Energetic activity for depression in young people aged 13-17 years: the READY feasibility RCT

Article history

The contractual start date for this research was in October 2019. This article began editorial review in June 2023 and was accepted for publication in August 2024. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The Health Technology Assessment editors and publisher have tried to ensure the accuracy of the authors’ article and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this article.

Copyright statement

Copyright © 2024 Smith et al. This work was produced by Smith et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, distribution, reproduction and adaption in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.

2024 Smith et al.

Design

The full READY trial methodology is reported in the protocol paper24 and this manuscript is reported in line with the CONSORT (Consolidated Standards for Reporting Trials) reporting guidelines for parallel-group randomised trials. 41

We conducted a three-arm, multisite, 12-week feasibility cluster RCT with 6-month follow-up to determine the feasibility of undertaking a full randomised trial to evaluate the clinical effectiveness and cost-effectiveness of exercise as an intervention for treating depression in young people. The design had an embedded process evaluation to examine the acceptability, delivery and adherence of the intervention. An economic evaluation tested the feasibility of collecting resource use data and intervention costs.

Setting

The trial was conducted in Hertfordshire, Bedfordshire and Norfolk. Young people seeking support for depression were identified from general practitioner (GP) practices, NHS Single Point of Access (SPA), Child and Adolescent Mental Health Services (CAMHS) or schools or self-referred to the study. Interventions were delivered by registered exercise professionals (REPs) employed by local physical activity providers (e.g. Watford Football Club Community and Sports Trust, Active Luton, Norwich City Football Club Community Sports Foundation) in local community venues (e.g. sports facilities and community halls). At the time the study was designed, Sport England was funding a national network of exercise providers, often supported by football clubs. By the time the study was established, much of this funding had been withdrawn. However, there remains a range of providers that are committed to engaging with local communities to provide interventions of this kind (e.g. charitable trusts, sports clubs and local community organisations). While this removes access to a national network, it does mean that there is much more flexibility for local solutions. The study was conducted between October 2020 and August 2022.

Eligibility criteria

Intervention

The intervention was delivered to groups in 24 sessions over 12 weeks, with assessment at baseline (prior to the first session) and follow-up at 14 weeks and 26 weeks from baseline. Sessions were each delivered by a REP, supported by a mental health support worker (MHSW).

Young people were randomly assigned to one of the following three groups:

1. high-intensity physical exercise of alternating training sessions42,43 (e.g. basketball, football, circuit training to music, boxing drills)

2. low-intensity physical exercise of alternating training sessions44 (e.g. chair-based exercises and multiactivity sessions)

3. social active control including activities such as quizzes, computer-based games and group discussions, with the activities agreed on by the group.

Interventions were adapted due to COVID-19 restrictions (see Report Supplementary Material 5 and Report Supplementary Material 6). One main change was to ensure that participants did not share any equipment. Therefore, multiactivities were included, such as circuit training (high-intensity exercise) and chair-based exercises (low-intensity exercise). This also meant that participants were able to participate remotely if they were isolating at home. Furthermore, in the low-intensity group, to maximise attendance in the light of concerns about mental health and COVID-19, a hybrid option was offered. Young people in person were given an activity to engage in and young people logging in online were provided instructions on how to use common household objects such as stuffed socks to engage in games themselves that would mirror the activity in person. Original and alternative adapted exercises are shown in Report Supplementary Material 1 and Report Supplementary Material 2.

All intervention groups were offered a behaviour change ‘healthy living’ component to begin each session, to facilitate exercise adherence during 12 weeks of intervention, and to maintain physical activity over the longer term. 24 Group sessions were run in community settings or online where necessary and led by a REP who had been trained by a health psychologist and researcher with expertise in behaviour change (Angel Chater and Neil Howlett) for the study.

The minimum planned group size was nine participants to ensure that each group was of sufficient size to facilitate delivery. Each group was planned either as a single-sex group or to include at least two girls or two boys. Owing to the nature of the intervention, neither participants nor delivery staff were blinded to group allocation. All participants in the study continued to receive usual care.

