Results

Parts of this chapter have been reproduced from Wright et al.1 © 2022 The Authors. Journal of Child Psychology and Psychiatry published by John Wiley & Sons Ltd on behalf of Association for Child and Adolescent Mental Health. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. The text below includes minor additions and formatting changes to the original text.

Parts of this chapter have been reproduced from Hayward et al.42 © 2022 Hayward et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. The text below includes minor additions and formatting changes to the original text.

Parts of this chapter have been reproduced from Wang et al.2 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. The text below includes minor additions and formatting changes to the original text.

Main trial results

Internal pilot (stop/go assessment)

In July 2018, the Health Technology Assessment programme considered the performance of ASPECT against the stop/go criteria for the internal pilot (see Chapter 2, Stop/go criteria); to summarise:

• The trial had recruited 82 participants across sites (target, n = 81).
• Sixteen participants had provided data at a final 6-month follow-up point. This equated to 80% of those eligible for the 6-month follow-up, which exceeded the specified target of 70%.
• The DMEC had no concerns regarding trial safety and was supportive of its continuation.

Having successfully met the stop/go criteria, the trial was allowed to progress to the full definitive trial.

Recruitment

Participant recruitment commenced on 30 June 2017. The original recruitment period was due to run for 24 months and be based within five NHS trusts. However, lower than expected recruitment resulted in the period being extended by 7 months, with recruitment continuing until 31 January 2020, at which point the revised target (see Chapter 2, Sample size recalculation) had been exceeded. Alongside this, the number of recruitment sites was increased from the original five to 12 NHS trusts, encompassing 26 CAMHS sites, three third-sector voluntary services and one university-based CYP well-being service. The trial’s follow-up period ran from December 2017 to August 2020.

Participant flow

Figure 2 shows the CONSORT flow of participants through the trial. In total, 274 participants consented and six withdrew prior to randomisation. Two hundred and sixty-eight participants were randomised, of whom 197 (74%) had some 6-month follow-up data and 149 (56%) had primary outcome data. The number of participants recruited exceeded our revised target of 246. More participants were retained at 6 months than our target (197, compared with a target of 178); however, because of the COVID-19 pandemic, in the last 5 months of the trial follow-up period participants had 6-month follow-up data collected remotely and were not able to complete the primary outcome. Table 3 displays reasons for withdrawal in the case of the 71 participants who withdrew post randomisation.

FIGURE 2 The CONSORT flow diagram showing participant flow. DNA, did not attend. Reproduced with permission from Wright et al.1 © 2022 The Authors. Journal of Child Psychology and Psychiatry published by John Wiley & Sons Ltd on behalf of Association for Child and Adolescent Mental Health. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. The figure includes minor additions and formatting changes to the original figure. (Figure not reproduced here; see the original article.)

Baseline characteristics

The characteristics of the randomised participants are given in Table 4 and are well balanced between the groups. Thirteen sites comprising 12 NHS trusts (including 26 CAMHS sites and three affiliated voluntary agency services) and one university-based CYP’s well-being service recruited to the study, contributing between 1 and 78 participants per site. There were more girls (62%) than boys and participants were generally white British (96%). The average age of participants was 12 years. Almost half (47%) of the participants had no treatment preference and slightly more participants had a treatment preference for CBT than for OST (29% vs. 24%). The baseline characteristics of the parents/carers of all randomised participants are given in Table 5; they were more likely to be female (90%) and almost all (97%) were the parents of the participants.

The baseline assessments and patient-reported outcome measures (PROMs) for the participants are given in Table 6. The measures were generally well matched across the treatment groups, and the majority of participants scored either 7 or 8 on the ADIS {median [interquartile range (IQR)]: 8 [7–8]}, meaning that they were assessed as having a severe phobia. The median number of steps completed on the BAT at baseline was three, over half of the CYP completed either 0 or all 10 steps (IQR 0–10) and the mean BAT score was slightly higher in the OST group than in the CBT group. Baseline characteristics and assessments are presented by missing data status (missing or complete primary outcome data at the 6-month follow-up) in Appendix 10, Tables 49 and 50. The characteristics were similar between participants with and participants without missing data; baseline anxiety measures (CAIS and RCADS) were slightly higher in participants with missing 6-month primary outcome data.

Cognitive–behavioural therapy and one-session treatment intervention summaries

Table 7 shows a summary of the CBT and OST interventions. Excluding participants who received no treatment, participants in the CBT group received an average of 6.1 hours of treatment over 5.9 sessions; excluding sessions that occurred after the 6-month follow-up, this reduces to an average of 5.4 hours of treatment. Participants in the OST group had an average of 2.2 sessions conducted over 3.8 hours. For the OST group, 63% received treatment PP, 6% were deemed as having too much treatment (either too long or too many sessions; see Chapter 2, Analysis populations, for the OST PP definition) and 31% did not attend any treatment sessions. Our estimates, based on service closure dates and our multisource data information system (PROSPECT), suggest that COVID affected six (14%) participants’ treatment; three received between one and three sessions of CBT, and three received no treatment sessions. For the CBT group, 57% received treatment PP (at least four sessions), 16% received between one and three sessions and 27% did not receive treatment (Figure 3). Our estimates, based on service closure dates and PROSPECT, suggest that for the approximately 14% (n = 3) of those in the CBT group who received between one and three sessions, this was a direct result of COVID-19, alongside 8% (n = 3) who received no treatment sessions. These figures are likely to be underestimates, as some clinicians did not provide this information.

FIGURE 3 Number of treatment sessions attended before the 6-month follow-up by treatment group. (a) CBT; and (b) OST. (Figure not reproduced here; see the original article.)

Characteristics of the per-protocol population

There was little evidence of differences in baseline site, child and parent/carer demographics, and child measures between compliance groups, based on within-group multinomial logistic regression (see Chapter 2, Statistical methods). Only the EQ-5D-Y was found to be associated with compliance in the CBT group; it was highest for those who received treatment PP (mean 0.787) and lowest for those who did not attend any sessions (mean 0.661). EQ-5D-Y and ADIS composite scores were associated with compliance in the OST group; ADIS composite scores were higher in the PP group and ‘too much treatment’ group than in the group that received no treatment (means of 7.69, 7.75 and 7.24, respectively). For OST, EQ-5D-Y scores were lowest in the ‘did not attend’ (DNA) group.

Therapist summaries

Table 8 shows the characteristics of the 85 therapists who delivered CBT and OST. Similar numbers of therapists expressed a preference for OST (35%) as expressed a preference for CBT (29%). Most therapists worked in CAMHS (85%) and there was a varied range of roles and qualifications.

Primary outcome analysis

The distribution of BAT steps shown in Figure 4 is similar for ITT and PP populations and a larger number of participants achieved 10 steps at 6 months than at baseline in both treatment groups; the majority of participants improved or achieved the same number of steps, even if not receiving treatment (Figure 5). The mean improvement is shown in Figure 6; a marginally larger improvement is seen for the PP group in both treatment groups.

FIGURE 4 Behavioural Avoidance Task last steps by treatment group, time point and analysis population. (a) ITT, baseline (n = 149); (b) ITT, 6 months (n = 149); (c) PP, baseline (n = 149); and (d) PP, 6 months (n = 113). (Figure not reproduced here; see the original article.)

FIGURE 5 Change in BAT last steps over time by treatment group and analysis population. (a) PP; and (b) ITT. (Figure not reproduced here; see the original article.)

FIGURE 6 Bubble plot of BAT last steps by treatment group for the ITT population. (a) CBT; and (b) OST. Reproduced with permission from Wright et al.1 © 2022 The Authors. Journal of Child Psychology and Psychiatry published by John Wiley & Sons Ltd on behalf of Association for Child and Adolescent Mental Health. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. The figure includes minor additions and formatting changes to the original figure. (Figure not reproduced here; see the original article.)

Among the participants completing the BAT at 6 months, the number of steps taken was similar across treatment groups. The mean BAT last steps at 6 months in the CBT group was 7.1 for ITT and 7.4 for PP; in the OST group it was 7.4 for ITT and 7.6 for PP (adjusted mean difference for CBT compared with OST –0.46, 95% CI –1.43 to 0.51 for ITT; mean difference –0.73, 95% CI –1.83 to 0.37 for PP) (Table 9). In each of these analyses the effect is larger for the PP group than the ITT group. On the standardised scale, 95% CIs for both ITT and PP were wholly below the standardised non-inferiority limit of 0.4 (Figure 7) and so there is evidence that OST is non-inferior to CBT.

FIGURE 7 Primary and sensitivity analyses of BAT steps between treatment groups on the standardised scale. Reproduced with permission from Wright et al.1 © 2022 The Authors. Journal of Child Psychology and Psychiatry published by John Wiley & Sons Ltd on behalf of Association for Child and Adolescent Mental Health. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. The figure includes minor additions and formatting changes to the original figure. (Figure not reproduced here; see the original article.)

A number of sensitivity analyses were undertaken on the primary outcome (BAT last steps at 6 months) with regard to excluding mistimed measurements, multiple imputation for missing data and CACE analyses to explore treatment adherence. In all cases, the treatment difference remained similar and the CI was wholly below the non-inferiority limit (Figure 8). For mistimed measurements and multiple imputation, the treatment difference was closer to zero than the other analyses that were marginally in favour of OST.

FIGURE 8 Primary and sensitivity analyses of BAT steps between treatment groups on the raw scale. (Figure not reproduced here; see the original article.)

Secondary outcomes analysis

The comparison of secondary outcomes at 6 months is presented in Table 10. All treatment differences were small, with their corresponding CIs crossing zero, and hence were in keeping with the results for the primary outcome; the CBT and OST treatments appear comparable for the secondary outcomes. Some point estimates were slightly in favour of CBT and others were slightly in favour of OST; none were statistically significant.

The number of participants considered to still have specific phobia at 6 months (based on the ADIS CSR) is presented in Table 11; 73% of the ITT population were considered to still have a phobia diagnosis at 6 months, this was the same across treatment groups. Fewer PP participants had specific phobia at 6 months, slightly more in the CBT group than in the OST group (odds ratio 1.41, 95% CI 0.60 to 3.32).

Subgroup analysis

The potential moderating effects of four predefined subgroups were explored by including an interaction between treatment and subgroup in the mixed-effects regression model. Results of the subgroup analyses are presented in Table 12 and summarised in a forest plot in Figure 9. No reliable statistical evidence of subgroups effects or interactions were found between the treatment groups. The adjusted mean difference for treatment in BATs where the stimulus had to be simulated (for example BII) was similar to the BAT real stimuli, and both CIs were wholly below the non-inferiority limit. The adjusted mean difference appeared larger in favour of OST when the participant preferred OST and when the therapist preferred OST.

FIGURE 9 Mean 6-month BAT last steps score by subgroup and treatment group (n = 149). (Figure not reproduced here; see the original article.)

Assessment of post-treatment follow-up time in relation to outcome

The number of weeks between treatment finishing and outcome assessment was likely to be different between the treatment groups, as OST is completed in fewer sessions. Figure 10 compares the distribution of days between a participant’s last session and their outcome assessment. Thirty participants in the CBT group had recorded treatment sessions after outcomes were collected compared with three participants in the OST group. These show as negative values on the plot where the last session received was after the primary outcome measure. The second panel was capped at zero; therefore, if the last time was negative it was classed as zero. The rightmost panel shows the time from the last session received before the primary outcome measure. Adjusting for days between last session that occurred before 6 months’ follow-up in the PP analysis (so participants who had no sessions are excluded from the analysis) had minimal impact on the estimation of the difference in BAT scores between the treatment groups (changed the PP treatment difference to 0.67, 95% CI –0.60 to 1.95). Including an interaction between treatment and days between session and follow-up in the mixed-effects analysis was not statistically significant.

FIGURE 10 Days between last session and follow-up by treatment group. (a) Days between last session and follow-up; (b) days between last session and follow-up, capped at 0; (c) days between last session and follow-up within 6 months. An outlying observation of –324 has been excluded from (a) for readability. (Figure not reproduced here; see the original article.)

Baseline anxiety and depression

The potential moderating effect of baseline anxiety (measured by RCADS child total anxiety score) and depression (measured by RCADS child depression score) was explored. The primary analysis model was repeated with the addition of an interaction term between RCADS and treatment group; the coefficients for the interaction terms are given in Table 13. There was weak evidence of a negative relationship between RCADS total anxiety score and 6-month BAT last steps, which was more marked for the CBT group, as is shown in Figure 11.

FIGURE 11 Baseline anxiety and 6-month BAT last steps by treatment group. Best-fit lines for CBT and OST from linear regression. (Figure not reproduced here; see the original article.)

Impact of site and therapist

The impact of site on treatment difference is presented in Table 14 and Figure 12. Sites that had at least five participants per treatment group with outcome data were included in the analysis. The treatment difference across sites was broadly similar although based on small numbers at most sites.

FIGURE 12 Mean 6-month BAT last steps score by site. (Figure not reproduced here; see the original article.)

Therapists saw too few participants each on average to make meaningful comparisons (between 1 and 14 participants per therapist in the study). The ICC for the clustering effect of therapist in the primary analysis model reflected this (therapist ICC < 0.00001). The ICC was slightly increased when repeated on therapists who saw at least two participants (ICC < 0.001), indicating that there was little clustering within therapist. The therapist characteristics and their moderating effect on treatment are presented in Table 15 and Figure 13. There is some evidence that the number of years of experience delivering psychological interventions to CYP has a moderating effect on treatment difference (p = 0.019); for therapists with 0–2 years of experience, the mean treatment difference is in favour of CBT, and the two subgroups with more years of experience are increasingly in favour of OST.

FIGURE 13 Mean BAT last steps at 6 months by therapist subgroup. (Figure not reproduced here; see the original article.)

