing limited insight into recurring content patterns and contextual factors that may influence educational value. Therefore, this study aimed to evaluate and compare the quality, reliability, content characteristics, and audience engagement of MBS videos on TikTok, YouTube, and Facebook using validated assessment instruments and complementary qualitative thematic analysis. By identifying platform-specific strengths and limitations, this study seeks to provide evidence that may inform patient education strategies and improve the dissemination of accurate MBS information through social media.
Methods
Study Design
This cross-sectional observational content analysis was conducted to evaluate the quality, reliability, and audience engagement of publicly available social media videos related to bariatric surgery. The study examined content available on three major social media platforms and assessed both informational quality and user engagement characteristics. Because the study involved analysis of publicly accessible online content and did not include human participants, patient data, or identifiable personal information, institutional review board approval was not required according to local regulations governing research involving publicly available data.
Platform Selection
TikTok, YouTube, and Facebook were selected for analysis because they represent major sources of online health information and differ substantially in content format, audience characteristics, and communication style. TikTok primarily hosts short-form video content, whereas YouTube and Facebook support longer educational and informational videos. Together, these platforms provide a broad representation of contemporary social media environments through which patients may obtain information regarding bariatric surgery.
Search Strategy and Video Selection
Video retrieval was performed on 13 January 2026 using a dedicated browser profile without prior browsing history, search history, or account personalization. Searches were conducted independently on TikTok, YouTube, and Facebook using the terms “bariatric surgery,” “gastric sleeve,” “gastric bypass,” and “weight loss surgery.” Search results were reviewed according to each platform’s default relevance ranking to reflect the information environment encountered by typical users.
Duplicate videos identified across search terms within the same platform were removed. Videos were eligible for inclusion if they primarily focused on bariatric surgery, were available in English or contained English subtitles, were publicly accessible at the time of retrieval, and had accumulated at least 1,000 views. Videos were excluded if they represented duplicate uploads from the same creator, consisted solely of promotional or advertising content without educational information, were inaccessible at the time of data collection, or were live-stream recordings that were not available as complete videos. For each platform, the first 15 eligible videos meeting the inclusion criteria were included, resulting in a final sample of 45 videos comprising 15 TikTok videos, 15 YouTube videos, and 15 Facebook videos.
Data Extraction
A standardized data extraction form was developed prior to data collection. For each video, information regarding platform, title, upload date, duration, and URL was recorded. Engagement metrics, including total views, likes, comments, and shares when available, were collected at the time of retrieval. Uploader characteristics, including account name, follower or subscriber count, verification status, and uploader type, were documented. Uploaders were categorized as healthcare professionals, healthcare institutions or hospitals, official media organizations, self-media creators, or other content producers. For subsequent analyses, uploaders were additionally classified as professional or non-professional.
Content characteristics were also recorded, including video category, primary educational purpose, and intended audience. Videos were categorized as educational, patient experience, procedure demonstration, news or media content, or other. The presence of a healthcare professional on screen, visual aids, on-screen text, and spoken explanations was documented. Overall video tone was classified as positive, neutral, or negative. A complete list of extracted variables is provided in Table 1.
Table 1.
Data Extraction Variables
| Domain | Variables Extracted |
|---|---|
| Video Metadata | |
| Platform | TikTok, YouTube, or Facebook |
| Title | Full title as displayed |
| Upload date | Day/month/year |
| Duration | Total length in seconds |
| URL | Full video URL for verification |
| Engagement Metrics | |
| Total views | Cumulative view count at time of retrieval |
| Views per 30 days | Estimated 30-day rolling view rate |
| Total likes / likes per 30 days | Cumulative and rolling likes |
| Total comments / comments per 30 days | Cumulative and rolling comments |
| Total shares / shares per 30 days | Cumulative and rolling shares (where available) |
| Uploader Characteristics | |
| Uploader ID / channel name | As displayed on platform |
| Number of followers / subscribers | At time of retrieval |
| Certification/verification status | Verified or Not verified (platform badge) |
| Uploader type | Self-media, Hospital, Healthcare Institution, Official media, Doctor/Individual Medical Professional, Media Organisation, Media Personality |
| Professional status | Professional (healthcare organisation or physician) vs. Non-professional (lay/self-media creator) |
| Content Characteristics | |
| Video category | Patient experience, Educational, Procedure/Transformation Demonstration, News/media, Other |
| Primary educational aim | As coded from content (e.g., Outcomes and expectations, Procedure explanation, General awareness, Nutrition and diet) |
| Secondary educational aim | Where applicable |
| Target audience | General public, Mixed, Healthcare students/professionals, Specific patient subgroup |
| Video Features | |
| Medical professional on screen | Yes / No |
| Visual aids used | Yes / No (e.g., animations, diagrams, slides) |
| On-screen text used | Yes / No (captions, subtitles, overlaid text) |
| Spoken explanation | Yes / No (narration or verbal description of content) |
| Overall tone | Positive, Neutral, or Negative |
All variables were coded independently by each rater. Discrepancies were resolved by consensus discussion
Quality Assessment
Video quality was evaluated using the Video Information and Quality Index (VIQI) and the Global Quality Score (GQS). The VIQI assesses four domains: flow of information, accuracy of information, overall quality, and precision. Each domain is scored using a five-point Likert scale, with higher scores indicating superior informational and production quality. A composite VIQI score was calculated as the mean of the four domain scores [8].
The GQS was used as a global measure of educational quality and usefulness. This instrument consists of a single five-point item that provides an overall assessment of the quality of health-related video content. Higher scores indicate greater educational value and usefulness to viewers [9].
The DISCERN instrument was used to assess the reliability and completeness of treatment-related health information. DISCERN consists of 15 items evaluating the quality of information presented regarding treatment choices, together with a single overall quality rating item. Each item is scored on a five-point scale, with higher scores reflecting more reliable and comprehensive information. Mean DISCERN scores and mean overall DISCERN ratings were calculated for each video [10].
Rater Training and Assessment
All videos were independently evaluated by four raters, each with more than 10 years of experience in bariatric surgery. Prior to formal assessment, raters participated in a structured calibration exercise using pilot videos that were not included in the final sample and provided with a rating guide contains all data regarding the rating criteria and tools attached to supplementary file. During this process, scoring criteria and operational definitions for the VIQI, GQS, and DISCERN instruments were reviewed to ensure consistency in interpretation and application. Following calibration, raters independently evaluated all included videos and remained blinded to the assessments of other raters throughout the study.
