ABSTRACT
Background and Objective Pediatricians and pediatric subspecialists worldwide have reported a marked increase in functional (conversion) disorders with tic-like behaviors during the COVID-19 pandemic. These patients often report frequent viewing of Tourette Syndrome (TS) TikTok videos, suggesting disease modeling. We aimed to evaluate tic phenomenology in videos posted on TikTok.
Methods The 100 most-viewed videos under #tourettes in TikTok were randomly assigned to two primary reviewers (n=3; <2 years independent practice), all pediatric neurologists specializing in movement disorders, for extraction and classification of tic phenomenology. Initial disagreements were solved by consensus. If not resolved, a senior reviewer (n=5; >2 years independent pediatric movement-disorder practice) served as tiebreaker. In addition, two primary and one senior reviewer rated each video on a Likert scale from 1= “All the tics are typical of TS” to 5= “None of the tics are typical of TS”. Mean scores and Spearman correlation between primary and senior reviewers were calculated.
Results Six videos without tic-like behaviors were excluded. Most videos depicted coprophenomena (coprolalia: 53.2%; copropraxia: 20.2%), often with unusual characteristics. Frequently, videos demonstrated atypical phenomenology such as very strong influence by the environment (motor: 54.3%; phonic: 54.3%), aggression (19.1%), throwing objects (22.3%), self-injurious behaviors (27.7%), and long phrases (>3 words; 45.7%). Most videos portrayed atypical, non-tic behaviors (Median [IQR] Likert ratings: Primary 4.5 [4-5]; Senior 5 [3-5]). Primary vs. senior rater scores demonstrated moderate agreement (r = 0.46; p<0.001).
Conclusions TS symptoms portrayals on highly-viewed TikTok videos are predominantly not representative or typical of TS.
Highlights
Many teenagers with functional tic-like disorder have been reported during the COVID-19 pandemic.
These patients report increased viewing of Tourette Syndrome TikTok videos, a popular social media platform, and present with similar tic-like behaviors.
Current TikTok videos are poorly representative of Tourette syndrome and more consistent with functional tic-like behavior.
We provide a detailed description of the phenomenology tics and tic-like behavior portrayed in TikTok.
INTRODUCTION
Tourette Syndrome (TS) affects nearly 1% of the population and is characterized by persistent motor and phonic tics.1 TS most commonly presents in boys and has a gradual onset, with tics appearing predominantly before the age of 10 years. Tics are brief, repetitive, patterned, non-rhythmic movements or vocalizations, sometimes performed in response to premonitory urges. Tics are divided in simple (e.g., blinking, grunting) and complex (e.g., gestures, words and phrases) Most tics have no clear external triggers and appear out of context.1 Generally, the emergence of tics follows a rostrocaudal distribution. Concentration, focusing on an activity or performing purposeful actions in the same muscles commonly decreases tics, limiting their interference with voluntary actions. Most TS patients can voluntarily suppress their tics, albeit only momentarily.1
Functional tic-like disorder (FTLD) is a subtype of functional movement disorder characterized by tic-like behaviors – sounds or movements that resemble tics. In contrast to TS, FTLD more commonly presents with multiple symptoms in a short time interval in adolescent and young adult females.2 Compared to tics, functional tic-like behaviors involve the arms or trunk more than the face, interfere more with voluntary actions, are less suppressible, and are more suggestible.2, 3 FTLD can be seen in combination with other functional neurologic symptoms and has a poor response to common tic medications.2 FTLD can also emerge in persons with TS, and in some cases, phenomenological features overlap between TS and FTLD.2
A rise in tic-referrals to movement disorders clinics has occurred during the COVID-19 pandemic. TS patients report worse symptoms during the pandemic,4 possibly related to increased anxiety and confinement.5 In parallel, several groups have noticed a dramatic increase in teenagers, mainly females, presenting with sudden onset FTLD.6-9 Although a general increase in functional movement disorders has also been reported,10 there has been a disproportionate increase in FTLD. Until recently FTLD represented only a minority of functional movement disorders.11 This brings the question “Why, specifically, tics?”.12
The platform TikTok (www.tiktok.com) has become very popular among teenagers.13 The use of social media and the viewing of the #tourettes TikTok webpage has significantly increased during the pandemic.6, 14 The increase use of social media may have had a detrimental effect on mental health during the pandemic.14 Many patients with FTLD report watching TS videos or posting videos of their movements in TikTok. This suggests a potential association between TikTok and the rise in FTLD.6-9 The accuracy of medical videos in social media is highly variable. In a previous study, 66% of videos portraying movement disorders were qualified as functional by experts in the field.15 However, there is limited information on the characteristics of the current videos available in TikTok. The goal of this study is to characterize and evaluate the phenomenology of tics and tic-like behavior portrayed in TikTok.
