#threadlift: Information Quality Analysis on YouTube, Instagram, and TikTok

Introduction

The human face changes with age. This change is characterized by thinning of the skin, loss of elasticity, loss of volume, and the face sagging downward in the direction of gravity. The deepened nasolabial sulcus, sagging cheeks, and non-defined jawline create an aged and tired appearance [1–3]. Although surgical methods are successful in the treatment, there is a search for alternative minimally invasive procedures to surgery due to its complications and the long and difficult recovery period [2, 3]. Thread lifting is a minimally invasive method that aims to improve the signs of aging via subcutaneous placement of sutures. The short- and long-term results of this treatment are still controversial [4, 5].
The demand for cosmetic procedures is increasing. Patients prefer minimally invasive procedures because of their short recovery period after treatment [6]. Social media has also a significant impact on the demand for such procedures. Many people considering cosmetic procedures use online resources, such as informations posted by patients and pysycians as well as advertisements, and educational materials [7, 8]. The rapid increase in the use of social media has caused changes in the methods of obtaining information and patients’ treatment preferences [9].

Objective

This study aimed to examine the quality of health professionals’ YouTube, Instagram, and TikTok content on thread lifting, reveal whether there is a difference between platforms and specialties, and reveal an approach to health professionals’ information sharing.

Material and methods

Search and sampling method

Within the scope of the study, the videos were accessed in January 2024 by using the search bar of three social media platforms: YouTube, Instagram, and TikTok, with the hashtag #threadlift. In the search results, all videos in the language that the researchers sufficiently understood were analyzed. To adequately evaluate the informative features of video content, videos with a duration of over 1 minute were included in the evaluation. All uploaded videos were submitted by healthcare personnel. Videos sharing personal experiences, containing only pictures and animations and advertisements, videos not containing informative content were excluded from assessment.

Quality of information

The number of likes, comments, views, and video uploaders were recorded during the review. The information provided in the video content was analyzed by two dermatologists using the 16-question DISCERN form and the Global Quality Scale (GQS); both scored 1–5 [10, 11]. The DISCERN scale evaluates a publication in 3 sections: reliability, quality of information about treatment modalities, and overall evaluation. The GQS evaluates a video in terms of the quality of the information it contains, the flow of the video, and the usefulness of the video for patients. After scoring each question on the DISCERN form, an assessment was made as a sum of the scores in each section and the total DISCERN score. In the evaluation based on the score, the scores were divided into five categories, and the video quality was classified as 1 “very poor”, 2 “poor”, 3 “good”, 4 “very good” and 5 “excellent”.

Statistical analysis

Descriptive data are presented as the mean, standard deviation, minimum, and maximum. Kappa (κ) and intraclass correlation coefficients (ICC) were used to assess the agreement between assessors. At a 95% confidence interval for the κ coefficient, a κ value of 0–0.20 indicates “none”, 0.21–0.39 “minimal”, 0.40–0.59 “weak”, 0.60–0.79 “moderate”, 0.80–0.90 “strong” and above 0.90 “almost perfect” agreement. Intraclass correlation coefficient (ICC) analysis was performed with 95% confidence intervals (CI), taking into account the two-way random model, mean rating (k = 2) consistency, and Pearson’s correlation method. At a 95% confidence interval, ICC coefficient values below 0.5 mean “Poor” reliability, values between 0.5 and 0.75 mean “Moderate”, values between 0.75 and 0.90 mean “Good” and values above 0.90 mean “Excellent” reliability. Spearman’s correlation test was used to analyze the correlation between the overall characteristics of the videos. The Kruskal-Wallis test was used to examine differences in video parameters depending on the video source, and post-hoc analysis was performed using the Mann-Whitney test. A p-value less than 0.05 was considered statistically significant, and statistical analysis was performed using SPSS version 21.0 for Windows (SPSS Inc., Chicago, IL, USA).