Inter-Rater Reliability
Inter-rater reliability was assessed using the intraclass correlation coefficient (ICC), calculated using a two-way random-effects model with absolute agreement. This model was selected because all videos were independently assessed by multiple raters and absolute agreement between raters was of primary interest. ICC values were interpreted according to established recommendations, with values below 0.50 indicating poor reliability, values between 0.50 and 0.75 indicating moderate reliability, values between 0.75 and 0.90 indicating good reliability, and values greater than 0.90 indicating excellent reliability. Fleiss’ kappa statistics were additionally calculated for categorical variables.
Derived Variables
Composite quality measures were calculated by averaging ratings across raters. The VIQI composite score was calculated as the mean of the four VIQI domains and subsequently averaged across all raters. The GQS score was calculated as the mean overall GQS rating across raters. Mean DISCERN scores were calculated by averaging the 15 DISCERN items and subsequently averaging across raters.
Total engagement was calculated as the sum of likes, comments, and shares. Engagement rate was calculated by dividing total engagement by total views, thereby normalizing engagement according to audience reach. Video age was defined as the number of days between the upload date and the date of data collection.
Qualitative Content Analysis
In addition to quantitative assessment, a qualitative analysis of free-text rater comments was conducted to explore contextual aspects of video content that were not fully captured by the VIQI, GQS, and DISCERN instruments. The qualitative corpus comprised comments recorded by raters during video assessment, including observations regarding content characteristics, educational value, information completeness, presentation style, and perceived strengths or limitations of individual videos.
Thematic analysis was performed using an inductive approach based on the framework described by Braun and Clarke [11]. All substantive comments were reviewed repeatedly to achieve familiarity with the data, after which initial codes were generated and grouped into broader themes. Themes were iteratively reviewed and refined to ensure coherence and relevance to the study objectives. Particular attention was given to recurring patterns across videos and platforms, including information completeness, educational content, promotional characteristics, patient-centered communication, and disclosure of risks and treatment alternatives.
To enhance the credibility of findings, comments from multiple raters were examined for convergence and divergence of observations. Qualitative findings were subsequently integrated with quantitative results to provide additional context and facilitate interpretation of platform-specific differences in video quality and informational content.
Measures to Reduce Bias
Several measures were implemented to reduce potential sources of bias. All searches were conducted using a clean browser profile without prior browsing history or account personalization to minimize algorithmic influence on search results. The same search terms, retrieval procedures, and eligibility criteria were applied consistently across all platforms. Raters underwent calibration prior to formal assessment and remained blinded to one another’s evaluations throughout the study. All videos were identified and assessed within a single data collection period to reduce temporal variation in content availability and engagement metrics.
Statistical Analysis
Statistical analyses were performed using R software (version 4.4.1; R Foundation for Statistical Computing, Vienna, Austria) and IBM SPSS Statistics (version 26.0; IBM Corp., Armonk, NY, USA) [12, 13]. Continuous variables were summarized using means and standard deviations or medians and interquartile ranges, as appropriate, while categorical variables were reported as frequencies and percentages. Statistical significance was defined as a two-sided p-value < 0.05.
Because quality scores and engagement metrics demonstrated non-normal distributions, non-parametric statistical methods were used for inferential analyses. Comparisons among TikTok, YouTube, and Facebook were performed using the Kruskal–Wallis test. When statistically significant differences were identified, pairwise Mann–Whitney U tests with Bonferroni adjustment were conducted for post hoc comparisons. Effect size for pairwise comparisons was reported using rank-biserial correlation.
Categorical variables were compared using the chi-square test or Fisher’s exact test when expected cell counts were less than five. Effect size for categorical analyses was reported using Cramér’s V. Associations between quality scores, engagement metrics, and video characteristics were assessed using Spearman’s rank correlation coefficient.
Results
Sample Characteristics and Video Metadata
A total of 45 videos were analyzed: 15 from TikTok, 15 from YouTube, and 15 from Facebook Supplementary Table S1. Video duration differed significantly across platforms (H = 26.18, p < 0.001). TikTok videos were the shortest (mean 13.9 ± 6.0 s; median 12.0 s, IQR 9.5–18.0 s), followed by Facebook (mean 168.9 ± 239.7 s; median 108.0 s) and YouTube (mean 899.0 ± 1485.9 s; median 301.0 s). TikTok vs. YouTube: U = 0, pBonf < 0.001, r = 1.00; TikTok vs. Facebook: pBonf = 0.002, r = 0.73; YouTube vs. Facebook: pBonf = 0.051.
All TikTok uploaders were unverified self-media creators (100%; 0% professional). All YouTube and Facebook uploaders were professional verified healthcare organizations or physicians (100%). All TikTok videos depicted patient-experience content with a positive tone, targeting the general public. YouTube content was predominantly educational (87%). A single negative-tone video, a news documentary on adverse bariatric surgical outcomes, appeared in the Facebook sample Table 2.
Table 2.
Baseline Characteristics of Videos by Platform (N = 45)
| Characteristic | TikTok (n = 15) | YouTube (n = 15) | Facebook (n = 15) | Overall (N = 45) |
|---|---|---|---|---|
| Video Metrics | ||||
| Duration, mean ± SD (s) | 13.9 ± 6.0 | 899.0 ± 1485.9 | 168.9 ± 239.7 | 360.6 ± 876.4 |
| Duration, median [IQR] (s) | 12.0 [9.5–18.0] | 301.0 [140.5–755.0] | 108.0 [45.5–186.0] | — |
| Total views, median [IQR] | 3,000,000 [1,650,000–5,100,000] | 859,287 [496,794–1,185,408] | 48,000 [16,000–62,000] | 802,000 |
| Total engagement, median [IQR] | 305,300 [140,250–442,450] | 8,965 [1,600–11,800] | 147 [52–571] | 14,700 |
| Engagement rate, median [IQR] | 10.3% [5.9–13.0%] | 0.64% [0.41–1.05%] | 0.30% [0.23–0.93%] | — |
| Followers, median [IQR] | 32,300 [19,100–69,000] | 56,500 [30,500–3,666,500] | 22,500 [7,400–37,250]* | — |
| Uploader Characteristics | ||||
| Verified, n (%) | 0 (0%) | 15 (100%) | 13 (87%) | 28 (62%) |
| Professional, n (%) | 0 (0%) | 15 (100%) | 15 (100%) | 30 (67%) |
| Uploader type | Self-media (100%) | Hospital (53%) Official media (33%) Doctor (13%) | Healthcare institution (73%) Medical professional (13%) Other (13%) | — |
| Content Characteristics | ||||
| Video category | Patient experience (100%) | Educational (87%) Other (13%) | Educational (47%) Patient exp. (20%) Other (33%) | — |
| Tone: Pos / Neutral / Neg | 15 / 0 / 0 | 9 / 5 / 0 | 9 / 3 / 1 | 33 / 8 / 1 |
Engagement = total likes + comments + shares. Engagement rate = total engagement / total views. Professional = hospital, healthcare institution, official medical media, or individual physician. *, Available only for 11 videos
Inter-Rater Reliability
Reliability was evaluated using the Intraclass Correlation Coefficient (ICC, model 2,1) to assess absolute agreement among the four independent raters. ICC values demonstrated excellent reliability for the VIQI composite (ICC = 0.88) and GQS (ICC = 0.84), and good reliability for DISCERN items (ICC = 0.81).