In addition, the initial plan was for groups to have a minimum of nine participants. This was based on feedback from our early public involvement and engagement work with young people, who said that this sort of group size would be ideal as it would not be too large and overwhelming but also they would not be the sole focus of the deliverers. Where there were smaller group sizes during the feasibility trial, particularly in the high-intensity group, additional support was provided to the young people who did attend to ensure that they felt as comfortable as possible in smaller groups or on their own. This involved offering greater flexibility in the activities with which the young people could choose to engage.

Participants were followed up at 14 and 26 weeks with questionnaires on depression, quality of life, self-esteem and service use. Some young people and intervention staff were interviewed to understand their experiences of the exercise sessions and participation. Full details of the interventions and measures used were published as a protocol and in pilot work. 24,45

Screening, eligibility assessment and recruitment process

Potentially eligible young people were identified from CAMHS (tier 2 and tier 3), SPA caseloads and/or waiting lists, and searches of GP records. Young people (or parents/carers on their behalf) were also able to self-refer via the study website. The study was advertised on social media and on leaflets and posters displayed in GP surgeries, local community centres, schools and other venues. Members of the research team were also invited into schools to promote the study.

Those who met the initial consultation criteria (e.g. able to travel to the venue or attend online if necessary) and consented to their details being passed to the study team were scheduled an informed consent meeting to obtain consent (if they were aged ≥16 years) or assent (if they were aged <16 years) to participate, with consent from parents/carers if their child was <16 years old.

Screening measures were then completed online (Table 1). If eligible, the young person was assigned to a group, and if ineligible, they were referred back to their referring service and signposted to services, if appropriate.

Once a group of nine young people had been formed, the trial manager randomised the group as a whole to one of the three intervention arms using a web-based REDCAP (Research Electronic Data Capture) randomisation tool. Randomisation was allocated with a block size of three for each study site.

Baseline and follow-up measures

Young people and parents completed online measures at baseline and at 14 and 26 weeks. These included data collection on psychological, carer burden, physical activity and adverse events. Attendance, adherence and physical exertion data were recorded at each session by the REP. Heart rate data during the session were recorded using chest band monitors. All measures are listed in Table 1.

Owing to the nature of the intervention, it was not possible to blind the study participants or the delivery staff. However, all study assessments were completed online independently by the study participants, reducing response bias during assessment. All data analysis was undertaken blinded to allocation.

Feasibility assessment

Feasibility criteria were predefined based on the stop–go criteria for the full trial:24,58

• Referrals and recruitment –

  • Recruitment of three sites providing mental health services to young people.

  • Identification of 30–50 patients per month from record screening at CAMHS and GP practices at each site.

  • Referral of 16–20 patients per month at each site for eligibility screening.

  • Recruitment of > 10% of eligible young people.

  • Identification of an additional five sites in the main trial (based on referral data).

• Acceptability –

  • Acceptability of interventions and questionnaires to young people.

  • Young people’s attendance at sessions to be > 66%.

  • REP and MHSW acceptance of training and willingness to deliver sessions.

  • Adherence to the intervention protocol by deliverers.

• Completion of trial measures –

  • > 80% of main outcome measures completed at 14 weeks.

  • Successful completion of resource use data for 80% of patients at baseline and 26 weeks.

  • Feasibility of collecting average and peak heart rate and accelerometer data at 14 and 26 weeks.

Process evaluation

As a response to the low recruitment rate, interviews were conducted with recruitment staff to obtain a better understanding of the key barriers and to identify solutions for improving recruitment. Three online interviews were conducted with four participants from three sites to obtain a detailed description of the steps involved in identifying, determining eligibility and enrolling young people; the staff member’s perceptions of the understanding and expectations of carers and young people about trial participation; and to identify key points in the recruitment process where carers and young people declined to take part or were excluded.

We aimed to observe 5–10% of intervention sessions and used a log to record observations and ascertain fidelity to the training and potential contamination.

We conducted focus group/interviews with participants after they completed interventions. Semistructured topic guides were used to explore the acceptability of intervention and study methods and the barriers to and facilitators of participation/intervention delivery. These focus groups/interviews were audio-recorded.

The focus groups/interviews were transcribed verbatim. We aimed to thematically analyse these using NVivo (QSR International, Warrington, UK) software, but, due to the small number of participants, we changed our approach. The two core members of the physical education team reviewed all transcripts and made independent notes summarising the key themes and findings to address the physical education aims. The key themes and findings were then synthesised, along with the intervention observations, to explore implementation fidelity and identify any modifications the study needed.