Results

A total of 182 thread-lifting videos were evaluated. 54% of the videos (n = 98) were posted on YouTube, 27% on Instagram (n = 49) and 19% on TikTok (n = 35). 41% of uploaders (n = 74) were medical doctors (MD) without specialty provided, 25% plastic surgeons (n = 46), 21% nurses (n = 39), and 13% dermatologists (n = 23). The analysis of the videos according to platforms and uploaders is shown in figure 1. A significant difference was found between video uploaders and platforms (p = 0.004). Plastic surgery specialists preferred using YouTube more than nurses, whereas nurses preferred using TikTok more than plastic surgeons (p = 0.001, p = 0.001). No differences were detected for dermatologists and MDs. The average number of views is 83,205.68, likes 1175.44, comments 71.36. The number of views, likes, and comments on the videos are summarized in table 1 according to the platforms.
The inter-rater agreement of each item and the GQS score was statistically significant after evaluation using the DISCERN scale (p = 0.001). ICC ranged between 0.721 and 0.989, the Pearson correlation coefficient ranged between 0.797 and 0.978, and the agreement was found to be “good” and “excellent”. The κ coefficient ranged between 0.788 and 0.945, and the agreement was between “strong” and “almost perfect”. In evaluating the total score of the sections on the DISCERN scale, the ICC coefficient for section 1 was 0.945, the κ coefficient was 0.707, the ICC coefficient for section 2 was 0.985, and κ was 0.649. The agreement was “excellent” regarding ICC and “moderate” regarding κ. For the total score of the DISCERN scale, the ICC coefficient was 0.954, the Pearson correlation coefficient was 0.969, and “excellent” agreement was observed, while the κ coefficient was 0.488 and the agreement was “poor”. The ICC coefficient was 0.950, the Pearson correlation coefficient was 0.911, and the κ coefficient was 0.866 in the inter-rater agreement of the GQS results. The agreement was “excellent” in terms of ICC and “strong” in terms of κ (table 2).
In the video rating made by dividing the scores obtained after DISCERN scoring into five categories, for section 1 on the reliability of the video content, the video content is “very poor” and “poor” on Instagram (n = 49) and TikTok (n = 35). On YouTube (n = 98), 8.1% of videos are “very poor”, 88.7% of videos are “poor” and 3% of videos are “good” (R = 0.86, ICC = 0.922, κ = 0.846, p = 0.001). For section 2, which evaluates the quality of the information in the video content, 21.4% of YouTube videos were classified as “good”, 76.5% as “poor” and 2% as “very poor”; 4% of Instagram videos were classified as “good”, 71.4% as “poor” and 24.4% as “very poor”; 48.5% of TikTok videos were classified as “very poor” and 51.4% as “poor” (R = 0.9, ICC = 0.947, κ = 0.870, p = 0.001). In section 3, where the overall quality of the videos was evaluated, 7.1% of YouTube videos were classified as “very good”, 28.5% as “good”, 57.1% as “poor” and 7.1% as “very poor”. 10% of Instagram videos were classified as “good”, 38.7% as “poor” and 51% as “very poor”. Of the TikTok videos, 28.5% were classified as “poor” and 71.4% as “very poor” (R = 0.913, ICC = 0.912, κ = 0.823, p = 0.001). In the evaluation of the total DISCERN score, 13.1% of the videos (n = 182) are “very poor”, 82.6% are “poor” and 4.3% are “good”. In the GQS (n = 182), 54.3% of the videos are not useful, 30.7% are slightly useful, 8.7% are moderately useful and 6% are useful (R = 0.92, ICC = 0.957, κ = 0.908, p = 0.001).
In the GQS evaluation of Instagram (n = 49) and TikTok (n = 35) videos, video content is useless (95.9% and 100%, respectively). On Instagram, 4.1% of videos were rated as less useful. YouTube (n = 98) videos were found to be useful (11.2%), 16.3% moderately useful, 55.1% slightly useful and 17.3% not useful, according to the GQS score. The classification of the videos according to the total DISCERN score and GQS evaluation are summarized in table 3. A similar inter-rater agreement was observed in the evaluation results according to the platforms (p = 0.001).
There was a significant difference (p = 0.001) in DISCERN scoring between YouTube, Instagram, and TikTok in terms of section 1, section 2, and section 3 (table 4). The average YouTube video score was higher than that of other platforms. Instagram videos were rated higher than TikTok videos. The lowest score was observed for TikTok. In the quality assessment based on the video uploaders, a significant difference between the groups was determined in the total DISCERN and GQS score (p = 0.002, p = 0.021, respectively), and no difference was observed in the analysis made within the platforms (p > 0.05). As a result of further analysis, the GQS score between MDs and nurses, DISCERN score between MDs and plastic surgeons, and GQS and DISCERN scores between plastic surgeons and nurses were different (p = 0.011, p = 0.003, p = 0.05, respectively).
The analysis indicates weak correlations between the total number of video comments, the total DISCERN score, the total number of views, and the GQS. A weak correlation was observed between Instagram video comments and the DISCERN score (R = 0.353, p = 0.01) and between the total number of comments and the DISCERN score (R = 0.221, p = 0.01). A weak correlation was also detected between the total number of views and comments and the GQS (R = 0.174, p = 0.02; R = 0.184, p = 0.01, respectively).