Video Content Features
Medical professional on screen differed significantly across platforms (χ² = 24.12, df = 2, p < 0.001): absent in all TikTok videos (0%, 0/15), present in 46.7% of YouTube (7/15), and 92.3% of Facebook videos (12/13). Spoken clinical explanation showed an identical gradient: 0% TikTok, 86.7% YouTube, 100% Facebook (χ² = 35.75, p < 0.001). Visual aids were used in all TikTok videos (100%), 60.0% of YouTube, and 15.4% of Facebook (χ² = 23.61, p < 0.001). On-screen text: TikTok 86.7% vs. YouTube 40.0% vs. Facebook 46.2% (χ² = 11.50, p = 0.022; Table 3).
Table 3.
Video Content Features by Platform
| Feature | TikTok n/N (%) | YouTube n/N (%) | Facebook n/N (%) | χ² (df) | p |
|---|---|---|---|---|---|
| Medical professional on screen | 0/15 (0%) | 7/15 (46.7%) | 12/15 (80%) | 24.12 (2) | < 0.001 |
| Visual aids used | 15/15 (100%) | 9/15 (60.0%) | 2/15 (13.3%) | 23.61 (4) | < 0.001 |
| On-screen text used | 13/15 (86.7%) | 6/15 (40.0%) | 6/15 (40%) | 11.50 (4) | 0.022 |
| Spoken explanation | 0/15 (0%) | 13/15 (86.7%) | 13/15 (86.7%) | 35.75 (2) | < 0.001 |
Video Information Quality
VIQI Score
The VIQI composite, comprising Flow, Accuracy, Overall Quality, and Precision, differed significantly across platforms (H = 29.29, p < 0.001). TikTok (mean 4.14 ± 0.47; median 4.33) and Facebook (mean 4.13 ± 0.51; median 4.22) were statistically equivalent (U = 106, pBonf = 1.00, r = − 0.005) and both significantly outperformed YouTube (mean 2.33 ± 0.18; median 2.50) with maximal effect sizes (TikTok vs. YouTube: U = 225, pBonf < 0.001, r = − 1.00; Facebook vs. YouTube: U = 0, pBonf < 0.001, r = 1.00) (Fig. 1a) (Table 4). YouTube scored lowest on all subscales.
Fig. 1.
Panel A: Mean VIQI subscale scores by platform with SEM error bars. Panel B: Mean GQS scores by platform with significance brackets (*** p < 0.001, Bonferroni-corrected). Dashed line indicates scale midpoint
Table 4.
Comparison of Video Quality Scores Across Platforms (VIQI, GQS, and DISCERN)
| Scale / Item | TikTok (n = 15) | YouTube (n = 15) | Facebook (n = 15) | H (KW) | p |
|---|---|---|---|---|---|
| VIQI Score (4-Item, Range 1–5) | |||||
| Composite mean | 4.14 ± 0.47 [4.33; 3.83–4.46] | 2.33 ± 0.18 [2.50; 2.29–2.50] | 4.13 ± 0.51 [4.22; 4.08–4.48] | 29.29 | < 0.001 |
| Flow | 3.33 ± 1.03 | 2.15 ± 0.30 | 3.75 ± 1.09 | — | — |
| Accuracy | 4.20 ± 0.75 | 2.18 ± 0.81 | 4.18 ± 0.78 | — | — |
| Overall quality | 4.05 ± 0.76 | 2.20 ± 0.93 | 4.09 ± 0.76 | — | — |
| Precision | 4.25 ± 0.83 | 2.77 ± 0.78 | 4.16 ± 0.82 | — | — |
| High quality (≥ 4), n (%) | 10 (66.7%) | 0 (0%) | 12 (80.0%) | — | — |
| GQS Score (1 Global Item, Range 1–5) | |||||
| GQS mean | 4.03 ± 0.49 [4.00; 3.75–4.38] | 2.25 ± 0.13 [2.25; 2.25–2.25] | 4.09 ± 0.54 [4.25; 3.81–4.50] | 29.99 | < 0.001 |
| High quality (≥ 4), n (%) | 9 (60.0%) | 0 (0%) | 10 (66.7%) | — | — |
| DISCERN (Range 1–5) | |||||
| Overall rating | 3.73 ± 0.56 [3.75; 3.25–4.13] | 2.12 ± 0.35 [2.25; 2.13–2.25] | 3.66 ± 0.53 [3.75; 3.50–4.00] | 29.44 | < 0.001 |
| 15-item subscale mean | 3.16 ± 0.57 | 1.92 ± 0.10 | 3.26 ± 0.45 | 29.34 | < 0.001 |
| High quality (overall ≥ 4), n (%) | 5 (33.3%) | 0 (0%) | 5 (33.3%) | — | — |
KW Kruskal-Wallis. Bonferroni-corrected post-hoc Mann-Whitney U: TikTok and Facebook both exceeded YouTube on all composite scales (r ≥ 0.995, pBonf < 0.001); TikTok vs. Facebook non-significant (pBonf ≥ 0.56)
GQS Score (Global Quality Item)
The GQS global item (“How would you rate the overall quality and usefulness of this video?“) showed an identical pattern (H = 29.99, p < 0.001): TikTok mean 4.03 ± 0.49, Facebook 4.09 ± 0.54, YouTube 2.25 ± 0.13 (Fig .1b). TikTok and Facebook did not differ (U = 92, pBonf = 1.00, r = 0.13); both exceeded YouTube (all pBonf < 0.001). VIQI and GQS were very strongly correlated (rs = 0.934, p < 0.001) (Table 4), confirming convergent validity.
