Sections of this chapter have been reproduced from Hollis et al.,78 under licence CC-BY-4.0.
To our knowledge, ORBIT is the first adequately powered RCT of ERP BT for tics. ORBIT also represents the largest trial of any behavioural treatment for tics and the first trial to examine the effectiveness of an online, internet-delivered behavioural intervention for tics in CYP compared to an active control condition. A key driver of the ORBIT trial is the lack of access to face-to-face BT for CYP with tics due to a lack of highly trained expert therapists. The overarching aim of ORBIT was to develop a clinically effective and cost-effective digital online behavioural intervention to close this treatment gap.
Key findings
The trial results support the clinical and cost-effectiveness of the online delivery of therapist-supported ERP for tics. Analysis of the primary outcome (tic severity at 3 months post-randomisation) indicated a significant effect in favour of therapist-supported ERP compared to supported psychoeducation (effect size = −0.31). Importantly, the therapeutic effect was durable, with a slightly increased effect size at the 6-month follow-up (−0.36). Compared with the psychoeducation comparator, participants were twice as likely to show a positive treatment response with the ERP intervention, with just under half (47%) having responded positively by the 6-month follow-up. In the ERP arm the YGTSS-TTSS was reduced by 16% at 3 months and by 25% at 6 months. Consensus clinical opinion is that these reductions are clinically important, as is the doubling of clinical response within the online ERP arm compared to psychoeducation. The reduction in tics associated with the ERP behavioural intervention was similar in those with and without coexisting anxiety and ADHD diagnoses. Additionally, secondary outcomes indicate that online ERP for tics also had a beneficial impact on anxiety, low mood and overall functioning, although caution is needed in interpreting these secondary outcomes, which involved significant amounts of missing data.
Long-term follow-up studies of any treatments for chronic tics have been exceedingly rare. This is the first controlled study to demonstrate the long-term effectiveness (beyond 6 months) of ERP for tics. The extended naturalistic follow-up showed sustained long-term effects of the ERP intervention for tics for up to 18 months after the start of treatment. The effect size of online ERP compared with psychoeducation was larger at 12 months (−0.36) than at the end-of-treatment 3-month primary end point. At 12 months, 46% were classed as responders to online ERP, with this increasing to 55% at 18 months. In addition, extended follow-up showed those receiving online ERP compared with online psychoeducation had reduced scores for low mood and anxiety at 12 and 18 months and superior tic-specific QoL, with the largest effects seen at 18 months.
In the naturalistic follow-up beyond 6 months, participants were free to access tic medication and therapy as clinically required. However, very few participants (< 10%) started new treatments during the extended follow-up period, although at least half were classed as non-responders. This highlights the lack of availability of behavioural treatment for tics in the UK outside this trial, with so few receiving treatments during the extended follow-up, although many needed it. Hence, this naturalistic follow-up essentially represents an intention-to-treat, parallel-group, long-term 18-month follow-up of online ERP for tics in CYP.
The total therapist contact time in the current trial was approximately 2.5 hours compared to 9–10 hours in comparable evidence-based face-to-face BT for tics.
The health economic analysis showed that the ICER for the cost per QALY analysis falls below the NICE cost-effectiveness threshold commonly used in health technology assessments of £20,000 per QALY gained. There is a 65% probability that the ERP intervention is cost-effective at this threshold.
The process evaluation showed that the ERP intervention had high fidelity of delivery and was evaluated positively by CYP, although some participants suggested some minor improvements and reach may have been constrained by the nature of the RCT. Only parental chapter completion was independently associated with clinical improvement. No other mediators or moderators were identified for either reduction in tic severity or clinical improvement.
Study successes
The trial recruited ahead of time and target, reflecting a significant unmet treatment need in the population. Acceptability and safety of the intervention were high. Retention rates to the 3-month primary outcome end point (90%) and 6-month follow-up (> 80%) were excellent. Retention rates to the primary outcome measure remained high at 12 months (81% in both arms) and 18 months (> 79% in both arms). Treatment adherence was high in both groups, with 88% in the ERP group and 94% in the psychoeducation group achieving the minimum treatment ‘dose’ specified as completion of the first four chapters for both interventions.
