Poor Information Quality of Abdominal Aortic Aneurysm Videos on Chinese Social Media: A Cross-Sectional Study

Abstract

Abdominal aortic aneurysm (AAA) is a severe vascular disease. Given the high dependency of Chinese users on domestic social media platforms, this study aimed to evaluate the content completeness and information quality of AAA related videos on China’s dominant social media. We searched for the keyword “abdominal aortic aneurysm” on the Chinese social media TikTok and Bilibili. The top 100 search results for each platform were collected based on the default sorting. The video content was evaluated for completeness and information quality using abdominal aortic aneurysm specific score (AAASS), modified DISCERN tool (mDISCERN), and Global Quality Scale (GQS). The study included 140 social media videos. Among all videos, median scores (IQR) for AAASS, mDISCERN, and GQS were 3.00 (2.00-4.00), 2.00 (1.00-3.00), and 3.00 (2.00-3.00). Analysis of correlations between video duration and engagement metrics revealed a weak positive correlation between “shares” and duration (r = .221, P = .009). Associations between engagement metrics and quality scales showed only “shares” correlated with GQS (r = .216, P = .011), while video duration demonstrated positive correlations with AAASS (r = .211, P = .012) and GQS (r = .234, P = .005). In various uploader identities, radiologists had significantly lower GQS scores than vascular surgeons, medical institutions, and cardiothoracic surgeons (P < .01). This study shows that videos about abdominal aortic aneurysm on Chinese social media exhibited poor content completeness and information quality. This highlights an urgent need for platforms to improve quality control algorithms and for medical professionals to be guided in producing more comprehensive, evidence-based content.

Abdominal aortic aneurysm-related videos on Chinese social media exhibit poor content completeness and low informational quality.

Video engagement metrics showed limited correlation with content accuracy.

The study calls for platforms to prioritize high-quality content via algorithms and urges professionals to create evidence based videos integrating patient-centered narratives.

Introduction

Abdominal aortic aneurysm (AAA) is a life-threatening disease characterized by pathological dilation of the abdominal aorta diameter to ≥3 cm or ≥1.5 times the normal arterial diameter, and it most commonly occur in the infrarenal abdominal aorta where a study indicates the mean normal diameter is 16.0-24.0 mm in males and 14.3-22.0 mm in females.^1,2 A systematic review of Asian populations found the prevalence in general populations to be approximately 0.61%. ³ The primary clinical risk of AAA lies in progressive aneurysm expansion and potential rupture leading to fatal hemorrhage, resulting 150 000-200 000 AAA rupture related deaths around the world annually. ⁴ AAA management strategies include conservative approaches and surgical interventions, notably open surgical repair (OR) and endovascular aneurysm repair (EVAR).^1,5-7 Due to the complexity of AAA progression and individual differences, patients often demonstrate insufficient understanding of treatment choices and the necessity of long-term surveillance. Effective patient education is crucial for early detection, diagnosis, and treatment, thereby improving prognosis. ⁸ However, in China, particularly in remote regions, unbalanced development in economic and vascular surgery infrastructure frequently hinder access to accurate diagnosis and standardized management.

Today, the rapid evolution of internet technologies and digital devices has promoted the video based social media as indispensable tools in daily life. These platforms hold unique advantages in disseminating health information, as audiovisual content is generally more comprehensible and memorable than text-based formats.^9-11 Recognizing this potential, clinicians increasingly utilize video platforms for medical education, with growing numbers of certified healthcare professionals and institutions creating science communication content on social media.^12-14 Concurrently, patients increasingly prefer short video platforms to clinic consultations for health knowledge acquisition, highlighting the significance in enhancing health literacy and public medical education. ¹² Thus, AAA related content propagation via these platforms may empower patients to self-assess symptoms and seek timely, standardized health care.