Patient and public involvement

Young people were actively involved in developing the research proposal, research process and all aspects of the trial. 24 This included ongoing consultation and establishment of a dedicated READY young people’s advisory group (YPAG) involving young people aged 13–17 years from across the study area who had lived experience of depression (either personally or through siblings/friends). They were recruited from three young people’s groups: Young Healthwatch Central Bedfordshire, the Youth Council of Hertfordshire Partnership University NHS Foundation Trust and Breckland Youth Advisory Board from Norfolk. Regular online and face-to-face YPAG meetings were held in accordance with National Institute for Health and Care Research (NIHR) INVOLVE guidance59. Three parents/carers and two Public Involvement in Research group members were on the Trial Steering Committee.

Equality, diversity and inclusion

Equality, diversity and inclusion was built into the study design at every stage, as reflected in the study design, the public and patient involvement (PPI) programme and the process evaluation. Study locations were selected to reflect a range of deprivation levels, ethnic populations and city and rural locations. PPI included a wide range of young people from various backgrounds and the process evaluation planned to explore barriers to recruitment of and engagement from particular groups.

Data analysis

Health economics

The aim of the economic component of this feasibility study was to test the methods and data collection tools proposed for the economic evaluation in the main trial. 24 We evaluated the completeness of data collection from patient questionnaires completed at baseline and at 14 and 26 weeks for resource use, the EuroQol-5 Dimensions, five levels (EQ-5D-5L), the Child Health Utility 9D (CHU-9D) and an indicative cost estimate of the exercise intervention.

This included:

• data collected on staff and fees for using relevant facilities

• intervention costs in the three groups using study records and expert opinion

• participant attendance at sessions.

Health and social care service used a modified Client Service Receipt Inventory (CSRI)60 completed by the young person in consultation with the MHSW and with input from the parent/carer where necessary. Details of the proposed health economic evaluation have been published. 56

As recruitment in this feasibility study was lower than expected, we revised a few originally proposed health economic methods. First, we did not calculate a cost per person of the intervention; instead, we reported the costs and estimated an indicative cost based on a group size that might be more reflective of likely group sizes in practice. Second, numbers meant that cost estimates related to service use would not be informative. Instead, tables were produced showing average number of items for each resource category type. Estimates provided total numbers of items, not disaggregated. This demonstrates what resource items occur most frequently and enabled a judgement of the likely resource use items that would be important drivers of the total cost.

Third, we did not estimate quality-adjusted life-years and instead just reported EQ-5D-5L and CHU-9D values at the three time points (Table 2). Analyses were conducted in Microsoft Excel^® (Microsoft Corporation, Redmond, WA, USA) and IBM SPSS version 28 (IBM Corporation, Armonk, NY, USA). Because of the short follow-up, discounting was not used.

Ethical considerations

The trial was undertaken in accordance with the principles of ICH (International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use) Good Clinical Practice, and all relevant ethics and governance processes, including the Health Research Authority approvals. Processes to manage risk were put in place to ensure timely and appropriate clinical support for the young person where necessary. Parent/carer involvement was encouraged through engagement in the study assessments and by providing a study information pack. Ethics approval was received from the East of England – Cambridgeshire and Hertfordshire Research Ethics Committee (reference 20/EE/0047).

Trial registration

This trial is registered as ISRCTN66452702 (registered 9 April 2020, www.isrctn.com/ISRCTN66452702).

Referrals, recruitment and retention (objective 3)

The CONSORT diagram is shown in Figure 1 (recruitment) and Figure 2 (overall progression). Cumulative referral and recruitment is shown in Figure 3. There were 321 young people referred to the study, 173 assessed for eligibility, 15 recruited and 14 randomised. The sample size achieved was 14 (social control, n = 7; low intensity, n = 6, high intensity, n = 1). Four participants withdrew (before 14 weeks) during the study (social group, n = 2; low intensity, n = 2). The recruitment rate was 0.31 participants per month, per site.

FIGURE 1.

CONSORT study flow diagram: recruitment.

FIGURE 2.

CONSORT study flow diagram: overall progression.

FIGURE 3.

Referral and cumulative study recruitment.