The participants in this trial had a moderate to severe level of baseline tic severity (mean YGTSS-TTSS 28.4, SD 7.7), which is approximately 0.5 SD higher than reported in previous face-to-face behavioural treatment trials2,5 and is representative of caseloads seen in routine care. The trial design minimised the clinical comorbidity exclusions, resulting in a sample broadly representative of real-world clinical practice, and included participants with autism spectrum disorder, a group usually excluded in similar behavioural intervention trials. In the behavioural intervention group, just under a third had a coexisting anxiety disorder and just under a quarter had ADHD. A relatively small proportion of participants (13%) were concurrently receiving tic medication. A particular strength of the design was the inclusion of an active comparator arm controlling for non-specific effects of therapist contact, homework assignments and online access. PPI has been a crucial element of the ORBIT trial, which has been integral to the study’s design, delivery and dissemination.
Limitations
The study has several limitations. First, despite being the largest trial to date of a behavioural intervention for tics, it is also the first adequately powered trial of therapist-supported, online ERP, and further replications are required. Second, it is not possible in this trial to separate the effects of digital online delivery and ERP. In future, clinical and cost-effectiveness comparisons of digital online versus face-to-face ERP/CBIT will be needed. Third, there remains the question of the ‘digital divide’, whereby some people do not have sufficient access to the internet and smartphones. This could have potentially limited the reach of this internet-delivered ERP intervention. While this does not appear to be an issue in the UK, with 90% of households having access to the internet and 98% of young people owning a smartphone according to the Office for National Statistics,151 it may be an important consideration when generalising these findings to other countries or vulnerable groups with lower levels of online access. Fourth, whilst the sex distribution is typical of a tic disorder population, a large proportion of the sample was white, which may limit the generalisability of the findings with regards to ethnicity. Fifth, the levels of tic medication use and comorbid OCD diagnoses were lower than in comparable studies conducted in the USA, which may limit generalisability to these populations. Sixth, while the level of tic severity in ORBIT is higher than in comparable studies and is representative of tic severity seen in ‘real-world’ practice, the findings may not be generalisable to those young people with tics outside the severity range of this study population. Finally, it is important to note that the delivery of this study was via two study centres with expertise in tics. Although the therapeutic content was delivered online, it is important for safeguarding that there is adequate supervision of the therapists supporting the intervention. It would be important in the future to evaluate the best model for national roll-out, including the benefits and cost-effectiveness of decentralising the therapist role and providing training to each site to support local delivery or to operate as a central remote delivery hub accepting referrals from local services.
Results in context
The magnitude of effect of this online ERP is about half the size reported from previous superiority trials of face-to-face HRT/CBIT for tics.152 However, it is difficult to make direct comparisons of therapeutic efficacy with previous trials of face-to-face BT given that this trial had a higher level of baseline tic severity, fewer comorbidity exclusions, a lower proportion of participants receiving tic medication, longer follow-up and a potent active comparator. In practice, the direct comparison of efficacy may also be misleading with respect to implementation because the purpose is not to replace face-to-face therapy but to allow this scarce resource to be better targeted to those who need it most (e.g. children with the most severe tics such as eye poking) and to offer an effective digitally enabled intervention to a much larger population of children/adolescents who are currently unable to access any behavioural treatment for tics.
Major differences between online delivery and face-to-face BT for tics are the reduced amount of therapist time, required skill level of the therapist and cost.
Interpretation of findings
In summary, the findings from this trial suggest that online therapist-supported ERP is a clinically effective and cost-effective BT for reducing tic symptoms in CYP that has durable and sustained long-term benefits. These benefits extend over the longer term beyond tic reduction to improved mood, anxiety and psychosocial functioning. The intervention required approximately a quarter of the therapist time that is required for face-to-face therapy and less highly trained therapists. In health-care settings where tic treatments are difficult to access and with few highly trained expert therapists, therapist-supported online ERP could greatly increase the availability and reach of this durable and cost-effective behavioural treatment for CYP with tic disorders.
