Autism in children and young people with functional tic-like behaviours: a retrospective cross-sectional study Authors: Holan Liang, Chelsea Wu, Morvwen Duncan, Lila Simpson, Abbie Smith, Katie Harrold, Zoe Pearman, Amy Warren, Sara Shavel-Jessop, Tara Murphy, Isobel Heyman, Sacha Evans, Matteo Catanzano Published: 2026 ABSTRACT AbstractBackgroundFunctional tic-like behaviours (FTLB) are characterised by multiple, complex, vocal and/or motor tics, typically with adolescent onset. Emerging research highlights neurodevelopmental conditions as vulnerability markers for FTLB development, yet no studies have specifically studied this subgroup. This study (1) investigates the prevalence of autism in young people with FTLB, (2) compares demographic and clinical features between patients with FTLB with and without autism and (3) examines the diagnostic accuracy of the Development and Well-Being Assessment (DAWBA) in identifying autism in this patient group.MethodsThe study comprised 63 children presenting to the Great Ormond Street Hospital Tic Service between August 2020 and May 2022 with an FTLB diagnosis. Demographic and clinical variables were determined through retrospective chart review.ResultsIn the FTLB sample, 69.8% were diagnosed with autism and 37% had Tourette syndrome, both much higher than in the general population. Results indicated that young people with FTLB with and without autism have similar rates of reported impairment, as measured by the Strengths and Difficulties Questionnaire, but those with autism had higher rates of co-occurring attention-deficit/hyperactivity disorder (ADHD, 61%) and specific learning disorders (27%). The DAWBA demonstrated poor sensitivity (ranging from 0.09 to 0.26) in predicting likelihood of autism in this group.ConclusionsThere is a high prevalence of autism in patients with FTLB. Those with FTLB and autism may be more likely to have ADHD and specific learning disorder compared with those with FTLB alone. Better clinical screening for autism is essential in patients with FTLB as DAWBA is insufficient.Keywords: Adolescent Health, Child Psychiatry, Mental health, Child Health, Neurology WHAT IS ALREADY KNOWN ON THIS TOPIC Approximately a quarter of young people with functional tic-like behaviours (FTLB) are reported to be autistic. Approximately a quarter of young people with functional tic-like behaviours (FTLB) are reported to be autistic. Functional tic-like behaviours (FTLB) are more common in adolescents assigned female at birth. Functional tic-like behaviours (FTLB) are more common in adolescents assigned female at birth. WHAT THIS STUDY ADDS Rates of autism may be higher if specific autism enquiry/screening rather than general mental health screening is implemented. Rates of autism may be higher if specific autism enquiry/screening rather than general mental health screening is implemented. Those with FTLB and autism may have mixed neurodevelopmental needs including ADHD and specific learning disorder. Those with FTLB and autism may have mixed neurodevelopmental needs including ADHD and specific learning disorder. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY Clinics diagnosing FTLB should consider autism-specific enquiry/screening. Further research on mechanistic links between autism and functional symptoms is suggested. Clinics diagnosing FTLB should consider autism-specific enquiry/screening. Further research on mechanistic links between autism and functional symptoms is suggested. Introduction Functional tic-like behaviours (FTLB) are a functional neurological disorder (FND) subtype, characterised by rapid-onset complex tic symptoms causing impairment and distress, not better explained by other medical/mental conditions.1 FTLB is differentiated from neurodevelopmental tics through clinical characteristics: FTLB has a later onset, typically around 14 years,2 and is more prominent in females with a history of anxiety.3 In 2023, expert consensus guidelines were drafted for the diagnosis of FTLB.4 During the COVID-19 pandemic, there was an increase in FTLB.5 In specialist clinics, there was an increase from 2% in 2018 to 36% in 2021.6 Pandemic-related social isolation,7 rise in social media use8 and contemporaneous popularity of social media influencers with tics may have contributed to the onset of FTLB in susceptible individuals.9 Great Ormond Street Hospital Tic Service (GOSH TS) participated in an international youth registry of FTLB.10 Pooled registry data reported anxiety disorders as the most common