Popularity versus quality in TikTok educational videos on congenital nasolacrimal duct obstruction

Background: Congenital nasolacrimal duct obstruction is a common ocular condition in early infancy and may lead to neonatal dacryocystitis or severe infection if not treated promptly. Short-video platforms such as TikTok are used by young parents to obtain health information, but the quality of related videos remains unclear. This study evaluated the quality of TikTok videos on congenital nasolacrimal duct obstruction and its association with uploader type and user engagement.

Methods: We conducted a cross-sectional review of TikTok videos retrieved with predefined keywords. We included 108 videos, classified by uploader type, and extracted characteristics, engagement metrics, and coverage of six content domains. Two attending ophthalmologists independently rated each video using DISCERN, the Global Quality Score, and the Patient Education Materials Assessment Tool for Audiovisual Materials. Group differences and Spearman correlations were analyzed.

Results: Among 108 videos, 62 (57.4%) were uploaded by medical professionals and 18 (16.7%) by non-Profit organizations, with a median duration of 48.5 s and median numbers of likes, comments, favorites and shares of 68, 8, 13, and 24, respectively. Videos uploaded by non-Profit organizations and medical professionals achieved substantially higher DISCERN scores (about 59.22 and 50.00 vs. 37.50 and 19.40), Global Quality Scores (median 4.5 and 4.0 vs. 3.0 and 1.0), and Patient Education Materials Assessment Tool for Audiovisual Materials understandability (median 92.31 and 84.62% vs. 73.08 and 53.85%) and actionability (both 75.00% vs. 66.67 and 50.00%) than those uploaded by non-Medical individuals and for-profit organizations (all P < 0.001). Spearman analysis found no significant correlations between overall quality scores and engagement, but duration correlated weakly with some selected quality indicators (r = 0.17–0.21, P < 0.05).

Conclusion: TikTok videos on congenital nasolacrimal duct obstruction show marked heterogeneity in content completeness and educational quality, largely determined by uploader type. Non-Profit organizations and medical professionals produce more reliable, understandable, and actionable videos, but these are not consistently more popular than lower-quality content. Clinicians and institutions should develop guideline-concordant short videos and direct parents toward trustworthy channels, while platforms consider mechanisms to highlight professionally verified pediatric eye-health information.

1 Introduction

Congenital nasolacrimal duct obstruction (CNLDO) is one of the most common congenital ocular conditions in early infancy and represents the underlying pathology in the majority of cases with neonatal epiphora (1, 2). It affects approximately 6–20% of infants in the first year of life and usually presents with persistent epiphora and mucous discharge due to distal nasolacrimal duct blockage at the valve of Hasner (3, 4). Most cases resolve spontaneously with maturation of the nasolacrimal drainage system (5–7). However, persistent or unrecognized CNLDO may predispose to recurrent infection, with neonatal dacryocystitis representing an acute infection of the lacrimal sac that can further progress to preseptal or orbital cellulitis and even systemic sepsis if not promptly treated (8–10). Current reviews and clinical guidelines emphasize early identification of symptoms by caregivers and primary-care providers, appropriate use of lacrimal sac massage and topical/systemic antibiotics (11, 12), and timely escalation to probing or surgery when obstruction persists (13–15).

In parallel, short-video platforms such as TikTok have rapidly become major sources of health information for young parents. While these platforms provide convenient, visually intuitive content, emerging evidence suggests substantial variability in the quality and reliability of medical videos (16, 17). Despite the high prevalence of CNLDO and the potential severity of neonatal dacryocystitis, the quality of TikTok videos addressing these conditions has not been systematically evaluated. Parents searching for “CNLDO” or “newborn eye discharge” may therefore be exposed to content that incompletely explains the natural course of CNLDO, underemphasizes warning signs of serious infection, or promotes non–evidence-based home remedies. This cross-sectional study aims to assess the quality, reliability and understandability of CNLDO–related videos on TikTok using validated evaluation tools, and to explore the association between video quality, uploader type and user engagement metrics.

2 Materials and methods

2.1 Platform and search strategy

TikTok (Chinese mainland version) was selected as the data source. A unified search was conducted on 23 November 2025 using the in-app search function. To approximate real-world search behavior of parents, we predefined combinations of Chinese keywords, including but not limited to “congenital nasolacrimal duct obstruction (先天性泪管堵塞)” and “neonatal dacryocystitis (新生儿泪囊炎)”. The platform’s default “comprehensive” ranking mode was applied. The top-ranked 120 videos were consecutively collected until a large number of duplicate or clearly irrelevant videos appeared, forming the initial candidate pool. The detailed screening process is shown in Figure 1.

FIGURE 1

Figure 1. Flowchart of identification, screening, and classification of CNLDO-related TikTok videos.

2.2 Inclusion and exclusion criteria

Inclusion criteria were as follows: (1) videos primarily addressing issues related to neonatal or infant dacryocystitis or CNLDO, including any of the following content domains—definition, symptoms, risk factors, diagnostic examinations, treatment and management, and/or prognosis; (2) videos involving at least one of the above domains were eligible; (3) language in Mandarin Chinese or with Chinese subtitles to ensure assessors’ comprehension; and (4) publicly accessible complete videos.

Exclusion criteria were: (1) purely entertaining or humorous videos without any medical or health-related information; (2) videos mainly consisting of commercial advertisements or product promotion with no specific disease-related educational content; (3) videos unrelated to neonatal or infant ocular problems; (4) duplicate videos; (5) videos without sound and subtitles such that content could not be judged; and (6) videos that had been deleted, set to private, or could not be played normally by the platform. Two researchers independently screened all candidate videos according to these criteria. Disagreements were resolved through discussion or adjudication by a third researcher.

2.3 Video characteristics and uploader type

For each included video, data were extracted using a standardized data collection form. Video parameters: upload date; video duration (seconds); and engagement metrics including number of likes, comments, favorites, and shares, all recorded as displayed by the platform on the search date.