Implications for future research and health care
We identify four key areas of future research. First, further research is needed to determine the optimum care pathways with respect to the sequencing and integration of digital and face-to-face BT for tics in CYP and how best to integrate online BT for tics within treatment pathways. Within care pathway research we suggest two delivery models that would be worthy of further study. The first is a ‘stepped care’ model, in which digital/online delivery may be delivered as a first-line behavioural intervention with non-responders or poor responders being ‘stepped up’ to more intensive face-to-face therapy. It is important to note that such a model would not suit all patients, particularly those with complex needs or limited online access, and as such should not be a mandatory first-line treatment before escalation to face-to-face services. We also acknowledge that such a model would require appropriate oversight of dropouts or treatment failures. However, given that the majority of patients currently do not receive any therapeutic support, this model is likely to increase access to evidence-based care. The second model to evaluate is the ‘blending’ of online and face-to-face therapy for more complex cases, thereby reducing the overall number of face-to-face sessions required. These models could be evaluated for their clinical and cost-effectiveness using further RCT designs, including ‘field trials’ or real-world evaluations exploring implementation in clinical practice. Alongside this, it would be important to evaluate whether the models would best be delivered by a specialist centre accepting outside referrals or by training local sites to deliver the intervention.
Second, we did not provide participants with the opportunity to select how they interacted with their therapist. Further research designs could explore the impact of personalisation on effectiveness and engagement, providing opportunities for participants to choose the mode and frequency with which they engage with their therapist.
Third, the value of including parental chapters could be extended and explored in online CYP mental health interventions more widely beyond the ORBIT trial intervention, such as in interventions aimed at other mental health disorders such as low mood and anxiety. The value of including parental chapters was demonstrated by the process evaluation, which showed that parental chapter completion was the only independent predictor of clinical improvement within the ERP intervention arm. Future trials could compare a parent-supported versus non-parent-supported arm.
Fourth, the costs and benefits of online ERP for CYP with tic disorders are likely to extend beyond these 18-month trial data, and hence further health economic analysis is required to project the findings over a longer time horizon using decision modelling. This would also make it possible to include other comparators, such as face-to-face therapy, to evaluate any cost savings of relatively cheap but effective online behavioural interventions compared to ‘gold standard’ best practice. Given the costs of conducting a lengthy RCT, we suggest that this may be best explored using a budget impact assessment as part of a real-world evaluation.
Given the shortage of highly trained therapists with expertise in tic disorders and limited access to BT, online delivery of ERP for tics has the potential to greatly expand the reach of effective behavioural interventions. From a public health perspective, with more efficient use of therapist time it should be possible to treat four people for every one person treated with face-to-face therapy. In addition, the requirement for less experienced therapists to support online BT should expand the potential pool of therapists and thereby further extend the availability of online-delivered BT for tics. A further strength of the online delivery model is that fidelity of therapeutic content is built into the intervention, making transfer to real-world effectiveness much less susceptible than in traditional face-to-face therapy to therapeutic drift and to being impacted by the skill level of individual therapists.
Given that the intervention was developed initially by the Karolinska Team in Sweden and refined by our team in the UK there are implications regarding intellectual property. The ORBIT team, including Nottinghamshire Healthcare NHS Foundation Trust, University of Nottingham, Great Ormond Street Hospital Trust and Karolinska Institute (Sweden), are working with the University of Nottingham Technology Transfer Office to facilitate these matters. Background intellectual property (software and know-how) is held by the Karolinska Institute inventors, while the UK-based parties have all contributed variously to adapting ORBIT for UK users. All parties have agreed a plan leading towards the generation of a vehicle to make the intervention available within the UK.