Unblinding

Researchers conducting the outcome assessments were blind to treatment group allocation. In cases where treatment group was revealed between baseline and the 6-month follow-up, a different researcher was sought to conduct the 6-month outcome assessment. In many therapy sites that were distant from the study coordinating centres in York and Sheffield, this was not always possible because of no available alternative research assistant. Overall, 14 unblinding incidents (over 13 participants) occurred for the CBT group and 20 incidents (over 16 participants) occured for the OST group. The majority were accidental (91%) via disclosure from the participant (38%) or therapist (26%). Summaries of the unblinding incidents are presented in Appendix 3, Table 30.

Safety and harms

The AEs and SAEs are summarised in Table 16. Four SAEs occurred during the course of the study, all were inpatient hospitalisation and were assessed by the relevant site principal investigator as unrelated to phobia treatment.

Fidelity assessment results

The recording of sessions for the trial was much lower than expected, with 55 CBT sessions recorded (from 12 participants) and 15 OST sessions recorded (from 12 participants) out of a total of 889 sessions (from 203 participants). Of these sessions, 14 out of 55 of the CBT sessions were the first or second session and, therefore, were not included in the fidelity assessment (leaving 39 CBT sessions), and only 4 out of 15 of the OST sessions were exposure sessions; the CBT sessions were chosen randomly, and the four OST sessions were all reviewed and allocated randomly to the assessors.

The fidelity assessment of the audio-recordings was high in all recorded sessions; the average was 84.6% for OST and 76.9% for CBT, and ranged from 61.5% to 100% in both groups (see Appendix 4, Table 31). Multisession CBT may distribute therapeutic techniques across sessions and so any comparisons need to take this into consideration. Using established criteria from Borrelli et al.,65 assessed sessions were classified as having either high fidelity (8/15 sessions) or moderate fidelity (7/15 sessions), with no sessions classified as having low fidelity.

The fidelity assessment of competence showed that none of the therapists scored lower than 2 for any of the competencies assessed in the session recording, and the total score was higher than 50% for all recorded sessions, which means that all sessions ‘passed’ using the Stallard et al.63 pass criteria (see Appendix 5, Table 32). The average score for the OST assessed sessions was 56.5 (range 55–58) and for CBT it was 70.3 (range 55–83), indicating that the therapists in the CBT assessed sessions were more competent.

Health economic results

Introduction

The primary aim of the economic evaluation was to assess the cost-effectiveness of OST compared with CBT from the NHS and PSS perspective. An economic evaluation from the societal perspective was included in the sensitivity analysis, which included costs from the educational system, private out-of-pocket expenses and parental productivity costs reported in the trial.

Missing data

Availability of quality-of-life and cost data

A total of 340 CYP were recruited to the trial. After removing 72 ineligible CYP, 268 CYP with specific phobias were available for analysis (134 were allocated to OST and 134 to CBT). The complete case here refers to the CYP who completed EQ-5D-Y questionnaires and whose parents completed resource use questionnaires at any time point. Details of the QoL and cost data available at each data collection time point are shown in Table 17. As shown, QoL data (EQ-5D-Y) were fully available at any time point for 193 (72.0%) CYP, and cost data from the NHS and PSS perspective were available for 193 (72.0%) CYP. Overall, a total of 190 (70.9%) CYP had both EQ-5D-Y and resource use (from the NHS and PSS perspective) data at both collection time points. This sample constitutes the complete-case group, which is one of the two groups used for all the subsequent analyses. The second group (imputed-case or base-case group) is discussed in Multiple imputation.

Multiple imputation

To account for the data from the sample other than the complete-case group (29.1%), missing utility scores and costs were further imputed using multiple imputation via chained equations (assuming missing at random at each time point). The following variables were used in the imputation process to ensure best fit of the imputed results: trial arm, age, sex, site, underlying mental health conditions [ASD and attention deficit hyperactivity disorder (ADHD)], EQ-5D-Y utility scores and ADIS highest CSR scores at baseline. Given that it would be speculative to impute missing utility at baseline, one child or young person (0.4%) was excluded from the health economics study. The base-case (imputed) sample was 267 (133 allocated to OST and 134 allocated to CBT); this is the sample used for the primary cost-effectiveness analysis.

Baseline characteristics

The descriptive statistics of participating CYP characteristics and the additional predictors used in multiple imputation are presented in Table 18. Around one-third of the CYP in the OST and the CBT groups were male, and more than 50% of the CYP in both groups were of secondary school age (ranging from 11 to 16 years). Differences in the ADIS scores and EQ-5D-Y utility scores at baseline were marginal across arms and samples. Overall, the baseline characteristics are consistent across samples (base case and complete case) and with the main statistical analysis.

Cost

Two types of costs were considered in this study: the cost of intervention (obtained from the therapists) and the cost of service use (self-reported by the parents/guardians). All costs were expressed in 2018–19 Great British pounds.

Unit costs

Individual-level resource use together with unit costs were used to calculate the total services cost for each child and young person with a specific phobia. Unit costs were obtained from published sources (i.e. Reference Costs 201874 and the unit costs of health-care and social care from the PSSRU 201875). Appendix 6, Tables 33–41, presents the summary of key unit costs used in the study.

Intervention costs

Intervention costs include both training and intervention delivery costs. The training cost information was obtained from our study team, and included both ‘train the trainers’ and ‘therapist training’ costs. The estimated training costs were allocated to each session and each child and young person who received the intervention. In relation to the intervention delivery costs, the information was obtained directly from the therapists using the self-reported questionnaires. Intervention sessions were costed on the basis of the salary of the therapists involved (for both preparation and session delivery) and the additional resources required. Table 19 shows the total and average costs of each component that were used to deliver the OST and CBT sessions. As shown, the main cost driver of training costs was trainer fees, whereas the main cost drivers of intervention delivery costs were the costs for intervention preparation and delivery and the costs for stimuli. On average, the estimated intervention costs for OST were £184.26 (£62.19 for training and £122.07 for intervention delivery) and for CBT were £58.59 (£0 for training and £58.59 for intervention delivery).

Service use and costs

Table 20 shows the summarised service use per CYP in each treatment group based on the complete cases only. The difference in service use between the OST and the CBT groups varies across items at two time points. At baseline, CYP in the OST group seemed to have used slightly more NHS services and incurred more private expenses and productivity than those in the CBT group. At the 6-month follow-up, resource use in both treatment groups decreased and such decreases were more prominent in the OST group than in the CBT group. The medication use for both treatment groups at baseline and the 6-month follow-up remained largely unchanged; the majority of medicines used were for conditions such as allergy, asthma, diabetes and mental health.

The total costs broken down by perspective, type of service and treatment group before and after imputation are presented in Table 21. As shown, the costs substantially varied from one child or young person to the other, as illustrated by the wide 95% CIs.

The total service costs to the NHS providers (before imputation) were £513.49 (95% CI £351.10 to £675.89) for the OST group compared with £549.05 (95% CI £362.41 to £735.70) for the CBT group. The average costs for health-care services were similar in the two groups, except the CAMHS costs. The average cost for the use of CAMHS-related services was marginally higher for the CYP with specific phobias in the CBT group (£210.47 in the OST group and £277.34 in the CBT group). Such a cost difference was likely to have been driven by one patient in the CBT group who reported 25 psychotherapist appointments over the past 6 months and, thus, incurred a high cost. Given that this value was entirely plausible, we kept it in the data set without any adjustment.

The average total cost for the use of education services (before imputation) was £28.77 (95% CI £8.13 to £49.42) in the OST group compared with £16.24 (95% CI £2.97 to £29.51) in the CBT group. As with the costs of CAMHS-related services, the higher average cost in the OST group was likely to have been driven by two high-cost patients, the first of whom received 10 education welfare officer visits and 11 school nurse visits over 6 months. Again, we kept these values in the analysis, as the scenarios were plausible.

Children and young people in the OST group also incurred lower average costs in private expenses and parental productivity losses than those in the CBT group. Overall, CYP in the OST group incurred lower average costs in both NHS/PSS and societal perspectives. This is observed in both the complete case and the base case. However, owing to the high-cost cases, the cost differences need to be interpreted with caution.

Quality of life

EuroQol-5 Dimensions Youth version

Table 22 shows the mean EQ-5D-Y (three-level) utility scores in the two treatment groups at each time point when scores were not imputed (complete case) and when scores were imputed (base case). It can be seen that in both treatment groups there was no significant change in EQ-5D-Y scores from baseline to 6 months. The fluctuations between baseline and 6 months were small in both the base case and the complete case. After calculation using the area under the curve approach (see Chapter 2, Effectiveness), it was found that OST produced similar mean QALYs (0.43 QALYs) to CBT. Further details for the responses in each domain can be found in Appendix 7, Tables 42 and 43.

Child Health Utility 9D

The change in the mean CHU-9D scores between the two treatment groups at any time point is presented in Table 23. CYP in the CBT group had slightly higher utility scores at baseline than those in the OST group. As with the EQ-5D-Y results, the observed changes at the 6-month follow-up in either treatment group were marginal. Such marginal differences were observed in both the complete case and the base case. Consistent with EQ-5D-Y results, the OST produced similar mean QALYs (0.43 QALYs) to CBT. Further details for the responses in each domain can be found in Appendix 8, Tables 44 and 45.

Behavioural Avoidance Task

The change in the mean BAT scores between the two treatment groups at any time point is presented in Table 24. CYP in the CBT group had slightly higher BAT scores than those in the OST group at baseline. The observed reductions at the 6-month follow-up in either treatment group were between two and three score points. Such differences were observed in both the complete case and the base case. In total, CBT reduced mean BAT scores slightly more (0.28 score points in base case and 0.32 score points in complete case) than OST. These represent incremental physical approaches to the feared stimulus (e.g. entering a room and approaching and stroking a dog for a person with a dog phobia). These all occur on 10-point scales, so a three-point difference is a clinically significant movement (e.g. when someone has not been able to enter the room scoring zero).

Economic analysis

Primary analysis

Table 25 shows the ICERs based on the base case for the primary analysis. Before any adjustments, on average, CYP with specific phobias receiving OST incurred £273.08 less costs than those receiving CBT from the NHS/PSS perspective, and maintained similar QALY improvements, as measured by EQ-5D-Y.

To account for the uncertainty and adjust for the imbalanced EQ-5D-Y utility scores and health-care costs at baseline, the estimates of incremental costs and the QALYs from regression were bootstrapped to simulate 5000 pairs of net cost and net outcomes, as recommended by NICE for health technology appraisals.96 The regressions were controlled for age, sex, site, phobia type, EQ-5D-Y utility, health-care costs and ADIS CSR score at baseline. After bootstrapping, on average, CYP with specific phobias receiving OST incurred £302.96 (95% CI £28.61 to £598.86) less costs and gained 0.002 (95% CI 0.002 to 0.002) more QALYs than those receiving CBT. The calculated ICER was below the recommended threshold range specified for NICE decision-making in England and Wales (£20,000 to £30,000 per QALY gained), further confirming that OST is likely to be cost-saving.

Figure 14 shows the cost-effectiveness plane for OST compared with CBT based on 5000 bootstrapped estimates of incremental costs and incremental QALYs. The navy line represents the willingness-to-pay threshold of £20,000 per QALY gained. The simulated estimates were largely below the threshold line and sat in the fourth quadrant, suggesting that OST is likely to be cost-saving.

FIGURE 14 Base-case cost-effectiveness plane of OST compared with CBT (outcome measure: QALY measured by EQ-5D-Y, NHS/PSS perspective). WTP, willingness to pay. Adapted with permission from Wang et al.2 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. The figure includes minor additions and formatting changes to the original figure. (Figure not reproduced here; see the original article.)

The CEAC of OST compared with CBT is presented in Figure 15. The probability of OST being cost-effective is 98% if decision-makers are willing to pay £20,000 for one QALY gained and 99% if they are willing to pay £30,000 per QALY gained.

FIGURE 15 Base-case CEAC of OST compared with CBT (outcome measure: QALY measured by EQ-5D-Y, NHS/PSS perspective). WTP, willingness to pay. Adapted with permission from Wang et al.2 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. The figure includes minor additions and formatting changes to the original figure. (Figure not reproduced here; see the original article.)

Sensitivity analyses

To account for uncertainty in the costs and QALY estimations, a set of sensitivity analyses were conducted (Table 26). The mean incremental cost and QALY estimates from the complete case were along the line of the base-case scenario, yielding a negative cost per QALY gained. The same was observed in the group that received at least one intervention session during the trial follow-up period. Those who did not receive any intervention sessions owing to the national lockdown caused by the COVID-19 pandemic were excluded from the analysis. Other sensitivity analyses using different cost perspectives and a different instrument (CHU-9D) to measure QALYs, and excluding one-off OST training costs, were also conducted. In the above-mentioned sensitivity analyses, the mean ICER pairs lay below the recommended NICE threshold (£20,000–30,000 per QALY gained) (see Table 26), and the majority of the bootstrapped estimates sat in the fourth quadrant and below the NICE threshold line (Figure 16), suggesting that OST is likely to be cost-saving and cost-effective. The cost-effectiveness of OST is supported by a range of sensitivity analyses; however, this is subject to the caveat that results are derived from a very small incremental change in the effectiveness parameter. When the measured health outcome was the BAT score, the calculated ICER was £1848.82 per BAT score. This means that, on average, the NHS needs to pay £1848.82 to decrease the BAT score by 1 point.

FIGURE 16 Sensitivity analysis: cost-effectiveness plans of OST compared with CBT. (a) scenario 1, NHS/PSS + EQ-5D-Y (complete case); (b) scenario 2, NHS/PSS + EQ-5D-Y (CYP had at least one intervention); (c) scenario 3, NHS/PSS + CHU-9D; (d) scenario 4, societal + EQ-5D-Y; (e) scenario 5, NHS/PSS + BAT scores and (f) scenario 6, NHS/PSS + EQ-5D-Y (excluding OST training cost). WTP, willingness to pay. Adapted with permission from Wang et al.2 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. The figure includes minor additions and formatting changes to the original figure. (Figure not reproduced here; see the original article.)