TikTok and Bilibili, 2 dominant Chinese video platforms with over 1 billion users, host vibrant health education communities. ¹⁵ Their dedicated health channels attract numerous certified medical institutions and clinician accounts producing health content. However, such content remains non-peer-reviewed, raising concerns about quality, reliability, and accuracy. Access to high quality health information is one of the key determinants of health. ¹⁶ The quality of health information can be assessed across multiple dimensions. Frameworks such as the 13 dimension Clinical Information Quality (CLIQ) framework for digital health developed in previous studies and the comprehensive Health Information Quality (HIQ) theoretical framework, which integrates various assessment tools and standards, highlight both the importance and multifaceted nature of evaluating Internet-based health information quality.^17,18 Furthermore, developing digital health literacy helps leverage the potential of digital media to support health and well-being, as well as mitigate or eliminate the impact of misinformation and disinformation. ¹⁹ Recent studies have evaluated the quality of videos on platforms addressing breast cancer, liver cancer, thyroid nodules, gastroesophageal reflux disease, depression, and other diseases.^20-24 However, no prior research has systematically analyzed AAA related video content on Chinese video social media. While previous studies have evaluated AAA information on platforms like YouTube or other health topics on Chinese platforms, these findings have limited applicability to the mainland Chinese context. Due to legal restrictions and the cultural uniqueness of the Chinese internet community, the vast majority of Chinese users are unable to access information from YouTube, demonstrating a significant dependency on domestic platforms. Therefore, a significant knowledge gap exists regarding the specific information quality, completeness, and reliability of AAA content on China’s unique and dominant video based social media. This study aims to quantitatively assess this research gap and provide empirical evidence for health communication strategies in the Chinese context.

Methods

Data Source and Ethical Considerations

This study was conducted as a cross-sectional analysis of video content related to AAA. The study setting comprised 2 major Chinese social media platforms: TikTok (Chinese version) and Bilibili, which were selected based on their high user engagement and significant volume of health-related content in China. Our analysis was based on publicly available data. The study did not involve the reproduction or republication of the copyrighted video content. Instead, we analyzed the characteristics and quality of the videos, which aligns with fair use principles for academic research. All data reported are aggregated, and no personally identifiable or specific video-level information is disclosed in the manuscript.

Search Strategy

To avoid the influence of personalized recommendation algorithms on search results, we cleared the cache and browsing history on the TikTok and Bilibili applications and registered a new account specifically for data retrieval. The keyword “腹主动脉瘤” (AAA in Chinese) was used for the searches. We extracted the top 100 videos from each platform based on platform default algorithmic sorting. The sample size is comparable to those reported in previous, similar studies in this field. Given the rapid updates on online contents, all video retrievals were completed and recorded on February 10, 2025. Exclusion criteria included: (1) irrelevant content (non-AAA topics, advertisements, commercial promotions, or professional biomedical lectures); (2) non-Chinese videos; (3) non-informative videos (static images without narration or text); (4) duplicate videos.

Video Features and Information Quality Assessment

The basic characteristics of the videos included uploader identity, video duration, upload date, and engagement metrics (likes, comments, and shares). Video length, upload date, and engagement metrics were recorded at the time of retrieval. Uploaders were categorized as vascular surgeons, cardiothoracic surgeons, radiologists, general physicians, or non-practitioners including medical institutions, and uncertificated bloggers.

Video content was evaluated using the AAA-specific score (AAASS, Supplemental Table S1). ¹⁴ AAASS was modified by Radonjic et al in their evaluation of AAA video content quality on YouTube from the primary checklist used by Goldberg et al ²⁵ and Knops et al ²⁶ and covers diagnosis, treatment and postoperative management of AAA. To ensure the applicability of the AAASS to the Chinese social media context, the AAASS checklist was first reviewed and discussed by the 2 senior vascular surgeon authors to confirm its content validity covering all critical domains. 1 point was assigned to each item present in the checklist, with total scores categorized as: exceptional (17-20), very useful (13-16), moderately useful (9-12), poor (5-8), or very poor (0-4).

