Quality of ovarian cancer videos on TikTok,Rednote,and WeChat: A public health perspective

Abstract

Background

Ovarian cancer is one of the most common and lethal gynecological malignancies, with high mortality rates due to late-stage diagnoses and frequent recurrence. Despite its significant global health impact, public awareness of ovarian cancer remains low, contributing to delayed diagnosis and treatment. The rapid rise of short-video platforms, such as TikTok, Rednote, and WeChat, presents an opportunity to enhance public knowledge and early detection of ovarian cancer.

Method

This study analyzed ovarian cancer-related videos from TikTok, Rednote, and WeChat using two validated quality assessment tools: the Global Quality Scale (GQS) and the modified DISCERN (mDISCERN) tool. A total of 220 videos were examined for content quality, thematic coverage, and engagement metrics. The videos were categorized into themes such as epidemiology, etiology, symptoms, diagnosis, treatment, complications, and maintenance therapy. Additionally, user comments were analyzed to assess public sentiment toward the videos. The study also examined the correlation between video quality and audience engagement.

Results

Videos on TikTok demonstrated the highest quality in terms of accuracy and engagement, followed by Rednote, with WeChat videos showing the lowest quality. A significant gap was observed in the coverage of critical topics such as complications and maintenance therapy, with many videos providing only partial information. Specialist-created videos scored higher in quality compared to those created by non-specialists and individual users. Moreover, higher-quality videos were associated with greater audience engagement, including more likes, shares, comments, and collections. Positive sentiment in user comments was most strongly correlated with videos focusing on treatment and maintenance therapy.

Conclusion

This study reveals significant quality variation in ovarian cancer videos across Chinese platforms. To enhance public health communication, content accuracy must improve, with a focus on professional creators and key topics like complications and maintenance therapy.

Keywords

ovarian cancer short videos platforms information quality Rednote TikTok WeChat

1. Introduction

Ovarian cancer is one of the most prevalent and highly devastating malignant tumors among gynecological cancers.¹ Owing to the insidious early symptoms and the lack of specific screening methods,² its mortality rate ranks among the highest among gynecological malignancies.³ Statistics show that approximately 324,000 new cases of ovarian cancer¹ were diagnosed worldwide in 2022, with around 207,000 associated deaths.⁴ It is projected that the number of new global cases will increase to roughly 270,000 by 2025. In China, the incidence of ovarian cancer ranks third among gynecological malignancies,⁵ while its mortality rate tops the list of tumors affecting the female reproductive system.

Its hazards are further manifested in the dual challenges of a “high advanced-stage rate”⁶ and a “high recurrence rate”⁷: approximately 80% of patients are diagnosed at an advanced stage, with tumors having spread beyond the pelvic cavity. Among patients with advanced ovarian cancer who achieve remission after initial treatment, about 75% experience recurrence within two years, and the 5-year survival rate merely ranges from 10% to 40%,⁸ which severely threatens women’s life and health. Of greater concern is the generally low level of public awareness⁹ regarding ovarian cancer on a global scale. Most people lack essential health knowledge,¹⁰ such as the etiological factors of the disease, the characteristics of high-risk groups, and early warning signs (e.g., abdominal distension and dull pelvic pain). Consequently, some patients miss the optimal treatment window due to delayed medical consultation. Meanwhile, the insufficient dissemination of information related to the prevention, treatment, and follow-up of ovarian cancer has further exacerbated the disease burden.

With the rapid advancement of short - video platforms, they have now become a core channel for the public to obtain health - related knowledge¹¹ and hold enormous potential in improving public awareness of ovarian cancer and promoting early screening and diagnosis. TikTok, Rednote, and WeChat Channels boast a large user base in China¹¹ and feature substantial content concerning ovarian cancer. However, the credibility of the viewpoints presented in these contents remains questionable due to the diverse backgrounds of content creators.¹² Additionally, some of the content is fragmented, which hinders the public from forming a comprehensive understanding of ovarian cancer. As the three most mainstream short - video platforms in China, the dissemination of misinformation to the public through these platforms is likely to lead to inappropriate health - related decisions and ultimately result in adverse health outcomes.

Previous studies^13,14 have revealed that only a limited number of research efforts have focused on the quality of health - related short videos, and no specialized evaluation has been conducted targeting ovarian cancer, a disease with high mortality. Therefore, we intend to employ the GQS^15,16 and the mDISCERN^11,17 tool, both of which have been developed, to assess the quality of relevant short videos. We will collect data regarding the themes of the short videos, the characteristics and attitudes of the uploaders, and analyze the current video content to gauge public sentiment. Furthermore, viewers’ comments will be collected to evaluate the general public’s attitudes toward ovarian cancer.

Significance: Short-video platforms have emerged as the core channel for the public to access health knowledge regarding ovarian cancer. This study represents the first to concurrently evaluate ovarian cancer-related short videos across three major Chinese platforms—TikTok, Rednote, and WeChat —using validated tools, namely the Global Quality Scale (GQS) and modified DISCERN (mDISCERN). By correlating uploaders’ credentials with content quality¹⁸ and public sentiment,^15,17 the research yields three actionable practical insights: optimization directions for platform governance, intervention strategies for public health, and a framework paradigm for Mandarin-language health communication.