The high-intensity group with only one participant was included for ethical reasons. A number of participants withdrew before group randomisation, leaving only one young person who had agreed to be randomised as the recruitment for the study came to an end. As this young person had committed to the study, a decision was made to run the intervention for that person, accepting that there would only be one person in the group.

Descriptive statistics of participant characteristics and outcomes at baseline are given in Table 3. As there was only one participant in the high-intensity group, data from this group have not been reported. The low-intensity group was predominately male (5 of 6, 83.3%), whereas all participants in the social group were female. Seven of the 13 young people identified as White British (54%), with 1 from another white background, and 4 of 13 (31%) reported a non-white ethnicity. The CDI-2 score ranged from 16 to 31, and the young people reported a significant amount of time in moderate activity per week (610–1052 minutes).

Figure 3 summarises the cumulative recruitment of young people into the study. Initially the focus for identifying potential recruits was on the triage services (the SPA, or other regional variations) for referrals to mental health support into CAHMS. Although a large number of young people were referred through these triage services, this led to a small number of young people being randomised (2 from 126). Reasons for ineligibility were age (too young, n = 5) or a CDI-2 score that was too low (n = 9) or too high (n = 8). Alternative routes of recruitment via GPs and schools were also tested. Four GP practices sent 460 letters, referred 45 young people and screened 15, with 6 randomised (40% of referrals). Among schools, there were 45 referrals and 5 randomised (11%). For comparison, 2 from 46 referrals were randomised from referral agencies (4%).

Following randomisation, the overall retention rate was 71.4% (see Figure 2; Table 4). Attendance at sessions (Table 3) was 78.7% overall, ranging from 66.7% (week 4) to 90.5% (week 12). The young people in the high-intensity group (n = 1) appeared to show, overall, a consistently high attendance, with the social group (n = 7) overall showing lower attendance (lowest 57.1% at week 4). Table 4 shows that most sessions were delivered face to face, with 9.1% of the social control sessions delivered online due to COVID-19 infections. The low-intensity group was delivered in a hybrid format with 34.6% attendance online.

With regard to heart rate data (Table 5), the young people in the high-intensity group did not achieve the target heart rate, and three participants in the low-intensity group did achieve target (33.3% of non-missing values). The average and peak heart rate reported by group shows a higher peak and average heart rate in the high-intensity group than in the low-intensity group, indicating that there was a difference in the heart rate achieved by the study participants.

Data completeness at baseline (see Report Supplementary Material 2) is high, with all outcomes at baseline achieving 83.3–100% except the COM-B (Capability, Opportunity, Motivation-Behaviour change) motivation subscale, which was 66.7% in the low-intensity group.

Data completeness of outcomes at 14 weeks is given in Report Supplementary Material 3. All outcomes achieved 100% completion rate, except for the parental burden scale total (BSFC) in the social group (80.0%) and the accelerometer data (social group, 80.0%; low-intensity group, 50.0%). Data completeness of outcomes at 26 weeks is given in Report Supplementary Material 4. Completion rate of all outcomes at 26 weeks was 100% in the high-intensity group and 80.0% in the social group. For the low-intensity group, 100% was achieved for all outcomes, except the Youth Physical Activity Questionnaire at 75.0% and the accelerometer data at 50.0%. Efficacy outcomes at follow-up are not reported but can be supplied on application.

Process evaluation

In the process evaluation, we examined the feasibility of delivering the intervention across the three sites (objective 1) to address adherence to intervention protocol by exercise professionals, adherence to interventions by participants, acceptability, and barriers to recruitment and participation by young people.

The additional investigation into recruitment processes was undertaken while recruitment was ongoing. We found four main barriers to recruitment.

1. Length of time between identifying potentially eligible young people and trial participation commencing. The need to deliver group activities in each trial group meant that some young people were put ‘on hold’ until enough young people had been recruited to the allocated arm, leading to substantial delays (in some cases of >6 months). This meant that some young people might have lost interest in participating or did not want to participate if only a few other young people were in their allocated group.

2. Staff were unsurprised that young people suffering from low mood, and potentially from social anxiety, were reluctant to participate in a study requiring them to meet with other young people they did not know to carry out activities that they were perhaps not confident in participating in.