DISCERN Scores
The DISCERN overall rating differed significantly (H = 29.44, p < 0.001): TikTok 3.73 ± 0.56, Facebook 3.66 ± 0.53, YouTube 2.12 ± 0.35. TikTok and Facebook were equivalent (U = 104, pBonf = 1.00); both exceeded YouTube (pBonf < 0.001). DISCERN overall ≥ 4 (high quality): TikTok 5/15 (33.3%), Facebook 5/15 (33.3%), YouTube 0/15 (0%). The 15-item subscale mean showed the same pattern (TikTok 3.16 ± 0.57; Facebook 3.26 ± 0.45; YouTube 1.92 ± 0.10; H = 29.34, p < 0.001) (Table 4).
Despite the uniformly low proportion of videos meeting the predefined high-quality thresholds, the YouTube videos were not uniformly devoid of educational features. Most were classified as educational (13/15, 86.7%), with spoken clinical explanations present in 13/15 (86.7%), visual aids in 9/15 (60.0%), and a medical professional visible in 7/15 (46.7%). However, qualitative rater comments indicated that the videos generally emphasized postoperative weight loss and changes in body profile, with limited clinical context and incomplete discussion of complications, long-term outcomes, treatment risks, and alternative treatment options.
DISCERN Item-Level Completeness
Systematic gaps were identified across all platforms. The most critically deficient items were risk disclosure (overall mean 1.94/5), consequences of no treatment (2.14/5), treatment alternatives (2.27/5), and support for shared decision-making (2.49/5). YouTube scored lowest on every item. Facebook achieved the highest scores for clarity of aims (4.30/5) and achievement of aims (4.12/5; Table 5; Figs. 2 and 3).
Table 5.
Mean DISCERN Item Scores by Platform (1–5; Averaged Across 4 Raters)
| # | DISCERN Item | TikTok | YouTube | Overall | |
|---|---|---|---|---|---|
| 1 | Are the aims clear? | 4.03 | 2.68 | 4.30 | 3.66 |
| 2 | Does it achieve its aims? | 3.92 | 2.55 | 4.12 | 3.52 |
| 3 | Is information relevant to patients? | 3.87 | 2.47 | 4.04 | 3.44 |
| 4 | Are sources of information clear? | 3.32 | 1.85 | 3.52 | 2.88 |
| 5 | Is information balanced and unbiased? | 3.67 | 2.08 | 3.30 | 3.01 |
| 6 | Details of additional sources of support? | 2.83 | 1.63 | 3.29 | 2.57 |
| 7 | Does it refer to areas of uncertainty? | 2.90 | 1.83 | 2.77 | 2.49 |
| 8 | Does it describe how treatment works? | 3.62 | 2.00 | 3.50 | 3.03 |
| 9 | Does it describe benefits of treatment? | 3.33 | 2.38 | 3.82 | 3.16 |
| 10 | ⚠ Does it describe RISKS of treatment? | 2.43 | 1.12 | 2.29 | 1.94 |
| 11 | What happens if no treatment is used? | 2.15 | 1.85 | 2.45 | 2.14 |
| 12 | Does it describe effect on quality of life? | 2.55 | 2.02 | 3.25 | 2.59 |
| 13 | Is it clear there is more than one Tx choice? | 3.12 | 1.32 | 2.62 | 2.35 |
| 14 | ⚠ Does it support SHARED DECISION-MAKING? | 2.73 | 1.68 | 3.11 | 2.49 |
| 15 | ⚠ Does it provide information on Tx ALTERNATIVES? | 2.98 | 1.28 | 2.57 | 2.27 |
| — | DISCERN Overall Rating | 3.73 | 2.12 | 3.66 | 3.16 |
⚠ = critical patient-safety items with overall mean < 2.5/5. Highlighted rows (pale red) = DISCERN domains essential to informed consent and shared decision-making. Scores 1 (not addressed) to 5 (fully addressed)
Fig. 2.
DISCERN Item-Level Scores by Platform (Bar Chart)
Fig. 3.
DISCERN Item Completeness Heatmap
Engagement Metrics Across Platforms
Total views differed significantly (H = 33.84, p < 0.001): TikTok median 3,000,000 > YouTube median 859,287 (pBonf = 0.002, r = − 0.73) > Facebook median 48,000 (all pairwise pBonf < 0.001). Total engagement was even more divergent (H = 38.45, p < 0.001): TikTok 305,300 vs. YouTube 8,965 vs. Facebook 147 (all pairwise r ≥ 0.96, pBonf < 0.001). Engagement rate (H = 29.04, p < 0.001): TikTok 10.3% vs. YouTube 0.64% vs. Facebook 0.30%; TikTok exceeded both (pBonf < 0.001); YouTube and Facebook did not differ (pBonf = 0.54).
Professional vs. Non-Professional Uploaders
Non-professional uploaders (n = 15; all TikTok) achieved significantly higher VIQI scores (median 4.33 vs. 2.50; U = 104, p = 0.005, r = 0.52), GQS scores (median 4.00 vs. 2.50; U = 118, p = 0.013, r = 0.46), and DISCERN overall ratings (median 3.75 vs. 2.50; U = 106, p = 0.006, r = 0.51) compared with professional uploaders (n = 30) (Fig. 4). Conversely, professionals achieved dramatically lower engagement: total views (median 172,492 vs. 3,000,000; r = 0.87), total engagement (912 vs. 305,300; r = 1.00), and engagement rate (0.6% vs. 10.3%; r = 1.00) (Table 6).
Fig. 4.
Quality Scores by Uploader Professional Status
Table 6.
Professional vs. Non-Professional Uploaders: Quality and Engagement Comparison
| Variable | Professional Median (n = 30) | Non-Professional Median (n = 15) | U | p | r (effect size) |
|---|---|---|---|---|---|
| VIQI Score (1–5) | 2.50 | 4.33 | 104 | 0.005 | 0.52 |
| GQS Score (1–5) | 2.50 | 4.00 | 118 | 0.013 | 0.46 |
| DISCERN Overall (1–5) | 2.50 | 3.75 | 106 | 0.006 | 0.51 |
| Total Views | 172,492 | 3,000,000 | 30 | < 0.001 | 0.87 |
| Total Engagement | 912 | 305,300 | 0 | < 0.001 | 1.00 |
| Engagement Rate | 0.60% | 10.3% | 0 | < 0.001 | 1.00 |
r = effect size = 1 − 2U/(n1 × n2). Professional = hospital, healthcare institution, official medical media, or physician. All TikTok uploaders are non-professional; all YouTube and Facebook are professional — platform confounding should be considered
Correlations: Quality, Engagement, and Video Characteristics
VIQI, GQS, and DISCERN overall were mutually strongly correlated (VIQI vs. GQS: rs = 0.934; VIQI vs. DISCERN: rs = 0.884; GQS vs. DISCERN: rs = 0.855; all p < 0.001) (Fig. 5). No quality metric significantly predicted total views or total engagement (all p > 0.43). VIQI marginally predicted engagement rate (rs = 0.309, p = 0.049). All quality scales were inversely associated with video duration, follower count, and video age, largely reflecting platform-level confounding (Table 7).