2. Materials and methods

2.1. Ethical considerations

The study was performed in accordance with the ethical guidelines of the Declaration of Helsinki. The data used in this study were sourced from publicly available video content on platforms such as Rednote, WeChat, and TikTok. These data are publicly accessible and do not involve any personal or private information. The use of all data strictly adheres to the respective platforms’ terms of service and complies with ethical standards for academic research. All keywords and content were in Simplified Chinese, reflecting mainland-Chinese linguistic norms.

In the present study, short-video data were collected from three major platforms: Rednote, WeChat, and TikTok. To ensure comprehensive content capture, the data collection process utilized a combination of scientific terminology and colloquial expressions as search keywords. Videos were retrieved in accordance with each platform’s default sorting mechanism, which guaranteed consistency throughout the data selection process. For each platform, the top 100 videos displayed under the default “comprehensive” ranking were collected following the entry of the keyword “卵巢癌” (ovarian cancer including its colloquial variants) in the search bar. This default ranking relies on a proprietary composite scoring model that incorporates multiple engagement metrics—primarily views, likes, shares, and recency—though the specific weighting methodology is not disclosed by the platforms.

Videos were collected according to the default ranking order generated by each platform following keyword searches. We acknowledge that platform algorithms typically prioritize content based on engagement metrics (e.g., views, likes, comments, and shares), which may favor more popular or highly interacted videos. This sampling strategy was intentionally adopted to reflect a user-exposure perspective, as most users rely on default search results and rarely navigate beyond the first few pages. Therefore, algorithm-ranked videos are more likely to be encountered and consumed by the general public. From a digital health communication standpoint, evaluating highly exposed content may better represent the information environment that users actually experience.

To enhance reproducibility, all searches were conducted in incognito mode (logged out), with cached data and cookies cleared and location services disabled to minimize personalization effects. Screenshots of search result pages and time-stamped browser logs were archived to ensure traceability.

For the purpose of ensuring data stability and relevance, videos classified as advertisements or those uploaded within 7 days prior to the search date were excluded. Advertisements were operationally defined as videos that explicitly promoted specific medical institutions, commercial products, or paid services, or that contained direct marketing elements such as pricing information, discount offers, contact details for appointments, or explicit calls to purchase products or services. These exclusions ensured that the analysis focused on informational and educational content related to ovarian cancer rather than promotional material.

Videos uploaded within 7 days before data collection were excluded because newly published content on short-video platforms often undergoes rapid fluctuations in visibility and engagement metrics (e.g., views and likes) during the initial algorithmic exposure phase. Excluding recently uploaded videos helped ensure that engagement indicators had reached a relatively stable level, thereby improving comparability across videos and reducing bias related to early algorithm-driven amplification.

Given that platform rankings may exhibit temporal fluctuations, complete result lists were captured within a defined 48-hour window (November 1–2, 2025) to maintain temporal consistency across platforms. This cross-sectional design was intended to minimize variability during data collection and enhance the reliability and robustness of the findings (Figure 1).

Figure 1.

Search strategy for videos on ovarian cancer screening.

All videos were captured within a 48-hour period to ensure temporal consistency across platforms. Given the dynamic nature of short-video platforms, where rankings and engagement metrics may change frequently, restricting data collection to a short and defined timeframe minimized variability caused by algorithm updates and new uploads sampling process. This study was designed as a cross-sectional evaluation intended to provide a snapshot assessment of the content quality and reliability at a specific point in time rather than to examine longitudinal trends.

2.2. Video characteristics

A range of video-related metrics was systematically documented, including counts of likes, comments, shares, and saves, as well as the follower numbers of the content creators.

2.3. Uploaders characteristics

The uploaders of the videos were categorized into the following groups: Specialists, Non-specialists, and Individual user. The specialists include those in obstetrics and gynecology and oncology. Non-specialists mainly include doctors in other department such as hepatobiliary traditional Chinese medicine doctors, surgery, and radiologists.

2.4. Video review and categorization

Two reviewers (Xuan Zhou and YunTao Hao) independently assessed all retrieved videos and removed those that were duplicated or unrelated to the study focus. The content of each video was then classified into thematic categories, including epidemiology, etiology, symptoms, diagnosis, treatment, complications, and maintenance therapy. Because individual videos often encompassed more than one topic, the number of themes addressed per video was also documented. Videos that did not fall into any of these predefined categories were considered irrelevant and subsequently excluded.

2.5. Comment towards the ovarian cancer

To examine the association between ovarian cancer video themes and positive comment proportions across platforms, we first categorized each included video into one or more of five pre-defined themes: Symptoms, Treatment, Diagnosis, Complications, and Maintenance treatment (as described in Section 2.4). For each platform (TikTok, Rednote, WeChat), we calculated the proportion of positive comments per video. Positive comments were operationally defined as content expressing explicit support, active learning interest, emotional encouragement, or affirmative feedback toward ovarian cancer knowledge dissemination (e.g., “This professional explanation helps me understand ovarian cancer symptoms”, “Thank you for sharing these practical treatment insights”). Neutral comments referred to objective factual inquiries without emotional tendency, and negative comments included content expressing doubt, criticism, or disagreement with the video content; only positive comment proportions were included in subsequent analyses.