3. Staff argued that the thresholds of the trial (CDI-2 score of > 16) screening measures were too high for many of the young people who would benefit from taking part in the trial. Staff felt that several young people who were highly motivated to participate were excluded on this basis (n = 9).

4. The decision to participate in the trial was not always driven by the young person. Staff received referrals and had subsequent contact with carers who were anxious that the young person in their care receive some support. Without buy-in from the young people themselves, this ultimately affected whether they took part.

Two key solutions to overcoming these barriers were proposed by staff. First, staff felt that recruitment in schools might help to increase rates of recruitment, leading to shorter waiting times before commencing intervention sessions. Second, staff wondered if eligibility screening thresholds could be relaxed to enable participation of a wider pool of young people who had reported having low mood and were motivated to participate.

The main process evaluation was otherwise completed as planned. One focus group/interview took place per site. Separate sessions were arranged for young people and staff. Four young people from the social control group took part in a focus group. One young person from each of the low-intensity and high-intensity groups took part in separate interviews. Two staff from the social control condition took part in a focus group and five staff from the low-intensity group took part in a focus group. The staff from the high-intensity group were invited to participate in a focus group but did not respond to the invitation; two out of the three staff had previously taken part in a focus group for the social control group.

Independent observations were undertaken for 8 of the 72 intervention sessions (11%). This comprised three social control sessions, three low-intensity exercise sessions and two high-intensity exercise sessions. Most sessions were observed by two team members who completed independent observations, leading to 13 intervention observation logs completed. There were five team members in total who conducted observations. Where issues with fidelity to the intervention manual were noted by the observers, these were discussed within the team and brought to the attention of the REP where appropriate.

Health economics

The intervention was provided in three venues, with an average venue cost of £1061 for the 24 sessions over 12 weeks. Each session was attended by a REP and a MHSW (assistant psychologist). The REPs and MHSWs were costed assuming agenda for change grade 4. 61 It was assumed that 90 minutes was required to allow for travel for each staff type. This gave an estimate of £2663 for the 24 sessions for staff time. Including the costs of the venue gave estimates of total costs of £3749, £3698 and £3398 for the social activity, low-intensity and high-intensity groups, respectively. It is difficult to compare the cost per person for the three groups as numbers are likely to be higher in actual practice and the high-intensity group had only one participant. If we assume that in practice groups would be offered to eight people, then the approximate costs per person of the intervention would be £460, or approximately £20 per person per session.

Of the 14 individuals randomised, the CSRI was completed for 13 (social, n = 7; low intensity, n = 5; high intensity, n = 1). At the first follow-up 10 had completed CSRI (social, n = 5; low intensity, n = 4; high intensity, n = 1) and at the final follow-up the CSRI was completed for eight individuals (social, n = 4; low intensity, n = 3; high intensity, n = 1). Table 6 shows the quantities of the different resources reported in the CSRI at baseline, 14 weeks and 26 weeks. Common items of resource use were seeing a classroom assistant at school, secondary care contacts with CAMHS, seeing a GP, and having some form of counselling. Resource use appeared to be highest at baseline and during the first period (up to 14 weeks), with lower resource use in the second, follow-up period (up to 26 weeks), even allowing for the reduced numbers of participants followed up for 6 weeks.

For both the EQ-5D-5L and the CHU-9D, 13 were completed at baseline (social, n = 7; low intensity, n = 5; high intensity, n = 1), 10 were completed at 14 weeks (social, n = 5; low intensity, n = 4; high intensity, n = 1) and 9 were completed at 26 weeks (social, n = 4; low intensity, n = 4; high intensity, n = 1). The mean scores on the EQ-5D-5L and CHU-9D for the three time periods are given in Table 7. Where these instruments had been completed, they were completed in full and there was a response for each question and hence overall scores for all participants who returned these instruments could be calculated. Scores for the two instruments were generally similar. Mean values were generally higher for the CHU-9D than for the EQ-5D-5L, although this difference had narrowed at the 26-week follow-up. Correlations between all 32 responses for these instruments at all three time points were calculated. The EQ-5D-5L and the CHU-9D appeared highly correlated (0.821, significant at the 0.01 level). Correlations between these preference-based instruments and the CDI-2 were –0.537 and –0.565 for the EQ-5D-5L and the CHU-9D, respectively, both significant at the 0.01 level. These correlations are negative, as decreases in both the EQ-5D-5L and the CHU-9D indicate worse health, whereas decreases in the CDI-2 represent improved health. These results indicate cautious support for the use of either the EQ-5D-5L or the CHU-9D in future studies of depression in adolescents.