The modified DISCERN tool (mDISCERN, Supplemental Table S2)^12,15,27 and Global Quality Scale (GQS, Supplemental Table S3)^22,24 were employed to assess reliability and overall quality. The mDISCERN, evaluates clarity, reliability, bias, reference supplementation, and areas of uncertainty for video-based health information. ²⁸ The GQS, a 5-point Likert scale, is commonly used to assess the quality of health information on online video, with 1 point for poor quality and 5 for excellent quality. ²⁹

Following the inclusion and exclusion screening process, a pilot test was conducted. The 2 independent vascular surgeons scored 20 videos using a standardized checklist. Any discrepancies between the reviewers were resolved through comparing scores, discussing discrepancies, and refining their interpretation of the scoring criteria until a consensus was achieved. After this calibration, the same 2 reviewers independently reviewed all videos extracted from the 2 platforms.

Statistical Analysis

Data analysis and visualization were performed using R software (v4.3.2) and GraphPad Prism (v10.0). Interrater reliability between reviewers was calculated by intraclass correlation coefficient (ICC) estimates and its 95% confidence intervals (CI), consistency, 2-way random model. Normally distributed data are presented as mean and standard deviation (SD), while non-normally distributed data expressed as median and the interquartile range (IQR). Group comparisons utilized Mann-Whitney U tests for whether video quality and engagement differed between the 2 platforms, and Kruskal-Wallis H tests for whether the uploader’s professional identity significantly impacted the information quality. A P < .05 was considered statistically significant.

Result

An initial sample of top 100 videos per platform was extracted from TikTok and Bilibili (total N = 200). Videos that were non-Chinese, duplicated, without substantial content, or irrelevant to the study were excluded. A total of 140 videos were included in the study (Figure 1). Interrater reliability assessments demonstrated excellent agreement across all evaluation scales: AAASS (ICC = 0.88, 95% CI 0.84-0.92), mDISCERN (ICC = 0.93, 95% CI 0.90-0.95), and GQS (ICC = 0.89, 95% CI 0.85-0.92). Any discrepancies between reviewers were resolved through discussion to generate a final consensus score for subsequent analysis.

Figure 1.

Flowchart of the selection of videos for analysis.

General Features of Videos

The study included videos uploaded between 14 August 2020 and 9 February 2025. Video metrics demonstrated a median duration of 73 s (IQR 45.00-121.25), with median engagement counts of 50 likes (IQR 13.25-157.50), 4 comments (IQR 1.00-12.75), and 7 shares (IQR 2.00-28.50). In the 140 videos, platform-certified medical professionals constituted the primary uploader group: vascular surgeons contributed 95 videos (67.86%), followed by radiologists (13 videos, 9.29%), cardiothoracic surgeons (11 videos, 7.86%), and general physicians (4 videos, 2.86%). A smaller proportion of videos were uploaded by medical institutions (8 videos, 5.71%) and uncertified bloggers (9 videos, 6.43%). This highlights that the most relevant specialists, vascular surgeons, were the dominant content creators, responsible for over two-thirds of the AAA videos analyzed. AAASS scoring revealed that 101 videos (72.14%) mentioned AAA diagnosis, 94 videos (67.14%) discussed treatment topics, and 25 videos (17.86%) covered postoperative management. While diagnosis and treatment were commonly discussed, postoperative management, a critical component of long-term AAA care, was markedly underrepresented in less than one-fifth of the videos. Detailed video characteristics are presented in Table 1.

Table 1.

General Features of Videos From Different Sources.