Two independent reviewers (Xuan Zhou and YunTao Hao) conducted double-blind sentiment coding for all collected user comments (12,864 total). Prior to formal coding, the reviewers completed unified calibration training using 200 randomly selected comments to standardize the judgment criteria for positive/neutral/negative comments. Inter-rater reliability for sentiment coding was quantified using Cohen’s kappa coefficient (κ), with an excellent agreement achieved (κ= 0.88, P < 0.001). Any coding discrepancies were resolved through discussion and consensus with a third arbitrator (ChunHui Wang).

We then computed point-biserial correlation coefficients (r) to quantify the linear relationship between the presence of each theme (binary indicator: 1 = theme present, 0 = theme absent) and the positive comment proportion of videos on each platform. A heatmap was constructed to visualize these correlation coefficients, with color intensity and hue mapped to r values (blue = stronger positive correlations; orange = weaker correlations) for intuitive interpretation.

2.6. Videos assessments

The quality and reliability of the collected videos were systematically evaluated using two well-established assessment tools: the DISCERN instrument and the Global Quality Scale (GQS). The DISCERN tool, originally designed to assess the quality of written health information, was adapted for video content. It evaluates five key dimensions: clarity of purpose, relevance of information, accuracy of evidence, balance and bias, and overall reliability. Each dimension is scored on a 5-point Likert scale (1 = poor, 5 = excellent), with higher scores indicating higher quality(Table S2). The GQS provides a global assessment of video quality, also using a 5-point scale (1 = poor quality, 5 = excellent quality), focusing on the overall usefulness, clarity, and educational value of the content(Table S1).

Two independent reviewers (Xuan Zhou and YunTao Hao), both with expertise in public health research methodology and clinical ovarian cancer practice (dual professional background), conducted the evaluations. To ensure consistency and minimize bias, the reviewers underwent a training session to familiarize themselves with the scoring criteria and application of the tools, and then completed a pre-evaluation of 20 ovarian cancer-related videos for scoring criterion calibration, unifying the judgment scale for all evaluation dimensions. A third arbitrator (ChunHui Wang) assigned the final score if the two raters’ scores were inconsistent. Inter-rater reliability was assessed using Cohen’s kappa coefficient (κ), which measures the level of agreement between reviewers. A Cohen’s kappa value of 0.85 was achieved, indicating excellent inter-rater agreement. Any discrepancies in scoring were resolved through discussion and consensus. All evaluations were conducted item by item in accordance with the structured quantitative scoring rules of GQS and mDISCERN, without open subjective judgment, to further reduce the interpretation bias of the evaluators.

The evaluation results were interpreted based on the mean scores for each dimension of the DISCERN tool and the GQS. Videos with an average DISCERN score ≥ 4.0 and a GQS score ≥ 4.0 were classified as high-quality, while those with scores below 3.0 were considered low-quality. Scores between 3.0 and 3.9 indicated moderate quality. Previous studies have validated the aforementioned tools, particularly in the context of social media platforms for health information quality evaluation.^14,15,18

2.7. Statistical analysis

As the data did not follow a normal distribution, continuous variables were summarized using the median and interquartile range (IQR). Differences between two groups were examined with the Mann–Whitney U test, whereas comparisons across three or more groups were assessed using the Kruskal–Wallis H test. Agreement between the two independent reviewers was quantified with Cohen’s kappa statistic (κ), interpreted as follows: <0.20, poor; 0.21-0.40, fair; 0.41-0.60, moderate; 0.61-0.80, good; and >0.81, excellent. P values were adjusted using the Bonferroni method.

Spearman’s rank correlation was applied to explore associations between video-related variables (e.g., duration, likes, comments, shares, saves, and creator follower count) and video quality scores. Point-biserial correlation was used to analyze the relationship between binary video theme indicators and continuous positive comment proportions, as this approach aligns with the underlying distribution characteristics of the study variables. A two-tailed p-value <0.05 was considered statistically significant.

Effect size was evaluated using Cramer’s V, which ranges from 0 to 1, with the following thresholds: 0.00-0.10, weak; 0.10-0.30, moderate; 0.30-0.50, strong; and≥0.50, very strong. Spearman’s rank correlation was applied to explore associations between video-related variables (e.g., duration, likes, comments, shares, saves, and creator follower count) and video quality scores. A two-tailed p-value <0.05 was considered statistically significant.

All statistical procedures were carried out using IBM SPSS Statistics version 25, and visualizations were generated with GraphPad Prism version 9.

3. Results

3.1. Video characteristics

Our study included 65 WeChat videos, 72 from Rednote and 83 from TikTok, after excluding similar and irrelevant content (Figure 1). All videos on WeChat, Rednote and TikTok were in Chinese or had Chinese subtitles. Te characteristics of the videos from WeChat, Rednote and TikTok are detailed in Table 1. The analysis focused on several key metrics, including likes, collections, comments, shares, video duration, and followers. The median video duration varied significantly across platforms. WeChat has the least number of comments at 11 (9.75-197.25), significantly lower than Rednote [194 (32-1251.5)] and TikTok (n = 83)[7330.5 (28.5-12210.25)]. Compared with the other two platforms, TikTok has consistently demonstrated a higher level of engagement in terms of likes, collections, comments, shares, video duration and the number of followers.

Table 1.

Characteristics of videos about ovarian cancer on WeChat/Rednote/TikTok.