Impact of patient and public involvement input

At the pre-application consultation stage, young people with depression and experience of CAMHS (aged 14–15 years) identified the importance of emotional support for young people participating in the study, leading to the addition of MHSWs to co-facilitate the activity groups alongside the REPs. A LGBTQ+ group of young people (16–19 years old) and an ethnically mixed group of young women from Luton (aged 15–17 years) helped us consider the complexity of gender in relation to exercise and depression, and to enhance inclusion and diversity in our PPI. In addition, the Public Involvement in Research group at the University of Hertfordshire provided advice to the team at different stages of study development and reviewed research ethics documents.

In total, we held 12 READY YPAG meetings (a mix of face-to-face and online meetings), involving 46 young people and 10 youth leaders, between November 2019 and March 2023. YPAG members provided valuable advice to the research team regarding key stages of the research, including recruiting and engaging with schools; using social media relevant to young people; reviewing research ethics documents and data collection materials; providing advice on moving the intervention online (during the pandemic) and post-COVID-19; and co-authoring a research article about the READY YPAG. In addition, three parents of young people are on the Trial Steering Committee, and they have contributed valued advice to the research team.

Feasibility outcomes (objective 3)

The study tested the feasibility of delivering a full RCT of the effectiveness and cost-effectiveness of exercise to treat depression among young people. It examined recruitment, retention, adherence to interventions by young people and to protocol by intervention staff, data completion, acceptability, and barriers to and facilitators of engagement of young people.

Table 7 summarises the feasibility stop–go criteria. The most important finding is that the trial failed to recruit to target, achieving a sample size of 14 in total against a reduced target of 27. However, other aspects of the feasibility study indicate more positive outcomes. Considerable numbers of young people were identified, with more than 20 identified per month on average in two study sites. In addition, retention in the study was strong (71%), and attendance at intervention sessions remained high (> 66%). Qualitatively the exercise professionals and health support workers providing the intervention were willing to engage and reported a positive experience overall. Some issues were identified with delivery of the behaviour change element of the intervention, which can be addressed in a future study. Where young people remained in the study, data completion was very high for most data, apart from some measures of heart rate that proved challenging to collect, partly due to the nature of the process of data collection, and potentially due to COVID-19-related restrictions. The young people wore the accelerometers, and data on physical activity were successfully collected at baseline and 26 weeks. The interest from other study sites was strong, although given the challenges in recruiting study participants engagement with additional sites was not pursued. In general, if engagement with young people can be addressed, it is possible that a large-scale randomised trial could be completed in the future.

Summary of findings

We report on our stated objectives to examine the feasibility of delivering the intervention and to establish the feasibility of conducting a RCT. The COVID-19 pandemic disrupted the recruitment and delivery of interventions. Young people can be identified through a wide range of NHS and non-NHS sources. Although the study had a good number of referrals, the original target for recruitment was reduced from 81 to 27, and only 14 (8% screened) were randomised; of these, only one person was randomised to the high-intensity group. Therefore, the study did not recruit sufficient young people to demonstrate feasibility to proceed to a full-scale trial. The study did provide valuable insights and demonstrated that the intervention was acceptable, and that, once engaged, young people do complete the study. Retention rates (71%) and data completion rates (> 80%) were good and give a good indication that the key barrier to successfully delivering a study is getting the young people to engage at the outset. Through strong PPI engagement we were able to develop some recruitment strategies and identify key influencing factors to enhance engagement with and delivery of interventions. Owing to the limited number of randomised young people, there are limits to our understanding of the reach and representativeness in the study sample.

Lessons learned

Future research and implications for decision-makers

Developing appropriate recruitment strategies via NHS, GP practices, schools and social media and early engagement with the local CRN to support the process would help recruitment. Collaborations between the NHS services and sports delivery partners could help to increase capacity to deliver interventions. However, in-person contact rather than remote consultations and delivery are likely to be important.

