Abstract
Objective
Potentially malignant oral disorders (PMODs) can progress to oral cancer, making the quality of available health information especially relevant. This research aimed to analyze TikTok content related to potentially malignant oral disorders (PMOD), considering aspects such as source, duration, and content quality.
Materials and methods
A cross-sectional observational study was conducted to analyze TikTok videos related to potentially malignant oral disorders (PMODs), using the hashtags #leukoplakia, #lichen planus, #actinic cheilitis, #oral lichenoid lesions, and #erythroplasia. English- and Spanish-language videos were included; duplicates, unrelated, advertising, or silent videos were excluded. Video characteristics, engagement (views, likes, comments, saves, shares), viewing and interaction indices, and type of uploader were recorded. Two evaluators scored the content with DISCERN and the Global Quality Scale (GQS).
Results
A total of 96 videos were analyzed. The largest group of creators were nano-influencers (1–10,000 followers; 58.3%). The DISCERN score had a median of 25 (IQR: 22–29), and most of the videos obtained very low Global Quality Scale (GQS) scores (median 2; IQR: 1–2). A very strong correlation was observed between DISCERN and GQS (r = 0.890, p < 0.001). Videos addressing actinic cheilitis obtained slightly higher scores, although the differences were not statistically significant.
Conclusion
The overall quality of TikTok videos on potentially malignant oral disorders was low. DISCERN and GQS scores showed a strong correlation, and quality was not related to popularity or engagement metrics. These findings highlight the importance of healthcare professionals contributing accurate information to social media, and the need to adapt or develop evaluation tools for video-based content.
Keywords: Discern, Global Quality Scale (GQS), Leucoplakia, Oral lichen planus, Tiktok
Introduction
Potentially malignant oral disorders (PMODs) encompass a variety of conditions affecting the oral mucosa, characterized by an elevated risk of progressing to malignancy [1–3]. Oral cancer represents a significant public health concern, with a rising incidence particularly among young adults. Often, oral cancer is preceded by PMODs, providing a crucial window for early diagnosis and the potential to decrease both morbidity and mortality rates [4–6]. The global prevalence of PMODs is estimated to be around 4.5%, though this figure varies considerably across different geographical regions [2].
The most frequently encountered types of PMODs in clinical settings are leukoplakia (OL), lichen planus (OLP), oral lichenoid lesions (OLL), oral submucous fibrosis (OSF), proliferative verrucous leukoplakia (PVL), erythroplasia, and actinic cheilitis. Less common pathological conditions include oral discoid lupus erythematosus and graft-versus-host disease [1, 4].
PMODs are related to numerous etiological factors, both genetic and environmental. Solar radiation, tobacco use, alcohol use, and betel nut chewing, have been associated with a higher frequency of the development of specific lesions. OL has been consistently linked with the consumption of tobacco and alcohol, while betel nut chewing is associated with OSF, which is highly prevalent in Asian populations [5]. On the other hand, OLP and PVL do not show a constant association with known environmental agents, so that it is believed that genetic factors or other causes that have yet to be identified may intervene in their etiopathogenesis [2, 6].
Access to medical information has changed: the general population increasingly resorts to the use of the internet, due to its easy access and broad dissemination [7]; TikTok is an extremely popular social media platform, with more than one billion users, and a growing source of information. It allows reaching many people, generate influence, or become part of group trends [8, 9].
TikTok offers a notable potential as a communication tool for health-related information, but there are important concerns about the precision and reliability of the content disseminated [7–11]. Anyone can upload videos without having professional training in health, which creates a scenario in which the quality of information is not guaranteed. This could derive into the dissemination of messages that are inexact or confusing, which negatively affect the public perception and the attitudes towards important health issues. There is no system of censorship or peer review, which negatively affects the reliability of the content [11–14]. In this context, it is fundamental for health professionals and institutions to critically analyze the content available in platforms such as TikTok and to assess their usefulness for health education [15, 16].
Particularly, the growing number of videos related with potentially malignant oral disorders on the platform raises the urgent need to examine its content [17–20]. Although these videos could be a useful way to increase awareness about these disorders, until now, the quality or the educational validity of the information presented in TikTok has not been evaluated. The research question guiding this study was: What is the quality, reliability, and scientific accuracy of TikTok videos related to potentially malignant oral disorders?
The objective was to answer this question by analyzing TikTok videos, assessing their source, duration, quality (DISCERN and GQS), and engagement metrics.
Materials and methods
A cross-sectional analysis of the content on potentially malignant oral lesions was performed on TikTok. Given that the content of the TikTok videos are considered public data, it was not necessary to obtain approval from the local research ethics committee.