Uploader account characteristics: account name and verification status. Based on account verification information and profile descriptions, uploader type was categorized into four groups: (1) Medical (professional medical users, such as ophthalmologists or pediatricians); (2) Non-Medical (non-professional users, such as patients, parents, or science communicators); (3) Non-Profit organizations (for example, medical institutions or news organizations); and (4) For-Profit organizations (accounts primarily engaged in product marketing, such as postpartum care centers or maternal-and-child stores).

2.4 Video quality and credibility assessment tools

To ensure scientific rigor in the evaluation, we referred to relevant articles published in journals of the American Academy of Ophthalmology (AAO) as content evaluation standards (8, 18). Three validated instruments were used to comprehensively assess video quality and patient education value. All raters jointly reviewed the scoring manuals before formal assessment to ensure consistent understanding.

DISCERN instrument: The classic 16-item DISCERN tool was used to evaluate the reliability and quality of treatment-related information. Each item is scored from 1 to 5 (1 = “severely inadequate”; 5 = “excellent”), yielding a total score ranging from 16 to 80. Higher total scores indicate more reliable and higher-quality information on treatment options (16).

Global Quality Scale (GQS): The overall educational quality and viewing experience of each video were assessed using the 5-point GQS (1 = very poor quality, not helpful; 5 = excellent quality, highly recommended), integrating information accuracy, structural coherence, and practical usefulness (19).

Patient Education Materials Assessment Tool for Audiovisual Materials (PEMAT-A/V): The PEMAT-A/V was employed to evaluate understandability and actionability. Each item was rated as “yes,” “no,” or “not applicable.” Scores were converted to percentage values, with higher percentages indicating that patients would find the material easier to understand and act upon (20).

2.5 Rating procedure and inter-rater reliability

All included videos were independently rated by two attending ophthalmologists, each with approximately 5 years of clinical experience, who had received structured training in health education material assessment. During scoring, each rater watched the entire video in a quiet environment (including audio, subtitles, and on-screen images) and recorded item-by-item scores for each quality instrument.

To reduce potential bias, uploader identities were anonymized using coded identifiers, and raters did not view or record follower counts or other engagement metrics that might influence subjective judgments. The two raters completed their ratings independently at different times. Inter-rater reliability for each scoring tool was evaluated using the intraclass correlation coefficient (ICC). If the ICC for any instrument indicated poor agreement, the research team revisited the scoring criteria, discussed discrepant items, and refined the scoring instructions accordingly.

2.6 Statistical analysis

All data were entered into electronic spreadsheets by two researchers and cross-checked for accuracy. Statistical analyses were performed using IBM SPSS Statistics version 27.0 (IBM Corp., Armonk, NY, United States). Normality of continuous variables was assessed with the Shapiro–Wilk test. Normally distributed continuous variables were expressed as mean ± standard deviation, whereas non-normally distributed variables were presented as median and interquartile range [M (P25, P75)]. For comparisons among groups, one-way analysis of variance (ANOVA) or the Kruskal–Wallis test was used for continuous variables, depending on variance homogeneity and distribution. When overall differences were significant, post hoc pairwise comparisons were conducted using Bonferroni or Dunn–Bonferroni corrections as appropriate. Spearman or Pearson correlation analyses were used to examine associations between quality scores and user engagement metrics. All statistical tests were two-sided, and a P < 0.05 was considered statistically significant.

3 Results

3.1 Analysis of basic video characteristics

A total of 108 TikTok videos related to congenital nasolacrimal duct obstruction were finally included. Most videos were uploaded by medical professionals: ophthalmologists uploaded 58 videos (53.7%) and pediatricians 4 videos (3.7%), for a total of 62 videos (57.4%). Non-Profit organizations (including hospitals and news media) uploaded 18 videos (16.7%), of which 15 (13.9%) were from hospitals and 3 (2.8%) from news media. Non-Medical individual users (science communicators and patients or family members) also uploaded 18 videos (16.7%), including 8 (7.4%) by science communicators and 10 (9.3%) by patients or caregivers. For-Profit organizations uploaded 10 videos (9.3%) (Table 1). Overall, the median numbers of likes, comments, favorites, and shares for the included videos were 68.0 (24.5, 228.0), 8.0 (1.25, 26.75), 12.5 (3.0, 55.25), and 23.5 (5.25, 102.75), respectively. The median video duration was 48.5 s (32.5, 72.75), indicating a moderate level of user engagement and dissemination for CNLDO-related content on the platform (Figure 2).

TABLE 1

Table 1. Basic characteristics and engagement metrics of CNLDO-related TikTok videos by uploader type.

FIGURE 2

Figure 2. Distribution of uploader types for CNLDO-related TikTok videos.

Marked differences in engagement metrics were observed across different uploader types. Videos uploaded by Non-Medical individual users received the highest engagement, with median numbers of likes, comments, favorites, and shares of 629.5 (167.5, 4491.5), 170.5 (27.5, 471.0), 95.5 (19.75, 443.25), and 154.5 (29.0, 285.25), respectively. Videos from Non-Profit organizations showed intermediate levels of engagement, with 175.0 (13.5, 313.75) likes and 19.5 (9.0, 55.0) shares. In comparison, videos from Medical individual users had slightly lower exposure and engagement than those from Non-Medical individual users, but still achieved a certain dissemination effect, with 52.0 (30.75, 115.50) likes and 24.5 (7.75, 92.50) shares. Videos uploaded by For-Profit organizations had the lowest engagement, with a mean of only about 7.60 ± 1.71 likes, median values of 1.0 for both comments and favorites, and a median of 0 shares, suggesting limited attention to their educational content. Regarding video duration, Non-Medical individual users produced the longest videos on average (76.17 ± 50.03 s), followed by Non-Profit organizations at 58.0 s (49.5, 120.75). Videos from Medical individual users had a moderate duration of 47.0 s (35.0, 64.0), whereas those from For-Profit organizations were the shortest (22.50 ± 4.88 s) (Table 1). Overall, videos uploaded by Medical individual users and Non-Profit organizations tended to cluster around 1 min in length, which is more consistent with the “short and concise” pattern favored for health education on short-video platforms (Figure 3).