Variables	Total (n = 220)	WeChat (n = 65)	Rednote (n = 72)	TikTok (n = 83)	P	Cramer’s V	Post-hoctest
Likes, M (Q₁, Q₃)	53 (16-452)	103 (13.12-1184)	306.5 (98.5-4218)	631.5 (89.75-7280.3)	0.002	0.35	T>R>W
Collections, M (Q₁, Q₃)	56 (16.25-452)	106.5 (8.25-527)	133 (67-730.25)	579.5 (22.25-1961.5)	0.001	0.12	T>R>W
Comments, M (Q₁, Q₃)	43.5 (16-84.26)	11 (9.75-197..25)	194 (32-1251.5)	7330.5 (28.5-12210.25)	0.010	0.27	T>R>W
Shares, M (Q₁, Q₃)	59 (15-158.5)	87 (14-330)	264 (32-1285.3)	9380 (184.25-12630.5)	<0.001	0.32	T>R>W
Video duration(s), M (Q₁, Q₃)	77.5 (43-117.75)	36.47 (26-136.62)	68.5 (42.5-94.25)	85.85 (49.75-151.5)	0.046	0.23	T>W; R>W
Followers, M (Q₁, Q₃)	11.6 (0.88-98.32)	10.13 (1.26-16.75)	28.4 (13.75-135.3)	45.65 (11.75-534.4)	0.025	0.11	T>W; R>W

3.2. Video content

The distribution of ovarian cancer–related themes is summarized in Table 2. Over half of the videos provided complete information on epidemiology (52.73%), while 30.91% offered partial coverage. Etiology was frequently mentioned but often superficially, with 48.64% presenting partial details and 39.09% delivering full explanations.

Table 2.

Completeness of 220 ovarian cancer related-short videos content on WeChat/Rednote/TikTok.

Video content	Not involve(0 point)	Partial explanation(1 point)	Full explanation(2 point)
Epidemiology, n (%)	36 (16.36)	68 (30.91)	116 (52.73)
Etiology, n (%)	27 (12.27)	107 (48.64)	86 (39.09)
Symptoms, n (%)	3 (1.36)	206 (93.64)	11 (5.00)
Diagnosis, n (%)	5 (2.27)	184 (83.64)	31 (14.09)
Treatment, n (%)	13 (5.91)	167 (75.91)	40 (18.18)
Complications, n (%)	125 (56.82)	65 (29.55)	30 (13.64)
Maintenance treatmentn, n (%)	142 (64.55)	28 (12.73)	50 (22.73)

Symptoms were the most commonly addressed topic (98.64%); however, nearly all descriptions were partial (93.64%), with only 5.00% providing comprehensive information. Diagnostic content appeared in 97.73% of videos, though full explanations were limited to 14.09%. Treatment was mentioned in 95.09% of videos, but detailed therapeutic information was presented in only 18.18%.

In contrast, complications and maintenance therapy were substantially underreported. More than half of the videos omitted complications entirely (56.82%), and 64.55% did not address maintenance therapy. Only 13.64% and 22.73% of videos, respectively, provided complete information in these two domains.

3.3. Video quality assessment

First, inter-rater reliability for video ratings was assessed. Statistical analysis revealed a Cohen’s kappa coefficient of 0.67, indicating good inter-rater agreement between the two evaluators and confirming the analytical validity of the data. Video quality varied across platforms, as shown in Table 3 and Figure 2. TikTok consistently scored the highest, with median GQS scores for content accuracy and engagement quality both at 4 (range 1-5). Its mDISCERN scores for clarity of purpose and overall reliability were also 4 (range 1-5). In contrast, WeChat received the lowest scores, with a median GQS for content accuracy of 2 (range 1-5) and mDISCERN scores for clarity and reliability both at 2. Rednote scored between the two, with GQS scores for content accuracy and usefulness at 3 (range 1-3) and mDISCERN scores for clarity of purpose and overall reliability at 3 (range 1-5).

Table 3.

Quality assessment of videos about ovarian cancer on WeChat/Rednote/TikTok (with subscale scores).

Platform	GQS (Content accuracy)	GQS (Usefulness)	GQS (Engagement quality)	mDISCERN (Clarity of purpose)	mDISCERN (Relevance of information)	mDISCERN (Overall reliability)
WeChat	2 (1-5)	2-4	3	2 (1-5)	2-4	2
TikTok	4 (1-5)	2-4	4	4 (1-5)	2-4	4
Rednote	3 (1-3)	2-4	3	3 (1-5)	2-4	2

Figure 2.

Quality assessment of videos on ovarian cancer on three platforms.

Overall, TikTok outperformed both WeChat and Rednote in video quality, particularly in content accuracy and engagement, while WeChat had the lowest quality scores. The results suggest that videos created by specialists and non-specialists generally exhibited higher quality in terms of engagement and information reliability than those uploaded by individual users (Table 4).

Table 4.

Quality comparison between the videos uploaded by specialists/non-specialists/individual user.