A- search strategy
For each hashtag, the first 100 videos returned by the TikTok algorithm were screened, following established methodologies for social media studies [12, 20] (Fig. 1).
Fig. 1.
Flow diagram
B- eligibility criteria
The search hashtags were defined: #leukoplakia, #lichenplanus, potentially malignant disorders (PMOD), #actiniccheilitis, #orallichenoidlesions (OLL), and #erythroplasia (EP).
Inclusion criteria were: (i) English- or Spanish-language videos, (ii) directly addressing PMODs, and (iii) accessible to the public. Exclusion criteria were: (i) duplicates, (ii) videos without sound or narration, (iii) languages other than English or Spanish, (iv) unrelated content, and (v) advertising or commercial videos.
C- data collection
The following variables were recorded Characteristics of the videos, including duration (in seconds), time elapsed since publication, number of views, number of “likes,” shares, comments, and times the video was saved. Additionally, two indices were calculated [11, 16]:
Viewing index: calculated by dividing the number of views by the number of days since the video was uploaded.
Engagement index : determined by adding the number of “likes,” comments, and times saved, and then dividing by the total number of views.
For clarity, the variables were grouped into the following categories:
-
(i)
Video characteristics: duration, time since publication, language, and presence of sound.
-
(ii)
Viewing and engagement index.
-
(iii)
Content creator characteristics: type of uploader (healthcare professional vs. non-professional) and influencer classification (nano, micro, medium, macro, mega).
-
(iv)
Content quality measures: DISCERN and Global Quality Scale (GQS) scores.
Content creators were also classified according to their type of influencer, based on their follower count:
Nano-influencers: 1 to 10,000 followers; Micro-influencers: 10,000 to 50,000; Medium-influencers: 50,000 to 500,000; Macro-influencers: 500,000 to 1 million; and Mega-influencers: 1 to 5 million followers [12, 15].
D- evaluation and scoring systems
The DISCERN instrument, developed by Charnock et al. [21] (Table 1), was used to assess the reliability and quality of the information. It consists of 16 items: questions 1–8 evaluate reliability, questions 9–15 assess the quality of the information, and question 16 provides an overall rating. Each item is scored from 1 to 5, and the total score is classified as Excellent (≥ 63), Good (51–62), Acceptable (39–50), Poor (27–38), or Very Poor (< 26).
Table 1.
The 16 DISCERN questions, to be responded using a 1–5 score
| 1. Are the aims clear? |
| 2. Does it achieve its aims? |
| 3. Is it relevant? |
| 4. Is it clear what sources of information were used to compile the publication (other than the author or producer)? |
| 5. Is it clear when the information used or reported in the publication was produced? |
| 6. Is it balanced and unbiased? |
| 7. Does it provide details of additional sources of support and information? |
| 8. Does it refer to areas of uncertainty? |
| 9. Does it describe how each treatment works? |
| 10. Does it describe the benefits of each treatment? |
| 11. Does it describe the risks of each treatment? |
| 12. Does it describe what would happen if no treatment is used? |
| 13. Does it describe how the treatment choices affect overall quality of life? |
| 14. Is it clear that there may be more than one possible treatment choice? |
| 15. Does it provide support for shared decision-making? |
| 16. Based on the answers to all of the above questions, rate the overall quality of the publication as a source of information about treatment choices. |
Videos were also rated on a 1–5 scale, where scores of 1–2 indicated low quality with limited patient information, 3 indicated medium quality with partially useful content, and 4–5 indicated good to excellent quality with clear and educational content. Scores < 3 were considered low educational value, while scores > 3 were considered good educational value [11, 21–23] (Table 2).
Table 2.
General characteristics of the videos analyzed
| Characteristics | N | % | |
|---|---|---|---|
| Group | Oral Lichen Planus | 26 | 27.1 |
| Oral Leukoplakia | 24 | 25.0 | |
| Potentially Malignant Oral Disorders | 21 | 21.9 | |
| Actinic Cheilitis / Actinic Keratosis | 25 | 26.0 | |
| Content creator | Nano influencer | 56 | 58.3 |
| Micro influencer | 17 | 17.7 | |
| Mid-tier influencer | 16 | 16.7 | |
| Macro influencer | 4 | 4.2 | |
| Mega influencer | 3 | 3.1 | |
| Uploader | Healthcare professional | 56 | 58.3 |
| Non-professional | 40 | 41.7 | |
| Video length | 0–15 s | 10 | 10.4 |
| 16–30 s | 21 | 21.9 | |
| 31–45 s | 12 | 12.5 | |
| ≥ 46 s or more | 53 | 55.2 | |
| GQS | Score 1 | 49 | 51.0 |
| Score 2 | 41 | 42.7 | |
| Score 3 | 5 | 5.2 | |
| Score 4 | 1 | 1.0 | |
The scientific accuracy of the content was evaluated against World Health Organization (WHO) guidelines on potentially malignant oral disorders [1, 6].