Sources	Videos, n (%)	Duration (s), median (IQR)	Likes, median (IQR)	Comments, median (IQR)	Shares, median (IQR)
Total	140 (100)	73.00 (45.00-121.25)	50.00 (13.25-157.50)	4.00 (1.00-12.75)	7.00 (2.00-28.50)
Platforms
TikTok	86 (61.43)	65.00 (41.75-96.00)	97.50 (43.75-207.75)	6.50 (3.00-16.00)	8.00 (2.75-34.25)
Bilibili	54 (38.57)	87.50 (52.25-177.25)	7.00 (1.00-45.25)	1.00 (0.00-3.25)	5.00 (0.00-13.00)
Uploader identities
Vascular surgeons	95 (67.86)	71.00 (44.00-104.00)	55.00 (12.00-160.00)	5.00 (1.00-10.00)	6.00 (2.00-20.00)
Cardiothoracic surgeons	11 (7.86)	60.00 (45.00-122.00)	155.00 (88.00-444.00)	16.00 (5.00-45.00)	30.00 (2.00-174.00)
Radiologists	13 (9.29)	45.00 (18.00-144.50)	25.00 (14.50-121.50)	3.00 (0.00-12.00)	4.00 (2.00-21.00)
General physicians	4 (2.86)	358.50 (130.50-361.75)	9.00 (3.25-13502.75)	1.50 (1.00-665.00)	12.50 (9.00-1105.00)
Medical institutions	8 (5.71)	79.00 (63.75-127.00)	20.00 (1.00-38.00)	1.00 (0.00-3.25)	15.00 (0.50-70.50)
Uncertificated bloggers	9 (6.43)	100.00 (66.50-229.50)	44.00 (31.00-171.00)	6.00 (0.50-22.50)	10.00 (1.50-69.00)

Platform analysis indicated significant differences. Bilibili contained significantly longer videos than TikTok (P = .0024), while TikTok videos demonstrated superior engagement metrics, with significantly higher median likes (P < .0001), comments (P < .0001), and shares (P = .0143). Pearson correlation coefficient analysis between video duration and engagement metrics identified a weak positive association only for shares (r = .221, P = .009). This finding suggests that user engagement metrics are poor indicators of overall video quality.

Platform-Based Video Quality Analysis

We analyzed video quality based on the platform (Figure 2). Among the 140 videos, 1 Bilibili video attained the highest AAASS score (14/16) and GQS score (5/5). The highest modified mDISCERN score of 5/5 was achieved by 5 videos. Median quality scores across all content were AAASS: 3.00 (IQR 2.00-4.00), mDISCERN: 2.00 (IQR 1.00-3.00), and GQS: 3.00 (IQR 2.00-3.00), corresponding to classifications of very poor, low reliability, and moderate quality, respectively (Table 2).

Figure 2.

AAASS, mDISCERN and GQS scores between TikTok and Bilibili. P-value was calculated by Mann-Whitney U tests.

Note. ns = not significant.

Table 2.

AAASS, mDISCERN and GQS Scores of Videos From TikTok and Bilibili.

Scales a	Total	TikTok	Bilibili	P-value b
AAASS	3.00 (2.00-4.00)	3.00 (2.00-5.00)	2.50 (1.00-4.00)	0.185
mDISCERN	2.00 (1.00-3.00)	2.00 (1.00-3.00)	2.00 (1.00-3.00)	0.612
GQS	3.00 (2.00-3.00)	3.00 (2.00-3.00)	3.00 (2.00-3.00)	0.437

^aThe scoring results are represented by the median (IQR).

^bCalculated from the comparison between TikTok and Bilibili.

Recognizing the potential variation in user profiles and content focus across Tiktok and BiliBili, we utilized the independent sample Mann-Whitney U test to compare the quality differences between videos on TikTok and Bilibili. Although the median AAASS score for TikTok videos was higher than that for Bilibili videos (3 vs 2.5), no statistically significant differences were observed between the 2 platforms across all 3 assessment scales. The correlation between engagement metrics and the quality assessment scores were further investigated, showing a positive correlation between shares and GQS (r = .216, P = .011). From a patient education perspective, this means both platforms are equally unreliable, with neither offering a demonstrably superior source of quality information.