METHODS
Search Strategy
Our video search was performed on a single day on March 27th, 2021. Before starting the search, the computer’s cache/cookies and search history were cleared to prevent the effects of previous searches on our sample. We looked for all the videos under #tourettes. We chose this hashtag because it is commonly used in videos portraying TS and recently had an increase in popularity.
We selected the 100 most-viewed videos. This number represents a reasonable sample of the videos that would be seen by a person searching for TS in TikTok while maintaining the feasibility of evaluating a large number of videos. Duplicated videos or videos that did not show tics were excluded. The number of comments and “likes” were recorded for each video. Of note, some of the videos portrayed the same subject at different times. We did not limit our search to only one video per subject because we wanted our sample to realistically represent all available videos in TikTok. We did not group the videos by the portrayed subject.
Video review
All reviewers were pediatric neurologists specializing in movement disorders. Primary reviewers (AZV, AB, and JG) had less than 2 years of independent practice. Senior reviewers (SW, LT, RL, PB, and DG) were more experienced specialists. Based on a literature review and clinical experience, an initial list of common, atypical, and severe tic-phenomenology was generated for use in quantifying phenomenology in videos. We rated the presence or absence of each feature in the entire video, allowing for the possibility of multiple tics per video. A pilot was conducted with 10 videos. After analysis, primary reviewers discussed difficulties in data extraction. As a result, some tic-like characteristics judged too difficult to reliably and accurately assess in short and edited videos were eliminated. The final list of tic phenomenology and definitions used for data extraction is presented in the Supplementary Materials.
We utilized the construct “context-dependent tics” (CDT) in our data extraction to classify behaviors that appeared to be strongly influenced by the environment. This included two features that frequently occurred together and were difficult to assess separately based only on short videos. The first are behaviors that are very specific to the external situation of the subject. For example, involuntary phrases that followed the content of an ongoing conversation or complex motor behaviors using objects in the environment. The second are tics that are triggered by an external stimulus, such as clapping specifically when seeing a dog; these have been called stimulus-bound or reflex tics.16 Similar to other studies, we excluded ecophenomena (repeating other people’s words or actions) and paliphenomena (repeating one’s own words or actions) from this category.16
Each video was randomly assigned to two primary reviewers who independently quantified and extracted the characteristics of tics or tic-like behaviors, blinded to each other’s initial assessment. Both primary reviewers discussed disagreements and tried to come to a consensus. If this was not achieved, a senior reviewer was the tiebreaker. Any movement or sound that was portrayed as a tic and is likely to be interpreted as a tic by a lay audience was included. Qualitative descriptions of other patterns of tics in these videos were also obtained.
In addition to evaluating tic phenomenology, all videos were rated globally using a Likert scale from 1= “All of the movements or sounds portrayed are typical of a primary tic disorder such as Tourette syndrome” to 5= “None of the movements or sounds portrayed are typical of a primary tic disorder such as Tourette syndrome”. Each video was randomly assigned to be rated by 2 primary reviewers, who also evaluated tic phenomenology, and 1 senior reviewer, all of them blind to the others’ assessment.
Ethical Considerations
All videos included were publicly accessible in social media, therefore considered part of the public domain. Consistent with other studies of publicly available material in social media, consent from creators and ethics approval were not obtained.17, 18 No identifiable information is included in the manuscript.