Inter-rater agreement
Two evaluators independently assessed each video using the DISCERN and Global Quality Scale (GQS) tools. The level of concordance between evaluators was calculated using Cohen’s Kappa coefficient (κ). Agreement was high (κ = 0.89), and any discrepancies were resolved through consultation with a third reviewer.
E- data analysis
The data were processed using the JAMOVI statistical software, version 2.4.8 (R Core Team, 2022; The jamovi Project, 2023) [24, 25]. Descriptive analyses were performed, reporting medians and interquartile ranges (IQR) for quantitative, and frequencies and percentages for qualitative variables. The Kolmogorov-Smirnov test indicated that the data did not follow a normal distribution and variance homogeneity was not observed across groups. Consequently, Spearman’s Rho was applied for the correlation analysis. The Mann-Whitney U test was used to compare the results on the type of uploader, and the Kruskal-Wallis test was used to compare the results according to the group of videos with respect to the lesions. The contingency tables were created to compare the sets of qualitative variables. The level of significance was established according to p < 0.05. The effect size was defined by following the recommendations by Domínguez-Lara [26].
Results
A total of 96 videos related with potentially malignant oral lesions were analyzed. These corresponded to oral lichen planus (27.1%), oral leukoplakia 25% (Table 1) Most of the videos were created by nano-influencers (58.3%) followed by micro-influencers (17.7%).
The analysis of the quantitative metrics showed that the number of followers of the content creators had a median of 27,500 (IQR: 4,169–84,100) (Table 3). The number of views per video was low, with a median of 23 (IQR: 8–83).
Table 3.
Metrics analyzed and overall evaluation (Median and Interquartile Range, IQR)
| Variable | Median (IQR) |
|---|---|
| Number of followers | 27,500 (4,169–84,100) |
| Number of views | 23 (8–83) |
| Number of likes | 20 (12–2,021) |
| Number of comments | 10 (1–23) |
| Number of saves | 21 (20–2,020) |
| TOTAL DISCERN | 25 (22–29.5) |
| Global Quality Score (GQS) | 2 (1–2) |
The number of times the videos were saved reached a median of 21 (IQR: 20–2,020). Regarding the quality of information, the TOTAL DISCERN score had a median of 25 (IQR: 22–29.5), and the Global Quality Score (GQS) presented a median of 2 (IQR: 1–2).
Correlations between quality scores and perceived evaluation were also examined (Table 4). The perceived overall evaluation showed a significant positive correlation with DISCERN (r = 0.655, p < 0.001) and with GQS (r = 0.614, p < 0.001). In contrast, most engagement metrics (views, likes, comments, followers) did not correlate with either DISCERN or GQS, except for a weak negative correlation between DISCERN and the number of times videos were saved (r = − 0.227, p < 0.05). Finally, DISCERN and GQS themselves showed an extremely high correlation (r = 0.890, p < 0.001), confirming the consistency between both scales (Fig. 2).
Table 4.
Correlations. TOTAL DISCERN score and the Global Quality Score (GQS) with the metrics of the published content
| TOTAL DISCERN | Global Quality Score | |
|---|---|---|
| Number of followers of uploader | -0.145 | -0.128 |
| Days since upload | -0.010 | -0.050 |
| Duration (sec) | 0.081 | 0.112 |
| Number of views | -0.093 | -0.141 |
| Number of likes | -0.097 | -0.123 |
| Number of comments | -0.100 | -0.117 |
| Saves | -.227* | -0.193 |
| Engagement Index | -0.048 | -0.079 |
| Viewing Index | -0.099 | -0.112 |
| Overall evaluation | .655** | .614** |
| TOTAL DISCERN | .890** |
Global Quality Score: GQS; *p<0.05; **p<0.001
Fig. 2.
Distribution of DISCERN scores across GQS categories
Table 5 provides an overview of the characteristics of the videos according to the type of lesion analyzed. It was observed that with respect to the type of content creator, no statistically significant differences were found in the distribution of the types of influencers among the different groups of lesions (χ² [12] = 13.3, p = 0.344; V = 0.215). Likewise, after analyzing the profile of the uploader, no significant differences were observed if it was a health professional or not (χ² [3] = 6.14, p = 0.105; V = 0.253). The duration of the videos did not vary significantly between the groups either (χ² [9] = 5.82, p = 0.758; V = 0.142). Lastly, the quality of the videos, evaluated through the Global Quality Score, did not show statistically significant differences between the groups (χ² [9] = 9.41, p = 0.400; V = 0.181).