Scores	Specialists (N=94)			Non-specialists (N=73)			Individual users (N=53)			P-value
	M	Min-Max	Q1-Q3	M	Min-Max	Q1-Q3	M	Min-Max	Q1-Q3
GQS	3	2-5	3.75, 5	3	2-5	2,4	3	1-5	2, 3	0.54
mDISCERN	3	1-5	2, 4	3	1-5	2,4	2	1-5	1, 3	0.25

3.4. Correlation analysis (Table 5)

Spearman correlation analysis demonstrated significant positive associations between video quality and multiple engagement metrics (Table 5). Both GQS and mDISCERN scores showed moderate correlations with likes, collections, comments, and shares. Specifically, GQS was correlated with likes (r = 0.234, p < 0.001), collections (r = 0.230, p < 0.01), comments (r = 0.282, p < 0.001), and shares (r = 0.210, p < 0.001). Similar patterns were observed for mDISCERN, which demonstrated even stronger associations with likes (r = 0.350, p < 0.001), comments (r = 0.320, p < 0.001), and shares (r = 0.270, p < 0.001). By contrast, video duration showed only weak or negligible correlations with both GQS (r = 0.150, p < 0.05) and mDISCERN (r = 0.013, p > 0.05), indicating that longer videos were not necessarily of higher quality. Engagement metrics (likes, collections, comments, shares) were strongly interrelated, suggesting that videos generating one form of audience response were likely to generate others. Overall, higher-quality videos—those with better accuracy, clarity, and reliability—tended to receive greater audience engagement, highlighting the potential impact of well-produced content on viewer interaction.

Table 5.

Spearman correlation between video quality and audience interaction on WeChat/Rednote/TikTok.

	GQS	mDISCERN	Likes	Collections	Comments	Shares	Video duration
GQS	1
mDISCERN	0.826***	1
likes	0.234***	0.35***	1
collections	0.23**	0.211***	0.734***	1
comments	0.282***	0.32***	0.81***	0.786***	1
shares	0.21***	0.27***	0.742***	0.839***	0.756***	1
Video duration	0.15*	0.013	0.007	0.01	0.054	0.021	1

3.5. Correlation between video themes and positive comments (Figure 3)

A heatmap was constructed to visualize point-biserial correlation coefficients between ovarian cancer video themes and positive comment proportions across three major social media platforms: TikTok, Rednote, and WeChat. Rows represent individual platforms, while columns correspond to five core video themes: Symptoms, Treatment, Diagnosis, Complications, and Maintenance Treatment. The color gradient (right panel) encodes the strength of correlation: darker blue tones indicate stronger positive correlations, whereas orange hues denote weaker associative relationships.

Figure 3.

Correlations between video themes and positive comment proportions across platforms (point-biserial coefficients). Heatmap of point-biserial correlation coefficients (r_pb). Rows represent the target platform, and columns represent ovarian cancer video themes. The color intensity and hue reflect the value of the correlation coefficient, as shown in the scale bar, with blue indicating strong correlations and orange indicating weak correlations.

The strength of the theme-positive comment association varied significantly across platforms. On TikTok, the Treatment theme exhibited the strongest correlation with positive user feedback (r = 0.8), followed by Maintenance Treatment (r = 0.72). This suggests that content centered on therapeutic topics resonated most profoundly with the audience on this platform. For Rednote, correlations were moderate across all themes, with no themes achieving a strong association threshold. In contrast, all theme-positive comment correlations on WeChat were weak, with the highest correlation observed for Symptoms (r = 0.4) and the lowest for Complications (r = 0.25), indicating that video themes exerted limited influence on positive engagement within this platform.

These findings highlight platform-specific patterns in user responsiveness to ovarian cancer-related content. TikTok users demonstrated the most positive engagement with treatment-focused material, whereas positive feedback from WeChat users was less tightly linked to thematic content. This inter-platform variation may reflect differences in user demographics, content consumption behaviors, or algorithmic prioritization of health information—insights that can inform targeted strategies for disseminating high-quality ovarian cancer educational content across social media channels.

4. Discussion

The rapid expansion of social media has profoundly reshaped public access to health information. Short-video platforms, in particular, have emerged as influential communication channels due to their visual appeal, rapid dissemination speed, and broad audience reach.¹⁹ Prior studies²⁰ have demonstrated that social media can enhance health literacy, promote preventive behaviors,²¹ and strengthen public engagement in disease awareness campaigns.²² Platforms such as TikTok, Rednote, and WeChat Channels have become primary sources of health information for Chinese users,^23,24 especially for younger audiences²⁵ who rely heavily on digital media for medical knowledge. However, the open nature of these platforms also increases the risk of disseminating low-quality or misleading information, which may negatively influence health-related decision-making. These findings suggest that ovarian cancer–related short videos achieve substantial reach and engagement across platforms. However, while engagement metrics indicate user exposure and interaction, they do not directly measure changes in health knowledge or behavior. Therefore, the findings should be interpreted as reflecting potential informational influence rather than confirmed public health impact.

This study analyzed 220 ovarian cancer–related short videos from TikTok, Rednote, and WeChat and revealed substantial variation in content completeness, quality, and reliability across platforms. Overall, epidemiology, etiology, symptoms, diagnosis, and treatment were frequently mentioned, whereas essential topics such as complications and maintenance therapy were notably underrepresented. These findings align with previous studies showing that while short videos often capture public interest, they tend to prioritize high-traffic topics at the expense of comprehensive medical education. Platform differences were significant: TikTok consistently produced higher-quality videos with superior GQS and mDISCERN scores, indicating greater content accuracy, clarity, and reliability. Rednote ranked moderately, while WeChat demonstrated the lowest overall quality. These results parallel earlier findings from cervical cancer^11,18 and lung cancer studies,¹⁴ where TikTok videos also outperformed other platforms in quality and engagement. Uploader identity was an important determinant of video quality. Videos created by specialists achieved the highest scores, whereas individual users produced the least reliable content. This pattern mirrors prior research across multiple cancer types,^26,27 emphasizing the pivotal role of professional contributors in improving the scientific rigor of online health information.