Fig. 5.
Information Quality vs. Engagement Rate by Platform
Table 7.
Spearman Rank Correlations Between Quality Scales, Engagement, and Video Characteristics (N = 45)
| Variable Pair | VIQI rs | p | GQS rs | p | DISCERN rs | p |
|---|---|---|---|---|---|---|
| VIQI vs. GQS | 0.934 | < 0.001 | — | — | — | — |
| vs. DISCERN Overall | 0.884 | < 0.001 | 0.855 | < 0.001 | — | — |
| vs. Total Views | -0.087 | 0.575 | -0.121 | 0.434 | -0.094 | 0.543 |
| vs. Total Engagement | 0.013 | 0.931 | — | — | 0.038 | 0.807 |
| vs. Engagement Rate | 0.309 | 0.049 | 0.234 | 0.141 | 0.297 | 0.059 |
| vs. Video Duration | -0.470 | 0.001 | -0.442 | 0.003 | -0.439 | 0.003 |
| vs. Follower Count | -0.539 | < 0.001 | -0.533 | < 0.001 | -0.521 | < 0.001 |
| vs. Video Age (days) | -0.589 | < 0.001 | -0.648 | < 0.001 | -0.611 | < 0.001 |
rs Spearman rank correlation. Red = significant at α = 0.05 (two-tailed). “—” = not computed. Negative associations with duration/age/followers largely reflect platform-level confounding
Qualitative Analysis of Rater Notes
Approach
A thematic analysis was conducted on all substantive rater comments extracted from rater notes and free-text fields across four raters. Comments rated as trivially empty were excluded, yielding 47 substantive comments. Inductive analysis identified five overarching themes (Table 8).
Table 8.
Qualitative Themes from Rater Comments: Summary and Representative Excerpts
| # | Theme | Platform(s) | Description | Prevalence | Representative Rater Excerpts |
|---|---|---|---|---|---|
| 1 | Visual storytelling without clinical depth | TikTok | Silent visual-only format; no spoken clinical explanations; animations may be inaccurate | 15/15 TikTok | “no audio, just diagrammatic appearance of RYGB”; “The animations provide totally wrong expectations" |
| 2 | Surgical transformation as sole message | YouTube & TikTok | Videos reduced to weight/body-profile changes; no clinical context | All 15 YouTube; most TikTok | “Only changes in weight and profile reported” (Rater 4); “motivation — no any informations” (Rater 1) |
| 3 | Incomplete risk & outcome disclosure | All platforms | Absence of complications, long-term outcomes, and post-operative risks | 10 + videos across platforms | “No complications provided”; “Deficit in long-term complications”; “only 4 months follow-up" |
| 4 | Promotional and advertorial framing | Physician self-promotion, surgical tourism marketing, lack of balanced clinical perspective | 4–5 Facebook videos | “advertising for surgical tourism”; “just doctor advertisement”; “advertising way" | |
| 5 | Accessibility and emotional appropriateness | TikTok & Facebook | Non-English content; patient distress footage; graphic surgical content without warnings | 2–3 videos | “not in English — could not evaluate”; “fear of the patient — not suitable for all patients" |
Themes
- Theme 1: Visual Storytelling Without Clinical Depth (TikTok)
- All 15 TikTok videos relied exclusively on visual content — photographs, transformation timelines, and surgical animations — without spoken clinical commentary. Raters noted: “no audio, just diagrammatic appearance of RYGB”; “very good for patient motivation, but no clinical information”; “The animations at some points provide totally wrong expectations.” A recurring structural observation: “Should be divided into smaller clips” (noted independently for four separate videos).
- Theme 2: Surgical Transformation as Sole Message (YouTube / TikTok)
- Rater 4 provided near-identical commentary for all 15 YouTube videos: “Only changes in weight and profile were reported.” Rater 1 similarly described TikTok content as “motivational to patients without information” or “pure motivation — shape changes only, no any informations.” This convergence across independent raters confirms that YouTube videos in the sample functioned as visual testimonials rather than clinical education, despite institutional origins.
- Theme 3: Incomplete Risk and Outcome Disclosure (All Platforms)
- Multiple raters across platforms flagged absent complications, risks, and long-term outcomes: “No complications provided” (TikTok); “only preoperative and after 4 months — no long-term follow-up” (YouTube); “Deficit in long-term complications” (TikTok, two separate videos). The sole negative-tone Facebook documentary was noted as “perfect in describing the journey of complications or leak”, implicitly marking it as the exception rather than the norm.
- Theme 4: Promotional and Advertorial Framing (Facebook)
- Several Facebook videos were identified as promotional: “advertising for surgical tourism — journey, enjoying”; “just doctor advertisement — very simple and short video”; “Advertising way.” At least one Facebook video featured a “carpenter/tool analogy” to explain bariatric surgery — described by a rater as creative but potentially misleading.
- Theme 5: Accessibility and Emotional Appropriateness (TikTok / Facebook)
- Two sub-themes emerged. First, language barriers: “not in English so could not evaluate.” Second, emotional appropriateness: “the fear of the patient was a significant issue — not suitable for all patients.” Additionally: “pure surgery and its details — not for all patients to see” (noted on two TikTok videos featuring graphic surgical footage without discretion warnings).
Discussion
Video content on MBS has rapidly expanded over the past decade, paralleling the increasing global uptake of these procedures. This growth has been further amplified by the use of such content as a form of advertisement and patient engagement [14, 15]. Consequently, there is a pressing need for systematic evaluation of this material to ensure its reliability, as it plays a significant role in shaping patient decisions and attitudes toward surgery. Our comprehensive study provides a cross-platform evaluation of MBS-related video content on TikTok, YouTube, and Facebook, integrating quantitative quality assessment with qualitative thematic analysis to better understand how surgical information is presented and consumed within contemporary digital environments.
A total of 45 videos were analyzed, evenly distributed across TikTok, YouTube, and Facebook. Video duration differed significantly across platforms (H = 26.18, p < 0.001), with TikTok videos being markedly shorter (mean 13.9 ± 6.0 s) compared to Facebook (168.9 ± 239.7 s) and YouTube (899.0 ± 1485.9 s). Despite their brevity, TikTok videos achieved the highest engagement, with a median of 3,000,000 views and an engagement rate of 10.3%, significantly exceeding YouTube (859,287 views; 0.64%) and Facebook (48,000 views; 0.30%) (all p < 0.001). Content characteristics also differed substantially: all TikTok videos were non-professional, patient-experience–based, and positively framed, whereas all YouTube and Facebook videos were produced by professional healthcare sources, with predominantly educational intent.