Summary

Compared with CBT, OST marginally decreased the service use costs and maintained similar QALY improvements. This is evident in both primary and sensitivity analyses, which considered costs derived from various perspectives and QALYs measured by different instruments.

Qualitative results: the feasibility and acceptability of one-session treatment

Intervention acceptability is a key priority in the development, implementation and evaluation of therapeutic initiatives. In applied health research, acceptability is typically defined as ‘a multi-faceted construct that reflects the extent to which people delivering or receiving a health-care intervention consider it to be appropriate, based on their anticipated or experienced cognitive and emotional responses to the intervention’.97 Nesting qualitative research within RCTs is recommended to explore intervention acceptability fully.

Acceptability is generally perceived to have an important, although not exclusive, influence on treatment uptake, engagement and outcomes. In ASPECT, intervention uptake (i.e. the proportion of CYP receiving one or more sessions of their allocated treatment) was 69% and 73% for OST and CBT, respectively. Quantitative data reporting clinicians’ and families’ baseline treatment preferences are presented in Chapter 4, Baseline characteristics. Although informative, summary statistics can act only as proxy indicators of intervention acceptability and are limited in their ability to elucidate essential process learning.

Three qualitative studies were nested within ASPECT. The first two explored the acceptability of OST from the perspective of CYP who received it and their parents, and the third explored the feasibility and acceptability of delivering OST from the perspective of trial clinicians (see Chapter 2, Qualitative methods, for the methods for each of these studies).

Acceptability, when defined as a cognitive–emotional construct, has both prospective and retrospective components and can be influenced by cognitions or emotions occurring before, during and after intervention delivery. The extent to which these converge or diverge is ultimately an empirical question for the researcher.

Cognisant of this possibility, our findings are presented in line with the objectives of this study, namely to explore CYP, parent and clinician experiences prior to participation in ASPECT along with their experiences of OST delivery, and their subsequent reflections on intervention impact and acceptability. These objectives provided a basic, deductive framework for our analysis. Thorough descriptions of the inductive interpretive themes relevant to each objective are presented in turn, supported by direct quotations from the data. Participants are assigned a unique identifier rather than a name or pseudonym in the text, and relevant demographics are included for each. Additional data are included in the summary coding tables in Appendix 9, Tables 46–48.

Study 1: acceptability of one-session treatment to trial participants

The aim of the first study was to explore the acceptability of OST from the perspective of CYP participating in the trial.

Representativeness of the sample

All CYP who received OST and provided consent to be contacted for interview were eligible for inclusion (34 were approached and 27 participated). Baseline characteristics for the participating CYP are presented in Table 27. Data for the randomised ASPECT participants (n = 268) are provided for comparative purposes.

Children’s views

Children’s experiences of phobia prior to ASPECT

The qualitative analysis interpreted four themes from the data relating to CYP’s experiences of phobia prior to ASPECT. These themes highlighted (1) divergence and commonalities in child-led concepts of phobia, (2) widespread child and family burden, (3) a history of ineffective support and (4) a rising recognition of treatment need.

Divergence and commonalities in child-led concepts of phobia

Divergence in CYP’s conceptualisations of phobia led to the delineation of two groups: those who attributed their difficulty to a discrete incident and those who had no clear memory of how or when their phobia emerged. Phobia was more readily rationalised as a logical response to a triggering stimulus in those who could pinpoint a seminal event:

I was at the beach when I was little and then my dad was trying to take a photo of me. And when my dad was trying to take a photo of me, a dog came up to me and knocked me over then licked me, so I kind of started with it like that.

Common to all children was a desire to direct their energy towards conquering the functional challenges of phobia rather than to further their understanding of its origin or cause. A shared narrative among CYP endorsed a relatively simplistic, developmentally appropriate concept of phobia, in which an exaggerated fear response was conceived as an unwelcome entity that imposed unnecessary restrictions and generated feelings of sadness and shame:

I think that was kind of the main thing was that I was just getting quite embarrassed about not being to have it. Or going down and them all being – questioning me on why I’m not having it or feeling a bit like ‘oh why am I so afraid?’. You do just kind of question like ’cause I didn’t, you don’t really understand, make sense of it.

Widespread child and family burden

The interviewed CYP described moderate to high levels of burden, affecting both the individual and their wider networks. A communal discourse delineated a growing sense of academic and peer exclusion, with access to specific classes or social events prohibited by phobias. Among older CYP, in particular, the avoidance of phobic stimuli had often led to the initiation of safety behaviours, with emotional reassurance derived from the long-term maintenance of self-imposed routines:

I refused to go places because of it, especially during the winter months when the bugs were going round. I wouldn’t leave the house because I’d be scared of catching something.

Irrespective of age, most CYP described a growing reliance on behavioural adaptations that extended beyond their own social spheres to those of their family members. They described a heavy reliance on kin support for stimuli avoidance and acknowledged that, in this context, their immediate family absorbed a lot of the practical and emotional impact:

Well, I couldn’t watch certain TV [television] programmes. I couldn’t go into some shops. And they even appeared on websites too, which means, so it meant we had to be careful where we looked.

Consistent in CYP’s narratives was the notion that forced exposure by informal carers was unhelpful. Some CYP recounted specific incidents in which parents had deliberately introduced them to the stimuli without fully understanding the impact that this would have. Most CYP interpreted these attempts as well-meaning, although collusion and coercion on the part of family members was occasionally identified, engendering a sense of mistrust in those affected:

Well, my mum and dad tried to help me go under the high roofs of like the bits I was scared of, to try to overcome my fear of it. But it didn’t really work.

A history of ineffective support

Children frequently alluded to a tendency for parents and peers to underestimate the severity of their phobia, precipitating a sense of cognitive dissonance (i.e. a misalignment with thoughts, beliefs and behaviour) in some participants. In older CYP in particular, awkwardness and embarrassment had emerged. Persistent minimisation of phobia led some to question the legitimacy of their condition and the credibility of their treatment need:

I had to go out school sometimes to go to the counselling things and then and they were a bit like ‘where are you going?’. I’m like ‘oh yeah I’ve got a phobia of needles’ and they’re like ‘why don’t you just have it? It isn’t that bad’, and then you kind of think, ‘am I just being dramatic?’. You kind of feel a bit bad about it. Yeah, yeah a bit kind of embarrassed I feel like I should be able to have this, why can’t I have this? It just doesn’t feel too great.

Only a minority of CYP conceived their phobia as a diagnosable condition sufficient to warrant health service intervention. Previous support accessed by this subgroup of CYP included generalised anxiety management, hypnotherapy and play therapy, but each of these interventions had left them disappointed. CYP attributed their lack of effect to either a failure in problem recognition or the intervention content. Exposure and response prevention was regarded as a crucial element in phobia treatment and one that was lacking in non-specialist interventions. A shared discourse among CYP revealed a striking lack of awareness of phobia-specific treatment options:

No, ’cause I didn’t know it was available, so I didn’t even know it was a thing. So, I’d never gone to anyone. Even people I’d spoken to about it, I hadn’t even realised it was a thing.

Symptom minimisation served to both delay formal help-seeking and discourage onward referral. A small number noted that their phobia had been dismissed or downplayed by health-care professionals in the past, which they had found to be upsetting. Very few described physical or financial barriers to phobia treatment, most choosing to focus instead on attitudinal and emotional deterrents. Alongside professional and social attitudes, CYP described a wealth of internal barriers to treatment, including their own fears and anxieties about what formal treatment would involve:

I don’t think you want to go because of what you’re talking about and it’s a sensitive subject when you’re scared of it.

Rising recognition of treatment need

CYP’s apprehension towards formalised phobia treatment contrasted sharply with their desire for effective intervention. Participants described an urgent and growing demand for symptom improvement, alluding to a potential turning point at which the anticipated benefits of treatment began to outweigh the costs. Adopting an ego-centric view, CYP were motivated to access treatment predominantly to ameliorate internal distress or to counter imminent health risks and offset risk behaviours:

For my meningitis injection, I could only get the polio and tetanus ones, and not the meningitis injection which is quite important. So, it was a pretty big and urgent matter to get, to sort it out.

Well ever since I was little, I’ve been running into roads to avoid them. And it was, it was getting to the point where it probably needs sorting out, otherwise something dangerous could have happened.

The optimal outcome of OST, from the CYP’s perspective, was a meaningful reduction in anxiety, with distal benefits for daily functioning and QoL. Although younger CYP tended towards expectations of a cure, older individuals were typically more cautious. Cognisant of the length of time that they had already lived with their phobia, older participants sought to moderate (rather than eliminate) their fear, dampening it down to a level more conducive to their daily routine:

Well I sort of like the idea of getting the phobia at least somewhat under control, ‘cause that genuinely that would take a massive hassle out of everyday life.

Children and young people consistently expressed a hope that OST would work, but acknowledged that prior to the treatment they did not understand its primary mechanism of action. Some queried the extent to which an engrained phobia could be separated from the self, while others questioned how readily their stimuli could be replicated and how realistic this would be:

I sort of went in being like this isn’t going to help you know ’cause obviously they can’t get a really giddy dog because they can’t have one that you think might hurt you and whatnot.

In some instances, CYP’s expectations were directly influenced by their previous treatment experiences. Those who were able to draw on prior experiences of CBT believed that OST would have a beneficial impact, while those with experiences of less-structured therapies were often more sceptical:

I knew from CBT very recently that it’s not just going to completely fix a problem and I wasn’t expecting that at all. But I was expecting it to sort of like ease the like impact it had on my life and be able to like give me methods to cope with it rather than just avoidance behaviours, ‘cause that isn’t really healthy and it gets in the way of things I want to do.

I was just more like oh this probably gonna be like the play thing and probably it’s – probably is gonna go and do everything I’m probably not gonna have anything done from it but obviously I did.

Common to all CYP’s narratives was the notion that deciding to engage in OST had been guided less by a guarantee of effectiveness and more by a lack of opportunity costs. Expectations of more efficient delivery under the OST model minimised the cognitive and emotional barriers that had historically prevented CYP from accessing treatment, leaving them with sufficient impetus to engage in this intervention:

I think going into the 3-hour sessions I was thinking like it’s going to be better than having to like do a session and then work and like – kind of in between that space like panic about it and just getting done.

Children’s experiences of one-session treatment delivery

Three interpretative themes were derived from CYP’s accounts of their experiences of OST delivery within the trial. These themes identified three key elements critical to treatment acceptability: (1) session format, (2) expert facilitation and (3) partnership working.

Session format

Most CYP recognised and appreciated the stepped format of OST, accepting this as a largely suitable way to treat phobias. Specifically, graded exposure was viewed as an adaptable and secure way to challenge their fears and manage their anxiety responses:

I think because it’s so gradual you don’t ever feel so out of your comfort zone that it’s impossible. Because it’s just something you’ll have just done, it’s just like a tiny little step of what you’re about to do.

Some commented on the length and intensity of the session, noting that the intervention left them feeling emotionally exhausted. These views did not appear to be age or phobia dependent however, clustering predominantly in CYP with additional learning needs (e.g. ASD). A strong counter-narrative identified session length as a beneficial feature of OST, advocating the continuity and efficiency of intervention delivery as a specific advantage of this treatment modality:

It was a lot and I did kind of get overwhelmed towards the end of it and I was like, I need to stop now ’cause it’s just too much.

If you were to go through the CBT, you’d be going back and forward and then there’s petrol and then there’s work and then there’s school. And all these things that need to be taken into consideration. But, with the one-session treatment, it’s half a day. You’re done.

Some children acknowledged that having a maximum of two sessions was less intrusive on their daily lives, minimising travel burden and opportunity costs and maximising their propensity for therapy engagement. Most, however, focused on emotional gains, closely aligning their in-session experiences with their original expectations of treatment. The condensed format of OST was appraised positively by most CYP, who appreciated not having to revisit previous sessions. A common view was that single-session therapy minimised the build-up of pre-treatment anxiety, thereby reducing their temptation to revert to safety behaviours or drop out of treatment completely. Although a direct comparison with other therapy models was not always possible, CYP hypothesised that the continuity of OST had increased the efficiency of their exposure, enabling them to capitalise on early progress and improve their outcomes:

Like you know because say if you didn’t enjoy this, then you got to go back for 12 other sessions or how long it was, it just seems a bit like you dread it a bit, and it doesn’t really be as effective.

. . . it’s better to sit there for 3 hours and try and feel comfortable with the thing you don’t like than dread the week, the day of the week you have to go.

Expert facilitation

Most CYP expressed positive views about the therapeutic environment, with their experiences particularly influenced by physical features of the clinical setting and, more frequently, the OST facilitators’ communication style. CYP consistently described the OST therapists as kind, friendly and relaxed, and perceived their gentle encouragement and non-judgemental attitudes as an important determinant of intervention success:

Even though there was the white colour scheme and the grey bland colour scheme, it still made me feel welcomed because of the rugs, the soft chairs, the nice people.

I think she understood me, so she’d know when I was a bit like ‘no!’, and then she’d be able to encourage me when she thought I could do a bit more, so it’s the right balance.

Therapist competency, as well as their confidence, was identified as a core ingredient of effective facilitation, placing OST firmly within the remit of a trained health professional’s role. Considerable value was placed on flexible, child-centred approaches, with different narratives highlighting the potential for individual therapist effects. Although a small number reported feeling rushed through their treatment, the majority described an intervention that was coherent and empowering, with progress determined at an appropriate pace:

It was definitely a well-thought-out, more like, well-designed intervention. And the actual like getting, having more control over it, meant it felt more like personalised and you were in control.