FIGURE 3

Figure 3. Engagement metrics of CNLDO-related TikTok videos by uploader type.

3.2 Video content analysis

The content of CNLDO-related videos was categorized into six core domains: disease definition, disease classification, clinical symptoms, risk factors, diagnostic examinations, and treatment and management. Among all 108 videos, clinical symptoms were the most frequently mentioned domain, covered in 84 videos (77.78%). Treatment and management and disease definition were tied for the second highest coverage, each appearing in 78 videos (72.22%). Diagnostic information was provided in 68 videos (62.96%), risk factors were mentioned in 63 videos (58.33%), and disease classification had the lowest coverage, being addressed in only 58 videos (53.70%). Overall, CNLDO-related TikTok videos tended to focus on “symptoms and management,” whereas systematic explanations of disease types and risk factors were relatively insufficient (Table 2; Figure 4).

TABLE 2

Table 2. Content coverage of CNLDO-related TikTok videos across six core domains stratified by uploader type.

FIGURE 4

Figure 4. Content coverage of six core domains in CNLDO-related TikTok videos by uploader type.

As shown in Table 2, content coverage varied substantially across uploader types. Medical users (n = 62) performed best overall in all domains. In particular, 58 videos (93.55%) provided a relatively clear disease definition. High coverage was also observed for treatment and management (83.87%, 52/62), symptoms (80.65%, 50/62), diagnosis (79.03%, 49/62), and risk factors (74.19%, 46/62), while classification was mentioned in 70.97% of videos (44/62). Non-Profit organizations (n = 18, including hospitals and news media) also showed strong content integrity: coverage rates for symptoms and treatment and management both reached 88.89% (16/18), while definition and diagnosis were each present in 83.33% (15/18), risk factors in 77.78% (14/18), and classification in 66.67% (12/18), suggesting a certain advantage in systematic science popularization (Figure 4).

In contrast, videos from Non-Medical individual users (n = 18) and For-Profit organizations (n = 10) were clearly less comprehensive. Although Non-Medical individual users showed a relatively high coverage of symptoms (77.78%), coverage of definition (22.22%), treatment and management (16.67%), diagnosis and risk factors (both 11.11%), and classification (5.56%) was markedly inadequate. Videos from For-Profit organizations mainly focused on treatment and management (70.00%) and symptoms (40.00%), whereas coverage of definition, diagnosis, risk factors, and classification was only 10.00–20.00%. These findings indicate that, compared with Medical users and Non-Profit organizations, Non-Medical individual and commercial accounts have substantial gaps in delivering key disease information, which may lead to incomplete understanding or even misconceptions among parents (Figure 4).

Among the 108 included videos, the most common clinical pitfalls were lack of follow-up/re-evaluation advice (38/108, 35.19%) and lack of stepwise management and indications for procedures (37/108, 34.26%), followed by poor actionability/unclear caregiver instructions (32/108, 29.63%). Notably, these pitfalls were concentrated in the Non-Medical and For-Profit groups, with the highest counts observed for stepwise-management omissions (18 and 10 videos, respectively) and missing follow-up guidance (16 and 10 videos, respectively) (Table 3).

TABLE 3

Table 3. Frequency of common clinical pitfalls and potentially harmful omissions identified in TikTok videos on CNLDO.

3.3 Video quality analysis

For Videos reliability, Treatment choice, DISCERN scores, Overall quality score, Understandability, Actionability, and GQS, the intraclass correlation coefficients between the two raters were 0.868, 0.878, 0.842, 0.853, 0.845, 0.828, and 0.819, respectively (all P < 0.001) (Figure 5). Video quality among different uploader types was then compared based on the DISCERN instrument, PEMAT-A/V, and GQS (Table 4).

FIGURE 5

Figure 5. Inter-rater reliability of quality assessment for CNLDO-related TikTok videos.

TABLE 4

Table 4. Comparison of video quality scores among different uploader types based on DISCERN, PEMAT-A/V, and GQS.

Overall, videos from Non-Profit organizations had the highest quality indicators: the Videos reliability score was 32.39 ± 1.85, the Treatment choice score was 23.00 ± 2.11, the median Overall quality score was 4.00 (3.00, 4.00), and the DISCERN scores were 59.22 ± 2.51. Medical individual users ranked second, with a median Videos reliability score of 27.00 (27.00, 28.00), a median Treatment choice score of 19.00 (18.00, 21.00), a median Overall quality score of 3.00 (2.00, 4.00), and a median DISCERN scores of 50.00 (49.00, 51.00). Non-Medical individual users showed clearly lower scores in all domains, whereas For-Profit organizations had the lowest scores: their Videos reliability and Treatment choice scores were only 8.60 ± 1.67 and 9.40 ± 1.27, respectively, with a DISCERN scores of 19.40 ± 1.90 and a median Overall quality score of 1.00 (1.00, 2.00). Kruskal–Wallis tests indicated statistically significant differences among the four uploader types in Videos reliability, Treatment choice, Overall quality score, and DISCERN scores (H = 38.17–88.15, all P < 0.001) (Figure 6a).

FIGURE 6

Figure 6. Distribution of quality indicators of CNLDO-related TikTok videos by uploader type.(a) Reliability, treatment choice, understandability and actionability; (b) Overall quality score, DISCERN scores and GQS.