Information quality demonstrated a consistent platform-dependent pattern. The VIQI composite score differed significantly (H = 29.29, p < 0.001), with TikTok (4.14 ± 0.47) and Facebook (4.13 ± 0.51) outperforming YouTube (2.33 ± 0.18). A similar trend was observed across GQS (H = 29.99, p < 0.001) and DISCERN overall scores (H = 29.44, p < 0.001), where TikTok and Facebook achieved comparable ratings (GQS ≈ 4.0; DISCERN ≈ 3.7), while YouTube consistently ranked lowest (GQS 2.25; DISCERN 2.12). High-quality content (GQS ≥ 4) was present in 60.0% of TikTok and 66.7% of Facebook videos, but absent in YouTube (0%). Strong correlations were observed between all quality metrics (rs ≥ 0.855, p < 0.001), confirming internal consistency.
However, critical informational gaps were identified across all platforms. DISCERN item-level analysis revealed poor coverage of essential patient-centered domains, including risk disclosure (mean 1.94/5), consequences of no treatment (2.14/5), and treatment alternatives (2.27/5). Notably, no quality metric significantly predicted total views or engagement (all p > 0.43), indicating that content visibility operates independently of informational quality. Furthermore, non-professional content (all TikTok) demonstrated significantly higher quality scores than professional content (p ≤ 0.013), while simultaneously achieving dramatically greater reach and engagement. These findings highlight a pronounced mismatch between content dissemination, perceived value, and the delivery of comprehensive, clinically relevant information.
This mixed-methods cross-platform analysis provides a comprehensive evaluation of MBS related video content on TikTok, YouTube, and Facebook, revealing several critical insights regarding information quality, reliability, and audience engagement. The findings demonstrate a striking divergence between informational quality and user engagement, highlight substantial deficiencies in clinically essential content domains, and underscore the influence of platform-specific content structures on how bariatric surgery information is communicated and perceived.
A principal finding of this study was the lack of a clear association between audience reach and informational quality. TikTok videos achieved substantially higher numbers of views, total engagement, and engagement rates than YouTube and Facebook videos; however, total views and total engagement were not significantly associated with VIQI, GQS, or DISCERN scores. VIQI showed only a weak positive association with engagement rate (rs = 0.309, p = 0.049). This aligns with prior literature demonstrating that social media engagement is driven primarily by entertainment value, emotional resonance, and visual appeal rather than informational accuracy or completeness [16]. The absence of correlation between quality metrics and engagement reinforces concerns that widely disseminated content may not reflect high-quality or evidence-based information. In the context of MBS, this disconnect is particularly important, as patient expectations and decision-making may be shaped by highly visible yet incomplete or misleading content.
Interestingly, TikTok videos demonstrated high VIQI and GQS scores despite the absence of spoken explanations and the exclusive reliance on visual storytelling. This finding appears counterintuitive when compared with traditional assumptions that longer, professionally produced educational content provides superior informational value. However, the VIQI instrument partially captures flow, visual clarity, and presentation quality, which may favor concise and visually structured content typical of TikTok [17]. Nevertheless, qualitative analysis revealed that these videos frequently lacked clinical depth, omitted essential information, and occasionally conveyed unrealistic expectations regarding surgical outcomes. This discrepancy highlights a limitation of quantitative scoring systems when applied to short-form media and emphasizes the importance of complementary qualitative assessment.
The dominance of patient-experience narratives on TikTok and, to a lesser extent, YouTube reflects a broader trend in social media health communication, where personal storytelling is often prioritized over evidence-based education. While such narratives may enhance relatability and patient engagement, they inherently lack generalizability and may introduce bias through selective outcome presentation. Previous studies have shown that patient testimonials often emphasize positive outcomes while underreporting complications and long-term challenges, potentially leading to unrealistic expectations among viewers [18]. In the present study, this phenomenon was consistently observed, with transformation-focused content serving as the primary message across many videos.
The underrepresentation of shared decision-making elements further highlights a gap between available online content and contemporary patient-centered care principles. High-quality health information should support patients in understanding available options, weighing risks and benefits, and participating actively in treatment decisions. However, the majority of videos analyzed in this study did not adequately address these aspects, suggesting that current social media content may not effectively support informed decision-making processes. This finding is particularly relevant given evidence that educational interventions, including video-based tools, can significantly enhance patient understanding and engagement during the informed consent process [19–21].
Platform-specific differences in content characteristics also provide important insights. YouTube and Facebook videos were predominantly produced by healthcare professionals and institutions; yet, paradoxically, they demonstrated lower quality scores than TikTok videos. This result may be explained by the nature of the included content, which often consisted of transformation-focused or promotional material rather than structured educational presentations. Prior research has similarly noted that the presence of a healthcare professional does not guarantee high-quality or comprehensive information, particularly in unregulated online environments [22].
Facebook content, while less engaging overall, demonstrated a higher prevalence of healthcare professionals on screen and spoken explanations, suggesting a more traditional educational format. However, qualitative analysis identified a subset of videos with clear promotional intent, including surgical tourism advertising and physician self-promotion. The presence of such content raises concerns regarding conflicts of interest and the potential for biased information dissemination. These findings are consistent with broader concerns about commercialization in online health information, where promotional objectives may overshadow educational value [23].
The inverse relationship observed between video duration and quality scores further reflects the influence of platform-specific characteristics. Shorter videos, particularly on TikTok, were associated with higher quality scores, whereas longer YouTube videos demonstrated lower scores. This may be attributed to differences in content structure, viewer retention strategies, and algorithmic preferences. Short-form content often necessitates concise and visually optimized messaging, which may enhance perceived clarity and flow [24]. However, this brevity inherently limits the ability to convey complex medical information comprehensively, reinforcing the need for cautious interpretation of quality metrics in this context.
The qualitative findings provide critical context to the quantitative results and reveal recurring patterns that are not fully captured by standardized scoring systems. The theme of “visual storytelling without clinical depth” underscores the limitations of visually driven content in conveying nuanced medical information. Similarly, the identification of “incomplete risk and outcome disclosure” across platforms highlights a systemic issue in online bariatric surgery content. These findings emphasize that while quantitative tools such as VIQI, GQS, and DISCERN are valuable, they should be complemented by qualitative analyses to achieve a more comprehensive evaluation.