Table 5.
Characteristics of videos published by health professionals and those made by non-professionals
| Oral Lichen Planus | Oral Leukoplakia | Potentially Malignant Oral Disorders | Actinic Cheilitis / Actinic Keratosis | |||||
|---|---|---|---|---|---|---|---|---|
| N | % | N | % | N | % | N | % | |
| Content creator (χ2(df) = 13.3(12); p = 0.344; V = 0.215) | ||||||||
| Nano influencer | 14 | 53.8 | 11 | 45.8 | 13 | 61.9 | 18 | 72.0 |
| Micro influencer | 6 | 23.1 | 6 | 25.0 | 3 | 14.3 | 2 | 8.0 |
| Mid-tier influencer | 6 | 23.1 | 3 | 12.5 | 3 | 14.3 | 4 | 16.0 |
| Macro influencer | - | - | 2 | 8.3 | 2 | 9.5 | - | - |
| Mega influencer | - | - | 2 | 8.3 | - | - | 1 | 4.0 |
| Uploader (χ2 (df) = 6.14(3); p = 0.105; V = 0.253) | ||||||||
| Healthcare professional | 13 | 50.0 | 19 | 79.2 | 12 | 57.1 | 12 | 48.0 |
| Non-professional | 13 | 50.0 | 5 | 20.8 | 9 | 42.9 | 13 | 52.0 |
| Video length (χ2 (d) = 5.82(9); p = 0.758; V = 0.142) | ||||||||
| 0–15 s | 3 | 11.5 | 4 | 16.7 | 1 | 4.8 | 2 | 8.0 |
| 16–30 s | 5 | 19.2 | 7 | 29.2 | 6 | 28.6 | 3 | 12.0 |
| 31–45 s | 4 | 15.4 | 3 | 12.5 | 2 | 9.5 | 3 | 12.0 |
| 46 s and more | 14 | 53.8 | 10 | 41.7 | 12 | 57.1 | 17 | 68.0 |
| GQS (χ2 (df) = 9.41(9); p = 0.400; V = 0.181) | ||||||||
| Score 1 | 15 | 57.7 | 12 | 50.0 | 12 | 57.1 | 10 | 40.0 |
| Score 2 | 11 | 42.3 | 9 | 37.5 | 9 | 42.9 | 12 | 48.0 |
| Score 3 | - | - | 2 | 8.3 | - | 0.0 | 3 | 12.0 |
| Score 4 | - | - | 1 | 4.2 | - | - | - | - |
Global Quality Score: GQS Note: χ2: chi squared; df: degrees of freedom; V: Cramer’s V; low effect: 0.10; medium effect: 0.30; large effect: 0.50
An analysis was performed to verify if there were differences between the videos published by health professionals and those published by non-professionals (Table 6). With respect to the type of influencer (p = 0.065) or duration of the video (p = 0.089), no significant differences were observed. As for the quality of the information (GQS), no statistically significant differences were found (χ² [3] = 1.89, p = 0.0595; V = 0.140). Nevertheless, the health professionals obtained a more favorable distribution, with higher proportions of GQS scores of 3 (7.1%) and 4 (1.8%).
Table 6.
Comparison of quantitative metrics by type of oral disorder (Median and Interquartile Range, IQR) and Kruskal–Wallis test results
| Healthcare professionals | Non-professionals | |||
|---|---|---|---|---|
| N | % | N | % | |
| Content creator (χ2 (df) = 8.84(4); p = 0.065; V = 0.303) | ||||
| Nano influencer | 33 | 58.9 | 23 | 57.5 |
| Micro influencer | 6 | 10.7 | 11 | 27.5 |
| Mid-tier influencer | 10 | 17.9 | 6 | 15.0 |
| Macro influencer | 4 | 7.1 | - | - |
| Mega influencer | 3 | 5.4 | - | - |
| Video length (χ2 (df) = 6.51(3); p = 0.089; V = 0.260) | ||||
| 0–15 s | 8 | 14.3 | 2 | 5.0 |
| 16–30 s | 11 | 19.6 | 10 | 25.0 |
| 31–45 s | 10 | 17.9 | 2 | 5.0 |
| 46 s and more | 27 | 48.2 | 26 | 65.0 |
| GQS (χ2 (df) = 1.89(3); p = 0.0595; V = 0.140) | ||||
| Score 1 | 27 | 48.2 | 22 | 55.0 |
| Score 2 | 24 | 42.9 | 17 | 42.5 |
| Score 3 | 4 | 7.1 | 1 | 2.5 |
| Score 4 | 1 | 1.8 | - | - |
Global Quality Score: GQS: χ2: chi squared; df: degrees of freedom; V: Cramer’s V; low effect: 0.10; medium effect: 0.30; large effect: 0.50
When comparing the metrics related with the videos, as a function of the type of oral lesion addressed (Oral Lichen Planus, Oral Leukoplakia, Potentially Malignant Oral Disorders –PMOD-, and Actinic Cheilitis) (Table 7), In the comparison of metrics by type of oral disorder using the Kruskal–Wallis test, no statistically significant differences were found in the number of followers, views, or saves (p > 0.05). However, significant differences were observed in the number of likes (p = 0.005; η²H = 0.131) and comments (p = 0.002; η²H = 0.156), with small-to-moderate effect sizes, suggesting that these variables may vary depending on the type of lesion presented in the videos.