In terms of Understandability and Actionability, Non-Profit organizations again performed best, with median scores of 92.31 (84.62, 92.31) and 75.00 (75.00, 100.00), both clearly above the 70% threshold commonly considered acceptable. Medical individual users also achieved favorable levels, with median Understandability and Actionability scores of 84.62 (76.92, 84.62) and 75.00 (66.67, 75.00), respectively. Although Non-Medical individual users showed a moderate Understandability score of 73.08 (59.62, 76.92), their Actionability score was only 66.67 (50.00, 66.67), close to or slightly below the acceptable threshold. For-Profit organizations had the lowest scores for both indicators, with Understandability of 53.85 (53.85, 55.77) and Actionability of 50.00 (50.00, 50.00). Regarding GQS, Non-Profit organizations and Medical individual users achieved median scores of 4.50 (4.00, 5.00) and 4.00 (4.00, 4.00), respectively, whereas Non-Medical individual users and For-Profit organizations scored only 3.00 (2.00, 3.00) and 1.00 (1.00, 2.00). Differences in Understandability, Actionability, and GQS among uploader types were all statistically significant (H = 48.49–65.98, all P < 0.001) (Figure 6b).

Post hoc pairwise comparisons (Tables 5, 6) further demonstrated that videos uploaded by Non-Profit organizations were significantly superior to those from Non-Medical individual users and For-Profit organizations in Videos reliability, Treatment choice, Overall quality score, DISCERN scores, Understandability, Actionability, and GQS (multiple comparisons, P < 0.001 for many contrasts). Compared with Medical individual users, Non-Profit organizations also showed advantages in Videos reliability, Treatment choice, and DISCERN scores. Medical individual users had significantly higher Overall quality, Understandability, and GQS than Non-Medical individual users and For-Profit organizations (multiple comparisons, P < 0.05). In contrast, differences between Non-Medical individual users and For-Profit organizations were not significant for several indicators, including Videos reliability, Treatment choice, Understandability, Actionability, and GQS (P > 0.05). Taken together, these findings suggest that the quality of CNLDO-related videos on TikTok is highly dependent on uploader type: information provided by Non-Profit organizations and Medical professionals is relatively more reliable, easier to understand, and more actionable, whereas videos from commercial accounts and some Non-Medical individuals are generally of low quality (Figure 7).

TABLE 5

Table 5. Pairwise comparisons of DISCERN-based quality indicators between different uploader types.

TABLE 6

Table 6. Pairwise comparisons of understandability, actionability, and GQS between different uploader types.

FIGURE 7

Figure 7. Comparison of quality indicators of CNLDO-related TikTok videos among different uploader types. (a) Reliability; (b) treatment choice; (c) overall quality score; (d) DISCERN scores; (e) understandability; (f) actionability; (g) GQS. *P < 0.05, **P < 0.01, ***P < 0.001, and ns P≥0.05.

3.4 Correlation analysis

Spearman correlation analysis was used to explore associations between video quality indicators and dissemination metrics. As shown in Figure 8, within the set of quality-related indicators, Videos reliability, Treatment choice, Overall quality score, DISCERN scores, Understandability, Actionability, and GQS were all moderately to highly positively correlated with one another (r ≈ 0.42–0.90, all P < 0.05). Among these, Videos reliability showed the strongest correlation with DISCERN scores (r ≈ 0.86), and was also highly correlated with GQS (r ≈ 0.76), indicating that more reliable videos tend to have higher levels of evidence and better overall subjective quality. The correlation coefficient between Treatment choice and DISCERN scores was close to 0.90, suggesting strong consistency between the completeness of treatment information and the overall evidence-based quality. Understandability and Actionability were also significantly and positively correlated (r ≈ 0.51), implying that videos that are easier to understand are more likely to provide parents with actionable recommendations.

FIGURE 8

Figure 8. Correlations between quality indicators and engagement metrics of CNLDO-related TikTok videos. (a) Spearman correlation matrix with significance marks (asterisks indicate P < 0.05). (b) Heatmap of Spearman correlation coefficients for quality and engagement indicators.

In contrast, correlations between quality indicators and engagement metrics (likes, comments, favorites, and shares) were generally weak, with most correlation coefficients close to 0 and not statistically significant (r ≤ 0.12, P > 0.05). Only video duration showed a low positive correlation with some quality indicators (for example, r ≈ 0.17–0.21 with Videos reliability and DISCERN scores, P < 0.05). This suggests that on TikTok, engagement metrics do not have a simple linear relationship with the health education quality of videos. On the other hand, very strong correlations were observed among likes, comments, favorites, and shares (r ≈ 0.69–0.88, all P < 0.001), indicating that these engagement metrics mainly reflect a common dimension of “popularity” rather than professional quality per se. Overall, the professional quality of CNLDO-related videos appears to be determined primarily by the content itself and is only weakly associated with platform engagement levels (Figure 8).

4 Discussion

To our knowledge, this is the first study to systematically evaluate the content and quality of TikTok videos on congenital nasolacrimal duct obstruction and to compare these parameters across different uploader types. Overall, three major findings emerged. First, the number of CNLDO-related videos on the platform was limited, and most were uploaded by ophthalmologists, pediatricians, or medical institutions; however, information on symptoms and management was far more common than basic knowledge such as disease definition, classification, and risk factors, indicating an unbalanced content structure. S, video quality was highly dependent on uploader type: videos from Non-Profit organizations and Medical professionals clearly outperformed those from Non-Medical individuals and For-Profit organizations in terms of reliability, treatment information, Understandability, Actionability, and GQS. Third, strong correlations were observed among the various quality indicators themselves, whereas correlations between quality indices and engagement metrics (likes, comments, favorites, and shares) were weak, suggesting that popularity does not equate to health information quality.

For young parents who rely on short-video platforms for health information (21), whether videos convey key diagnostic and management checkpoints accurately is directly related to the risk of delayed consultation or inappropriate self-management (22). In this study, although more than 70% of videos mentioned disease symptoms and management strategies, only about half provided a systematic description of disease classification and risk factors. Some videos confused neonatal dacryocystitis with other causes of neonatal ocular discharge, or downplayed “red-flag” signs such as acute dacryocystitis and systemic infection. These findings contrast sharply with the integrated educational goals of CNLDO clinical guidelines, which emphasize the natural history, warning signs, and appropriate timing of intervention (23). If parents mainly encounter short videos that stress messages such as “most cases resolve spontaneously” or “just massage more and it will be fine” without adequately explaining high-risk features and clear indications for medical evaluation, recognition of serious complications may be delayed, imposing unnecessary burdens on affected children and their families (24, 25).