Data Analysis
Tic phenomenology was summarized by percentages of videos representing each feature. In addition, qualitative descriptions are provided. Likert scale ratings of both primary reviewers were averaged for comparison with the senior reviewer. Likert scale ratings were summarized using medians and interquartile ranges. Correlation between mean primary reviewers’ score and senior reviewer’s score was calculated using Spearman correlation.
RESULTS
Videos
The 100 videos in the initial search were posted by 38 different users. On average, each video had 2,060,379 “likes” and 47,922 comments. None of the videos were duplicated and 6 did not show tics. In addition, 2 videos were removed from the TikTok webpage before all reviewers could assign a Likert score. Therefore, we included 94 videos for the assessment of tic phenomenology and 92 videos for Likert scale scoring.
Phenomenology
Table 1 shows the frequencies of the portrayed tic-like phenomenology and table 2 presents illustrative examples. Many videos showed a very high number of different tics in a single day (>10 tics; 52.1%). Coprophenomena was frequently portrayed (copropraxia: 20.2%; coprolalia: 53.2%). Context-dependent coprophenomena and coprolalia consisting of long phrases (>3 words) were the most common atypical characteristics noted. Many subjects reacted positively to coprophenomena and used it for a comedic effect. In some videos, coprophenomena was directed at a specific person. Some subjects replaced words in a sentence or song lyric with obscene words while keeping the same number of syllables and/or rhythm of the sentence or song.
CDT were also frequently portrayed (motor: 54.3%; phonic: 54.3%). Commonly, CDT were complex actions and resulted in frustration or embarrassment (e.g., were ego-dystonic). Phonic CDT were often portrayed as unwanted, occasionally insulting, verbal responses that were consistent with the context of an ongoing conversation. Other common patterns of CDT include complex motor actions while verbally describing such action, or referring to popular characters (e.g, mentioning a popular farm animal cartoon while cooking that specific type of meat). In extreme cases, videos showed subjects almost constantly reacting involuntarily to the environment, giving the appearance of a complete loss of inhibition.
Aggression toward other people or objects, present in 19.1% of videos, was often non-stereotyped and, occasionally, accompanied by complex verbal behavior. Similarly, self-injurious behavior (SIB), present in 27.7% of videos, was commonly non-stereotyped. Putting non-edible or injurious material inside the mouth or on the tongue was seen frequently. Throwing objects, present in 22.3% of videos, was commonly seen in the context of cooking or baking. A “challenge” consisting of holding an egg for as long as possible was portrayed by multiple subjects. Often, the challenge ended with the subject throwing the egg after a few seconds. Long phrases (>3 words) were commonly seen (45.7% of videos) and were frequently context-dependent. In some cases, tic-like behavior consisting of prolonged phrases referenced a previous behavior, such as denying that the previous behavior was a tic. We found one example of a motor tonic tic that was portrayed as lasting ∼30 minutes causing injuries to the subject.
We qualitatively noted that tic-like behaviors more frequently involved the trunk and extremities, but most videos also showed tic-like behaviors involving the head and neck. Since our analysis is based on entire videos and not individual tic-like behaviors, the predominance of behaviors involving the trunk and extremities is not apparent in table 1. We did not extract the complexity of tic-like behaviors due to difficulty in assessing it accurately in all videos. However, we noted that many videos showed a mixture of simple and complex tic-like behavior with a preponderance of complex behaviors.
Likert score
The primary reviewers’ median Likert scale score, after averaging both primary reviewers, was 4.5 (IQR:4-5) and the senior reviewers’ median Likert score was 5 (IQR: 3-5). The Spearman correlation coefficient between primary and senior reviewers was 0.46 (p<0.001; Figure 1).
Scatter plot of the Likert scale scores between senior reviewers (y-axis) and primary reviewers (x-axis). The Likert scores from primary reviewers was averaged. The size of the circles represents the number of videos at each point. The bar plots for senior and primary reviewers are shown in the corresponding margins.
