If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
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.
The 100 most-viewed videos under #tourettes in TikTok were randomly assigned to two of three primary reviewers (<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, one of five senior reviewers (>2 years independent pediatric movement disorder practice) served as a 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”. Median scores and Spearman correlation between primary and senior reviewers were calculated.
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, nontic behaviors (median [IQR] Likert ratings: 5 [4-5]). Primary vs. senior rater scores demonstrated moderate agreement (r = 0.46; P < 0.001).
TS symptom portrayals on highly viewed TikTok videos are predominantly not representative or typical of TS.
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, nonrhythmic movements or vocalizations, sometimes performed in response to premonitory urges. Tics are divided into 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.
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 patients with TS can voluntarily suppress their tics, albeit only momentarily.
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.
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.
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” was 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.
All reviewers were pediatric neurologists specializing in movement disorders. Primary reviewers (A.Z.V., A.B., and J.G.) had less than 2 years of independent practice. Senior reviewers (S.W., L.T., R.L., P.B., and D.G.) were more experienced specialists. An initial list of common, atypical, and severe tic phenomenology was generated for use in quantifying phenomenology in videos. Typical features were defined as those commonly seen in TS, whereas atypical features were defined as those seen more frequently in FTLD. This was based on a prior review by Ganos et al.
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 behaviors that are triggered by an external stimulus, such as clapping specifically when seeing a dog; these have been called stimulus-bound or reflex tics in the TS literature.
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. In cases where disagreements occurred, reviewers were then unblinded and able to review and discuss the other's interpretation and attempted to reach 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. Since this is a subjective rating, all reviewers were kept blind to each other's assessment and no tiebreaker process was used for this part. As a result, each video had 3 Likert scores assigned, two from primary reviewers and one from a senior reviewer. We used a Likert scale, instead of a dichotomous variable, recognizing that a complete differentiation of these types of behaviors is impossible based on short videos and that there is phenomenology overlap between TS and FTLD.
All videos included were publicly accessible in social media and 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.
No identifiable information is included in the article.
Tic phenomenology was summarized by percentages of videos representing each feature. In addition, qualitative descriptions are provided. An additional analysis based on frequency of videos posted by the same username is presented in the Supplementary Materials.
Each Likert score was included in the analysis independently. Each video had 3 Likert scores assigned by different reviewers. These were summarized using medians and interquartile range (IQR). An additional analysis was performed to assess the Likert scores per level of experience. For this analysis, Likert scale ratings of both primary reviewers were averaged for comparison with the senior reviewer. Correlation between mean primary reviewers' score and senior reviewer's score was calculated using Spearman correlation.
The 100 videos in the initial search were posted by 37 different users. The median of the number of videos posted by an individual username was 1 (IQR: 1-2, range: 1-34). 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.
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.
TABLE 1.Phenomenology of Tic-Like Behavior in TikTok Videos
Ego-dystonic tics are tics that appeared to be intrusive and not in line with the individual's values and resulted in frustration or embarrassment.
Saying “Mix it with your glasses” while stirring a mixture of slime using the subject's glasses. Then saying “wear them again” while putting the glasses back without cleaning them. The subject appears frustrated.
Nonstereotyped, complex, subject narrates the ongoing motor behavior, ego-dystonic.
Ego-dystonic tics are tics that appeared to be intrusive and not in line with the individual's values and resulted in frustration or embarrassment.
Changing multiple parts of the lyrics of the nursery rhyme Old McDonald. For example, changing “Everywhere a moo-moo” for “Everywhere a BigMac” or adding “called Peppa” after “and on his farm he had a pig”. Subject appears frustrated.
Nonstereotyped, follows the tune of the song, refers to popular characters (Peppa is a cartoon pig), ego-dystonic.
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 nonstereotyped and, occasionally, accompanied by complex verbal behavior. Similarly, self-injurious behavior (SIB), present in 27.7% of videos, was commonly nonstereotyped. Putting nonedible 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.
In our qualitative assessment, we noted that many videos contained a greater number of occurrences of tic-like behaviors involving the trunk and extremities than of the head and neck. Since our analysis is based on the presence or absence of a feature in the entirety of a video, this unexpected 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.