The role of community engagement (social media, public health agencies, community groups) needs to be further explored. Strong PPI and engagement via young people’s advisory groups (including YPAGs) is important to ensure that research is relevant to young people.

A range of sources suggest that a key task is to build engagement and trust with young people seeking help. Hard-pressed services with little time tend to resort to remote forms of communication that may not be helpful in establishing engagement with young people. An important question then remains around understanding how to provide the available resources that would allow successful engagement and improve recruitment into research studies.

The available evidence suggests that the intervention merits further evaluation, assuming that the shortcomings related to recruitment and engagement can be addressed to enhance participation in a trial. This would include building on our process evaluation and conducting in-depth face-to-face work with young people and their families to better understand their concerns. The evidence base suggests that the trial processes as otherwise designed worked well, and the interventions were largely delivered as envisaged. There remains a significant need to provide evidence-based support for this group of young people.

Key learning points

A large-scale RCT is not feasible based on this study, and further work is needed to develop screening and recruitment routes for young people. Our study has highlighted the difficulty in recruiting young people aged 13–17 years into this type of research study. Working through referral agencies, primary care practices and schools enabled significant numbers of young people to be identified, but this did not always lead to recruitment into the study. Anecdotally, from contact with youth workers and observations by recruiting staff, face-to-face contact enables trust to be established with the young people and could be key to successful engagement. There is a need to identify resources for a dedicated team to work with young people and their families to promote engagement. This was further supported by our PPI and process evaluation, undertaken during the trial, indicating that young people and their families respond more fully when contacted in person. The representativeness of our sample is uncertain as we did not achieve our intended sample size to establish feasibility. Future studies need to consider lower thresholds on the CDI-2 screening score to encourage young people with low mood to participate. Furthermore, although this study could have been feasible in the absence of the pandemic, complexities highlighted around the format of hybrid session delivery and the reasons for participant withdrawal or the low uptake among those referred to the study conducted during the pandemic need to be further examined.

What this adds to existing knowledge

Although we know that recruiting young people to mental health research studies can be challenging,70–72 our study highlights the disruptive impact of the COVID-19 pandemic on conducting mental health research in this group. It identifies the mechanisms to be considered around wider recruitment strategies and emphasises the importance of face-to-face communications with young people and their families. Our PPI activities and engagement of young people enabled us to plan our research through the changing phases of the pandemic and helped to highlight ways of building trust with young people and their families and potentially overcoming barriers to recruitment had circumstances been more favourable.

CRediT contribution statement

Megan Smith (https://orcid.org/0000-0002-1482-2350): Conceptualisation, Data curation, Project administration, Resources, Software, Supervision, Validation, Visualisation, Writing – original draft, Writing – reviewing and editing, Other contributions.

Ryan James (https://orcid.org/0000-0002-5448-5424): Resources, Visualisation, Writing – original draft, Writing – reviewing and editing.

Neil Howlett (https://orcid.org/0000-0002-6502-9969): Conceptualisation, Funding acquisition, Methodology, Project administration, Resources, Supervision, Visualisation, Writing – original draft, Writing – reviewing and editing.

Silvana Mengoni (https://orcid.org/0000-0002-9431-9762): Conceptualisation, Formal analysis, Funding acquisition, Methodology, Visualisation, Writing – original draft, Writing – reviewing and editing.

Julia Jones (https://orcid.org/0000-0003-3221-7362): Funding acquisition, Visualisation, Writing – original draft, Writing – reviewing and editing, Other contributions.

Erika Sims (https://orcid.org/0000-0002-7898-0331): Conceptualisation, Investigation, Methodology, Project administration, Supervision, Visualisation, Writing – original draft, Writing – reviewing and editing.

David Turner (https://orcid.org/0000-0002-1689-4147): Conceptualisation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Visualisation, Writing – original draft, Writing – reviewing and editing.

Kelly Grant (https://orcid.org/0000-0001-5319-8127): Formal analysis, Investigation, Project administration, Resources, Supervision, Visualisation, Writing – original draft, Writing – reviewing and editing.

Allan Clark (https://orcid.org/0000-0003-2965-8941): Conceptualisation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Visualisation, Writing – original draft, Writing – reviewing and editing, Other contributions.