co-occurring conditions in FTLB (66%), while 28% had comorbid depressive disorders.10 Autism and attention-deficit/hyperactivity disorder (ADHD) were seen in almost a quarter of FTLB cases. Non-pooled autism prevalence data demonstrated autism prevalence varied by clinic site, with GOSH TS demonstrating higher rates. It is unclear if this stems from differences in population groups or autism detection methods. The GOSH TS used the Development and Well-Being Assessment (DAWBA; 11) at preassessment, followed by universal clinical screening for autism at initial assessment. This involved active clinical enquiry into past and present social communication and peer interaction differences. Where evidence from this line of enquiry led clinicians to suspect assessment was warranted, full autism assessment was recommended. Anecdotally, clinicians noted that many patients reaching clinical threshold for warranting autism assessment were not being identified by DAWBA screening. As many tic services do not offer universal clinician enquiry about autism, instead relying on screening instruments like the DAWBA, it is important to explore the sensitivity of the DAWBA as an autism screening tool in this population. Research on aetiological models for FND informs understanding of FTLB. Keynejad et al suggested the stress-diathesis model whereby individual genetic risk and early adversity (maltreatment and trauma) can increase susceptibility to FND.12 They hypothesise that interaction between genetic predisposition and/or childhood adversity influences future response to stress and vulnerability to FND. Those with low genetic risk for FND and minimal childhood adversity require a severe stressor to precipitate FND, but for individuals with high genetic risk and/or childhood adversity, a mild stressor may be sufficient to trigger FND. Autism as a risk factor for FND fits with this model. Research indicates a high prevalence of autism in FND.10 13 Autism is a genetic condition with links to mental and physical health problems including anxiety, depression and neurological problems (such as epilepsy and headache).14 It is associated with alexithymia and sensory hypo/hypersensitivity which have established links to FND.15 Therefore, potential mechanistic reasons exist for why autism is a vulnerability factor to consider in a stress-diathesis model of FND and FTLB. At present, there is limited research addressing autism in FTLB. This study conducts secondary data analysis using anonymised datasets collected from GOSH TS, a specialist tic clinic, to (1) determine the prevalence of autism in a clinical sample of children and adolescents with FTLB; (2) compare demographic and clinical characteristics between patients with FTLB with autism spectrum disorder (FTLB+ASD) and without (FTLB−ASD) autism; (3) explore the utility of the DAWBA in identifying autism in patients with FTLB; and (4) reflect on findings with respect to mechanistic models of FND. Methods Design This is a retrospective cross-sectional secondary data analysis conducted using anonymised data collected by the GOSH TS. Setting and patients The original dataset contained 175 patients seen consecutively at GOSH TS from August 2020 to May 2022, collected for clinical audit. The inclusion criteria for this study were age <18 at assessment and diagnosis of FTLB. After excluding patients who failed these criteria as they did not have FTLB (all patients were <18), our sample consisted of 63 individuals. This sample size is underpowered for between-group comparisons, so statistical tests are not reported, but CIs are given. However, FTLB is a rare condition and it was felt that basic data reporting would indicate areas for further research. Main outcome measures Demographic variables Demographic data including age, sex assigned at birth and ethnicity were extracted through electronic health records. Clinical variables The DAWBA11 was completed electronically by parents and/or youth prior to initial clinical assessment. It is a standardised diagnostic tool used for examining common mental health conditions in youth. All parents were requested to complete the parent DAWBA; the self-report DAWBA was offered to all youth aged 11+ attending the clinic. The DAWBA generates scores denoting the probability of children and adolescents meeting criteria for psychiatric and neurodevelopmental diagnoses (ie, anxiety, depression, post-traumatic stress disorder (PTSD), obsessive-compulsive disorder (OCD), autism, ADHD) based on International