The GQS and mDISCERN evaluations showed that ovarian cancer–related videos displayed variable scientific accuracy and reliability. TikTok videos achieved the highest median scores across all quality dimensions, including content accuracy, usefulness, engagement quality, and overall reliability. In contrast, WeChat videos showed substantial quality deficiencies, with many lacking essential disease-related information. The identity of the content uploader emerged as a key determinant of quality. Specialists—primarily gynecologic oncologists and obstetricians—produced videos with significantly higher GQS and mDISCERN scores than non-specialists and individual users. This is consistent with findings in throat cancer short-video analyses,¹⁵ all of which reported that professional backgrounds are strongly associated with better-quality online health content. The predominance of low-quality content from individual users underscores the ongoing concern that non-expert sources may inadvertently spread misinformation or oversimplified narratives about complex diseases. The underrepresentation of complications and maintenance therapy further highlights a critical gap in the current digital landscape. These areas are essential for long-term disease management, yet often omitted due to their complexity or low perceived popularity. As seen in prior studies, oversimplification or omission of key health topics in short-video formats may contribute to fragmented public understanding and misaligned health expectations.²⁸

Our correlation analysis revealed that higher-quality videos received significantly greater user engagement, including likes, shares, collections, and comments. Both GQS and mDISCERN were moderately correlated with engagement metrics, suggesting that audiences tend to interact more with videos that provide clearer, more accurate, and more reliable information. These findings support the broader literature^29,30 indicating that high-quality medical content is more likely to gain public traction and contribute to effective health communication. In cervical cancer and other disease-related studies, similar positive associations between content quality and engagement have been observed. Interestingly, video duration had only weak correlation with quality scores, suggesting that longer videos are not inherently superior. This indicates that the effectiveness of short-video health communication depends more on content accuracy and clarity than on video length.

This study possesses several strengths. First, it is the first to systematically assess ovarian cancer–related short videos across three major Chinese platforms using validated tools (GQS and mDISCERN), filling an important research gap in cancer-related digital health communication. Second, the multi-platform design allows for cross-platform comparisons, offering insights into structural differences in content quality and dissemination mechanisms. Third, by analyzing uploader identity alongside content quality and user engagement, the study highlights the crucial contribution of medical professionals in improving digital health literacy. Practically, the findings carry significant implications for policymakers, platform regulators, and healthcare providers. Short-video platforms may consider implementing stricter authentication requirements for medical content, integrating medical fact-checking mechanisms, or prioritizing content from qualified professionals. Healthcare organizations may leverage short-video formats to deliver evidence-based information on ovarian cancer prevention, early detection, treatment modalities, and survivorship care. The study also establishes a transferable evaluation framework that can be applied to other diseases and languages.

In addition to overall engagement, the correlation analysis between video themes and the proportion of positive comments further highlighted platform-specific patterns. On TikTok, positive sentiment was most strongly associated with videos focusing on treatment and maintenance therapy, and to a lesser extent with complications, diagnosis and symptoms. This suggests that TikTok users particularly value concrete information on therapeutic options and long-term management, and are more likely to respond favorably when videos address actionable issues such as treatment decision-making and survivorship care. On Rednote, correlations between themes and positive comments were generally moderate, indicating a more diffuse pattern of audience interest across topics. By contrast, WeChat showed consistently weaker correlations, implying that thematic content on this platform is less predictive of positive audience feedback. Taken together, these findings underscore that not only video quality but also thematic focus and platform ecology jointly shape how viewers emotionally respond to ovarian cancer–related content, and that tailored communication strategies³¹ may be needed for different short-video platforms.

Several limitations should be acknowledged. First, videos were sampled based on platform default ranking algorithms rather than random sampling. Platform algorithms tend to prioritize highly engaged content, so our sample may overrepresent popular videos while underrepresenting less-viewed or newly uploaded content. This algorithm-driven sampling approach may limit the generalizability of the findings. Second, the cross-sectional design within a 48-hour window provides only a snapshot, and content rankings may shift over time due to algorithm updates and user engagement. Third, the analysis focused exclusively on Chinese-language videos from three platforms, which may limit generalizability to other cultural or linguistic contexts. Fourth, although GQS and mDISCERN are widely used, their application still involves some subjective judgment despite good inter-rater reliability. Fifth, this study did not systematically validate video content against established clinical guidelines for ovarian cancer; thus, subtle inaccuracies or deviations from evidence-based recommendations may not have been fully detected. Sixth, viewer demographic data (age, gender, educational level) were unavailable due to platform restrictions, limiting subgroup analyses across different population groups. Finally, engagement metrics (views, likes, comments) reflect user exposure but do not directly measure changes in health knowledge, attitudes, or behaviors; thus, conclusions regarding public health impact should be interpreted with caution.