Discussion

In our study, which analyzed content shared by healthcare professionals on YouTube, Instagram, and TikTok on thread lifting, plastic surgeons and dermatologists constituted only half of those who created content in this field of cosmetic dermatology. Conversely, a comparatively larger volume of content was produced by healthcare professionals who lacked specialized training in this field. Additionally, YouTube and Instagram were the most preferred platforms by doctors, while nurses favored TikTok. In their study on global social media trends, Braun et al. also found that, on many topics, the majority of posts are made by healthcare professionals other than dermatologists and plastic surgeons [12]. Competent and experienced plastic surgeons and dermatologists lag in online visibility, contributing to a gap in the dissemination of accurate and reliable information [13, 14].
In the evaluation of video content, the overall quality was low across all platforms. Although the videos are useless for patients according to the GQS, they are insufficient in terms of the reliability and quality of the content according to the DISCERN criteria. According to the GQS, the videos are generally “useless” for the patients, and they also fall short in reliability and content quality based on the DISCERN criteria. While YouTube stands out slightly for higher video quality, the best content only reaches a “good” level, and high-quality videos can be counted on one hand. TikTok content performs the worst in this regard. Gürler et al.’s study on the reliability of social media videos found videos on YouTube to be more reliable. Similarly, Zargaran et al.’s study on TikTok concluded that TikTok was not a reliable source [15, 16]. The number of likes, comments, and views of videos shows a poor correlation with quality ratings, suggesting that popularity often accompanies misinformation. Osman et al. reported that the number of views and likes of YouTube videos should not be taken as an indicator of the quality of the information presented [17].
Since minimally invasive procedures are preferred today, thread lifting has become popular for rejuvenating the face and pulling anatomical planes upward. However, thread lifting has disadvantages and complications as with any medical procedure applied to humans. The short- and long-term results remain controversial [18–20]. Therefore, this procedure, promoted as a miracle, should be discussed in detail with the patient. The exchange of information between patients and physicians has changed with the introduction of the Internet and social media platforms. Much information about the health field is currently discussed on platforms such as YouTube, Instagram, and TikTok. These platforms have also become sites where unproven or controversial treatment methods on issues concerning human health are marketed by blending them with informative content about health, but without adequately discussing all aspects [21]. In this area, where everything can be freely discussed, physicians and non-physicians contribute freely without limitations. However, whether this contribution contains accurate and reliable information and the quality of the content remains uncontrolled. Thus, misinformation enters a vicious circle and spreads rapidly among people [22].
It is unreasonable to discuss the quality and reliability of the content created by non-health professionals as it reflects their personal experiences. In contrast, higher quality is expected from content created by health professionals. Afful-Dadzie et al. reported that public health platforms and lay users significantly contribute to the poor quality of social media content and emphasized the need for health professionals to play a greater role in providing accurate information [23]. However, Mehta et al. state that despite being experts, health professionals often produce low-quality, unreliable content primarily aimed at attracting patients and marketing [24]. This finding is similar to the finding in our study that plastic surgeons and dermatologists had higher scores for cosmetic procedures but did not significantly outperform other health professionals. Moreover, the presence of persuasive claims about procedural results in some content implies a focus on promotional appeal rather than informational accuracy.
Although there are no precise scales to help determine the quality of video content, the scope of the DISCERN and GQS can help in this regard. In addition, established criteria for high-quality health information — such as currency, objectivity, clarity, evidence-based content, relevance, balance, and the absence of bias — may serve as useful guidelines [25]. When uploading videos to social media, it might be helpful to outline the video topic and what is not covered due to time constraints, mention additional information sources, cover all treatment aspects without promising outcomes, and discuss alternative treatments while addressing risks and benefits.
This study is limited by its exclusive focus on English-language content and the use of assessment tools not specifically designed to evaluate social media material.

Conclusions

After examining thread-lifting videos, we found that much of the content is of low quality and inadequate, produced primarily for marketing purposes, as previously noted in the literature. It appears that content production is dominated by healthcare professionals other than plastic surgeons and dermatologists, despite the latter having more extensive training in facial rejuvenation procedures. It is crucial to emphasize that cosmetic procedures like thread lifting are medical interventions, not commercial products. Today, it has become an ethical necessity for plastic surgeons and dermatologists, experts in this field, to be at the forefront and emphasize quality in the content.

Funding

No external funding.

Ethical approval

Not applicable.

Conflict of interest

The authors declare no conflict of interest.

References