Partnership working

High importance was placed on partnership working, with many CYP regarding the OST assessment session as an invaluable opportunity to meet their therapist and input directly into their treatment session. They reported deriving a breadth of benefits from collaborative treatment planning; one participant rationalised that it was easier to progress through treatment when the steps were known in advance and another suggested that the use of self-directed challenges maximised treatment relevance and meaning:

You feel like you’re more involved in the actual therapy, you don’t feel like it’s just been planned by someone else like for you. You feel like it’s more personalised to your own, like, fears and what and how far you actually want to go.

Successful collaboration was predominantly defined in terms of a synergistic relationship established between the OST therapist and the child. Very few children expressed the need to have a parent or carer present, although, where this occurred, it was positively received, helping children to articulate their thoughts and providing reassurance:

I mean for yeah for me it was useful kind of having someone there ’cause of course I don’t know [therapist name] that well I don’t – didn’t know the nurse that well so having someone that I know really well as support is something really useful for me anyway.

Children’s reflections on the impact and acceptability of one-session treatment

The qualitative analysis interpreted a final three themes from CYP’s post-therapy reflections. These themes were (1) children’s perceptions of treatment outcomes, (2) a potential role for participant suitability criteria and (3) possible modifications to treatment delivery.

Children’s perceptions of treatment outcomes

CYP’s narratives revealed a tendency for pre-treatment expectations and post-therapy reflections to diverge. Those who reported feeling anxious before therapy recounted treatment experiences that were much more positive than they had anticipated, CYP’s appraisals of treatment outcomes were also largely positive, with many perceiving that they had either achieved or exceeded their intervention goals:

. . . it’s always worse in your head than it is when you’re actually going through it. And I was really nervous going in, but when I was actually having the treatment I was kind of like ‘Why was I scared? What was there to be scared of?’.

I had become so used to this life of living with phobias that I thought that that was what life is. And so when I was kind of like oh this isn’t what – I don’t know have to worry about everything like every single second of a day. It was kind of – it was a nice surprise you know. It was a nice surprise to not have to worry so much.

Overall, CYP reported feeling less anxious after OST. Valuing the realism of the stimuli that were encountered, they consistently praised the creativity and interactivity of OST and the learning opportunities embedded within OST. OST’s capacity to mimic real-world scenarios was frequently highlighted as an important element of the intervention, increasing CYP’s sense of self-efficacy and providing them with self-management skills:

Even when we came out afterwards there were some dogs on the street and, mum was amazed that I could just walk past them.

In some instances, OST was reported to have had secondary benefits, having taught skills that were transferable to other developmental challenges and aspects of life. In particular, older CYP emphasised the ‘spillover’ effects of treatment, retrospectively identifying the positive effects of OST for both themselves and their network members:

I still get quite worked up about come quite distressed about my mocks because of my GCSEs [General Certificate of Secondary Education] and they’re stressful. But I was doing some breathing this morning as well kind of just going up on my way to school ’cause getting quite worked up and it is it was the kind of the same thing where I went, I was quite nervy, no before my exam, I went through my exam it wasn’t as bad as I thought it would be and then I’ve calmed myself down at that point so it yeah, useful things.

Yeah, I feel a lot happier in myself as a person and around other people. And I think people feel more relaxed around me – like they don’t have to constantly be like – like worrying about what they say or anything.

Some children reported that the language used by the OST therapists had helped them to understand the origins of their phobia and enhanced their knowledge of physiological responses to phobic stimuli. Although not explicitly prioritised as one of their goals at therapy outset, CYP upheld this benefit as a valuable element of intervention delivery, acknowledging that it had played a central role in normalising their experience, reducing their embarrassment and nurturing a greater sense of internal control:

Because it helped me to understand why I was scared of them. It might sound weird but I never really knew why I was scared of them. So I was scared of them, up until now I didn’t know why.

A minority of CYP reported that OST had not enabled them to reach their treatment goals. Although disappointed, these individuals acknowledged that there were few opportunity costs to OST participation, accepting it as a helpful platform for additional intervention. Consequently, nearly all recommended OST an acceptable support mechanism for other CYP, albeit with the caveat that progress could not be guaranteed:

I’d explain how it worked for me, how my life has changed and just like, say like, yeah, ‘It’s helped me a lot. It might help you. You can always give it a go – there’s no harm in trying’, sort of thing. Obviously it might not work for everybody, but if there are people it could help and it could work for, that’s always a plus.

A potential role for participant suitability criteria

Reflecting on the potential variability in intervention outcomes, CYP highlighted a need for services to consider individual requirements when instigating a pathway of care. Older CYP expressed caution regarding the suitability of OST for younger individuals, although younger CYP themselves did not uphold this view. Developmental stage emerged as a stronger determinant of intervention acceptability than chronological age, with treatment engagement dependent on cognitive capacity and educational maturity:

I do think that like for teenaged years it definitely works. I don’t know with young children ‘cause I feel like 3 hours, they wouldn’t be able to like, they’d just get bored and start getting a bit like agitated, I feel like once you’re older you’re a bit more mature and you can actually like be in a room doing the therapy for multiple hours. It means you’re then, you’re more likely to actually gain something from it and not like just get fidgety and distracted.

A shared discourse emphasised a need for OST participants to be able to maintain sufficient concentration for a prolonged appointment. One child expressed a preference for multiple shorter sessions of CBT, rationalising this in terms of their need to process information in shorter segments. Another, who lived with ADHD, suggested that his ability to engage in OST had most likely been affected by his capacity to focus:

After like an hour and a half to 2 hours I got really like ‘cause I’ve, with me having ADHD and only sitting like I get where he can’t sit for ages . . . and with us sitting there for a long time I get, I kept walking around the room so I just trying to distract myself.

Children and young people’s acceptance of their phobia and the intensity of their fear were also raised in the context of suitability. Remaining open to the challenge of an exposure intervention and having sufficient resilience to cope with this over a sustained period were often highlighted as discrete prerequisites for OST:

I would only recommend it, if the person had kind of accepted their phobia and understood their phobia and felt ready to kind of deal with it. Obviously ’cause I feel like with the one-session treatment you have to be aware of how you’re feeling and you have to not be in denial about how you are.

Extrinsic influences, such as intervention context and timing, were also considered important, with life circumstances and family support emerging as two core features influencing treatment uptake and engagement:

It’s hard when you’ve got a lot going on in your life – it’s hard to focus your time and energy into overcoming the phobia because it does take a lot out of you. ’Cause all the energy you’d usually be using to worry about it you’re kind of having to force yourself to go ‘no, you’re OK – you don’t have to worry’, you’re having to kind of like actively think about it. So, it does take a lot, so I think it has to be at the right time.

If my mum wasn’t worried about me I probably wouldn’t of done anything to do with this. But you need the support from the people around you to actually be able to – [laughs] to almost shove you in it. Almost as much as, you’ve got to want to do it yourself. But you need someone like a bit of encouragement because it’s hard.

Suggested modifications to treatment delivery

Although session length was recognised as a defining feature of OST, this was also something that CYP often felt could be changed. Although a few CYP advocated for a longer session to accommodate more reticent individuals, the majority suggested that shortening the session, or at least providing a comfort break, could be advantageous:

I feel like a 10-minute break or something in the middle. An hour and a half or 2 hours or something then break in the middle and get a drink, or snack or stuff like that and then carry on.

Some CYP also suggested that a follow-up session would improve their OST experience. This was typically to enable a practitioner to check on their progress or provide them with another supervised opportunity to consolidate their skills. In rare instances, the addition of a follow-up session was advocated as an opportunity for health service development and quality improvement:

. . . having a follow-up if it was about 2 or 3 weeks after just to kind of go through what I’ve been through, but not at the same length of time, not for 3 hours. But, just to kind of go back through it so you’ve still got it in your brain.

Because it would help him understand erm – maybe how erm – if they give him feedback then, then for his next person he could use that feedback.

Summary of children’s views

Overall, CYP’s narratives upheld OST as a credible and engaging intervention that aligned with their treatment needs. CYP initially approached treatment with some trepidation, but confidence in their own ability to participate in the intervention increased with treatment experience. Qualitative data suggested that individuals perceived few opportunity costs to OST participation, although their own intrinsic readiness emerged as an important influence on treatment motivation. Improved information and orientation may have a role to play in ensuring that CYP are emotionally and cognitively prepared to receive OST.

Therapists’ competence, collaborative goal-setting and therapeutic rapport were identified as core features critical to the success of OST. Having a therapist who was knowledgeable, empathic and trustworthy was important to CYP. Most reported that OST had developed their understanding of their phobia and given them the skills to manage its symptomatic and functional effects. Transferable learning, in the form of improved anxiety management, was also reported by some.

The efficiency of single-session delivery was largely acknowledged to have a positive impact on therapeutic experience and outcome. However, CYP’s narratives raised an important issue around its universal suitability, suggesting that patient choice and personalisation should be retained as important features in referral pathways. Child-centred variables (e.g. developmental stage and additional needs) may best be considered in conjunction with family-centred variables (e.g. support availability) and phobia characteristics. Some additional modifications to OST delivery were advocated, specifically the addition of a follow-up session to improve individuals’ experiences of treatment closure and provide an opportunity for progress review.

Study 2: acceptability of one-session treatment to parents

Our second qualitative study aimed to explore the acceptability of OST from the perspective of the parents and caregivers of CYP who were allocated to receive OST.

Representativeness of the parent sample

All of the parents/guardians who were interviewed were female, were aged between 30 and 59 years and had CYP who had also participated in interviews. Characteristics of these CYP and their phobias are presented in Table 27.

Parents’ views

Akin to study 1, our findings for study 2 are laid out in line with the objectives of this work, namely to explore caregivers’ experiences of phobia prior to their participation in ASPECT, their experiences of OST delivery and their subsequent reflections on intervention impact and acceptability. These objectives provided a basic deductive framework for our analysis. Thorough descriptions of the inductive interpretive themes relevant to each objective are presented in turn, supported by direct quotations from the data. Participants are assigned a unique identifier in the text. Child age, sex and phobia type are provided. Additional data are included in the summary coding tables in Appendix 9, Tables 46–48.

Caregivers’ experiences of childhood phobias prior to ASPECT

The qualitative analysis interpreted four themes from the data relating to caregivers’ experiences prior to ASPECT. These themes highlighted (1) minimisation of childhood phobias, (2) suboptimal coping, (3) a rising need for intervention and (4) insecurity over treatment efficacy.

The minimisation of childhood phobias

When reflecting on their understanding of phobia prior to ASPECT, parents’/guardians’ understanding of the severity and legitimacy of phobia varied. Many reported being dismissive of their children’s phobia initially, framing it as either a difficult developmental phase or a challenging and unnecessary behaviour:

We genuinely thought it would be something she’d just grow out of. And then as it came to a time for her to have vaccinations at school it became very apparent she wasn’t growing out of it.

No, and it took a while for her to – to kind of talk about it. So it was coming across at the beginning that she just was being really difficult. Erm and – and so it took a bit of unpicking to finally get to the bottom of it.

Often parents alluded to a protracted period over which they felt that their child’s phobia had become engrained. Some were able to identify a specific time at which their child had been frightened by their phobic stimulus. These parents were more likely to rationalise their child’s behaviour and more likely to conceptualise it as a chronic yet coherent response:

I think it probably stemmed from when she was very small and we’d gone to see Disney on Ice [(an ice-skating show) The Walt Disney Company, Burbank, CA, USA] and then on the way back her friend projectile vomited in the car, which was obviously an enclosed space. And then it happened again with her sister, and then with her friend. So I think it stemmed from a very young age.

Few parents had sought professional help prior to the study. Specific phobias were commonly located outside lay constructs of mental health and were afforded less salience than other health issues, physical or mental, that their children encountered. This lack of recognition of phobia was typically compounded by a lack of awareness of viable treatment options, discouraging timely help-seeking and effective response:

I mean it didn’t even occur to me to go like to a doctor and ask for help. It hadn’t occurred to me at all, I think we were trying to manage it ourselves and I think it’s why it’s that – having that frustration and then realising actually it’s, it’s beyond what we thought.

Everyone knows that CAMHS is there to support mental health, but you don’t consider a phobia to be part of the mental health.

Occasionally, health services were held accountable for reinforcing a lack of parental action. A small number of caregivers who had tried to seek help alluded to both financial and attitudinal barriers to treatment access. These parents recognised NHS funding limitations for CYP’s mental health provision and a lack of validation of childhood phobias in primary care as two factors devaluing treatment need:

But I was also aware that the NHS don’t offer huge amount of support to kids as well. It’s not that they don’t want to, it’s just that there aren’t enough resources and I know that – that [child name]’s anxiety needs would not be huge compared to other kids. So I thought she would be at the bottom of a list and it wouldn’t really be a priority so I didn’t have huge expectations.

I think once we went to the GP and regarding to her phobia but they said – they said they can’t really a do really anything because it’s something like most kids are afraid of and she will grow up out of it.

Suboptimal coping

Suboptimal recognition and help-seeking for childhood phobias meant that many families had become accustomed to challenging scenarios and behaviours. Accommodation was the most common coping strategy reported by the parents referred to OST, typically manifesting as either the intense negotiation or the complete avoidance of a phobia stimulus. Both forms of accommodation were purported to precipitate family stress. Tension was particularly evident in families living with health-related phobias (e.g. BII phobias), for which parents archetypally conceded to coercive rather than collaborative approaches. Irrespective of phobia type, conflict in parental responsibilities emerged, a shared parental narrative highlighting the guilt that caregivers incurred from trying to manage their children’s phobias:

It’s almost felt like it’s – they’re there to pin her arm down rather than to help her get over it. I think the more people get involved probably scarier it is, ’cause she is getting out of control, she’s pinned down and, you know, her arm’s pulled around my back and held in place so she can’t move and make it worse, but in her own mind that probably makes it a million times worse.