Future studies may consider expanding sample sizes, adopting random sampling strategies, including additional platforms, and performing longitudinal or multilingual analyses to improve representativeness, generalizability, and temporal validity. To reduce subjective bias in content evaluation, future research can use blind evaluation by masking uploader identity, platform source, and engagement metrics, recruit multidisciplinary researchers, and apply consensus scoring; combining manual review with artificial intelligence–based natural language processing and machine learning to convert GQS and mDISCERN criteria into algorithmic models will further enhance objectivity and reproducibility.

It should be noted that engagement metrics such as views, likes, and comments reflect only user exposure and interaction, not actual improvements in health knowledge, attitudes, or behavioral outcomes. Future research is therefore needed to explore public perception, trust formation, and real-world behavioral impacts of ovarian cancer short-video education, rather than drawing conclusions about health effects solely from interactive data.

5. Conclusion

This study reveals significant quality variations in ovarian cancer-related videos across TikTok, Rednote, and WeChat. TikTok videos generally provided higher-quality content, while WeChat videos were less reliable. Specialist-created content was more accurate than that from non-specialists or individual users. Key topics like complications and maintenance therapy were underrepresented. The findings suggest the need for stricter content verification and improved quality control on short-video platforms. Enhancing the accuracy of health information on these platforms could improve public awareness and health outcomes for ovarian cancer.

Supplemental material

Supplemental material - Quality of ovarian cancer videos on TikTok, Rednote, and WeChat: A public health perspective

Supplemental material for Quality of ovarian cancer videos on TikTok, Rednote, and WeChat: A public health perspective by Xuan Zhou, YunTao Hao, ChunHui Wang, ShuYu Han, Xin Sun, YaQian Lin, and XiaoHua Wang in Digital Health.

Supplemental material

Supplemental material - Quality of ovarian cancer videos on TikTok, Rednote, and WeChat: A public health perspective

Footnotes

Acknowledgements

The authors would like to express their gratitude to the video uploaders for their contributions to public health.

ORCID iD

XiaoHua Wang

Author contributions

ZX conceived and designed the study. HT collected the top 300 videos. WH collected the characteristics of the videos and authors. HY was responsible for reviewing, classifying, and scoring the videos. SX and LQ analyzed the data. ZX wrote the original draft. HT reviewed and edited the manuscript. WH critically revised the manuscript for intellectual content. All the authors contributed to manuscript writing and editing and approved the fnal draft for submission.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Youth Science and Technology Research Projects of the Hebei Provincial Health Commission (NO. 20232701)

Declaration of conflicting interests

No datasets were generated or analysed during the current study.

Data Availability Statement

The data sets generated during and/or analyzed during this study are available from the corresponding author on reasonable request.*

Supplemental material

Supplemental material for this article is available online.

Appendix

References

Caruso

Weroha

Cliby

. Ovarian Cancer: A Review. Jama 2025; 334(14): 1278–1291. https://doi.org/10.1001/jama.2025.9495

Sideris

Menon

Manchanda

. Screening and prevention of ovarian cancer. Med J Aust 2024; 220(5): 264–274. https://doi.org/10.5694/mja2.52227

Smolarz

Biernacka

Łukasiewicz

, et al. Ovarian Cancer-Epidemiology, Classification, Pathogenesis, Treatment, and Estrogen Receptors' Molecular Backgrounds. Int J Mol Sci 2025; 26(10): 4611. https://doi.org/10.3390/ijms26104611

Loizzi

Cerbone

Arezzo

, et al. Contraception as chemoprevention of ovarian cancer in BRCA1 and BRCA2 women. Hormones (Athens) 2024; 23(2): 277–286. https://doi.org/10.1007/s42000-023-00519-6

Zhang

Lian

, et al. Global status and attributable risk factors of breast, cervical, ovarian, and uterine cancers from 1990 to 2021. J Hematol Oncol 2025; 18(1): 5. https://doi.org/10.1186/s13045-025-01660-y

Arora

Mullangi

Vadakekut

, et al. Epithelial Ovarian Cancer. StatPearls. Treasure Island (FL) ineligible companies. Disclosure: Sanjana Mullangi declares no relevant financial relationships with ineligible companies. Disclosure: Elsa Vadakekut declares no relevant financial relationships with ineligible companies. Disclosure: Manidhar Reddy Lekkala declares no relevant financial relationships with ineligible companies: StatPearls Publishing Copyright © 2025, StatPearls Publishing LLC.; 2025.

Baradács

Teutsch

Váradi

, et al. PARP inhibitor era in ovarian cancer treatment: a systematic review and meta-analysis of randomized controlled trials. J Ovarian Res 2024; 17(1): 53. https://doi.org/10.1186/s13048-024-01362-y

Monk

Barretina-Ginesta

Pothuri

, et al. Niraparib first-line maintenance therapy in patients with newly diagnosed advanced ovarian cancer: final overall survival results from the PRIMA/ENGOT-OV26/GOG-3012 trial. Ann Oncol 2024; 35(11): 981–992. https://doi.org/10.1016/j.annonc.2024.08.2241

Papalois

Fotopoulou

. Infrastructural and public health awareness gaps for the diagnosis and treatment of ovarian cancer: A literature review. Arch Gynecol Obstet 2024; 309(5): 1807–1813. https://doi.org/10.1007/s00404-024-07371-y

10.