AAASS scale was used to evaluate the quality of AAA-related content on TikTok and Bilibili, revealing significant omissions and content imbalance on both platforms. (Figure 3). In the diagnosis part, both platforms provided limited accurate definitions of AAA, with Bilibili slightly outperforming TikTok (18.52% vs 10.47%). Also, Bilibili showed superior coverage of AAA risk factors (27.78% vs 15.12%). Nearly half of the videos on both platforms discussed the risks of leaving AAA untreated. Both platforms focused heavily on treatment topics with a severe content bias, particularly EVAR (60.47% for TikTok vs 40.74% for Bilibili), while OR was mentioned less frequently (25.58% for TikTok vs 24.07% for Bilibili). Moreover, information essential for surgical decision making was almost entirely absent. Fewer than 20% of videos on each platform addressed conservative management of AAA or provided information regarding candidate eligibility for OR and EVAR. Postoperative management received limited attention, with TikTok having a slightly higher proportion than Bilibili. Notably, neither platform included content addressing 30-day mortality rates associated with either surgical approach, forming a complete information vacuum that is a critical data point for patient consent.

Figure 3.

Analysis of AAA-specific content comprehensiveness on TikTok and Bilibili. (A) Content comprehensiveness related to diagnosis of AAA on TikTok and Bilibili. (B) Content comprehensiveness related to treatment of AAA on TikTok and Bilibili. (C) Content comprehensiveness related to postoperative management of AAA on TikTok and Bilibili.

Uploader-Based Video Quality Analysis

We examined whether the uploader’s professional identity, serving as an indicator of professional expertise, significantly impacted information quality. Further analysis of uploader-based quality disparities revealed distinct patterns (Figure 4). While the data showed the potential value of expertise, that the maximum AAASS (14/20) and GQS (5/5) scores were both achieved by a single vascular surgeon-uploaded video. But the top quality content was not exclusive to experts. The secondary AAASS (13/20) was attained by 1 video each from a cardiothoracic surgeon and an uncertified blogger. A total of 5 videos received the highest mDISCERN score (5/5), with 3 uploaded by vascular surgeons, and 1 each by a cardiothoracic surgeon and an uncertified blogger.

Figure 4.

Evaluation of AAASS, mDISCERN and GQS scores among different uploader identities.

The median AAASS scores for all uploader categories were considered as very poor. For the mDISCERN scale, the median scores of vascular surgeons, cardiothoracic surgeons, and uncertified bloggers were rated as less reliable, while those of radiologists, general physicians, and medical institutions were divided as unreliable. On the GQS scale, the median scores of vascular surgeons, cardiothoracic surgeons, medical institutions, and uncertified bloggers were rated as moderate quality, whereas those of radiologists and general physicians were rated as generally poor quality. It suggests that while a patient might find an excellent video from a surgeon, the typical video they mostly encounter is of very poor completeness and low reliability.

Kruskal-Wallis testing identified significant intergroup differences in GQS scores (P < .01), with radiologists underperforming vascular surgeons, medical institutions, and cardiothoracic surgeons. The lower GQS scores among radiologists may reflect their focus on diagnostic imaging rather than comprehensive patient education. No other statistically significant differences were observed across the remaining scales among the groups (Table 3).

Table 3.

AAASS, mDISCERN and GQS Scores of Videos Among Different Uploader Identities.

Identities	AAASS a	mDISCERN	GQS
Vascular surgeons	3.00 (2.00-4.00)	2.00 (1.00-3.00)	3.00 (2.00-3.00)
Cardiothoracic surgeons	4.00 (2.00-6.00)	2.00 (1.00-2.00)	3.00 (2.00-3.00)
Radiologists	2.00 (1.00-5.00)	1.00 (1.00-2.00)	2.00 (1.00-3.00)
General physicians	1.50 (1.00-4.25)	0.25 (1.00-1.75)	2.00 (1.25-2.75)
Medical institutions	2.00 (1.00-6.00)	1.00 (1.50-2.75)	3.00 (3.00-3.00)
Uncertificated bloggers	4.00 (1.00-8.00)	2.00 (0.50-3.00)	3.00 (1.50-3.00)
P-value	.6016	.1450	.0178*

^aThe scoring results are represented by the median (IQR).

^*P < .05.