Table 7.
Comparison of quantitative metrics by type of oral disorder (Median and Interquartile Range, IQR) and Kruskal–Wallis test results
| Variable | Oral Lichen Planus | Oral Leukoplakia | Potentially Malignant Oral Disorders | Actinic Cheilitis | H statistic | p-value | η²H |
|---|---|---|---|---|---|---|---|
| Number of followers | 27,500 (2,070–63,150) | 23,500 (3,726–80,775) | 12,878 (3,994–76,425) | 59,700 (33,050–108,225) | 4.65 | 0.199 | 0.022 |
| Number of views | 23 (21–39) | 23 (19–239) | 9 (5–23) | 21 (6–2,023) | 7.65 | 0.054 | 0.062 |
| Likes | 20 (10–20) | 19 (11–25) | 2,022 (20–2,024) | 20 (1–2,021) | 12.80 | 0.005* | 0.131 |
| Comments | 23 (10–23) | 9 (5–23) | 1 (0–8) | 10 (0–23) | 14.67 | 0.002* | 0.156 |
| Saves | 20 (20–2,023) | 285 (20–2,022) | 20 (10–38) | 24 (20–50) | 4.53 | 0.210 | 0.020 |
| TOTAL DISCERN | 25 (22–29) | 24 (21–28) | 25 (22–30) | 25 (23–32) | 3.10 | 0.377 | 0.015 |
| Global Quality Score (GQS) | 2 (1–2) | 2 (1–2) | 2 (1–2) | 2 (1–2) | 1.27 | 0.737 | 0.008 |
Medians and IQR in parentheses. Kruskal–Wallis H statistic shown with corresponding p-values and eta squared (η²H). p < 0.05 considered statistically significant
Although no statistically significant differences were detected, videos addressing actinic cheilitis showed slightly higher DISCERN scores compared with the other groups (25; IQR: 23–32 vs. 24–25 in the others). Regarding the GQS, median values were homogeneous across all lesion types (2; IQR: 1–2).
No statistically significant differences were observed in the quantitative metrics analyzed according to uploader profile. The medians of followers, views, likes, comments, saves, DISCERN, and GQS were similar between professionals and non-professionals (all p > 0.05), and the effect sizes were small (rrb < 0.20).
Discussion
TikTok is gaining popularity worldwide, and many research studies have been published on it. The results from the present study show that the quality of the information about PMOD available on TikTok is low, both in terms of reliability (DISCERN scores) and overall quality (GQS). The DISCERN score had a median of 25 (IQR: 22–29) and most of the videos obtained very low GQS scores. This could be attributed, in part, to the absence of editorial or review mechanisms in the digital platforms, which allows for the massive circulation of content without scientific filters [7, 8, 11].
Many studies coincide that short-video digital platforms, such as TikTok, are a dynamic, accessible, and plural environments that facilitate the dissemination of medical knowledge to broad and heterogeneous audiences [12–15]. These tools play an important role in the collective awareness on diverse health problems, among which we find potentially malignant oral disorders (PMOD), which comprise a clinically-diverse groups of entities that due to the lack of adequate management, can progress to become an oral squamous cell carcinoma (OSCC) [1–3].
Managing PMOD presents a challenge due to the lack of specific early markers that reliably indicate malignant transformation. In this sense, the monitoring of habits, an early clinical diagnosis, rigorous follow-up, and opportune therapeutic interventions, are fundamental elements for reducing the risk of progression (4–5). Therefore, having clear, accessible, and evidence-based information is essential for promoting an adequate understanding by the population and to contribute to the prevention of OSCC [2, 6].