Consistent with TikTok-based studies of cardiovascular, endocrine, and other ocular diseases (26–31), we also observed pronounced differences across uploader types. Non-Profit organizations (including hospitals and news media) achieved the highest scores for Videos reliability, treatment information, Overall quality score, DISCERN scores, and the PEMAT-A/V domains of Understandability and Actionability. Their quality levels were comparable to those reported for high-quality short videos on coronary heart disease, thyroid nodules, cataract, and dry eye (26, 29–31). Medical individual users performed slightly less well than institutional accounts, particularly in the rigor of treatment descriptions and the degree of structured presentation, which may reflect constraints on personal creators in terms of time and resources. In contrast, videos from Non-Medical individuals and For-Profit organizations frequently showed information gaps or excessive simplification. In this study, both uploader types scored significantly lower than Medical users and Non-Profit organizations on DISCERN, PEMAT-A/V, and GQS, and some videos even recommended folk remedies or commercial services lacking evidence support. This pattern aligns closely with findings from short-video analyses of mitral valve regurgitation, coronary heart disease, and myopia (32–34), which indicate that commercial accounts tend to highlight products or services while neglecting comprehensive disease background and risk communication. For clinicians and public health practitioners, these results underscore the need to actively guide parents toward content produced by hospitals, professional societies, or verified expert accounts, while remaining alert to potential information bias from marketing-oriented channels (35, 36).

It is noteworthy that we found moderate-to-strong positive correlations among the video quality indicators, but no significant correlations between quality and engagement metrics such as likes, comments, favorites, and shares; only video duration showed a weak positive association with several quality indices. Similar phenomena have been reported in multiple studies on YouTube and TikTok (37–39), where high-quality health information does not necessarily achieve more likes or views, whereas “eye-catching” titles, dramatic visuals, or emotionally charged narratives are more likely to drive user behavior. Thus, treating engagement data as a simple proxy for “high-quality content” may systematically underestimate truly evidence-based educational materials. From an algorithmic perspective, recommender systems tend to favor content with high engagement and long viewing time, which may inadvertently amplify the spread of low-quality or even misleading medical information (40, 41). This suggests that medical professionals creating short videos should not only ensure accuracy and clear structure but also draw on strategies from digital health communication and behavioral science (42, 43), such as narrative storytelling, visualized presentation, and action-oriented conclusions, to improve viewing and sharing without compromising scientific rigor (44, 45).

Our findings support the incorporation of standardized evaluation tools, such as DISCERN and PEMAT-A/V, into short-video health education to guide content creation and quality control (46). By checking scripts against these instruments item by item before filming, creators can systematically verify whether they have adequately explained disease definition, risk factors, available treatment options, benefits and risks, and specific actions that parents can take. In addition, medical institutions could establish “official recommended accounts” and themed playlists to aggregate high-quality videos that have undergone peer or multidisciplinary review, thereby reducing parents’ screening burden in an information-saturated environment. For platform governance, it may be feasible—without making explicit medical judgments—to introduce structural quality signals, such as whether risk warnings are listed, whether clear consultation suggestions are provided, and whether sources are cited, so that recommendation algorithms can assign more weight to potentially high-quality medical content (47, 48). At the same time, further enhancing the visibility of professional identity verification and nudging users toward following credentialed physicians or medical institutions may help improve the overall quality of health information that parents encounter (49).

This study has several limitations. First, it was a cross-sectional analysis conducted at a single time point. The number, ranking, and engagement metrics of TikTok videos change dynamically; therefore, our findings represent only a “snapshot” of the search date and cannot capture temporal trends in content evolution. Second, we examined only videos on the TikTok platform and did not include other languages or platforms; thus, the generalizability of our conclusions to different cultural and platform contexts may be limited. Third, although the “Non-Profit” uploader category may comprise heterogeneous entities (e.g., hospital-based institutional accounts vs. news media accounts), the news media subgroup was small in our sample. Further stratification would likely reduce statistical power and yield unstable estimates; Fourth, although we used validated tools such as DISCERN and PEMAT-A/V and adopted a dual-rater approach to enhance consistency, assessment of video quality inevitably retains some degree of subjectivity. Fifth, the present study primarily focused on the completeness and presentation of information and did not verify each statement against the latest clinical guidelines on an item-by-item basis. Future work could perform more fine-grained quantitative analyses of content accuracy and transparency of risk communication. Sixth, we did not analyze comment sections or user interaction content, which may further influence parental perceptions and behaviors. These limitations highlight the need for future multicenter studies conducted at multiple time points, across different platforms, and with larger samples, ideally incorporating behavioral outcomes, to more comprehensively evaluate the real-world impact of short-video health information on CNLDO management.

5 Conclusion

This study systematically evaluated the content and quality of TikTok videos related to congenital nasolacrimal duct obstruction. Video quality was closely associated with uploader type: videos posted by Non-Profit organizations and Medical professionals demonstrated markedly higher reliability, treatment information quality, overall quality, and better Understandability and Actionability than those from Non-Medical individuals and For-Profit accounts, whereas correlations with engagement metrics were weak—“popular” did not necessarily mean “high-quality” (50). In the future, clinicians and public health professionals should be encouraged to actively participate in creating authoritative educational content, to guide parents toward information from trustworthy accounts, and to advocate for further optimization of platform mechanisms for professional verification and quality signaling (51).

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

Ethics statement

The studies involving humans were approved by the Ethics Committee of Quzhou Affiliated Hospital of Wenzhou Medical University (Quzhou People’s Hospital). The studies were conducted in accordance with the local legislation and institutional requirements. All procedures were performed in accordance with the ethical principles laid down in the Declaration of Helsinki. Written informed consent for participation was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and institutional requirements. The authors declare that they have read and confirmed that they strictly adhered to TikTok’s terms of use for social media research during the study. All data used are derived from publicly available short video content and corresponding public interaction data on the TikTok platform.

