Outcome in Pediatric Functional Tic Disorders Diagnosed During the COVID-19 Pandemic

Initial screening discovered 90 patients diagnosed as having FND; 29 of these patients had functional tics diagnosed by a board-certified pediatric neurologist and therefore met the inclusion criteria. All patients had at least one follow-up clinical encounter. Twenty dyads completed telephone interviews and were included in comparative analyses of the impressions of parents or guardians; at least one member of five dyads could not be contacted despite multiple attempts, at least one member of three dyads declined to complete the interview, and one dyad withdrew from the study after the guardian survey but before the patient survey. Descriptive statistics are provided in Table S1 in the online supplement. A majority of the patients were teenage girls (mean±SD age=15.8±1.4 years) with preexisting anxiety or mood disorders who reported life stressors and had been exposed to tics through social media (55%) or personal contacts (31%). Reported races and ethnicities were White (86%), Black (7%), Hispanic (3%), and “other” (3%). The most common phenomenological features included trunk or limb involvement, complex movements, a varied repertoire, context-dependent movements, and tics occurring during the examination (see Figure S1 in the online supplement). Features typical of Tourette syndrome, such as worsening of tics when alone, were rare. Prior to diagnosis, many patients had medications prescribed to treat presumed primary tic disorder and had already been receiving psychotherapy (see Table S1 in the online supplement). New treatments prescribed after diagnosis included cognitive-behavioral therapy and comprehensive behavioral intervention for tics.

The primary outcome measures of parent- and guardian-rated CGI-I scores are provided in Table 1 with other outcome measures. At the final follow-up, 21 of 29 patients (82%) reported at least some improvement since diagnosis. Three of the patients who reported no improvement since the diagnosis had improved before the diagnosis. Seventeen of 29 patients (59%) were much or very much improved. Despite these improvements, ongoing interference was common. Examples of reported interference included having tics interrupt sentences and engaging in fewer social activities because of tics.

The clinical course of each patient, including symptom onset, acute care visits, diagnosis, initial improvement, and follow-ups, is shown in Figure 1A. The median date of symptom onset was January 1, 2021 (interquartile range [IQR] from July 30, 2020, to July 26, 2021), nearly 8 months after COVID-19 was declared a national emergency in the United States (Figure 1B). Nearly half (48.3%) of the initial presentations among our sample were to emergency departments or urgent care clinics. The median date of diagnosis was June 7, 2021 (IQR from May 5, 2021, to September 21, 2021), with median time from symptom onset to diagnosis of 197 days (IQR=54–456). Although some patients improved before receiving a diagnosis, a majority of patients improved rapidly after receiving a diagnosis, with a median time until initial improvement of 21 days (IQR=0–61) after diagnosis (Figure 1C).

Univariate regression models assessing factors predicting improvement within 1 month following diagnosis identified age at time of diagnosis as a positive prognostic factor, with each additional year of age increasing the probability of improving by a factor of more than 2 (odds ratio=2.11/year, 95% CI=1.03–4.32, p=0.041). An inverse association between time from symptom onset to diagnosis and improvement within 1 month of diagnosis was suggested with each month delay decreasing odds of recovery (odds ratio=0.89/month, 95% CI=0.78–1.00, p=0.055). No factors predicted final CGI-I scores, although age at diagnosis and patient agreement with diagnosis were borderline-positive correlates. The full results of regression analyses are available in Table S2 in the online supplement.

This single-center, cross-sectional cohort study demonstrated rapid improvement but not resolution of functional tics after diagnosis for a majority of pediatric patients, although many patients continued to note long-term impairment from tics.

Our findings corroborate recent reports based on clinician-rated tic severity scales by Howlett et al. (16). In addition to clinician ratings, we evaluated patient and parent impressions of outcome, time to recovery, and factors predicting recovery. To do so, we chose CGI-I score as the primary outcome measure to incorporate patient and guardian impressions of recovery. This scale is recommended for use by the FND Core Outcome Measures group (17) and allows for more emphasis to be placed on disability associated with tics rather than the features of tics themselves.

Rapid improvement was more likely for older patients and patients with a shorter time between onset and diagnosis. The importance of early diagnosis in prognosis in FND has been well established (13), and our data provide additional support for this notion in the context of pediatric functional tics. Alternatively, the critical factor may be that a patient who experiences symptoms for a shorter duration at the time of diagnosis may be more likely to improve spontaneously. As many of our patients improved soon after the diagnosis of functional tics, this may highlight the importance of clear communication of the diagnosis early in the disease course. Most of our patients agreed with the diagnosis, which may have contributed to their overall improvement. In addition to patient education, ongoing treatment of functional movement disorders may include cognitive-behavioral therapy (18) and physiotherapy (19). Although data are lacking, comprehensive behavioral intervention for tics may be considered for treatment of functional tics given the similarity of the element of motor retraining used in physiotherapy for other functional movement disorders (1, 6).

Our finding of an association between younger age and delayed recovery could indicate a difference in the association between age and prognosis in pediatric FND populations compared with adult FND populations. Studies evaluating adults with FND have noted an association between earlier age at onset and improved prognosis (13). This relationship is consistent with improved prognosis in pediatric FND (14, 15), including findings in Howlett et al.’s study of pandemic-associated functional tics (16). The influence of age on FND outcome within pediatric populations is less well studied. Although age did not influence prognosis in Raper et al.’s analysis of pediatric FND (15), we found that younger age was associated with lower likelihood of improvement at 1 month after diagnosis. Given the lack of power to correct for multiple comparisons, this finding requires replication. However, one possible explanation of this finding is that a greater ability to understand the diagnosis of FND with cognitive maturity improves prognosis.

Although the finding was not statistically significant, compared with their guardians, the patients tended to report more lingering or severe long-term symptoms. Because difficulty diverting attention from symptoms is a component of FND pathophysiology (20, 21), patients may report more symptoms than their guardians. This potential relationship bears confirmation with further study.

Several recent reports of functional tics have noted associations with observing tics and tic-like movements on social media (12, 22). However, only 50% of our cohort reported viewing tics on social media; others reported personal contacts with tics and sometimes stated that they “caught” tics from close contacts. This finding suggests a possible role for sociogenic elements outside of social media.

The features of tics noted in our patients were consistent with previous descriptions of functional tics and those observed by others during the COVID-19 pandemic (9, 12, 23). As in any study of functional movement disorders, the distinction of functional from primary tic disorders is challenging (23), and we cannot exclude the possibility that some of our cohort had a primary tic disorder, particularly because 13% of our cohort reported a family history of tics.

Many patients with FND are lost to follow-up. Therefore, a strength of our study is that follow-up data, in the form of a clinical encounter or a study interview, were available for all patients. A weakness of our study is the small sample size, which did not provide the power to correct for multiple comparisons. Additionally, some phenotypic features may have been present but not documented in the EMR, and therefore, the frequency of particular features may have been underreported. Data abstraction from EMRs was not blinded to the study hypotheses, which could have biased the interpretation of EMR-based estimation of CGI-I scores. Certain elements of our study, such as measures of patient-reported time to recovery, may be prone to recall bias; we sought to address this by framing most questions in relation to current symptoms.

In our cohort of pediatric patients with functional tics, rapid but incomplete recovery was common. We found support for the importance of early diagnosis in the treatment of functional tic disorders and raise the possibility that younger age may be a poor prognostic factor among pediatric functional tic populations.

The authors thank Mark Quigg, M.D., for providing formative feedback.