Just as recent studies centered on the analysis of the health content on TikTok (11,16.), the present study used widely accepted tools to assess quality, such as DISCERN and Global Quality [26], observing an extremely strong correlation between the scores obtained in the TOTAL DISCERN and the Global Quality Scale (GQS) (r = 0.890, p < 0.001).
Many studies have pointed out that the quality of the information about potentially malignant oral disorders on YouTube™ is, in general, low. Thus, with respect to oral leukoplakia, Romano et al. (2021) [17] analyzed the quality of the videos available in this platform, finding that more than 35% contained misleading or potentially dangerous information. The videos from reliable sources, such as universities or official medical sites, obtained significantly higher scores. Similarly, Kovalski et al. [18] evaluated 28 English-language videos about oral leukoplakia on YouTube™. Most were classified as low quality, with limited usefulness and reliability.
Regarding oral lichen planus, Morais et al. [19] reviewed 481 YouTube™ videos on the topic, but only 37 met the criteria for inclusion. Most presented content with a poor-reliability and low quality. In addition, a positive relationship was observed between the duration of the video and its quality, which could indicate that the contents of longer duration allow addressing the topics more deeply and with greater clarity. Thus, the longer duration of the videos on YouTube, as compared to TikTok, allow developing the topics better, therefore offering contents with a more educational approach [11].
In our study, the quantitative metrics of popularity (views, likes, interaction) did not show relevant differences among the groups of lesions. However, videos related to actinic cheilitis obtained slightly higher TOTAL DISCERN scores, although the differences were not statistically significant.
The quality of the information was not aligned with popularity, which is concerning, as it can lead to patients trusting the inexact content solely due to its viral characteristic [10, 12–15]. Our results showed that the number of times that a video was saved was negatively correlated with the DISCERN score (r=-0.227, p < 0.05).
The increasing circulation of health content on the internet, in an environment without clear regulations, requires users to develop critical skills to recognize reliable information [7, 14, 15]. This situation is even more complex in specific sectors of the population, such as older adults, who must face significant barriers due to their low level of digital literacy and their limited access to technology [7–9].
In this context, the role of health professionals becomes crucial. It is necessary for them, along with education institutions and governmental organizations, to assume an active commitment in the production and dissemination of online content that is truthful, accessible, and evidence-based. In our study, no significant differences in video quality were found between those created by healthcare professionals and those made by non-professionals. This finding does not match what has been reported on other platforms, where professional content usually scores higher [12, 15, 26]. A possible explanation is that TikTok’s short format and informal style limit the ability to provide detailed information, thereby reducing the advantage that professionals might otherwise have over other creators.
The characteristics of the platforms themselves, such as the brevity of the format, the informal style, or the demands from the algorithms, impose important limitations when transmitting rigorous contents [27–31]. In addition to this, we find the frequent use of technical language, which can make the understanding of the general public more difficult. In some cases, the visibility and popularity of the content is prioritized over their educational value, which can compromise its usefulness from an educational point of view.
This study has some limitations. It was cross-sectional, and TikTok is a dynamic platform, so search results may change over time. The evaluation tools (DISCERN and GQS) are subjective, even if agreement between reviewers was high. The search was based on a limited set of hashtags, so other keywords could have produced different results. Finally, only English- and Spanish-language videos were included, which restricts the generalization of the findings.
Future studies could complement these tools with instruments like the Patient Education Materials Assessment Tool (PEMAT) [27, 32], which evaluates understandability and actionability, thus providing a more comprehensive assessment of the educational value of digital content.
Conclusions
The overall quality of TikTok videos about PMODs was low. Quality was not related to popularity or engagement metrics, indicating that greater reach does not necessarily mean better information. Although videos on actinic cheilitis achieved slightly higher scores, the differences were not significant. Taken together, these results highlight the importance of healthcare professionals becoming more involved in creating reliable and accessible content, so that the information reaching the public on widely used platforms such as TikTok is truly valuable.
Acknowledgements
Not applicable.
Abbreviations
- PMODs
Potentially Malignant Oral Disorders
- WHO
World Health Organization
- N/A
Not Applicable
Authors’ contributions
Pia Lopez -Jornet data curation, formal analysis, methodology, visualization, investigation, original draft; Layla Bennouna data curation, formal analysis, methodology, visualization, investigation; Francisco Parra-Perez visualization, investigation; Eduardo Pons-Fuster visualization, investigation. All authors have read and approved the manuscript.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.