Rising intervention need

Nearly all of the parents had witnessed their child’s phobic responses evolve and intensify, until such times that they could no longer uphold a developmental perspective. Parents described a rising need to alleviate the emotional burden of phobia on their children, identifying this as the primary reason that they had sought or accepted help. Most appeared motivated to reduce the anxiety that their children experienced when encountering a triggering stimulus. Occasionally, however, parents also verbalised an aspiration to ameliorate more chronic socioemotional effects. Some spoke of a desire to alleviate the shame and embarrassment that their children experienced, either to directly enhance their emotional well-being or to prevent social marginalisation by their peers:

I really wanted her to be able to find ways of managing anxiety, because she was upset by the fact that she’s scared of dogs.

It’s quite crushing really like – I think he was worried his friends would pick on him.

Additional drivers of help-seeking presented as parental concerns for childhood safety and physical health. Parents of CYP living with dog or insect phobias were often apprehensive that their children would flee into danger, while parents of children with blood or needle phobias were concerned that they would miss key vaccinations or subvert preventative health routines:

It’s distressing ’cause I don’t like to see him unhappy and then it creates that worry around – you know – there’s still a meningitis jab that he hasn’t had. So there’s that kind of worry, ’cause I’m sort of, trying to do the best for his health and that’s sort of a bit of a contradiction if he’s not had that vaccination isn’t it? So – and then the dental – the issue around the dental [the dentist] about making life much more complicated. ’Cause he had to have an operation under general anaesthetic which – you know, I don’t believe in unnecessary surgery, so that sort of goes against some values but at the same time we have to compromise sometimes because I couldn’t see what the other options were really.

Parental reports of CYP’s avoidance behaviours were common across the data set. A subset of parents highlighted a need to improve community and educational integration where possible. More consistently, however, parents were motivated to enhance family functioning, frequently advocating a need to improve parent–child relationships, alleviate stress on other family members and establish positive shared experiences:

In terms of outcomes it would – it would just be nice that I know that she could pick up a book that there might be a picture somewhere in it and that might be The Muppets [(group of puppet characters) The Muppets Studio, Los Angeles, CA, USA] and then she would be quite happy and wouldn’t be so scared she’d run off.

Against a backdrop of improving family interaction, parents hoped that OST would improve their own understanding and management of phobia, and facilitate constructive communication with their child:

I was just hoping that I’d get some advice on how to deal with [child name]’s phobia. And you know – what to do when she did see a dog.

I thought it would be a good opportunity for us to talk about things more openly. [Child name] is really good at talking about things openly anyway, so we had already had conversations around it. But I saw another opportunity to have that.

Insecurities over treatment efficacy

Parents reported scepticism when their child was referred to OST. A strong group narrative portrayed a cautious stance towards intervention efficacy, with caregivers being keen to protect themselves and their children from disappointment. Some ensured that their families also retained realistic expectations, while others worked hard to conceal their cynicism from their child:

My impression was I’m concerned that this is not going to meet his needs. I didn’t – I didn’t discuss any of that with [child name] I made sure that he was very focused on you know. And I think [child name] had really big expectations of it as well of what it was going to be for him.

I was quite realistic with it because I thought ‘You can’t cure it in 3 hours’. I understand the idea around it, but I was just hoping it would give him more confidence in how to deal with it. So, I was hoping that he would feel stronger . . . So, that’s what I wanted out of it, was coping mechanisms, but I didn’t have any – I didn’t want to set myself up and think ‘He’s gonna be absolutely fine when he walks out’ because that’s just upsetting for both of us. So I thought I’ll be practical and I’ll hope it gives him the tools he needs, but I didn’t think it would cure him.

Notions of a curative treatment were often discarded by parents and replaced with more pragmatic expectations of self-management support. Some doubted whether or not a time-limited intervention could ameliorate an engrained phobia, while others worried that OST would not align with their child’s needs:

I wondered how effective it would be, because knowing how [child name] reacted when he, you even mention the word blood, never mind needles and having blood taken, it would just send him into a spiral. You know if he thought he was going anywhere near a doctors surgery and there’s anything to do – you couldn’t even say the word blood test. We had to say ‘BT’ and abbreviate it for him – for him to be comfortable. So I just didn’t know how effective it was gonna be.

I thought it would help her, but from the other side I was a bit scared at how can you treat a phobia, especially the one she had, when it so common. And you are just like sort of – you are facing it every single day, because they are – obviously the flies everywhere. How can you treat it with just one session? So I was hoping it would help her, but I just – like I was in doubt [laughs] yeah that we might need some extra help after that.

Parents’ experiences of one-session treatment delivery

Three interpretative themes were derived from parents’ accounts of OST delivery. These themes centred on (1) the strengths of OST facilitation, (2) potential adaptations to support treatment effects and (3) the impact of the treatment delivery protocol.

The strengths of one-session treatment facilitation

Overall, parents reported that their children’s experiences of OST had been positive. Participants partially attributed this success to the high degree of control that CYP were afforded over treatment, acknowledging that collaborative treatment planning and child-led progression had enabled their child to set a comfortable therapeutic pace:

For me I think [therapist name] reiterated that [child name] has control all the way through the session, you know. And I think that gave her the confidence to move forward knowing that – you know if she’s uncomfortable she could speak out.

Most parents recognised the influential role that OST practitioners played in children’s treatment experiences. One parent suggested that the dialogue between the clinician and the child had sometimes come across as patronising, but more commonly caregivers commended the use of child-centred communication techniques. Emphasis was placed on the importance of clinicians making quick connections with OST participants; parents often upheld this as a crucial factor determining the quality of the therapeutic alliance and its outcome. For some families, the patience and consistency displayed by OST practitioners contrasted sharply with the reactive emotions expressed at home, providing a new and reassuring perspective for their child:

She had a really good understanding of children as well, she had a lovely manner with her, and she spoke in a language that [child name] understood, which put [child name] at ease as well.

We lost our patience and maybe that patience from the therapist it helped her. That somebody really actually sat down with her and explained everything about it. And let her understand why she’s afraid and how to manage it.

Parents acknowledged the benefits of having a skilled and knowledgeable therapist, recognising these as essential characteristics for optimal information transfer and the development of targeted, personalised treatment sessions. Equally salient, however, was the sense of objectivity that trained facilitators bestowed. OST practitioners were typically located outside parents’ and CYP’s social networks; the sense of independence and impartiality that this afforded was frequently recognised by parents as advantageous for therapeutic rapport:

Phobias are so specific to certain people; you need to understand the science and the right questions to ask somebody to get into the thing.

I’ve tried to do what I can with him, but it’s very different when you’re emotionally involved. It needs to be somebody sometimes I think, you’ve gotta accept that sometimes you need somebody who’s outside, who’s a professional, who is gonna be able to just walk away from it at the end of the day.

Adaptations to support treatment effects

Parents recommended some ways that OST delivery could be improved. Many articulated the need for a follow-up session, often highlighting this as a valuable, but currently missed, opportunity for clinicians to review a child’s progress and sustain their self-management skills. A small number conceptualised the follow-up session as a potentially shorter, parent-centred communication, which would enable caregivers who had not observed the therapy to better understand the intervention and support their child’s learning:

I know that 3-hour session was a one treatment session, but maybe another something similar afterwards to see how she’s, [child name]’s progressed from where she was then to dealing with a dog again. So, yeah I just feel like there should be a little something just to see how she is with the dog again.

Yeah, I think even if they’re given like you know a handout, or something like that to pass on. Yeah or – just like a follow-up phone call with like oh, this is you know, what we did, or you know, just some kind of follow-up information.

Parental involvement in the intervention varied and appeared to be determined largely by clinician preference. Those who had been present during therapy reported gaining a deeper understanding of childhood phobias and believed that they had developed some specific skills to support their child’s progress. In this context, parental inclusion was perceived as an important therapeutic ingredient and a potential determinant of sustained treatment effect:

For me it was really useful to see the techniques and the skills that she was being given because then in my head I was thinking at this precise moment in time she’s never going to walk into a GP’s surgery on her own and have an injection by herself. And, you know, for me to go in and just try and gently remind her of those skills and the breathing and the relaxation techniques, and what have you, is really useful.

Parents who were not able to observe OST confirmed this view. These individuals described their ideal of a shared appointment, noting that they would have welcomed additional guidance to help their child following treatment:

The organisation was a big thing ’cause, like I say – I feel like if I could have been there and known what strategies they put in place and you know, just the things that they did then I could have carried that on. Like you know, on the walk to school, or you know, that sort of thing.

The impact of the treatment delivery protocol

Limited access to transport, substantial travel distances and the available appointment times conflicting with family commitments were all cited as barriers that discouraged access to face-to-face treatment. Unique to OST, however, was the need to dedicate a continuous amount of time to therapy, which for some incurred significant opportunity costs:

You’ll have to take a bit of time off school, most kids won’t mind that. I suppose if – I work part time, so it wasn’t so much of an issue really for me, but perhaps some parents, if they work full time, it might be inconvenient to take their kids to some of them. But if it is an actual phobia, I would imagine it impacts on those families lives anyway. You know, if it’s significant enough to try and get help for it then, I would imagine you’re prepared to invest some time in it as well.

Competing participant narratives emerged regarding the acceptability of the OST delivery format. Some upheld the single session as logistically more manageable, manifesting in less interruption to school and work commitments. Many were keen to minimise the emotional burden of treatment and construed the lone appointment as less intimidating and anxiety-provoking for their children. Intervention efficiency was also acknowledged; parents suggested that there was less repetition or potential for regression with continuous therapy and more opportunity to evidence early improvement. Some advocated that this continuity was critical to intervention engagement, positing that their child would have been substantially less motivated and unable to sustain exposure if they had been referred to weekly CBT sessions:

I’d say a lot of it was the time constraints. You know, that’s got to be the honest reason, is the time constraints for myself and trying to manage her school.

She could really look back and just see how things had changed from before she went there and then when she’d left. Rather than it being something spread over so that she couldn’t so easy reflect on the difference.

Concomitantly, OST delivery was acknowledged to have its limitations. Perceptions of therapeutic intensity sometimes confounded those of session length, with emotional exhaustion recognised as a specific treatment risk. Several parents alluded to the cognitive impact of the intervention on caregivers and CYP. Some reported that their children needed a day off school to accommodate its acute effects:

She had no energy to do anything. She had no motivation or anything. She were just kind of like slumped as though there was a lot on her shoulders. But then it would be ’cause it was intense.

You’ve got to stop keeping them so long. ’Cause it’s exhausting for us and it’s exhausting for the child.

A shared discourse highlighted the importance of service organisation, both within treatment sessions and across the care pathway. Coordination of resources and effective communication with families was often a challenge. Parents tended to rationalise delays in treatment access that were incurred while sourcing stimuli, but were less sympathetic to setbacks caused by workforce shortages or procedural constraints. Several parents recommended that the interval between treatment planning and delivery needed to be shorter, with some noting that they had struggled unnecessarily when this was prolonged:

[We] had like real problems with getting the session arranged. The lady that we were allocated to actually arrange it, she was – she only worked a couple of days a week and she said like the dog that they had worked on Monday and they didn’t have a venue on a Monday that allowed dogs and things. So, it took quite a long time to arrange, well, all of the summer to be honest with you. I think it was, it was the – she was definitely back at school in the September, possibly the October when we actually got the treatment arranged and I did feel that we only actually managed to get it arranged because I contacted [trial manager name] . . . it just seemed to drag on far too, too long. And then when – when it was arranged it was arranged for a day when I was at work and we only got a week’s notice as well, so I couldn’t put in a holiday request.

Post-treatment reflections

The qualitative analysis interpreted two final themes from parents’ post-treatment reflections. These themes centred on (1) parents’ perceptions of treatment outcomes and (2) treatment suitability.

Parents’ perceptions of treatment outcomes

Parents generally felt that their children had benefited from OST. The most common benefits that parents reported for their children comprised increased knowledge and understanding of their phobia and their physiological responses to emotions; increased self-confidence; and reduced anxiety. Most described their children as feeling proud of what they had achieved in therapy and more self-efficacious following treatment, crediting this to the new knowledge and skills that they had acquired:

Her first thing she said when she came out was she was really proud of herself. So that was really nice.

Yeah definitely, it’s made a big difference, just for example like walking to school, she will run off through the park, you know, ahead of us. Whereas you know you wouldn’t go through the park initially and then to begin with, you’d hold her hands, but now you’re just carefree aren’t you?

Variation was apparent in the magnitude of effect that parents perceived. Although some parents reported marked improvements in their child’s behaviour, others reported a more moderate response. Typically, parents described partial progress, noting that, although some features of phobia were still present, their children were more likely to employ effective self-management strategies and less likely to engage in avoidant behaviours:

Well [child name] will still message me, she’ll still send me text messages if she’s feeling anxious about – Erm, because there’s been a lot of norovirus at the moment and a lot of children have been going off, schools have been closed and things so she has been quite anxious about catching that. So she will message me if, you know, a child’s been feeling unwell at college. Erm, but they’re so much shorter [laughs]. And, erm, I think she comes down a lot quicker. Her anxiety comes down so much quicker. And, and you can reason with her as well. And I don’t think, as far as I’m aware I don’t think she’s running out of rooms. You know, she’s managing to stay.

A small subset of parents described OST as a useful experience, but one that was ultimately disappointing in terms of the magnitude of effect that it conferred. Reflecting on the intervention, these parents indicated that that they might have entered therapy with high expectations that were not aligned with their child’s condition. Disenchantment largely, although not exclusively, clustered in parents of children with BII-type fears:

I just expected the result to be – that she’s over the phobia and she’d have a blood test or a needle any time. A bit presumptuous, really, and hoping and wishing, but that’s what I believed she would be like. And it clearly it wasn’t like that, ’cause even though she had the blood test, it took so much hard work from us and so much cooperation from the doctor, and for her to understand.