DISCUSSION
In this assessment of TikTok videos indexed in March 2021 by “#tourettes”, we found a high frequency of movements, vocalizations, and severe behavioral disturbances more characteristic of functional tic-like behaviors than of tics. Both primary and senior reviewers judged most videos as poorly consistent with a primary tic disorder (Likert score 4-5). Importantly, this video review is not equivalent to a clinical evaluation of the individuals in these videos. A clinical in-person assessment would include many elements not obtainable solely through video review. However, it does show that these videos could be misleading to the general public.
Coprophenomena, seen in ∼15% of TS patients,19 were overrepresented in these videos (copropraxia: 20.2%; coprolalia: 53.2%) and often had atypical features. Coprophenomena portrayed in these videos was frequently context-dependent, and included long phrases. Although urges to make insulting remarks about a person’s trait can be seen in TS20, coprophenomena in TS generally consists of short words or movements presenting out of context and without clear triggers.3 CDT were also very frequent (motor: 54.3%; phonic: 54.3%). Tic-like behavior that is very consistent with the situation of the subject is more frequently seen with FTLD than TS.21 Tics triggered by an external stimulus are present in ∼20% of TS patients but, unlike tics portrayed in these videos, they are more commonly simple and ego-syntonic.16
Rage attacks, which might include physical aggression, are present in 20-67% of patients with TS22, and SIB are seen in 35% of TS patients.23 However, these behaviors are not always tics and can also be caused by obsessive-compulsive thoughts, mood disorders, or impulsivity.22-26 Additionally, aggression to others due to different psychiatric disorders, including functional and factitious disorder, can be misdiagnosed as TS.27 We found a high number of videos showing aggression to others (19.1%) and SIB (27.7%) with atypical features for a tic. Although we were unable to define the exact cause of these behaviors, most videos implied or explicitly said that these were tics. This is particularly worrisome giving the modeling phenomenon that has been reported in recent FTLD patients.8
Other atypical phenomenology noted in the videos includes a very high number of different tics, preponderance of complex behaviors or behaviors involving the trunk and extremities, tic-like behaviors consisting of long phrases, and tic-like behaviors consisting of throwing objects. In TS, simple tics are more common than complex tics,28-30 and tics affecting the eyes, face, and neck are more common than tics affecting the rest of the body.31 Although phonic tics can include phrases,30 these are generally short. In our experience, tics consisting of throwing objects are very rare but we couldn’t find systematic evaluations of this feature.
A recent single-center analysis of TS TikTok videos was conducted to assess phenomenology. 32 Interestingly, the authors extracted videos the same week that we did therefore they likely analyzed similar sets of videos as the current study. They also found high rates of coprophenomena, complex tics, and self-injurious behaviors. In addition to these findings, our study added further details about atypical phenomenology and included overall impression of videos a using Likert scale by reviewers from multiple centers. Considering both studies, there is mounting evidence that multiple specialists from different centers, including adult and child neurologists, have come to similar conclusions about the accuracy of TS TikTok videos.
The present study was prompted by the authors’ clinical experience with an unexplained surge in adolescents presenting to our clinics with functional tic-like behaviors. The atypical phenomenology in the videos we reviewed has been frequently reported in patients with FTLD during the COVID-19 pandemic.7-9 Some patients present with a tic that is frequently seen in TikTok but otherwise would be rarely seen7, 9 and some unusual tics were shared by different subjects in these videos.32 All authors have noticed the same trends in our clinical practice.
In several recent studies, others have suggested a possible causal relation between TikTok TS videos and FTLD. 6-8, 10 One hypothesis is that the rise in FTLD is caused by “social contagion” or modeling.8, 12 While to date we do not have strong evidence to support this hypothesis, disease modeling is seen in functional neurologic disorders, and modeling of tics portrayed in social media has been reported in a previous case of mass functional (psychogenic) illness.33, 34 Alternatively, since some of the teenagers diagnosed with FTLD had a history of mild tics in childhood, echophenomena in genetically susceptible patients has also been postulated to explain some cases.7 Although infectious or post-infectious etiologies can present fulminantly, a recent review of movement disorders associated with proven COVID-19 infection found no cases of tics or tic-like events.35
In the treatment of Tourette Syndrome and tic-like behaviors, it is recommended to minimize attention (negative or positive) on tics or tic-like behaviors.36, 37 Increased attention to the abnormal behaviors can result in exacerbation and perpetuation of symptoms in both disorders but is more prominent in FTLD.2 These videos often show excessive attention to the person with tics. This effect can also extend to viewers reinforcing abnormal behaviors. We recommend that FTLD patients avoid watching TS TikTok videos. Additionally, the effect of these videos in TS patients has not been studied.