Declarations
Ethics approval and consent to participate
Not applicable. This study did not involve human participants or animals.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Warnakulasuriya S, Kujan O, Aguirre-Urizar JM, Bagan JV, González-Moles MÁ, Kerr AR, Lodi G, Mello FW, Monteiro L, Ogden GR, Sloan P, Johnson NW. Oral potentially malignant disorders: A consensus report from an international seminar on nomenclature and classification, convened by the WHO Collaborating Centre for Oral Cancer. Oral Dis. 2021;27(8):1862–80. 10.1111/odi.13704. [DOI] [PubMed] [Google Scholar]
- 2.Iocca O, Sollecito TP, Alawi F, Weinstein GS, Newman JG, De Virgilio A, Di Maio P, Spriano G, Pardiñas López S, Shanti RM. Potentially malignant disorders of the oral cavity and oral dysplasia: A systematic review and meta-analysis of malignant transformation rate by subtype. Head Neck. 2020;42(3):539–55. 10.1002/hed.26006. [DOI] [PubMed] [Google Scholar]
- 3.Warnakulasuriya S, Kerr AR. Oral Cancer Screening: Past, Present, and Future. J Dent Res. 2021;100(12):1313–20. 10.1177/00220345211014795. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Warnakulasuriya S. Oral potentially malignant disorders: A comprehensive review on clinical aspects and management. Oral Oncol. 2020;102:104550. 10.1016/j.oraloncology.2019.104550. [DOI] [PubMed] [Google Scholar]
- 5.Ganesh D, Sreenivasan P, Öhman J, Wallström M, Braz-Silva PH, Giglio D, Kjeller G, Hasséus B. Potentially Malignant Oral Disorders and Cancer Transformation. Anticancer Res. 2018;38(6):3223–9. 10.21873/anticanres.12587. [DOI] [PubMed] [Google Scholar]
- 6.Villa A, Lodolo M, Ha P. Oncological Outcomes of Patients With Oral Potentially Malignant Disorders. JAMA Otolaryngol Head Neck Surg. 2025;151(1):65–71. 10.1001/jamaoto.2024.3719. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Walsh-Buhi ER. Social Media and Cancer Misinformation: Additional Platforms to Explore. Am J Public Health. 2020;110(S3):S292–3. 10.2105/AJPH.2020.305949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Comp G, Dyer S, Gottlieb M. Is TikTok The Next Social Media Frontier for Medicine? AEM Educ Train. 2020;5(3). 10.1002/aet2.10532. [DOI] [PMC free article] [PubMed]
- 9.Calice MN, Bao L, Beets B, Brossard D, Scheufele DA, Feinstein NW, Heisler L, Tangen T, Handelsman J. A triangulated approach for understanding scientists’ perceptions of public engagement with science. Public Underst Sci. 2023;32(3):389–406. 10.1177/09636625221122285. [DOI] [PubMed] [Google Scholar]
- 10.Chen Y, Wang Q, Huang X, Zhang Y, Li Y, Ni T, Pan G, Luo D, Ni Y. The quality and reliability of short videos about thyroid nodules on BiliBili and TikTok: Cross-sectional study. Digit Health. 2024;10:20552076241288831. 10.1177/20552076241288831. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Cannatà D, Galdi M, Russo A, Scelza C, Michelotti A, Martina S. Reliability and Educational Suitability of TikTok Videos as a Source of Information on Sleep and Awake Bruxism: A Cross-Sectional Analysis. J Oral Rehabil. 2025;52(4):434–42. 10.1111/joor.13874. [DOI] [PubMed] [Google Scholar]
- 12.Ranpariya V, Chu B, Fathy R, Lipoff JB. Dermatology without dermatologists? Analyzing Instagram influencers with dermatology-related hashtags. J Am Acad Dermatol. 2020;83(6):1840–2. 10.1016/j.jaad.2020.05.039. [DOI] [PubMed] [Google Scholar]
- 13.Booth L, Aldaihani A, Davidson J, et al. Misinformation and Readability of Social Media Content on Pediatric Ankyloglossia and Other Oral Ties. JAMA Otolaryngol Head Neck Surg. 2025;151(2):143–50. 10.1001/jamaoto.2024.4211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Johnson-Arbor K. I Read Health Information Online—Can I Trust It? JAMA Intern Med. 2024;184(9):1138. 10.1001/jamainternmed.2024.0340. [DOI] [PubMed] [Google Scholar]
- 15.Vo K, Ezra N. Dermatology TikTok™ videos: applicable considerations. Clin Exp Dermatol. 2023;48(4):387–8. 10.1093/ced/llac080. [DOI] [PubMed] [Google Scholar]