Webb

Jordan

. Epidemiology of epithelial ovarian cancer. Best Pract Res Clin Obstet Gynaecol 2017; 41: 3–14. https://doi.org/10.1016/j.bpobgyn.2016.08.006

11.

Yao

, et al. Short video platforms as sources of health information about HPV vaccine: A content and quality analysis. Digit Health 2025; 11: 20552076251379340. https://doi.org/10.1177/20552076251379340

12.

Yuan

Wang

. Characteristics, hotspots, and prospects of short video research: A review of papers published in China from 2012 to 2022. Heliyon 2024; 10(3): e24885. https://doi.org/10.1016/j.heliyon.2024.e24885

13.

Sun

Zheng

. Quality of Information in Gallstone Disease Videos on TikTok: Cross-sectional Study. J Med Internet Res 2023; 25: e39162. https://doi.org/10.2196/39162

14.

Zhao

Yao

Sui

, et al. Current status of short video as a source of information on lung cancer: a cross-sectional content analysis study. Front Oncol 2024; 14: 1420976. https://doi.org/10.3389/fonc.2024.1420976

15.

Liu

Chen

Lin

, et al. YouTube/Bilibili/TikTok videos as sources of medical information on laryngeal carcinoma: cross-sectional content analysis study. BMC Public Health 2024; 24(1): 1594. https://doi.org/10.1186/s12889-024-19077-6

16.

Foster

Sarlashkar

Abraham

, et al. TikTok as a Source of Education and Misinformation in Lichen Sclerosus. J Low Genit Tract Dis 2025; 29(1): 68–71. https://doi.org/10.1097/LGT.0000000000000846

17.

Lai

Liao

, et al. The status quo of short videos as a health information source of Helicobacter pylori: a cross-sectional study. Front Public Health 2023; 11: 1344212. https://doi.org/10.3389/fpubh.2023.1344212

18.

Zhang

Yuan

Zhang

, et al. Short video platforms as sources of health information about cervical cancer: A content and quality analysis. PLoS One 2024; 19(3): e0300180. https://doi.org/10.1371/journal.pone.0300180

19.

Chew

Eysenbach

. Pandemics in the age of Twitter: content analysis of Tweets during the 2009 H1N1 outbreak. PLoS One 2010; 5(11): e14118. https://doi.org/10.1371/journal.pone.0014118

20.

Leong

Lee

Chien

, et al. Social Media-Delivered Patient Education to Enhance Self-management and Attitudes of Patients with Type 2 Diabetes During the COVID-19 Pandemic: Randomized Controlled Trial. J Med Internet Res 2022; 24(3): e31449. https://doi.org/10.2196/31449

21.

Chung

Vieira

Donley

, et al. Adolescent Peer Influence on Eating Behaviors via Social Media: Scoping Review. J Med Internet Res 2021; 23(6): e19697. https://doi.org/10.2196/19697

22.

Moreira

Mendes

Rama

, et al. AlergiaPT: A Portuguese media campaign to inspire people with allergies to make a positive change in their life. Porto Biomed J 2022; 7(1): e169. https://doi.org/10.1097/j.pbj.0000000000000169

23.

Chen

Wang

Huang

, et al. The quality and reliability of short videos about thyroid nodules on BiliBili and TikTok: Cross-sectional study. Digit Health 2024; 10: 20552076241288831. https://doi.org/10.1177/20552076241288831

24.

Xuan

Tian

Hao

. Quality assessment of short videos on health science popularization in China: scale development and validation. Front Public Health 2025; 13: 1640105. https://doi.org/10.3389/fpubh.2025.1640105

25.

Xie

Zhang

, et al. The effect of short-form video addiction on undergraduates' academic procrastination: a moderated mediation model. Front Psychol 2023; 14: 1298361. https://doi.org/10.3389/fpsyg.2023.1298361

26.

Ren

, et al. 5-Methylcytosine methylation predicts cervical cancer prognosis, shaping immune cell infiltration. J Int Med Res 2025; 53(4): 3000605251328301. https://doi.org/10.1177/03000605251328301

27.

Ren

, et al. Smoking exposure and cervical cancer risk: Integrating observational and genetic evidence. J Int Med Res 2025; 53(10): 3000605251383687. https://doi.org/10.1177/03000605251383687

28.

Liu

, et al. Quality assessment of health science-related short videos on TikTok: A scoping review. Int J Med Inform 2024; 186: 105426. https://doi.org/10.1016/j.ijmedinf.2024.105426

29.

Zheng

Tong

Wan

, et al. Quality and Reliability of Liver Cancer-Related Short Chinese Videos on TikTok and Bilibili: Cross-Sectional Content Analysis Study. J Med Internet Res 2023; 25: e47210. https://doi.org/10.2196/47210

30.

Wang

Sun

, et al. Evaluating the Quality of Health-Related WeChat Public Accounts: Cross-Sectional Study. JMIR Mhealth Uhealth 2020; 8(5): e14826. https://doi.org/10.2196/14826

31.

Steffens

Dunn

Leask

, et al. Using social media for vaccination promotion: Practices and challenges. Digit Health 2020; 6: 2055207620970785. https://doi.org/10.1177/2055207620970785