One interviewee recounted a wholly negative experience of OST, reporting a marked decline in their child’s well-being post treatment. In this case, the parent’s treatment expectations appeared more measured, but a lack of demonstrable progress and the exacerbation of existing symptoms meant that disappointment nonetheless ensued:

The whole experience was not what I expected because I did expect to have a little improvement and it wasn’t it went totally the other way and that really shocked me, so I guess I was a bit shocked with the reaction.

Treatment suitability

Variability in treatment outcomes precipitated a review of treatment suitability. Several parents questioned the universal applicability of OST, giving rise to a shared narrative that emphasised the importance of personalised care pathways and patient choice. The anxiety that children experienced between treatment allocation and onset was identified as one of the biggest challenges to intervention engagement by parents. Caregivers were consistent in articulating the optimal conditions for OST delivery, strongly emphasising a need to ensure that children were emotionally and cognitively ready for stimulus exposure:

If they’re already anxious about something, why would they want to put their self in that situation? You know, if someone’s scared of something, it takes some courage to go and actually face that fear and I think that’s what would probably stop most people. If they’re not gonna go there, it’s for that reason.

I think they’ve got to be, in their own head, they’ve got to be ready. They’ve got to sort of think to themselves, ‘Right, I’m ready to do this now’. Because it’s them that has to do it at the end of the day. So, it’s – it’s pointless me saying ‘[child name] you’re doing this’. She has to do it for herself. So, I didn’t wanna push her into doing anything she didn’t want to do, it could have been too soon.

Competing participant narratives considered whether chronological or developmental age should be used as an explicit criterion for participant referral. For some, chronological age was an acceptable marker of treatment capacity, but for others cognitive ability and individual needs took precedence:

I don’t imagine that we’d got, we would have got very far with a GP really when she was 7 or 8. Even if she’d had this therapy I don’t think she’d have been sort of really up for it, and now I’ve like, since she’s been about 9, like this last kind of year and a half, she’s really come on in a lot of, in a lot of ways just being able to cope with things.

Counter to a dominant discourse that emphasised the inherent efficiencies of OST, a minority of parents remained steadfast in believing that repeated treatment sessions were necessary to build an effective therapeutic alliance, expressing their preference for a more regular and intermittent approach. These parents typically cared for children with attention or information-processing deficits, acknowledging this as the primary reason why they felt that OST was unsuitable for their child:

I think it depends on the individual. For anybody with a phobia I think – it depends I suppose, on if the child’s got any additional needs or any other problems. They might find it too much in one 3-hour period. Might be a bit overwhelming.

I do think the smaller sessions would have probably worked better for her overall. Like I say, just to build that confidence and you know, maybe like you know, just going into a room with a dog and you know, eventually like just like a slower approach I guess of all the different steps.

The integration of multiple viewpoints nonetheless cautioned against adopting a fixed approach. Although one parent of a child with ADHD reported that they had found it difficult to sit for the duration of the OST session, another felt that it offered a more suitable format than CBT. For this reason, comorbidities were acknowledged to encompass a spectrum of need, with several parents emphasising the importance of clinicians and families working together to optimise treatment pathways and deliver child-centred care:

He was thrown in at the deep end to deal with such a lot for [child name]. But his sensory, his social, you know we understand [child name] and we understand his needs and these people OK, they’ve met him once before – to me I think you need to get to know the family and the child just a little bit more if you’re offering this session of what works for this child and this family. Because as you know autism is a massive spectrum and what works for one, might not work for the other.

Summary of parents’ views

A tendency for parents to minimise and accommodate phobia commonly contributed to a delay in help-seeking, which has been reported elsewhere.98 Phobias were often chronic and ingrained, and families sought help only when informal management became ineffective. Inadequate responses from health services contributed to symptom minimisation and limited treatment access. Historically, levels of service demand many not have equated to treatment need. Public health and education may have a role to play in facilitating access to timely, effective intervention.

Parents’ treatment goals aligned well with the remit and purpose of OST. Cautious perceptions of intervention efficacy at treatment referral may be ameliorated by enhancing parents’ understanding of the intervention and its underpinning evidence base.

Parents highlighted the critical role of clinicians’ competencies and rapport-building skills, as did CYP. Treatment planning and collaborative working were consistently highlighted as key components of OST in parents’ narratives. Enhanced follow-up and improved handover may benefit parents, facilitate home-based skills practice and contribute to the maintenance of treatment effects.

Parents perceived a range of treatment effects, highlighting a potentially important role for expectation management in treatment orientation. Patient choice in treatment mode should be retained, with an emphasis on shared decision-making at referral, aided by the provision of child suitability guidance. Reflections from parents highlighted some potential weakness in OST implementation. Intervention provision and coordination may benefit from greater standardisation and target-setting for treatment waiting times and communication.

Study 3: acceptability of one-session treatment to trial clinicians

The aim of the third study was to explore the acceptability of OST from the perspective of the clinicians delivering the intervention in the trial.

Representativeness of the sample

Seventy-seven clinicians were eligible to participate, of whom 16 expressed an interest and were subsequently interviewed. These individuals represented eight of the 13 sites that were open to recruit CYP to ASPECT. Table 28 summarises the characteristics of this sample. A comparison of 48 of the 77 practitioners who delivered interventions in ASPECT and disclosed their characteristics is provided.

Clinicians participating in the qualitative interviews held a variety of roles and worked in different services at different levels; the three most common roles of those who delivered OST in ASPECT (cognitive–behavioural therapist, CYP psychological well-being practitioner and CAMHS practitioner) were well represented in the qualitative study. The length of time spent delivering psychological interventions to CYP was also distributed similarly between the clinician sample in the acceptability study and the clinician sample in the trial.

Clinicians’ views

Clinicians’ views of one-session treatment at the outset of ASPECT

The qualitative analysis interpreted six themes from the data relating to clinicians’ views of OST at the outset of ASPECT. These themes explored (1) how OST aligned with professional roles, (2) clinicians’ perceptions of treatment demand, (3) clinicians’ motivations for trial participation, (4) the perceived effectiveness of OST, (5) perceptions of intervention feasibility and (6) the anticipated demand on OST practitioners.

The alignment of one-session treatment with the professional role

Clinicians identified dichotomous drivers of their participation in ASPECT. For some, particularly those who routinely worked with CYP experiencing mild to moderate anxiety and depression, their professional role was the main motivator for intervention delivery. These individuals perceived OST to align readily with their existing skill set, viewing the intervention to be conceptually and logistically coherent with their current role:

[B]ecause the role of a children’s well-being practitioner does work with children that have mild to moderate anxiety or depression, I could see that being able to do one-session treatment as part of our skills set. Having one-session treatment there would be beneficial and that would fit in normally within our practices.

For others, participation had been determined predominantly by their capacity to absorb additional work. These individuals were typically clinicians within wider service remits, most notably clinical psychologists and psychiatrists, whose professional remit normally extended to more complex presentations. These individuals were consistent in perceiving OST as a research intervention, mismatched with their normal role.

Therapeutic orientation also appeared to influence the ease of fit between OST and individual practitioner roles. Many of the interviewed clinicians had worked with CBT models throughout their careers and highlighted clear synergies between their prior experience of these interventions and the delivery requirements of OST. This alignment was evident irrespective of whether or not CBT practitioners had treated specific phobias before, with parallels drawn between the strategies used by OST and the techniques used with other anxiety-related disorders:

Maybe about 65% of my caseload for intervention at any one time would be behavioural experiments for symptoms of social anxiety. And that is really, fundamentally is a phobia. It’s just not – you know, it’s just not really seen as one. So, it was kind of not really any different in as much as for symptoms of social anxiety we do use behavioural experiments where we would think about a situation somebody’s avoiding that’s maintaining their anxiety, like raising their hand in class, and elicit a cognition – the thing they’re thinking about happening.

Perceived demand for phobia treatments

A common narrative among clinicians was an apparent lack of demand for phobia-specific therapies. Prior to ASPECT, practitioners reported that referrals for phobias had been low, with few cases accepted by local services. Proactive practitioners had searched service caseloads at study commencement but yielded few exemplars. One clinician reported an average of only two phobia cases per year.

Cross-case comparison across the qualitative data set revealed some opacity in these views. Although some individuals perceived a lack of treatment demand, others presented a more nuanced view. Rather than being absent from service referrals, specific phobias were sometimes acknowledged to have been overlooked by teams whose remit and culture had evolved to address more complex diagnoses. In these instances, specific phobias were known to present alongside other difficulties, but were unlikely be identified as ‘core business’ or prioritised as a treatment goal for CYP:

So it was usually part of something that was going on so there’d be some family issue and that was, sometimes it wasn’t even touched. They would mention it at some point, and it wouldn’t even get touched upon. ’Cause the main presenting goal they wanted was erm something different to – to, to look at the relationship with their parents.

Clinicians’ motivations for study participation

Professional development emerged as the primary driver of trial participation among those interviewed. Clinicians who followed systemic practices were curious to see how OST might work, while those who routinely delivered CBT-based treatment expressed a desire to expand their existing skill set. Newly qualified practitioners reported that OST competencies integrated well with and complimented the core training that they had recently completed:

I think there was a little bit of trepidation initially. I think I was also – yeah I was quite excited to see how it would pan out. And I think I just thought, I think I just felt it’s going to be a really worthwhile challenge to take on really.

I guess at the time that I did the training for the one-session treatment, I just – I can’t remember exactly at what point it was. But, I was either in the middle of the training year or just finished it. So I was still kind of in the mode of learning new interventions for anxiety. So from that perspective, I was looking forward to having another, I guess, method of being able to do that.

Clinicians were largely complimentary about OST training, advocating that the management of a protracted treatment session demanded dedicated training provision beyond the acquisition of generic CBT-based skills. Sessions were upheld as a necessary and invaluable opportunity to educate therapists, strengthen OST buy-in, and address individual and team concerns. Some particularly enjoyed the practical examples nested within the training, and others felt that the training format and the range of resources offered to aid learning were comprehensive.

More rarely, clinicians expressed a need for additional training, suggesting that this should include the dissemination of example treatment plans and additional skills practice. Although some interviewees posited that training needs were likely to be individualised and offset by clinical experience, others advocated the utility of universal skill development and review:

I already felt quite confident with things like CBT and I’ve delivered in the past, so maybe didn’t feel like I needed lots of in-depth – and training. So it felt right, for me. Perhaps if I was at the beginning of my training or something like that, then I would’ve wanted a bit more.

I think training’s always really good as like – even if you are a skilled practitioner, I think to have somebody stand in a room and go through the protocol and you know it – it just sort of reminds you of – it’s an opportunity really to just think out loud and maybe some difficult cases. So I would say I probably could have done but actually it, it’s been better practice because of having the training.

One participant mentioned a substantial delay between training and treatment delivery, which challenged the retention of learning and negatively impacted their confidence to deliver OST. A broader, supporting discourse explored the value of ‘top-up’ training sessions, with some suggesting that knowledge refreshment would have helped them to maintain intervention fidelity and sustain enthusiasm as an OST treatment provider.

Perceived effectiveness of one-session treatment

After completing their ASPECT training, most clinicians displayed a level of openness towards OST, drawing on a combination of evidence and tacit learning to enhance their confidence in its acceptability and effects. Irrespective of their professional role or baseline treatment preference, most hypothesised that OST would be effective for specific phobias, the main rationale being that there was a coherent link between theory and practice:

I think just looking at the intervention on a practical level I think it seems – yeah I suppose it feels like a common-sense approach. Yeah, I think I had confidence in it, in, certainly, alleviating some of the distress caused by the phobia.

I read one of the journal articles, one of the publications, and I thought ‘that sounds great’. And I had already heard about single session sort of treatments for adult phobias, sort of spider phobias and stuff like that, in a service I’d worked in before. And I’d heard that they can be really, really effective. And so I guess I may have – I may have come into it having, um, possibly having a belief that maybe sitting down and having the space and the opportunity to work up the hierarchy together, ensuring that a lot of the distractions in terms of seeking reassurance and things like that, that that could be quite carefully controlled. I had imagined that it might be more effective.

Despite this strong, communal enthusiasm for OST, several clinicians retained some core efficacy concerns. A common fear was that OST would enable CYP to overcome their anxiety in a supervised setting but that the intervention would ultimately lack transferability and sustainability in the real world. Some specifically questioned whether or not OST could offer sufficient support, suggesting that a single appointment might lack openings for generic psychoeducation and/or limit opportunities for CYP to consolidate their newly acquired skills:

It’s different to the traditional model of the kind of therapy that we provide which tends to be like weekly over a period of time, and in-between sessions that you’d expect there to be some kind of homework or more exposure work for the young person to carry out. Because I guess I worried that somebody might be able to overcome their anxiety in the room with me there with them, but then go out of the room and not be able to do that in in the wider world, really, when they got home.

I think with the CBT, um, we always tend to do an hour work psychoeducation and kind of mindfulness and teaching strategies and things. And it’s maybe a little bit broader. So, you’re teaching children sort of how their thinking impacts on their emotions and strategies and physical symptoms and the different ways and different ideas of managing that. And almost like a jigsaw, putting something together that works for them. So, specifically working on that phobia in one session – whether it would have the breadth and the longevity of impact that the CBT.

Perceptions of delivery feasibility

Although a restricted number of treatment sessions was hypothesised to limit intervention scope, it was the qualitative aspects of OST delivery that participants posited to be more likely to emerge as important influences on treatment engagement and outcome.