Our study has several limitations. First, behaviors due to obsessive-compulsive symptoms, disinhibition, or impulsivity can resemble tics and may have been mischaracterized by the reviewers. However, the included events were portrayed as tics and are likely to be interpreted as such by a lay audience. Second, the categorization of tics based on short videos is imperfect.38, 39 To limit subjectivity, multiple reviewers evaluated each video.
Finally, among several factors we could not assess by video and which therefore raise questions for future studies, we wish to suggest the possible role of secondary gain as a contributor to the high prevalence of severe, atypical features. A previous study found that negative portrayals of TS are more popular in social media.18 We note the possibility that the high number of views, comments, and likes for these videos may have resulted in psychological and/or economic benefits to the individuals posting the videos. While many individuals in these videos express an interest in increasing TS awareness, the present analysis suggests a risk of creating a highly inaccurate perception of TS. This problem may exist for other disorders, as a recent study reported that some TikTok videos intended to create eating disorders awareness could be interpreted as portraying pro-eating disorders messages.17
CONCLUSION
In sum, current TikTok videos are poorly representative of TS and could be misleading to the general public. The detailed description in this paper is intended to help pediatricians, child psychiatrists, and child neurologists recognize this atypical phenomenology to guide additional inquiry about social media exposure and treatment. Videos of tics are a notable example of the proliferation of misleading videos portraying medical conditions during the pandemic. However, similar trends have been reported in other neurologic and psychiatric disorders.17, 40 Although our findings could suggest an association between TikTok videos and the current spike in FTLD, our study was not designed to evaluate this. Further research is needed to define the relationship between social media and FTLD.
Data Availability
Data is available at request.
Footnotes
↵* Author moved institutions from Cincinnati Children’s Hospital Medical Center to Children’s National Medical Center in the course of this project.
Conflict of Interests: Dr. Zea Vera has received travel support from the Child Neurology Society. He has received research support from the Tourette Association of America, American Brain Foundation, and American Academy of Neurology. He is part of the editorial team of the Neurology Journal Resident & Fellow Section.
Dr. Lehman is on the Medical Advisory Board and receiving honoraria for giving educational lectures for the Tourette Association of America. She is/has been the PI on several studies regarding investigational therapies for Tourette Disorder for Emalex, Teva, Neurocrine. She was compensated for travel to an investigator’s meeting Neurocrine and received compensation for serving on a medical advisory board for Teva (Nov. 2019).
Dr. Wu has received research support from the National Institutes of Health. He is also the site principal investigator for EryDel clinical trial on ataxia-telangiectasia.
Dr. Gilbert has received honoraria and/or travel support from the Tourette Association of America/ Centers for Disease Control and Prevention, the Child Neurology Society, and the American Academy of Neurology. He has received compensation for expert testimony for the U.S. National Vaccine Injury Compensation Program, through the Department of Health and Human Services. He has received payment for medical expert opinions through Advanced Medical/Teladoc. He has served as a consultant for Applied Therapeutics and Eumentics Therapeutics. He has received research support from the NIH (NIMH) and the DOD. He has received salary compensation through Cincinnati Children’s for work as a clinical trial site investigator from Emalex (clinical trial, Tourette Syndrome) and EryDel (clinical trial, Ataxia Telangiectasia). He has received book/publication royalties from Elsevier, Wolters Kluwer, and the Massachusetts Medical Society.
All other authors report no conflict of interest.
Funding/Support: Self-funded
Clinical Trial registry name, registration number, and data sharing statement: NA
All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
Abbreviations
- TS
- Tourette Syndrome
- FTLD
- Functional tic-like disorder
- CDT
- Context-dependent tics
- SIB
- Self-injurious behavior