Classification of Diseases, 10th Revision/Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (ICD-10/DSM-5). We examine the DAWBA score for autism and the probability bands were grouped into categories: 0 (<3% probability), 1 (~15% probability), 2 (~50% probability), 3 (70%+ probability).11 The Strengths and Difficulties Questionnaire (SDQ; 16), parent reported and self-reported, is embedded in the DAWBA (self-report is only available to those aged 11+ years). It consists of 25 items, linked to five 5-item subscales, with an additional 4-item impact supplement scale. Advantages of the SDQ are its short nature and good psychometric properties, including good internal consistency (mean Cronbach’s α=0.73), validity and test–retest stability.17 All patients accessed clinical diagnostic assessment by a consultant-led multidisciplinary team (including psychiatrist, clinical psychologist or neurologist). This assessment included evaluation of co-occurring psychiatric and neurodevelopmental ICD-10/DSM-5 diagnoses (eg, anxiety, depression, PTSD, autism, ADHD, OCD and specific learning disorder (SpLD)). Previously established autism was corroborated, and where undiagnosed autism was suspected through clinical enquiry, clinicians referred to an internal autism clinic or the patient’s local National Health Service (NHS) autism pathway for diagnosis. Decisions to offer autism diagnostic assessment internally or refer to local autism services were made clinically. Youth already known to local mental health services were typically referred back to them to consider an autism assessment. Where assessments were conducted by GOSH TS, Autism Diagnostic Interview-Revised and Autism Diagnostic Observation Schedule-2 were both conducted by a multidisciplinary team as per National Institute for Health and Care Excellence guidelines. The timing of autism diagnosis was collected through review of electronic health records. The categories for this variable include: 0 (no diagnosis), 1 (diagnosed prior to TS assessment), 2 (diagnosed by TS service), 3 (diagnosed after TS assessment). Data analysis Descriptive statistics were used to describe the group’s features. The group was then divided into two groups, FTLB+ASD and FTLB−ASD, and group characteristics reported. Power calculations (online supplemental appendix 1) indicated that sample size was insufficient for significance testing. To examine the DAWBA’s ability to predict autism, the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) across the DAWBA probability bands (~15%, ~50%, 70%+) were calculated using methods from Trevethan.18 Results Prevalence of ASD Of 63 patients with FTLB, 23 (37%, 95% CI 25.7% to 48.9%) had Tourette syndrome (TSY), 44 (69.8%, 95% CI 57.6% to 79.8%) had autism and 17 (27%, 95% CI 17.6% to 39.0%) had both TSY and autism. Of the 44 patients with co-occurring autism, 29.5% (95% CI 18.2% to 44.2%) were diagnosed prior to initial assessment, 27.3% (95% CI 16.3% to 41.8%) were diagnosed by GOSH and 43.2% (95% CI 29.7% to 57.8%) were diagnosed following referral to local NHS autism pathways. For three patients, although a recommendation was made for autism assessment locally, the outcome of this assessment was not reported back to the clinical team, and for the purposes of this study, they have been considered as not autistic. Patients with FTLB without a history of TSY (n=40) demonstrated similar rates of autism (67.5%, 95% CI 52.0% to 79.9%), suggesting autism and FTLB are associated, independent of TSY. Demographic variables Mean assessment age was 14.2 years (SD=2.2); 14.2 years (SD=2.4) in those with autism and 14.4 years (SD=1.7) in those without autism. Table 1 shows sample descriptive statistics in relation to sex and ethnicity. FTLB is more prevalent in those assigned female at birth (84.1%). Expert consensus diagnostic criteria for FTLB4 allow a diagnosis of FTLB in children <12 years, where additional criteria are met. These include comorbid psychiatric symptoms which include anxiety and difficulties with social relationships, both of which are common in autism. In our sample, most children <12 years diagnosed with FTLB were autistic. Table 1Descriptive statistics for sex assigned at birth, ethnicity and age between groups (FTLB−ASD and FTLB+ASD)VariableSample size (proportion)Total (n=63)FTLB−ASD (n=19)FTLB+ASD (n=44)Sex assigned at birth Male10 (15.9%)2 (10.5%)8 (18.2%) Female53 (84.1%)17 (89.5%)36 (81.8%)Ethnicity White British41 (65.1%)13 (68%)28 (64%) Mixed white/black2 (3.2%)0 (0.0%)2 (4.3%) Asian Indian4 (6.3%)1 (5.3%)3 (6.8%) Black Caribbean1 (1.6%)0 (0.0%)1 (2.3%) Other black1 (1.6%)1 (5.3%)0 (0.0%) Other mixed1 (1.6%)0 (0.0%)1 (2.3%) Other ethnic1 (1.6%)0 (0.0%)1 (2.3%) Unknown12 (19%)4 (21%)8 (18%)Age 82 (3.2%)0 (0%)2 (4.5%) 92 (3.2%)0 (0%)2 (4.5%) 101 (1.6%)0 (0%)1 (2.3%) 111 (1.6%)1 (5.3%)0 (0%) 123 (4.8%)2 (11%)1 (2.3%) 1312 (19.0%)4 (21%)8 (18%) 147 (11.1%)2 (11%)5 (11%) 1515 (23.8%)3 (16%)12 (27%) 1614 (22.2%)6 (32%)8 (18%) 176 (9.5%)1 (5.3%)5 (11%)ASD, autism spectrum disorder; FTLB, functional tic-like behaviour. Descriptive statistics for sex assigned at birth, ethnicity and age between groups (FTLB−ASD and FTLB+ASD) ASD, autism spectrum disorder; FTLB, functional tic-like behaviour. Clinical variables SDQ scores Results of self-reported and parent-reported SDQ scores are shown in table 2. Both parent and self-report SDQ were available for 31 youth. Table 2Median self-reported and parent-reported SDQ by group (FTLB−ASD and FTLB+ASD) with 95% CIs (n=number with SDQ data available/number of cases)Parent-reported SDQFTLBn=53/63Difference*95% CI*FTLB−ASDn=12/19†FTLB+ASDn=41/44†Total18.0 (15.0–19.0)22.0 (16.0–25.0)−4.0−8.0 to 1.0Emotional8.0 (5.0–9.0)7.0 (6.0–9.0)1.0−2.0 to 1.0Conduct1 (1.0–3.0)2.0 (1.0–3.0)−1.0−2.0 to 1.0Hyperactivity7.0 (5.0–9.0)7.0 (5.0–9.0)0.0−2.0 to 1.0Peer3 (2.0–4.0)3.0 (2.0–6.0)0.0−3.0 to 0.00Prosocial9.0 (7.0–10.0)6.0 (4.0–9.0)3.00.00 to 3.0Impact6.0 (3.0–8.0)8.0 (7.0–10.0)−2.0−5.0 to −1.0Self-reported SDQFTLBn=34/63Difference*95% CI*FTLB−ASDn=9/19†FTLB+ASDn=25/44†Total21.0 (13.0–23.0)20.0 (14.0–23.0)1.0−7.0 to 4.0Emotional7.0 (5.0–8.0)6.0 (4.0–8.0)1.0−2.0 to 2.0Conduct2.0 (0.0–3.0)2.0 (1.0–3.0)0.0−2.0 to 1.0Hyperactivity7.0 (4.0–9.0)8.0 (5.0–9.0)−1.0−3.0 to 2.0Peer2.0 (1.0–4.0)3.0 (2.0–5.0)−1.0−3.0 to 1.0Prosocial9.0 (8.0–9.0)8.0 (6.0–8.0)1.00.00 to 3.0Impact6.0 (2.0–7.0)4.0 (1.0–8.0)2.0−3.0 to 3.0*Difference in medians with 95% CI (Wilcoxon rank-sum).†Median (IQR).ASD, autism spectrum disorder; FTLB, functional tic-like behaviour; SDQ, Strengths and Difficulties Questionnaire. Median self-reported and parent-reported SDQ by group (FTLB−ASD and FTLB+ASD) with 95% CIs (n=number with SDQ data available/number of cases) Difference in medians with 95% CI (Wilcoxon rank-sum). Median (IQR). ASD, autism spectrum disorder; FTLB, functional tic-like behaviour; SDQ, Strengths and Difficulties Questionnaire. FTLB+ASD and FTLB−ASD groups were broadly similar for both parent and self-reported SDQ measures (scatterplots are additionally available in online supplemental appendix 2). Unfortunately, our small sample size did not permit any valid significance testing. Co-occurring disorders Table 3 depicts the additional co-occurring mental health conditions within FTLB+ASD and FTLB−ASD groups. Although small sample sizes prevented the use of statistical testing, proportionally more patients with FTLB with autism had ADHD and SpLD than patients with FTLB without autism. Table 3Co-occurring disorders between groups (FTLB+ASD and FTLB−ASD)Co-occurring conditionFTLBn=63Difference95% CI*FTLB−ASDn=19FTLB+ASDn=44ADHD5.0 (26.3%)27.0 (61.4%)−35.1%−56.0% to −8%OCD4.0 (21.1%)7.0 (15.9%)5.2%−14% to 30%Specific learning disorder0.0 (0.0%)12.0 (27.3%)−27.3%−42% to −9%Anxiety12.0 (63.2%)34.0 (77.3%)−14.1%−39.3% to 9.1%Depression3.0 (15.8%)11.0 (25.0%)−9.2%−28% to 15.4%PTSD2.0 (10.5%)3.0 (6.8%)3.7%−10.4% to 25.5%Tourette syndrome6.0 (31.6%)17.0 (38.6%)−7%−30% to 19.5%*95% CIs calculated using Wilson score method.ADHD, attention-deficit/hyperactivity disorder; ASD, autism spectrum disorder; FTLB, functional tic-like behaviour; OCD, obsessive-compulsive disorder; PTSD, post-traumatic stress disorder. Co-occurring disorders between groups (FTLB+ASD and FTLB−ASD) 95% CIs calculated using Wilson score method. ADHD, attention-deficit/hyperactivity disorder; ASD, autism spectrum disorder; FTLB, functional tic-like behaviour; OCD, obsessive-compulsive disorder; PTSD, post-traumatic stress disorder. DAWBA sensitivity to ASD in the FTLB group According to the DAWBA screening tool results, 59.1% (95% CI 44.4% to 72.3%) of patients with FTLB+ASD had a probability of <3% of having ASD; 4.5% (95% CI 1.3% to 15.1%) had a probability of ~15%; 9.1% (95% CI 3.6% to 21.2%) had a probability of ~50%; and 6.8% (95% CI 2.3% to 18.2%) had a probability of 70%+. The remaining 20.5% had missing DAWBA autism data. DAWBA predictions were compared against clinical diagnosis rates. The specificity, sensitivity, PPV and NPV are reported in table 4. The DAWBA had high