We further analyzed the sub-item scores of the AAASS scale across different uploader identities (Figure 5). A major content gap was the systemic neglect of postoperative management. Overall, the descriptions of AAA provided by uploaders were incomplete, with most focusing on AAA diagnosis (72.14%) and treatment (67.14%), while postoperative management received significantly less attention (17.86%). Merely 13 videos (9.29%) comprehensively addressed all 3 domains, predominantly from vascular surgeons (10/13). Vascular surgeons provided the most comprehensive coverage of AAA related content, followed by cardiothoracic surgeons and radiologists. In contrast, general physicians performed the poorest, limiting their discussions to diagnosis only. Regarding diagnosis, most uploaders focused on symptoms and risks of untreated AAA, with uncertified bloggers surprisingly providing the most comprehensive diagnostic descriptions. Treatment discussions demonstrated critical imbalances that conservative management was notably neglected by all uploaders. Concerning surgical approaches, cardiothoracic surgeons and uncertified bloggers provided relatively balanced descriptions of both OR and EVAR, whereas other uploaders focused predominantly on EVAR. Postoperative care emerged as the most underserved domain with only 1 video, uploaded by a vascular surgeon addressing the 30-day mortality rates associated with OR and EVAR. This again suggests that while vascular surgeons performing the best in all uploaders, comprehensive videos were still the rare exception.

Figure 5.

AAA specific content comprehensiveness among different uploader identities.

Discussion

AAA is a severe chronic vascular disease, while ruptured AAA is a critical and rapidly progressing condition, with a mortality rate exceeding 60% and a high rate of misdiagnosis in emergency. ³⁰ However, there are currently limited drugs available to control the continued expansion of AAA or prevent its rupture.^31-34 Despite its severity, public awareness and understanding of AAA remain extremely inadequate. ³⁵ Given the widespread use of social media such as TikTok and Bilibili, an increasing number of patients are turning to these platforms for health-related information, making them potential tools for effective health communication.^20,22 Therefore, this study aimed to evaluate the completeness and reliability of AAA related content on TikTok and Bilibili. Our findings reveal that the completeness and reliability of AAA related videos are generally poor across platforms and uploader identities, making it difficult for patients to access accurate and comprehensive information on diagnosis, treatment, and postoperative management. These results are consistent with analysis of AAA videos on YouTube conducted by Radonjic et al, suggesting that low-quality health information is a global problem.

This study further validates the complex relationship between video engagement metrics and content quality in digital health communication.^21,36 Previous research has shown that user interactions on social media platforms are often uncorrelated or even negatively correlated with the completeness and reliability of medical information.^14,37 In this study, only shares showed a weak positive correlation with GQS scores, suggesting that patients may be more inclined to share content they perceive as “useful” but not necessarily “comprehensive.” Moreover, we found that AAASS and GQS scores were significantly positively correlated with video duration, indicating that high-quality videos often require longer durations to systematically explain technical terms and evidence-based content. However, the complex information and long duration of such content may reduce its appeal to general audiences.^38,39 We also found that videos with high engagement often focus on dramatic case stories, which are compelling but lack substantive scientific information. This finding has significant implications for health communication and patient understanding. It strongly suggests that platform algorithms that are engineered to prioritize high engagement content are ineffective at identifying and recommending high quality, reliable medical information. Patients searching for AAA knowledge are therefore more likely to be guided toward videos that are popular or feature drama rather than content that is accurate, comprehensive, and evidence-based. This mechanism is dangerous, as it makes the public susceptible to mistaking popularity for credibility, leading to a poor understanding of their condition and potentially flawed health decisions based on low quality information. For healthcare professionals and potential health video creators, it is essential to integrate evidence-based content with appropriate emotionally driven narratives to enhance the accessibility of health knowledge.