Cognisant of the OST protocol, many clinicians reflected on the potential impact of a protracted session, often accepting this as a pragmatic method of increasing treatment access by reducing opportunity costs. OST was repeatedly presumed to require a shorter time commitment from CYP overall, with frequent conjecturing that this would reduce school absences alongside increasing children’s engagement with and motivation for treatment. For older CYP especially, the OST format was conceived to align well with patient expectations, offering a coherent, person-centred intervention that could deliver fast results. Single-session delivery was considered advantageous, in terms of both capitalising on early contact and avoiding external distractions, two core features that clinicians perceived might expedite and amplify treatment effects:

My excitement was in that kind of adolescent process there is a lot about, ‘I want it, I want it now’. I think if they can see a movement to get rid of it they – the kind of motivation for carrying on, if you can push enough forward, I think motivation for carrying on, because they want this contact with their friends, they want to be seen, they, you know, that kind of thing, egotistical kind of thing. That age range love quick, you know, they are into that quick change, ‘let’s move it and let’s move it fast, I’m you know I’m important’ all of those kinds of things I thought it would fit into. So I was excited even at the training stage.

I thought it would be beneficial in, in terms of the saying ‘strike while the iron’s hot’. So if you’ve got a young person who is coming to the clinic ’cause they want to get over their phobia and they’re motivated actually sort of – I was thinking actually if you can use that motivation sort of whip up that motivation a bit more and then do that treatment in that one extended session then and keeping that energy going in the session, then my prediction was that that could be much more successful in terms of getting further up their hierarchy erm than if you did it over you know several months. And often over those several months something else can happen – so a family member might pass away or erm you’ve got exams come up or a holiday or illness or, and there’s all sorts of things that could come up.

Concern regarding the acceptability and feasibility of OST delivery was nonetheless strong. Evident in clinicians’ diffidence was an underlying concern that CYP would feel devalued and disenfranchised by the offer of one session, leading to a weaker therapeutic alliance and potential disengagement. Traditional notions of relationship-building and rapport seemed to be challenged by the OST model, with several clinicians expressing uncertainty as to whether or not sufficient trust could be established in a single session to ensure a meaningful outcome. A common pre-treatment belief was that shorter, more regular contact may be required to engage younger CYP and those living with neurodevelopmental disorders or dealing with mental health complexities or trauma:

I guess that was something I was –, I was maybe a bit unsure about. Because I suppose, when you’re doing the –, I suppose, when you’re delivering the low-intensity model over a number of weeks, you – you do get like more opportunity to build maybe rapport. Build a relationship which can, I feel, can have a lasting impact on you know, maintaining positive change. That kind of thing. So, I suppose in some respects, there was maybe a fear that that –, I think that – well an expectation that, therapeutic alliance might not be strong with OST.

I do have more anxieties about those young people who have got more comorbid presentations or some neurodevelopmental difficulties that actually, for those children, the children whose experienced significant trauma, who may really struggle to trust a therapist. That actually for those young people would it be as successful? And would – would you need more time for that engagement? Not necessarily the treatment session, but would you do more of an extended assessment to actually develop that young person’s trust in you as a therapist.

Anticipated demand on one-session treatment practitioners

A core concern for clinicians was the amount of distress that OST would invoke in CYP and the high responsibility that they faced in managing these challenging emotions over time. The risk of overwhelming CYP in a single, uninterrupted treatment session was something that several clinicians admitted to feeling quite daunted by, owing to not only the potential effect that it could have on CYP, but also the effect it would have on them:

I think obviously you’re dealing with somebody who’s kind of very anxious and kind of you know it’s, it’s that kind of constantly trying to help them manage that and staying calm when it can be – you know quite distressing.

Therefore, at treatment outset OST was often perceived as a potentially efficacious yet riskier model than traditional CBT. Single treatment appointments created the notion of a ‘one chance’ intervention and placed considerable weight on session planning and preparation. Clinicians contrasted this with the more intermittent approach of conventional CBT, which was perceived to offer greater opportunities to adjust and reflect on a treatment plan depending on how an individual responded. Some were unclear about how they would manage a stalled therapy appointment. Others were concerned about how easily families could access alternative provision should their own intervention prove unsuccessful. Individual narratives highlighted a lack of communally defined strategies for addressing poor treatment response or relapse:

There was just a bit of – I think a bit of misconception or a bit of confusion about the one-session treatment or CBT. The family were saying ‘What happens if the one-session treatment doesn’t work? What would happen then?’. The treatment is finished, so we would end treatment and be discharged.

Clinicians’ experiences of one-session treatment delivery

Three interpretative themes were derived from clinicians’ accounts of OST delivery. These themes explored (1) the work involved in delivering OST, (2) the value of parental involvement and (3) the logistical challenges encountered in intervention delivery.

The work involved in delivering one-session treatment

A shared discourse among clinicians acknowledged OST delivery as tiring. The cognitive load experienced by OST facilitators appeared to be high, with multiple challenges inherent in managing both the emotional and the temporal elements of the session. The ability to plan and pace treatment in a manner that balanced assessment, engagement, psychoeducation and exposure was crucial. Multiple narratives upheld the early phases of therapeutic exposure as the main ‘bottleneck’ in intervention delivery: progressing too slowly risked wasting treatment time and limiting treatment effects. In this context, clinicians frequently noted that professional confidence was key. The critical importance of balancing power dynamics was articulated by multiple clinicians, who candidly acknowledged the effort that was required to integrate adult-led facilitation into a predominantly child-led treatment approach:

And I think it’s just that kind of been switched on constantly across that – you know that period and been really aware of do we need to alter the steps were making? You know we’ve kind of set this before but actually do we need to kind of make some smaller steps in between? So yeah I think it was – it was good but it definitely was. I remember been very tired afterwards.

Clinicians often reported that their knowledge of a stimulus exerted an important procedural influence on the therapeutic outcomes. They frequently emphasised the need to prepare thoroughly before sessions, to acquire the expertise necessary to ensure that they could answer all of the CYP’s questions about a stimulus and to be able to model behaviour. Modelling behaviour was perceived to lend validity to the treatment and empower CYP to engage in exposure. In this context, practitioners emphasised the impact of their own self-efficacy and confidence on treatment efficacy; any doubt from the clinician was perceived to transfer to the CYP, which diluted the intervention potential:

I just think for the extended treatment session I think you just need to be super prepared beforehand and have everything there, everything ready and a real confidence in the approach as well. I think if you doubt it as a clinician that’s going to really come across to the young person or the parent. And children are you know – they’ve got these little antenna’s that they do pick up on any doubt in you as a therapist. Part of that was about ‘fake it ’til you make it’ belief in the model. You’ve really got to believe that this works and it and it does work.

A dominant narrative emphasised the value of establishing a strong therapeutic alliance, enriching the treatment experience with humour and enthusiasm. The importance of engaging and collaborating with CYP as equals was underscored in both intervention planning and delivery. Validating and normalising CYP’s feelings were perceived as crucial aspects of the clinician’s early role, eclipsed only by the importance of reviewing progress to ensure the sustainability of effects at the end of treatment:

To give them some validation and say how well they’d done to recognise you know – that they had a problem and they’d come and sorted it out. And we’ve come up with a plan and a way which to sort to manage that and then we can reflect back and see you know – those positive outcomes and actually you know. For someone at such a young age it’s a really valuable lesson or any problems they can come back in the future, with them and I think that’s a really, I think that’s a really nice message to give children, children of that, of that age really.

The value of parental involvement

Views were mixed regarding the inclusion of parents, with a tendency for clinicians trained in systemic therapies to view parental involvement as more important than those who took a purist approach. Those who reflected on parental presence as a positive experience acknowledged that it was often helpful during the preparation and goal-setting phases of OST, particularly when working with younger CYP or those with whom they were still establishing rapport:

They can explain or clarify or you know – whereas you wouldn’t challenge a child that you didn’t know. Whereas the parent can say ‘Yeah, but remember the time that you . . .’, I don’t know ‘. . . were near a dog’ or whatever. They can come from a place where they know the child, or they can maybe challenge or offer additional information, which is really helpful.

For younger CYP, parental presence was viewed as beneficial during OST to help the child to feel comfortable and maximise the continuity and magnitude of exposure. During treatment, parental involvement was optimised by ensuring that their role was clearly defined and that they were able to engage with and model positive behaviours with stimuli. Parental presence enabled clinicians to support caregivers in managing CYP’s anxiety and avoidance behaviours, a process that many clinicians perceived as fundamental to the facilitation of skills practice post intervention and the maintenance of therapeutic gains:

I’ve kind of said to Mum ‘She needs to keep practicing. You need to repeat it with another dog’ but because Mum wasn’t there, I had to almost explain all the stages and how we did it and everything to her. But Mum wasn’t able to for whatever reason. So, the gain or the progress wasn’t maintained.

Therapists tended to be more supportive of therapeutic independence for older CYP, reflecting that caregiver absence provided greater opportunities for young people to develop their own self-management skills. This was considered particularly important in instances where caregiver anxiety was high or where parental involvement supported a safety behaviour. Although parental involvement often contributed a layer of complexity to treatment management, clinicians acknowledged that it was sometimes helpful to see how and where phobic behaviours had been maintained. Parental presence offered therapists an opportunity to strengthen parents’ understanding and appreciation of a child’s or young person’s perspective. It facilitated education at the family level, encouraging parents to identify moments of unintentional collusion or reflect on the emotional language that they used:

I did just mention you know about the some of the comments were a bit – I felt were a bit unhelpful. But mums sort of narrative about the initial incident where you know you could argue that the phobia began from and that kind of thing. But some of that narrative could do with sort of changing a little bit and we did have some conversations about that.

Logistical challenges in treatment delivery

Clinicians reflected on the logistical issues that arose when delivering OST, particularly the suitability of the treatment setting. For those whose role it was to support CYP across a large geographical area, the importance of families being able to access treatment at a central location was paramount. Travel burden for practitioners was also acknowledged:

I mean if a client didn’t have transport via a parent that might have been a bit of an issue. Just from our service in terms of me then having to put you know more business use on my car insurance and stuff like that to have etc. But clients that I had, their parents had transport, so we were just arranging to meet at a location and that just simplified it.

School-based professionals positively evaluated their experiences of meeting CYP in school settings, accepting that this offered a unique opportunity to reduce accessibility barriers and opportunity costs. A more dominant discourse, however, underscored the importance of having a neutral therapeutic space. OST was often posited to be more difficult than traditional CBT to deliver in an educational setting owing to challenges inherent in managing CYP’s distress and maintaining confidentiality over a protracted period of time. A minority of clinicians cautioned against school delivery, believing that this carried the risk of CYP establishing an association between a phobia and the classroom environment:

[We] don’t often have a therapeutic space. Some schools can provide that, but not all schools. So yeah I guess there was that kind of, if this all goes wrong, and this child’s really distressed, and we’re in the middle of a school and the parents are not around, and all them other things. I suppose there was a little bit of anxiety around how do we best manage that.

Similar logistical challenges emerged in health-care settings. Therapists who were based in CAMHS settings reported constraints in securing therapeutic space, often highlighting a discrepancy between the 3-hour appointment times required for OST and the hourly room booking templates utilised by local administrative systems. Animal phobias were perceived as particularly difficult to accommodate, with additional permissions and space required to host live stimuli. Some clinicians highlighted practical challenges in ring fencing 3 hours of clinical time to facilitate OST and in securing the same dedication from third parties (e.g. dog handlers and community nurses), who were periodically needed to assist with exposure:

Other things that I found were, that make it more tricky, was dog phobia. OK sort of good to use a variation of dogs, it is, so say for example if you’ve got – if you’re using up to three dogs per 3-hour ASPECT session it, it worked wonderfully. But, where do you store the dogs? So what, what I’d – was fortunate enough to do was I was working on a Saturday when I did the OST for the dog phobias and I was fortunate enough to use different rooms within the CAMHS vicinity. So one room would have one dog in, another room would have another. Sort of which is how I got round but, in some services that might not be practical.

A shared narrative highlighted that a breadth of difficulties were encountered in obtaining and organising meaningful stimuli, with personalised treatment goals often necessitating substantial time investment to ensure that clinicians had sourced sufficient and appropriate resources. This time investment often compromised other elements of practice and was posited to be much longer and less manageable than traditional treatment preparation:

I found the, the intervention to be – I’ve enjoyed the sort of watching the progress of it. But I must admit I found the process of it to be quite stressful. Having to sort of juggle and arrange to get everything all into one place at one time. And that was probably one of the most sort of stressful parts of it. Whereas the CBT version of it was – although it was stressful getting the sort of the, the objects, it wasn’t working to a timetable in the exact same way.

Clinicians typically identified a service initiation phase during which the challenges of sourcing materials were amplified. Gradually, some clinicians established links with external providers and began to develop their own resources to draw on as they moved forward. Personal workarounds were common, but these relied heavily on the creativity of the individual or on proactive input from the ASPECT team. Therapists described a relatively high level of study support, encompassing both the provision of information and the physical and financial acquisition of treatment resources:

So, you know, there was some preparation, but it wasn’t impossible. But I think – I think just those initial ones, it sometimes took time to do that, you know. I think I put together a PowerPoint^® [Microsoft Corporation, Redmond, WA, USA] and it was videos and recording and things. But had we kept doing that – I think it’s like anything. It’s when you start, or when you initially do something, you put more resource and time and energy into it. But had you kept doing it, you would just repeat.

I found it really helpful like whenever I sort of e-mailed people to for resources of things like everybody sort of got back to me really quickly and gave me the things, or guided me to different resources, or passed different resources on to me. So that were all yeah kind of really, felt really supportive really.

Post-treatment reflections

The qualitative analysis interpreted six themes from the clinician’s post-treatment reflections. These themes centred on (1) clinicians’ perceptions of treatment outcomes, (2) case suitability, (3) the impact of phobia type, (4) reflections on session format and length, (5) possible protocol adaptions and (6) OST sustainability.