A further notable observation is the emotional framing of content, particularly on TikTok and Facebook. Videos highlighting dramatic transformations, patient fear, or surgical experiences may shape viewer perceptions through emotional engagement rather than objective clinical information. This aligns with existing literature suggesting that emotionally charged content is more likely to be shared and engaged with on social media platforms [25]. While such content may enhance awareness and outreach, it may also distort public understanding of the risks and benefits associated with bariatric surgery [26, 27]. In addition, Rukavina et al. reported frequent unprofessional behavior among healthcare professionals on social media, including breaches of patient privacy and blurred professional boundaries, alongside insufficiently robust regulatory guidelines governing online conduct [28]. Collectively, these factors may contribute to a distorted representation of bariatric surgery within social media environments.
The findings of this study have important clinical implications. Given the widespread use of social media as a source of health information, healthcare professionals must recognize its influence on patient knowledge and expectations. The observed deficiencies in content quality and completeness highlight the need for greater involvement of healthcare professionals in producing accurate, balanced, and accessible educational content. Additionally, there is a need for strategies to improve the visibility of high-quality information, potentially through platform-level interventions or collaboration with content creators.
Importantly, the health-information landscape is rapidly evolving, and patients are increasingly turning not only to social-media platforms but also to large language models (LLMs) and other artificial intelligence–based tools to obtain information about diseases, treatment options, and surgical procedures [29]. This emerging pattern introduces a new dimension to digital health communication, as patients may use LLM-generated responses alongside or instead of conventional online sources when seeking medical advice. Although LLMs may provide accessible and personalized explanations, their responses can also reflect limitations in the underlying training data and may contain inaccurate, incomplete, or insufficiently contextualized information. Therefore, the deficiencies identified in social-media content should be considered within this broader transition toward AI-mediated health information seeking. The observed deficiencies in content quality and completeness highlight the need for greater involvement of healthcare professionals in producing accurate, balanced, and accessible educational content [30]. Additionally, there is a need for strategies to improve the visibility of high-quality information, potentially through platform-level interventions, collaboration with content creators, and the development of reliable AI-assisted patient education resources [31].
From a research perspective, this study contributes to the growing body of literature evaluating online health information by providing a comparative analysis across multiple platforms and integrating both quantitative and qualitative methodologies. The use of validated assessment tools alongside thematic analysis enhances the robustness of the findings and allows for a more nuanced understanding of content characteristics.
Several limitations should be acknowledged. The sample size was limited to 45 videos, and only the top-ranked results were included, which may not fully represent the diversity of available content. Although we selected TikTok, YouTube, and Facebook because they are among the most widely used and influential social-media platforms for health-related content, other platforms were not evaluated, and their content may demonstrate different patterns of quality and engagement. Furthermore, some videos appeared to be produced by individuals who were not readily identifiable as bariatric surgeons or healthcare professionals. However, to minimize selection bias and reflect the content that users may realistically encounter on these platforms, videos were selected using a predefined random sampling approach rather than selectively excluding content based on the uploader’s professional background. Additionally, the perfect confounding between platform and uploader type limits the ability to isolate the independent effects of professional status. Finally, although validated assessment tools were used, their applicability to short-form video content may be limited, as previously discussed.
Future research should explore larger and more diverse samples, including additional platforms, and examine the impact of social-media content on patient knowledge, attitudes, and clinical outcomes. The integration of artificial intelligence–based content analysis may also provide new opportunities for large-scale evaluation of online health information.
Conclusion
This study demonstrates that MBS-related content on social media platforms is characterized by high engagement but variable and often insufficient informational quality. The lack of correlation between engagement and quality, combined with significant deficiencies in critical educational domains, underscores the need for improved content standards and greater involvement of healthcare professionals in digital health communication. Addressing these challenges is essential to ensure that social media serves as a reliable and effective tool for patient education in bariatric surgery.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The authors acknowledge the MIAX Research Lab for their valuable support in research activities.
Author Contributions
Conceptualization and study design were performed by M.Y.E.-M. and H.E.M. Data analysis was conducted by M.Y.E.M. and H.E.M. Video rating and data acquisition were performed by A.A., M.F.A.K., M.S.A.A., and H.E. Literature review, data curation, and initial drafting of the manuscript were carried out by N.S.A., P.E., N.G.A., and A.A.E. All authors contributed to manuscript writing, and critically reviewed the final version.
Funding
Open access funding provided by The Science, Technology & Innovation Funding Authority (STDF) in cooperation with The Egyptian Knowledge Bank (EKB). No funding was provided for this study.
Data Availability
No datasets were generated or analysed during the current study.
Declarations
Ethical Approval
This article does not contain any studies involving human participants or animals performed by any of the authors.
Informed Consent
Not applicable.
Artificial Intelligence Use
No artificial intelligence tools were used in the preparation of this manuscript.
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.
Ahmed Abokhozima and Mohamed Yasser El-Mezayen contributed equally to this work.
References
- 1.Sarwer DB, Heinberg LJ. A review of the psychosocial aspects of clinically severe obesity and bariatric surgery. Am Psychol. 2020;75(2):252–64. 10.1037/amp0000550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Mirghani H, Alamrani SAS, Alkonani AA, Al Madshush AM. The Impact of Bariatric Surgery on Weight Loss and Glycemic Control in Patients With Obesity and Type 2 Diabetes: A Systematic Review. Cureus. 2023;15(11):e49122. 10.7759/cureus.49122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Montemurro P, Cheema M, Hedén P. Patients’ and Surgeons’ Perceptions of Social Media’s Role in the Decision Making for Primary Aesthetic Breast Augmentation. Aesthetic Surg J. 2018;38(10):1078–84. 10.1093/asj/sjy021. [DOI] [PubMed] [Google Scholar]
- 4.Farag M, Bolton D, Lawrentschuk N. Use of YouTube as a Resource for Surgical Education-Clarity or Confusion. Eur Urol focus. 2020;6(3):445–9. 10.1016/j.euf.2019.09.017. [DOI] [PubMed] [Google Scholar]
- 5.Kirkpatrick CE, Lawrie LL. TikTok as a Source of Health Information and Misinformation for Young Women in the United States: Survey Study. JMIR infodemiology. 2024;4:e54663. 10.2196/54663. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Moorhead SA, Hazlett DE, Harrison L, Carroll JK, Irwin A, Hoving C. A new dimension of health care: systematic review of the uses, benefits, and limitations of social media for health communication. J Med Internet Res. 2013;15(4):e85. 10.2196/jmir.1933. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Zuckerberg M, Thompson B. An Interview with Meta CEO Mark Zuckerberg About AI and the Evolution of Social Media. Stratechery. 2025.