Author contributions

MZ: Conceptualization, Investigation, Methodology, Writing – review & editing. WX: Data curation, Investigation, Writing – review & editing. LX: Data curationl, Investigation, Writing – review & editing. YT: Data curationl, Investigation, Writing – review & editing. JZ: Conceptualization, Validation, Writing – review & editing. YG: Conceptualization, Validation, Writing – review & editing. ZZha: Formal analysis, Supervision, Writing – review & editing. ZZhu: Formal analysis, Supervision, Writing – review & editing. LJ: Formal analysis, Funding acquisition, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This study was supported by funding from 2024 Quzhou City “Southern Confucianism Elites” Special Training Program, Wenzhou Medical University Interdisciplinary Research Project, Quzhou City Science and Technology Project (No. 2025K003), Medical and Health Science Program of Zhejiang Province (No. 2025HY1387), Gusu Health Talent Program Project in Suzhou [No. (2022)192], and Jiangsu Provincial Young Science and Technology Talent Support Program (No. JSTJ-2025-816).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

1. Wladis EJ, Aakalu VK, Yen MT, Bilyk JR, Sobel RK, Mawn LA. Balloon dacryoplasty for congenital nasolacrimal duct obstruction: a report by the American Academy of Ophthalmology. Ophthalmology. (2018) 125:1654–7. doi: 10.1016/j.ophtha.2018.05.005

PubMed Abstract | Crossref Full Text | Google Scholar

2. Weatherby T. Relative frequencies of ophthalmia neonatorum and congenital nasolacrimal duct obstruction. Br J Gen Pract. (2021) 71:59. doi: 10.3399/bjgp21X714905

PubMed Abstract | Crossref Full Text | Google Scholar

3. Sathiamoorthi S, Frank RD, Mohney BG. Incidence and clinical characteristics of congenital nasolacrimal duct obstruction. Br J Ophthalmol. (2018) 103:527–9. doi: 10.1136/bjophthalmol-2018-312074

PubMed Abstract | Crossref Full Text | Google Scholar

4. Petris C, Liu D. Probing for congenital nasolacrimal duct obstruction. Cochrane Database Syst Rev. (2017) 7:CD011109. doi: 10.1002/14651858.CD011109.pub2

PubMed Abstract | Crossref Full Text | Google Scholar

5. Nakayama T, Watanabe A, Rajak S, Yamanaka Y, Sotozono C. Congenital nasolacrimal duct obstruction continues trend for spontaneous resolution beyond first year of life. Br J Ophthalmol. (2019) 104:1161–3. doi: 10.1136/bjophthalmol-2019-314980

PubMed Abstract | Crossref Full Text | Google Scholar

6. Pediatric Eye Disease Investigator Group Resolution of congenital nasolacrimal duct obstruction with nonsurgical management. Arch Ophthalmol. (2012) 130:730–4. doi: 10.1001/archophthalmol.2012.454

PubMed Abstract | Crossref Full Text | Google Scholar

7. Frick KD, Hariharan L, Repka MX, Chandler D, Melia BM, Beck RW, et al. Cost-effectiveness of 2 approaches to managing nasolacrimal duct obstruction in infants: the importance of the spontaneous resolution rate. Arch Ophthalmol. (2011) 129:603–9. doi: 10.1001/archophthalmol.2011.80

PubMed Abstract | Crossref Full Text | Google Scholar

8. Pediatric Eye Disease Investigator Group, Repka MX, Chandler DL, Beck RW, Crouch ER, Donahue S, et al. Primary treatment of nasolacrimal duct obstruction with probing in children younger than 4 years. Ophthalmology. (2007) 115:577–84.e3. doi: 10.1016/j.ophtha.2007.07.030.

PubMed Abstract | Crossref Full Text | Google Scholar

9. He X, Hu X, Zhang X, Yang W. Advances in imaging techniques for surgical management of common lacrimal drainage disorders. Int J Surg. (2025). doi: 10.1097/JS9.0000000000003972 [Epub ahead of print].

PubMed Abstract | Crossref Full Text | Google Scholar

10. Jia Q, Wang X, Li X, Xie C, Zhang Q, Mu J, et al. Analysis of research hotspots and trends in pediatric ophthalmopathy based on 10 years of WoSCC literature. Front Pediatr. (2024) 12:1405110. doi: 10.3389/fped.2024.1405110

PubMed Abstract | Crossref Full Text | Google Scholar

11. Asano M, Takeuchi M, Ohno T, Nakamura J, Mizuki N, Matsumura N. Lacrimal sac massage for congenital nasolacrimal duct obstruction: a multicentre randomised controlled trial. Br J Ophthalmol. (2024) 108:1281–5. doi: 10.1136/bjo-2023-324595

PubMed Abstract | Crossref Full Text | Google Scholar

12. Mohney BG, Sathiamoorthi S, Frank RD. Spontaneous resolution rates in congenital nasolacrimal duct obstruction managed with massage or topical antibiotics compared with observation alone. Br J Ophthalmol. (2021) 106:1196–9. doi: 10.1136/bjophthalmol-2021-318853

PubMed Abstract | Crossref Full Text | Google Scholar

13. Sathiamoorthi S, Frank RD, Mohney BG. Spontaneous resolution and timing of intervention in congenital nasolacrimal duct obstruction. JAMA Ophthalmol. (2018) 136:1281–6. doi: 10.1001/jamaophthalmol.2018.3841

PubMed Abstract | Crossref Full Text | Google Scholar

14. Repka MX. Timing of simple probing for congenital nasolacrimal duct obstruction: not so simple. JAMA Ophthalmol. (2018) 136:1286–7. doi: 10.1001/jamaophthalmol.2018.3855

PubMed Abstract | Crossref Full Text | Google Scholar

15. Lee KA, Chandler DL, Repka MX, Melia M, Beck RW, Summers CG, et al. A comparison of treatment approaches for bilateral congenital nasolacrimal duct obstruction. Am J Ophthalmol. (2013) 156:1045–50. doi: 10.1016/j.ajo.2013.06.014