Clinicians’ perceptions of treatment outcomes

Clinician’s post-treatment reflections were largely positive. Overall, therapists reported that they had enjoyed participating and for many the opportunity to observe CYP’s progress had been rewarding. OST was perceived to have both primary and secondary effects, with improvements in phobia reported alongside benefits to children’s family life, educational engagement and self-confidence. Clinical ‘spill over’ effects were also reported, with several clinicians reporting observable improvements in children’s comorbid presentations:

Not only has the treatment cured [child name]’s phobia – in inverted commas – but I believe it is largely responsible for [child name]’s overall improvement in his general confidence. He’s had a great year in school so far and he’s starting to shine, both teachers and family have commented on how much happier he seems.

Variability in clinician-reported effects was nonetheless apparent. In some cases, clinicians were surprised at how much progress had been made in one session, whereas others had been left disheartened. Drawing on this learning, some suggested that OST was the most effective for mild to moderate phobias. Others, however, took a more nuanced view. Expectation management emerged as a strong interpretative theme, with several advocating that the value of OST should be quantified through its capacity to deliver functional improvements rather than a clinical cure. In this context, OST was perceived to retain applicability across the severity spectrum, with clinicians assuming primary responsibility for the tailoring of meaningful and manageable goals:

Because there was another young person I saw that – actually didn’t even progress beyond the assessment appointment. It was kind of really severe – an impairing phobia. But it was never going to be realistic for that to have been cured in one session. But actually we could perhaps have set a, you know, a different goal that would’ve meant that the phobia was still around but more manageable.

Case suitability

Akin to CYP and parents, clinicians highlighted the need to consider an individual’s suitability for OST at referral based on an informal assessment of their treatment motivation, case complexity and needs. Consensus was that CYP needed to display sufficient intrinsic motivation to engage in the exposure work, although the potential impact of external motivators was also acknowledged. Where there was a time-sensitive reason for a child or young person to seek help (e.g. an upcoming school event or public health initiative), clinicians asserted that OST would most likely be preferable and advantageous for the child:

I think if there was – so say for example if we was against the clock like the patient was scared of needles but about to have an operation then in that sort of case then I would sort of strongly suggest they go down an OST-type route.

Having delivered OST, clinicians had no strong opinions regarding an optimal age range for treatment. Some retained their pre-treatment beliefs, suggesting that the length and intensity of treatment sessions could potentially challenge concentration in younger CYP and artificially limit its results. More frequently, however, clinicians reflected on the cognitive and behavioural demands of the intervention, confirming that the success of OST relied heavily on CYP’s capacity to concentrate and process information:

[O]bviously if we’re offering a one-session treatment, and you’ve got a spontaneous child with a lack of concentration, or a poor concentration span should I say, then actually it’s just going to make things a little harder.

Contradictory to their expectations, clinicians had varied opinions on whether or not OST was suitable for CYP with additional needs. The impact of neurodevelopmental disorders in particular was difficult to predict. Cognitive rigidity, a defining characteristic of ASD, was conceived as both a driver of goal achievement during exposure and a potential barrier to therapeutic rapport. A shared narrative highlighted the need for additional preparatory time to be allocated to CYP living with these disorders, but the notion that neurodevelopmental criteria should dictate treatment access was largely rejected in favour of providing additional service support:

I suppose thinking about what – the scaffolding around them and actually whether there is something that can be done to sort of allow that person to engage with the support. I think if there’s that scaffolding, then I think either would be OK to do.

The impact of phobia type

A dominant theme within the data set was the extent to which OST aligned with different phobia types. Clinicians revealed a level of uncertainty regarding the suitability of OST for BII- and vomit-induced fear. Key difficulties in these cases included both the practical challenges of coordinating exposure and the challenges inherent in replicating the physiological sensations or contextual influences necessary to trigger a fear response:

I guess with the needle phobia it’s a little bit different because although yes you can introduce them to the needles and have a little fake ‘ow’ etc., and some of that, . . . that just a little bit different [mm] to actually going into hospital and having that done so I think, yes, if you view all cases, apart from the needle therapy ones, I think, yeah, you’re satisfied that you’ve changed that [mm] and made a big difference just not quite as sure with the needle phobias really.

Multiple exposure sessions were considered more suitable for organising stimuli encounters across multiple settings and for securing timely access to BII- and vomit-related treatments. Clinicians reported that external permissions were often required to secure the commitment of allied health professionals, some of whom struggled to commit to a 3-hour therapy appointment in lieu of their own caseload. Exposure sessions were difficult to predict and collegial relationships were put at risk when OST appointments did not progress as planned:

I think in this case it was sort of having to locate and get a fake arm because I was doing needlestick phobia. Getting another member of staff who was a phlebotomist come in the session with me, so it was all these types of things that would in some ways it would make it a bit more tricky.

Repeat therapy appointments were advocated for both phobia types to address any delayed anxiety triggered by exposure to a potentially infectious stimulus. Clinicians accepted that the OST model did not offer this opportunity, noting that this posed a risk to CYP in the days following the treatment:

She was still really anxious after the OST ’cause she was waiting to get sick. So it was about how do you target that in a fear hierarchy in an OST? ’Cause it was you know – she didn’t have any rules about when it would happen. It was just she didn’t sleep for like a night and a day or something because she was so worried that she was going to get sick from it.

Reflections on session format and length

At study outset, many clinicians conceptualised OST as a ‘one chance’ intervention that placed considerable burden on a practitioner to achieve a positive outcome. Having delivered OST as part of ASPECT, several practitioners reflected on this concept, often reframing and refining their initial views.

A strong narrative within the data set revealed that, although it was perhaps logistically easier to implement conventional CBT, a single longer session of treatment arguably made more sense to clinicians from a theoretical point of view. Therapists largely felt that the therapeutic process of OST was clear and easy for families to understand. Most confirmed that the OST format was likely to maximise treatment engagement and outcome by removing educational attendance barriers, lessening pre-session fears and removing external distractions or treatment withdrawal temptations. Single-session delivery was also perceived to precipitate more efficient use of therapist and patient time, removing the need for between-session reviews and minimising the opportunities for the maintenance of safety behaviours:

I think it’s obviously more straightforward to just do treatment as usual, because it’s a similar setup to eight sessions for treatment of depression would be . . . but clinically, I just don’t feel as though, in the context of a phobia and exposure sort of stuff, I just don’t feel as though the week’s break is helpful. There could be you know – homework tasks in between sessional work stuff, but it’s very hard to know whether or not there were any maintaining behaviours going on, or any safety behaviours. And it’s, yeah – so I definitely think that, clinically, it’s a better fit to do the single-session treatment.

Some clinicians who had been concerned about session length at study outset described how treatment had progressed more quickly than they had expected, with CYP often capitalising on early success to progress further than they anticipated. Rather than being the inflexible treatment method that many clinicians had envisaged, OST was typically reframed as an adaptable model, with the additional time afforded by a longer session allowing CYP greater opportunities to move up and down their fear hierarchy in one continuous session. Thus, OST was perceived to enable more ‘side-stepping’ opportunities than conventional CBT, allowing CYP to repeat previous steps or vary them without losing treatment momentum:

The ability kind of in the moment to sort of move up and down that felt fitting for a young person you know. So you had the ability to do that, so you had that momentum you know whereby you could kind of move forward but equally you can kind of drop back in the moment if you felt that you needed to. Whereas of course in your separate sessions, you can do that but not in the same way that a one-session treatment allows.

Clinicians’ concerns regarding the constraints of the one session approach were most likely to be retained in instances where CYP had been unable to attain their treatment goals. Here, OST was at best considered a primer for further intervention and at worst a potentially disengaging experience for those who were left disappointed. A subset of clinicians supported protocol violations for more complex cases, having taken the personal decision to extend the intervention in these instances to ensure that treatment could be completed:

Kind of like what I was saying before, if there is complexities having that – that almost wiggle room to say actually we can do more.

Possible protocol adaptions

Clinicians identified several amendments to OST procedures that they had implemented locally. Some had reconfigured assessment proformas to make them more individualised, while others had developed additional homework sheets and resources to support CYP post treatment. Occasionally, photographs and videos were used to document progress and disseminated to families to aid skills practice and maintenance. Larger adaptations were also evident. In some areas, therapists had set up supervision groups to assist with session-planning and case reflection.

Suggestions for future delivery enhancements clustered around three themes: (1) the allocation of dedicated time for psychoeducation and anxiety management during treatment, (2) a corresponding division within the planning session to accommodate this, and (3) follow-up sessions to review progress, promote skills consolidation and support relapse prevention. The first of these requires careful consideration given that relaxation and anxiety management are not part of OST as defined. The latter addition was specifically recommended to improve both the CYP’s and the clinician’s experiences of treatment end:

Erm I think we would find out how things have gone [hmm] erm and if things have maybe sort of regressed a bit [Yeah]. Then encourage them to think about what they need to do in order to build up again and overcome things. So I’d use it as a, as a way of being able to recap on the learning [hmm] that they did in the OST and the previous, in the appointment before that.

I think it would be having a chat with them seeing if there was anything that they needed to consolidate. Would there be – you know would they want to kind of be in that situation again and just practice. Would it just be a case of having a walk and seeing if we could encounter some kind of challenges and how they’d manage. But I suppose it would be very much what, what the value would be for that person really.

One-session treatment sustainability

Clinicians reported that their competence in completing exposure work had increased and that they felt confident in being able to complete similar work in the future. Delivering OST appeared to precipitate a range of professional effects, including increased understanding of phobia-specific therapies, the proliferation and enhancement of CBT skills within a team and an increased awareness and enthusiasm for early intervention in this population:

I think it’s given me another string to my bow really in terms of being able to sort of deliver specific phobia treatment. It’s a really encouraging confidence building thing when it comes to people managing anxiety in different ways. It’s a good way of demonstrating it. I also think it’s really improved my competence in terms of actually working with people to do like the exposure work as well. I think it’s really helped.

The notion that childhood phobias fell on or outside the periphery of CYP’s mental health services was a malleable one. A direct but unintended consequence of ASPECT was that services were prompted to recognise and address this deficit in care. Involvement in the research appeared to increase practitioners’ appreciation of the depth of impairment induced by specific phobias and in some instances instigated a service-level shift in referral acceptance rates. Attitudinal changes among staff were advocated to dispel tacit barriers to treatment access and promote local care pathways and behaviour change:

. . . it’s increased the services awareness and I think the difficulty for our service before, ’cause we’re a specialist CAMHS service, it was the belief that we didn’t treat phobias. So – but we do actually have a clinical pathway for phobias and we should be treating phobias, so what was great about the trial was that it raised – we have six community CAMHS teams and it raised their awareness that actually we should be accepting referrals for phobias. Yeah so it’s been great for our service in that regard that we’ve probably been accepting more phobias that have come through as a consequence of raising people’s awareness of the clinical pathway.

Clinician’s views on whether or not they themselves would continue to use OST in practice were mixed. For some, particularly those who were used to working systematically, CBT continued to be the treatment of choice. This decision was largely based on individuals’ perceptions that OST offered fewer opportunities to build a strong therapeutic alliance and client rapport:

I think I would continue to deliver the CBT one, but I don’t think I would continue with the one-session one as it is now. If it was more relationship led if you like and person centred in that way then I would probably be happier to do it. But that was the missing part for me, and I think that really, makes a huge difference in any therapeutic piece of work and relationship.

Several organisational barriers to routine OST provision were identified. Clinicians reported that clinical and administrative procedures needed strengthening to ensure that phobia cases could be adequately tracked in referral systems. A dedicated, streamlined pathway for phobia treatment and resource allocation was perceived as essential to implementation success. Local service-level input was requested to ensure that adequate training, supervision and service support systems were established. Cross-pathway staffing was viewed as advantageous to promote team cohesion, enhance peer-to-peer learning and optimise team morale:

And actually the benefit that that then had for those practitioners who’ve then seen young people with say a simple phobia and they’ve got better, it’s boosted their motivation. So I think what it’s – what it’s now given us is something that we can go back to the managers and say actually this is you know, on lots of different levels, there’s a real incentive for actually having some presentation be seen much quicker. Even if there are only mild to moderate, it actually has a benefit for those clinicians in terms of their motivation and feeling like skilled therapists.

Summary of clinicians’ views

Overall, clinicians displayed a level of openness towards OST, accepting that it could reduce opportunity costs and increase treatment motivation and engagement for some CYP. The potential cognitive and emotional burden of treatment participation was acknowledged, as was the emotional and logistical demand that OST placed on therapists. Dedicated training for OST was upheld as important irrespective of professional role and experience. Some gaps in clinician’s understanding and preparation for OST delivery were evident; additional avenues for training expansion and tailoring are recommended.

Clinicians’ confidence in exposure treatment and their capacity to establish therapeutic rapport were perceived as critical process variables influencing treatment experience and success. Dual competencies in collaborative, child-centred working and effective time management were important. Clinicians displayed a level of openness towards third-party involvement in exposure therapy, although the greatest benefit of caregiver involvement was perceived to occur at OST assessment and treatment end.

Clinicians perceived OST to have utility in treating specific childhood phobias, although a diverse range of outcomes was acknowledged. Phobia type was perceived to impact on intervention feasibility and effects, with greater operational challenges present in vomit and BII phobias than other diagnostic categories. CYP readiness and developmental stage were also identified as potential influences on treatment suitability. Clinicians tended to support the development of intervention suitability guidance rather than the introduction of mandated eligibility criteria. The provision of additional treatment support may enhance OST accessibility.

One-session treatment implementation presented some organisational challenges. Greater resourcing and alignment with service administrative systems would probably support OST delivery in routine practice. Recognition and validation of the substantial time that clinicians spend preparing for OST would be advantageous.