This study also assessed differences in content completeness and quality across platforms and uploader identities. Platform comparisons revealed demographic discrepancies across various platforms influencing content availability. Bilibili had fewer AAA related videos than TikTok, likely due to its younger user profile, whereas AAA primarily affects older adults.^40-42 Nevertheless, there were no significant differences in quality scores between the 2 platforms. Although TikTok surpassed Bilibili in covering postoperative management, patients still face challenges in obtaining valuable health information within a short search time. Therefore, we recommend that platforms with health channels strengthen their algorithms to prioritize and recommend high quality content. Concerning uploader identities, videos with the highest scores across all scales were typically uploaded by vascular surgeons, who also provided the most comprehensive coverage of AAA related topics. This illustrates the expertise of vascular surgeons as specialists in AAA and highlights the need to encourage them to produce more high-quality educational content to improve public understanding of AAA. This suggests that while top experts in vascular medicine are capable of producing excellent content, the majority of them either fail to translate their expertise into accessible, comprehensive educational material or may oversimplify content in pursuit of engagement.

Significantly, the vast majority of videos (93.57%) did not cite authoritative clinical guidelines or research to support their content, which may raise concerns about their reliability. Most concerning was the near-total absence of information critical for patient decision making. In terms of surgical approaches, most videos disproportionately emphasized EVAR, neglecting the importance of individualized treatment and potentially misleading patients about the value of traditional open repair. The low coverage of conservative management and surgical indications may lead patients to believe that “all AAAs require immediate intervention.” The underrepresentation of postoperative management also constitutes a critical gap in AAA care education. Regardless of surgical approaches, systematic follow-up monitoring and complication awareness form essential components of the comprehensive care, which was regretfully omitted in these AAA related videos. This educational deficit calls the urgent need for multidisciplinary collaboration to provide accessible information to patients that could finally facilitate the development of evidence-based protocols that address surveillance schedules, anticoagulation therapy, and long-term cardiovascular risk management.

The impact of these content deficiencies is amplified within the specific sociocultural and medical context of China. Many videos emphasized the risks of untreated AAA, often using dramatic phrases like “a ticking bomb” to attract attention, potentially deviating from patient education. Most dangerous is the presence of active misinformation, such as the 3 videos incorrectly stating that AAAs exceeding 3 cm (rather than 5.0/5.5 cm) require surgical intervention. In a healthcare system where patients already face limited consultation times, this practice is highly likely to induce patient anxiety. As noted in our introduction, China faces a relatively closed internet community, leading to high patient dependency on domestic platforms. But the entire community lacks a unified, standardized quality assessment system or regulatory mechanism. This limited regulatory not only permits the spread of the low-quality information, but also fosters potential conflicts of interest, when content bias is inevitable. This aligns strongly with our study’s findings of a severe bias toward EVAR which is a costly minimally invasive procedure, and the neglect of open repair and conservative management. In this context, such misinformation poses a direct public health risk. It threatens to mislead patients, foster distrust in evidence-based medical advice, and ultimately contribute to unnecessary anxiety and potential costly and harmful overtreatment behaviors.

However, this study has several limitations. First, we only analyzed Chinese-language AAA related videos, although translated content in other languages may also be accessible to patients. Second, our focus on TikTok and Bilibili, the most popular platforms in China, may have overlooked representative content on other platforms. Third, analyzing only the top 100 videos retrieved by the default algorithm might introduces a selection bias. This approach inherently excludes videos that are ranked lower by the algorithm but may possess higher or lower quality, thus limiting the generalizability of our findings to the entirety of AAA content on these platforms. Finally, as a cross-sectional study, our findings reflect the current state of online information, which is subject to rapid and continuous updates.

Conclusion

This study compared and evaluated the content completeness and information quality of AAA related videos on Chinese social platforms TikTok and Bilibili. The results show that these videos on Chinese social media exhibited poor content completeness and information quality. TikTok and Bilibili videos showed no significant differences in AAASS, mDISCERN, and GQS scale scores. Vascular surgeons provided the most comprehensive and highest-quality videos, though overall quality remained unsatisfactory. These findings emphasize that platforms must improve their recommendation algorithms to identify and prioritize high quality, reliable health content. Additionally, medical professionals and institutions must be guided not only to produce accurate content, but also to ensure it is comprehensive and engaging to truly meet the public’s health information needs.