PubMed Abstract | Crossref Full Text | Google Scholar

16. Didier A, Fang L, Stiene J, Ud DS, Spencer C, Hamouda DM. Quality of educational YouTube videos related to pancreatic cancer: analysis of patient-focused and NCI-designated cancer center videos. JCO. (2023) 41:756–756. doi: 10.1200/jco.2023.41.4_suppl.756

Crossref Full Text | Google Scholar

17. Sun F, Zheng S, Wu J. Quality of information in gallstone disease videos on TikTok: cross-sectional study. J Med Internet Res. (2023) 25:e39162. doi: 10.2196/39162

PubMed Abstract | Crossref Full Text | Google Scholar

18. Feng ZX, Liu W, Li Z, Cui Y, Li L, Zhang C. Bony congenital nasolacrimal duct obstruction: a novel phenotype of aplasia of lacrimal and major salivary glands. Ophthalmology. (2023) 131:589–94. doi: 10.1016/j.ophtha.2023.12.003

PubMed Abstract | Crossref Full Text | Google Scholar

19. Barlas T, Ecem Avci D, Cinici B, Ozkilicaslan H, Muhittin Yalcin M, Eroglu Altinova A. The quality and reliability analysis of YouTube videos about insulin resistance. Int J Med Inform. (2023) 170:104960. doi: 10.1016/j.ijmedinf.2022.104960

PubMed Abstract | Crossref Full Text | Google Scholar

20. Shoemaker SJ, Wolf MS, Brach C. Development of the Patient Education Materials Assessment Tool (PEMAT): a new measure of understandability and actionability for print and audiovisual patient information. Patient Educ Couns. (2014) 96:395–403. doi: 10.1016/j.pec.2014.05.027

PubMed Abstract | Crossref Full Text | Google Scholar

21. Kirkpatrick CE, Lawrie LL. TikTok as a source of health information and misinformation for young women in the United States: survey study. JMIR Infodemiol. (2024) 4:e54663. doi: 10.2196/54663

PubMed Abstract | Crossref Full Text | Google Scholar

22. Frey E, Bonfiglioli C, Brunner M, Frawley J. Parents’ use of social media as a health information source for their children: a scoping review. Acad Pediatr. (2021) 22:526–39. doi: 10.1016/j.acap.2021.12.006

PubMed Abstract | Crossref Full Text | Google Scholar

23. Sasaki T, Matsumura N, Miyazaki C, Kamao T, Yokoi N, Fujimoto M, et al. Congenital nasolacrimal duct obstruction: clinical guideline. Jpn J Ophthalmol. (2024) 68:367–88. doi: 10.1007/s10384-024-01064-4

PubMed Abstract | Crossref Full Text | Google Scholar

24. Eneriz-Wiemer M, Liu SD, Chu MCY, Uribe-Leitz T, Rajani K, Sankar M, et al. Parents’ knowledge and education of retinopathy of prematurity in four California neonatal intensive care units. Am J Ophthalmol. (2018) 191:7–13. doi: 10.1016/j.ajo.2018.03.039

PubMed Abstract | Crossref Full Text | Google Scholar

25. Siegal AR, Ferrer FA, Baldisserotto E, Malhotra NR. The assessment of TikTok as a source of quality health information on varicoceles. Urology. (2022) 175:170–4. doi: 10.1016/j.urology.2022.12.016

PubMed Abstract | Crossref Full Text | Google Scholar

26. Xie X, Li Y, Chen Z, Zhou P, Jin X. Quality and accuracy of cardiopulmonary resuscitation teaching in short videos: an analysis across three major short video platforms. BMC Med Educ. (2025) 25:631. doi: 10.1186/s12909-025-06776-w

PubMed Abstract | Crossref Full Text | Google Scholar

27. Zhang Y, Huang C, Tong Y, Teng Y, Wan B, Huang J, et al. Quality assessment of spinal cord injury-related health information on short-form video platforms: cross-sectional content analysis of TikTok. Kwai, and BiliBili. Digit Health. (2025) 11:20552076251374226. doi: 10.1177/20552076251374226

PubMed Abstract | Crossref Full Text | Google Scholar

28. Ming S, Han J, Li M, Liu Y, Xie K, Lei B. TikTok and adolescent vision health: content and information quality assessment of the top short videos related to myopia. Front Public Health. (2023) 10:1068582. doi: 10.3389/fpubh.2022.1068582

PubMed Abstract | Crossref Full Text | Google Scholar

29. Cao J, Zhang F, Zhu Z, Xiong W. Quality of cataract-related videos on TikTok and its influencing factors: a cross-sectional study. Digit Health. (2025) 11:20552076251365086. doi: 10.1177/20552076251365086

PubMed Abstract | Crossref Full Text | Google Scholar

30. Huang M, Wang J, Wei J, Zhang R, Wang X, Gan J, et al. Assessing the quality of educational short videos on dry eye care: a cross-sectional study. Front Public Health. (2025) 13:1542278. doi: 10.3389/fpubh.2025.1542278

PubMed Abstract | Crossref Full Text | Google Scholar

31. Wang H, Zhang H, Cao J, Zhang F, Xiong W. Quality and content evaluation of thyroid eye disease treatment information on TikTok and Bilibili. Sci Rep. (2025) 15:25134. doi: 10.1038/s41598-025-11147-y

PubMed Abstract | Crossref Full Text | Google Scholar

32. Hao P, Liu G, Lian S, Huang J, Zhao L. Evaluating the quality of TikTok videos on coronary artery disease using various scales to examine correlations with video characteristics and high-quality content. Sci Rep. (2025) 15:9189. doi: 10.1038/s41598-025-93986-3

PubMed Abstract | Crossref Full Text | Google Scholar

33. Cui N, Lu Y, Cao Y, Chen X, Fu S, Su Q. Quality assessment of TikTok as a source of information about mitral valve regurgitation in china: cross-sectional study. J Med Internet Res. (2024) 26:e55403. doi: 10.2196/55403