specificity but low sensitivity to autism in this population. A 4×2 table of values showing the DAWBA diagnostic results against the gold standard assessment is available in online supplemental appendix 3. Table 4Sensitivity, specificity, PPV and NPV for DAWBA probability bands and 95% CIs in parenthesesDAWBA probability bandSensitivitySpecificityPPVNPV% correctly identifiedHigh (70%+)0.09 (0.03 to 0.22)1 (0.78 to 1)1 (0.44 to 1)0.26 (0.19 to 0.45)34.7 (22.9 to 48.7)Moderate (~50%)0.2 (0.1 to 0.36)1 (0.78 to 1)1 (0.65 to 1)0.33 (0.21 to 0.48)42.9 (30 to 56.7)Low (~15%)0.26 (0.14 to 0.42)1 (0.78 to 1)1 (0.7 to 1)0.35 (0.22 to 0.5)46.9 (33.7 to 60.6)DAWBA, Development and Well-Being Assessment; NPV, negative predictive value; PPV, positive predictive value. Sensitivity, specificity, PPV and NPV for DAWBA probability bands and 95% CIs in parentheses DAWBA, Development and Well-Being Assessment; NPV, negative predictive value; PPV, positive predictive value. Discussion Prevalence of autism in a clinical sample of young people with FTLB Our results show high rates (69.8%) of autism among adolescents with FTLB. Rates remain high, even when TSY is removed as a confounder. This supports high autism rates reported by the FTLB international registry but indicates rates may be much higher with better pick-up by clinician-led autism screening rather than with generic mental health screens. Our high rates of autism may call into question the possibility of ‘site specific overdiagnosis’; however, the majority (72.7%) of the autism diagnoses were made by community autism services from across the UK. Further, our reported prevalence rates concur with FND literature, where 69% meet clinical threshold for autism.13 Extant research also indicates patients with autism present with a significantly higher number of functional symptoms than non-autistic counterparts.13 Utility of DAWBA in identifying autism in patients with FTLB Our findings indicate the DAWBA has good specificity but poor sensitivity in predicting autism in a predominantly assigned female-at-birth sample. Most patients with FTLB with autism had not received an autism diagnosis prior to FTLB diagnosis. The FTLB population comprises primarily individuals assigned female at birth without global intellectual disability. Autistic individuals assigned female at birth are often diagnosed later than those assigned male or remain undiagnosed altogether.19 This may be due to increased ‘camouflaging’ in people with autism assigned female at birth20 and systems (families, education and healthcare) being less attuned to how autism presents in this group. Dworzynski et al21 found that in the absence of intellectual or behavioural issues, girls, compared with boys with comparable levels of autistic traits, are less likely to meet diagnostic threshold for autism when assessed by varying tools, including the DAWBA. This lack of sensitivity of the DAWBA for this specific population has been reported previously.22 Better screening for autism is required for young people assigned female at birth without global cognitive delay, as autism in this group is associated with poor mental health outcomes,23 which may include FTLB/FND. Comparison between FTLB cases with and without co-occurring autism Youth with FTLB and autism may experience difficulties with relationships and associated distress. This may be due to challenges that autistic youth face due to social communication differences, which can increase barriers to friendships.24 Autistic children may struggle with low friendship quality and low social status in friendship groups,25 leading to loneliness26 and bullying.27 Overall, levels of difficulty were not different from youth with FTLB without autism. Patients with FTLB with autism were more likely to have co-occurring ADHD and SpLD. There is evidence demonstrating shared genetic risk factors, for example, gene polymorphisms, between autism, TSY and ADHD.28 Twin studies demonstrate high co-occurrence among these conditions, suggesting potential genetic basis for comorbidity.29 Co-occurrence of multiple neurodevelopmental symptoms and disorders is well described.30 Reflection on findings with respect to mechanistic models of FND Our findings fit with the stress-diathesis model positing that genetic risk and early life events increase susceptibility to FND when later life stress is encountered.12 Autism (a genetic condition) is associated with elevated PTSD31 and anxiety,32 which is compounded in females at birth by autism being