- 8.Saglam K, Kayaalp C, Aktas A, Sumer F. Educational Video Addition to the Bariatric Surgery Informed Consent Process: a Randomized Controlled Trial. Obes Surg. 2020;30(7):2693–9. 10.1007/s11695-020-04552-x. [DOI] [PubMed] [Google Scholar]
- 9.Yeung AWK. Evaluation of Content Quality of Online Health Information by Global Quality Score: A Case Study of Researchers Misnaming It and Citing Secondary Sources. Publications. 2025;13(2):23. [Google Scholar]
- 10.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 Commun Health. 1999;53(2):105–11. 10.1136/jech.53.2.105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ahmed SK, Mohammed RA, Nashwan AJ, Ibrahim RH, Abdalla AQ, Ameen M. Using thematic analysis in qualitative research. J Med Surg Public Health. 2025;6:100198. 10.1016/j.glmedi.2025.100198. [DOI] [Google Scholar]
- 12.Chan BKC. Data Analysis Using R Programming. Biostatistics for Human Genetic Epidemiology. Cham: Springer International Publishing; 2018. pp. 47–122. [Google Scholar]
- 13.Yin LK, Lieng TC, Cheong WK. How To Analyse Your Research Data? Illustrations With Hands-On Exercises Using Spss. Malaysian family physician: official J Acad Family Physicians Malaysia. 2006;1(2–3):77–81. [PMC free article] [PubMed] [Google Scholar]
- 14.Paul B, Headley-Johnson SA. The Impact of Social Media on Health Behaviors, a Systematic Review. Healthc (Basel Switzerland). 2025;13(21). 10.3390/healthcare13212763. [DOI] [PMC free article] [PubMed]
- 15.Chirumamilla S, Gulati M. Patient Education and Engagement through Social Media. Curr Cardiol Rev. 2021;17(2):137–43. 10.2174/1573403x15666191120115107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Tellis G, Macinnis D, Tirunillai S, Zhang Y. What Drives Virality (Sharing) of Online Digital Content? The Critical Role of Information, Emotion, and Brand Prominence. J Mark. 2019;83:1–20. 10.1177/0022242919841034. [DOI] [Google Scholar]
- 17.Jiang L, Wang S, Liu X, Gan L, Wu Y, Wang Y. Assessment of content quality and reliability of short videos regarding myocardial infarction on TikTok and BiliBili: a cross-sectional study. Sci Rep. 2026;16(1):5053. 10.1038/s41598-026-35441-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.de Barra M. Reporting bias inflates the reputation of medical treatments: A comparison of outcomes in clinical trials and online product reviews. Soc Sci Med. 2017;177:248–55. 10.1016/j.socscimed.2017.01.033. [DOI] [PubMed] [Google Scholar]
- 19.Montori V, Ruissen M, Hargraves I, Brito J, Kunneman M. Shared decision-making as a method of care. BMJ Evidence-Based Med. 2022;28:213–7. 10.1136/bmjebm-2022-112068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Frcp FHMMMF. Shared Decision-Making in Patient Care: Advantages, Barriers and Potential Solutions. J Brown Hosp Med. 2024;3:13–5. 10.56305/001c.122787. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Keshavarz H, Wang T, Vardell E, Saeidnia H. Challenges that health professionals face to evaluate and trust online health information: the role of conscientiousness. Infosci Trends. 2024. 10.61186/ist.202401.01.05. [DOI] [Google Scholar]
- 22.Daraz L, Morrow A, Ponce O, Beuschel B, Farah M, Katabi A, et al. Can Patients Trust Online Health Information? A Meta-narrative Systematic Review Addressing the Quality of Health Information on the Internet. J Gen Intern Med. 2019;34:1884–91. 10.1007/s11606-019-05109-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Downing A, Perakslis E. Health advertising on Facebook: Privacy and policy considerations. Patterns. 2022;3. 10.1016/j.patter.2022.100561. [DOI] [PMC free article] [PubMed]
- 24.Luo C, Hasan NAM, Zamri bin Ahmad AM, Lei G. Influence of short video content on consumers purchase intentions on social media platforms with trust as a mediator. Sci Rep. 2025;15(1):16605. 10.1038/s41598-025-94994-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hudon A, Patterson É, Dagenais M, Cadeau S, Baillargeon C, Lam M, et al. Emotional and Psychosocial Correlates of Problematic Social Media Use Among Adults: Cross-Sectional Study. JMIR formative Res. 2026;10:e82098. 10.2196/82098. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.El-Masry H, Mahmoud MH, Mohamed BS, Elraggal D, Elkholy MH, Abokhozima A. The Nutritional Challenges Following Revisional Bariatric Surgery After Sleeve Gastrectomy: A Systematic Review and Meta Analysis. Obes Surg. 2025;35(12):5515–30. 10.1007/s11695-025-08325-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Scarano Pereira JP, Martinino A, Manicone F, Scarano Pereira ML, Iglesias Puzas Á, Pouwels S, et al. Bariatric surgery on social media: A cross-sectional study. Obes Res Clin Pract. 2022;16(2):158–62. 10.1016/j.orcp.2022.02.005. [DOI] [PubMed] [Google Scholar]
- 28.Rukavina V, Viskić J, Poplašen M, Relić D, Marelić M, Jokic D, et al. Dangers and Benefits of Social Media on E-Professionalism of Health Care Professionals: Scoping Review. J Med Internet Res. 2020;23. 10.2196/25770. [DOI] [PMC free article] [PubMed]
- 29.Foula MS, Bamalan OA, Foula A, Eldamati AM, Alratrout HA, AlQatari AA, et al. The Evolution of Bariatric and Metabolic Surgery in the Artificial Intelligence Era: A Comprehensive Systematic Review of Current Applications and Clinical Implications. Obes Surg. 2026. 10.1007/s11695-026-08908-7. [DOI] [PubMed] [Google Scholar]
- 30.El-Masry H, El-mezayen MY, Farouk B, Tawfik AM, Sharsher PS, Mohamed BH, et al. Performance of Large Language Models in Metabolic Bariatric Surgery: a Comparative Study. Obes Surg. 2026;36(2):538–45. 10.1007/s11695-025-08418-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Chong PL, Vaigeshwari V, Mohammed Reyasudin BK, Noor Hidayah BRA, Tatchanaamoorti P, Yeow JA, et al. Integrating artificial intelligence in healthcare: applications, challenges, and future directions. Future Sci OA. 2025;11(1):2527505. 10.1080/20565623.2025.2527505. [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
No datasets were generated or analysed during the current study.