- 16.Passos KK, Leonel AC, Bonan PR, Castro JF, Pontual ML, Ramos-Perez FM, Perez DE. Quality of information about oral cancer in Brazilian Portuguese available on Google, Youtube, and Instagram. Med Oral Patol Oral Cir Bucal. 2020;25(3):e346–52. 10.4317/medoral.23374. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Romano A, Lauritano D, Fiori F, Di Petrillo M, Hasan I, Lucchese A, Serpico R, Di Stasio D. Cross-sectional study on the quality of oral lichen planus videos on YouTube™. J Oral Pathol Med. 2021;50(2):220–8. 10.1111/jop.13128. [DOI] [PubMed] [Google Scholar]
- 18.Kovalski LNS, Cardoso FB, D’Avila OP, Corrêa APB, Martins MAT, Martins MD, Carrard VC. Is the YouTube™ an useful source of information on oral leukoplakia? Oral Dis. 2019;25(8):1897–905. 10.1111/odi.13161. [DOI] [PubMed] [Google Scholar]
- 19.Morais EF, Felix FA, dos Santos JLM, Martins HDD, Barboza CAG, Freitas RA. YouTube™ and oral lichen planus: an appraisal of the educational quality of information. Brazilian Oral Res. 2021;35:e006. 10.1590/1807-3107bor-2021.vol35.0006. [DOI] [PubMed] [Google Scholar]
- 20.Naseer S, Hasan S, Bhuiyan J, Prasad A. Current Public Trends in the Discussion of Dry Eyes: A Cross-Sectional Analysis of Popular Content on TikTok. Cureus. 2022; 14(2): e 22702. 10.7759/cureus.22702 [DOI] [PMC free article] [PubMed]
- 21.CharnockD SS, NeedhamG. GannR. DISCERN: an instrument for judging the quality ofwritten consumer health information on treatment choices. J Epidemiol Community Health. 1999;53(2):105–11. 10.1136/ jech.53.2.105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Meade MJ, Dreyer CW. Analysis of the Information Contained Within TikTok Videos Regarding Orthodontic Retention. J World Federation Orthodontists. 2022;11:170–5. [DOI] [PubMed] [Google Scholar]
- 23.Ceylan Şen S, Paksoy T, Göller Bulut D, Ustaoğlu G. Does YouTube™ provide reliable information on oral candidiasis? Oral Dis. 2023;29(1):290–9. 10.1111/odi.14338. [DOI] [PubMed] [Google Scholar]
- 24.The jamovi project. (2023). jamovi. (Version 2.4) [Computer Software]. Retrieved from https://www.jamovi.org
- 25.R Core Team. (2022). R: A Language and environment for statistical computing. (Version 4.1) [Computer software]. Retrieved from https://cran.r-project.org. (R packages retrieved from CRAN snapshot 2023-04-07).
- 26.Domínguez-Lara S. Effect size, a quick guide. Educación Médica. 2018;19(4):251–4. 10.1016/j.edumed.2017.07.002. [Google Scholar]
- 27.Paksoy T, Ceylan Şen S, Ustaoğlu G, Bulut DG. What do TikTok videos offer us about dental implants treatment? J Stomatol Oral Maxillofac Surg. 2023;124(1S):101320. 10.1016/j.jormas.2022.10.019. [DOI] [PubMed] [Google Scholar]
- 28.Oliveira Júnior JDA, Silva AVMV, Montalvany-Antonucci CC, Maia GA, Carvalho KRJ, Macari S. Are YouTube™ and TikTok™ videos useful as educational tool for patients with cleft lip and palate? Dent Press J Orthod. 2025;29(6):e2424151. 10.1590/2177-6709.29.6.e2424151.oar. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.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(1):9189. 10.1038/s41598-025-93986-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Basch CH, Donelle L, Fera J, Jaime C. Deconstructing TikTok Videos on Mental Health: Cross-sectional, Descriptive Content Analysis. JMIR Form Res. 2022;6(5):e38340. 10.2196/38340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Amin S, Chae SW, Washington PY, Okamoto SK, Youkey T, Pokhrel P. A mixed-method exploration of #vapingcessation videos on TikTok. J Addict Dis 2024 Oct 15:1–9. 10.1080/10550887.2024.2414139 [DOI] [PubMed]
- 32.Will J, Gupta M, Zaretsky J, Dowlath A, Testa P, Feldman J. Enhancing the Readability of Online Patient Education Materials Using Large Language Models: Cross-Sectional Study. J Med Internet Res. 2025;27:e69955. 10.2196/69955. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.