PubMed Abstract | Crossref Full Text | Google Scholar

34. Ming S, Han J, Yao X, Guo X, Guo Q, Lei B. Myopia information on TikTok: analysis factors that impact video quality and audience engagement. BMC Public Health. (2024) 24:1194. doi: 10.1186/s12889-024-18687-4

PubMed Abstract | Crossref Full Text | Google Scholar

35. Robbins R, Epstein LJ, Iyer JM, Weaver MD, Javaheri S, Fashanu O, et al. Examining understandability, information quality, and presence of misinformation in popular YouTube videos on sleep compared to expert-led videos. J Clin Sleep Med. (2023) 19:991–4. doi: 10.5664/jcsm.10520

PubMed Abstract | Crossref Full Text | Google Scholar

36. Mueller SM, Hongler VNS, Jungo P, Cajacob L, Schwegler S, Steveling EH, et al. Fiction, falsehoods, and few facts: cross-sectional study on the content-related quality of atopic eczema-related videos on YouTube. J Med Internet Res. (2020) 22:e15599. doi: 10.2196/15599

PubMed Abstract | Crossref Full Text | Google Scholar

37. Wang H, Yan C, Wu T, Zhang X, He J, Liu Z, et al. YouTube online videos as a source for patient education of cervical spondylosis-a reliability and quality analysis. BMC Public Health. (2023) 23:1831. doi: 10.1186/s12889-023-16495-w

PubMed Abstract | Crossref Full Text | Google Scholar

38. Zhang H, Zhang C, Mo J, Wang W, Wu Z. Quality of information in osteoporosis videos on TikTok: a cross-sectional study. Arch Osteoporos. (2025) 20:115. doi: 10.1007/s11657-025-01597-2

PubMed Abstract | Crossref Full Text | Google Scholar

39. Chahal R, Garg T, Khosla AA, Singh R, Karass SH, Sharma RK, et al. Quality assessment of information on chimeric antigen receptor T-cell therapy on YouTube: an analysis. JCO. (2023) 41:e13633. doi: 10.1200/jco.2023.41.16_suppl.e13633

Crossref Full Text | Google Scholar

40. Milli S, Carroll M, Wang Y, Pandey S, Zhao S, Dragan AD. Engagement, user satisfaction, and the amplification of divisive content on social media. PNAS Nexus. (2025) 4:gaf062. doi: 10.1093/pnasnexus/pgaf062

PubMed Abstract | Crossref Full Text | Google Scholar

41. Guess AM, Malhotra N, Pan J, Barberá P, Allcott H, Brown T, et al. How do social media feed algorithms affect attitudes and behavior in an election campaign? Science. (2023) 381:398–404. doi: 10.1126/science.abp9364

PubMed Abstract | Crossref Full Text | Google Scholar

42. Osman W, Mohamed F, Elhassan M, Shoufan A. Is YouTube a reliable source of health-related information? a systematic review. BMC Med Educ. (2022) 22:382. doi: 10.1186/s12909-022-03446-z

PubMed Abstract | Crossref Full Text | Google Scholar

43. Gaikwad A, Rachh P, Raut K, Nadgere J. Critical appraisal of YouTube videos regarding peri-implant diseases and conditions: a cross-sectional analysis. Clin Oral Implants Res. (2024) 35:498–509. doi: 10.1111/clr.14246

PubMed Abstract | Crossref Full Text | Google Scholar

44. Wieland ML, Vickery KD, Hernandez V, Ford BR, Gonzalez C, Kavistan S, et al. Digital storytelling intervention for hemoglobin A1c control among hispanic adults with Type 2 diabetes: a randomized clinical trial. JAMA Netw Open. (2024) 7:e2424781. doi: 10.1001/jamanetworkopen.2024.24781

PubMed Abstract | Crossref Full Text | Google Scholar

45. Nagarkar A, Martin G, Sadaniantz K, Iyengar S, Wisniewski HC, Denu MK, et al. Storytelling for health promotion: a scoping review. Am J Health Promot. (2025) 40:187–209. doi: 10.1177/08901171251365366

PubMed Abstract | Crossref Full Text | Google Scholar

46. 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) 53:105–11. doi: 10.1136/jech.53.2.105

PubMed Abstract | Crossref Full Text | Google Scholar

47. Ghalavand H, Nabiolahi A. Exploring online health information quality criteria on social media: a mixed method approach. BMC Health Serv Res. (2024) 24:1311. doi: 10.1186/s12913-024-11838-8

PubMed Abstract | Crossref Full Text | Google Scholar

48. Aslam R, Gibbons D, Ghezzi P. Online information on antioxidants: information quality indicators, commercial interests, and ranking by Google. Front Public Health. (2017) 5:90. doi: 10.3389/fpubh.2017.00090

PubMed Abstract | Crossref Full Text | Google Scholar

49. Yao L, Li Y, Lian Q, Sun J, Zhao S, Wang P. Health information sharing on social media: quality assessment of short videos about chronic kidney disease. BMC Nephrol. (2022) 23:378. doi: 10.1186/s12882-022-03013-0

PubMed Abstract | Crossref Full Text | Google Scholar

50. Jiang L, Wu W, Jiang X, Jiang F. Evaluating the quality of educational TikTok videos on diabetic retinopathy: a cross-sectional study. Front Public Health. (2025) 13:1718587. doi: 10.3389/fpubh.2025.1718587

PubMed Abstract | Crossref Full Text | Google Scholar

51. Gbedemah ZEE, Fuseini MN, Fordjuor SKEJ, Baisie-Nkrumah EJ, Beecham REM, Amissah-Arthur KN. Readability and quality of online information on sickle cell retinopathy for patients. Am J Ophthalmol. (2023) 259:45–52. doi: 10.1016/j.ajo.2023.10.023

PubMed Abstract | Crossref Full Text | Google Scholar