unrecognised.33 We suggest that autistic youth were particularly susceptible to FTLB/FND during the COVID-19 pandemic. There was evidence that although the pandemic was a stressful time for all adolescents (eg, due to changes in routine, heightened parental stress, increased time at home and family conflict34), it was a particularly difficult time for autistic youth who experienced higher levels of anxiety and emotional dysregulation.35 A second mechanistic model of FND is the integrated aetiological summary model. This proposes a pathway where FND symptoms develop and are maintained through biopsychosocial factors.36 It suggests that a history of anxiety/trauma/physical illness predisposes to FND. It then proposes that vulnerable individuals are predisposed to increased awareness of bodily sensations which are misinterpreted as health problems, exacerbating anxiety and perpetuating physical symptoms. This cycle continues, reinforced by health scaffold (beliefs and expectations of illness, influenced by personal knowledge and past experiences) and eventually forms FND. Symptoms are maintained by family, school and society’s responses to ill health, incorporating the sick role and secondary gain. Autism as a risk factor for FND again fits well with this model. Sensory hyper-responsivity or hyporesponsivity is common in autistic individuals.1 Deficits in emotional recognition and regulation (alexithymia) are frequently observed in both autism and FND.15 Interoception, the ability to accurately perceive one’s internal state and to assimilate physiological sensations with emotional cues and judgement, is reportedly low in individuals with autism and with FND.37 38 These factors may predispose an autistic individual to misinterpret physical and emotional cues as physical illness symptoms. Reduced cognitive flexibility in autistic youth39 may increase fixation on physical symptoms, reinforcing functional symptoms. Further, autistic individuals may exhibit greater challenges in mainstream education (incorporating sensory overstimulation and social difficulties) and therefore reluctance to attend school.40 As reduced school attendance is often a consequence of ill health/functional symptoms, this can be considered a secondary gain. Limitations Small sample size restricted ability to undertake statistical analyses. The participants were predominantly ‘White British’ and assigned female at birth, limiting the generalisability of results. The use of clinical samples referred to a national specialist service may lead to an oversampling of patients with more severe symptoms, further limiting generalisability. Due to the cross-sectional design of this study, comments on mechanisms are speculative. Conclusion High prevalence of autism is found in youth with FTLB and this fits with existing FND explanatory models. Increased co-occurring neurodevelopmental conditions may be seen in patients with FTLB with additional autism. General mental health screening questionnaires do not adequately identify autism in this group, and future research should prioritise better autism screening instruments. Supplementary material Acknowledgements Research at Great Ormond Street Hospital (GOSH) NHS Foundation Trust and the UCL GOSH Institute of Child Health is made possible by the NIHR GOSH Biomedical Research Centre. The views expressed are those of the author(s) and not necessarily those of the NHS, GOSH Children’s Charity, NIHR or the Department of Health. Footnotes Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors. Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Ethics approval: This study involves human participants and was approved by the Great Ormond Street Hospital for Children NHS Foundation Trust’s Clinical Audit Team as part of a registered service evaluation (audit reference number: 3370). As the collection and analysis of data were deemed not to require review by an NHS research ethics committee, the project proceeded as a service evaluation; however, care was taken to ensure ethical standards were met. No personally identifiable information was included. All patients attending the clinic are informed that their anonymised data may be used for service evaluation and audit. Data availability statement All data relevant to the study are included in the article or uploaded as supplementary information. References Associated Data Supplementary Materials Data Availability Statement All data relevant to the study are included in the article or uploaded as supplementary information.