The median Likert scale score was 5 (IQR: 4-5) (Fig). The distribution of Likert scores was similar between different levels of experience of reviewers (Supplementary Figure). The Spearman correlation coefficient between primary and senior reviewers was 0.46 (P < 0.001).
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.
were over-represented 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 TS,
Tics triggered by an external stimulus are present in ∼20% of patients with TS, but, unlike tics portrayed in these videos, they are more commonly simple and not ego-dystonic (i.e., tics that appear intrusive and not in line with the individual's values and resulted in frustration or embarrassment).
We found a high number of videos showing aggression to others (19.1%) and SIBs (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 patients with FTLD.
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,
Interestingly, the authors extracted videos the same week that we did; therefore, they likely analyzed similar sets of videos as the current study. The authors in this study focused on the most popular influencers. They noted a higher frequency of female subjects. They also found high rates of coprophenomena, complex tics, tics involving extremities, tics causing interference with actions, and SIBs. In addition to these findings, our study added further details about atypical phenomenology and included overall impression of videos using a 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.
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.
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.
Although similarities between the atypical features in these videos and current patients with FTLD have been noticed, it is unclear if the most common atypical features seen in our analysis are also frequently present in most subjects with FTLD. Extrapolating our findings to clinic patients should be done with caution. There have been no studies directly comparing the features seen in these videos and the phenomenology of patients with TS and FTLD in clinic. Such study would provide valuable information of this phenomenon.
Given the recent reports of a high number of patients presenting with FTLD misdiagnosed with TS, patients with atypical characteristics should be evaluated by experienced clinicians to make sure the diagnosis is correct. If a diagnosis of FTLD is made, then we recommend stopping unnecessary tic medications and focusing on functional interventions. Patients should resume regular school and pleasurable activities and work on healthy habits including healthy eating, sleep hygiene, good hydration, and regular exercise. Underlying mental health problems such as anxiety, especially social anxiety, mood disorders, and recent stressors should be assessed and addressed. Behavioral interventions such as cognitive behavioral therapy and habit reversal therapy can also be effective in managing symptoms.
Finally, in the treatment of TS and tic-like behaviors, it is recommended to minimize attention (negative or positive) on tics or tic-like behaviors.
These videos often show excessive attention to the person with tics. This effect can also extend to viewers reinforcing abnormal behaviors. We recommend that patients with FTLD avoid watching TS TikTok videos or videos showing neurologic symptoms for at least 4 weeks. Additionally, the effect of these videos in patients with TS 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.
To limit subjectivity, multiple reviewers evaluated each video. Third, we only found a moderate correlation between reviewers of different experience. This is likely due to mild discrepancies in scoring (e.g., score of 4 vs score of 5). This discrepancy also speaks to the difficulty in fully differentiating TS and FTLD based on short videos. Fourth, we did not separate our analysis by specific subject portrayed which could result in an over-representation of the most popular influencers. However, this parallels the experience of the general population as they are likely to encounter the most popular videos more frequently.
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.
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.
In sum, current TikTok videos are poorly representative of TS and could be misleading to the general public. The detailed description in this article 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.
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.
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 receives honoraria for giving educational lectures for the Tourette Association of America. She is/has been the principal investigator on several studies regarding investigational therapies for Tourette Disorder for Emalex, Teva, and Neurocrine. She was compensated for travel to an investigator's meeting for 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.
Financial support: This is a self-funded study.
Clinical Trial registry name, registration number, and data sharing statement: NA.
Author contributions: Dr. Zea Vera conceptualized and designed the study, collected data, carried out analysis, drafted the initial manuscript, and reviewed and revised the manuscript. Dr. Bruce conceptualized and designed the study, collected data, and reviewed and revised the manuscript. Dr. Garris conceptualized and designed the study, collected data, and reviewed and revised the manuscript. Dr. Tochen collected data and reviewed and revised the manuscript. Dr. Bhatia collected data and reviewed and revised the manuscript. Dr. Lehman collected data and reviewed and revised the manuscript. Dr. Lopez conceptualized the study and critically reviewed and revised the manuscript for important intellectual content. Dr. Wu conceptualized and designed the study, collected data, and critically reviewed and revised the manuscript for important intellectual content. Dr. Gilbert conceptualized and designed the study, collected data, and critically reviewed and revised the manuscript for important intellectual content. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.