Jamie Murdoch (https://orcid.org/0000-0002-9021-3629): Conceptualisation, Formal analysis, Funding acquisition, Methodology, Visualisation, Writing – original draft, Writing – reviewing and editing.

Lindsay Bottoms (https://orcid.org/0000-0003-4632-3764): Conceptualisation, Funding acquisition, Methodology, Visualisation, Writing – original draft, Writing – reviewing and editing.

Jonathan Wilson (https://orcid.org/0000-0002-5279-6237): Conceptualisation Funding acquisition, Methodology, Project administration, Supervision, Visualisation, Writing – original draft, Writing – reviewing and editing.

Shivani Sharma (https://orcid.org/0000-0002-7682-2858): Funding acquisition, Visualisation, Writing – original draft, Writing – reviewing and editing.

Angel Chater (https://orcid.org/0000-0002-9043-2565): Conceptualisation, Funding acquisition, Methodology, Project administration, Resources, Supervision, Visualisation, Writing – original draft, Writing – reviewing and editing.

Cecile Guillard (https://orcid.org/0009-0002-8844-9955): Conceptualisation, Data curation, Investigation, Project administration, Resources, Software, Supervision, Visualisation, Writing – original draft, Writing – reviewing and editing.

Timothy Clarke (https://orcid.org/0000-0002-3901-9601): Methodology, Visualisation, Writing – original draft, Writing – reviewing and editing.

Andy Jones (https://orcid.org/0000-0002-3130-9313): Funding acquisition, Methodology, Supervision, Visualisation, Writing – original draft, Writing – reviewing and editing.

Lee David (https://orcid.org/0000-0001-5319-6156): Funding acquisition, Methodology, Visualisation, Writing – original draft, Writing – reviewing and editing.

Solange Wyatt (https://orcid.org/0009-0007-0829-4436): Conceptualisation, Funding acquisition, Methodology, Project administration, Supervision, Visualisation, Writing – original draft, Writing – reviewing and editing.

Claire Rourke (https://orcid.org/0000-0002-7631-9275): Conceptualisation, Data curation, Project administration, Resources, Software, Supervision, Validation, Visualisation, Writing – original draft, Writing – reviewing and editing.

David Wellsted (https://orcid.org/0000-0002-2895-7838): Conceptualisation, Funding acquisition, Methodology, Project administration, Supervision, Visualisation, Writing – original draft, Writing – reviewing and editing, Other contributions.

Daksha Trivedi (https://orcid.org/0000-0002-7572-4113): Conceptualisation, Funding acquisition, Methodology, Project administration, Supervision, Visualisation, Writing – original draft, Writing – reviewing and editing, Other contributions.

Karen Irvine: Conceptualisation.

Antony Colles: Conceptualisation.

Maria Leathersich: Conceptualisation, Data curation Investigation, Project administration, Resources, Software Supervision.

Martin Pond: Data curation, Resources, Software.

Sue Stirling: Formal analysis.

Ann Marie Swart: Funding acquisition, Methodology, Other contributions.

Jonathan Hill: Other contributions.

Jonathan Sinclair: Other contributions.

Garry Tew: Other contributions.

Wendy York: Other contributions.

Chris Armitage: Other contributions.

Jen Beer: Other contributions.

Sarah Byford: Other contributions.

Roshan Jayalath: Other contributions.

Marion Cowe: Other contributions.

Chris Foreman: Other contributions.

Louise Cobb: Other contributions.

Emma Lovell: Other contributions.

Angela Greenhalgh: Other contributions.

Disclosure of interests

Full disclosure of interests: Completed ICMJE forms for all authors, including all related interests, are available in the toolkit on the NIHR Journals Library report publication page at https://doi.org10.3310/KWNH4507

Primary conflicts of interest: Julia Jones was a member of the National Institute for Health and Care Research (NIHR) Research for Patient Benefit Programme and the East of England Regional Advisory Committee 2019–23. Jamie Murdoch was a member of the Health and Social Care Delivery Research Associate Board (2016–18). Andy Jones was a member of the NIHR Public Health Research Research Funding Board (2014–20).

Data-sharing statement

All data requests should be submitted to the corresponding author for consideration. Access to anonymised data may be granted following review.

Ethics statement

Ethical approval was received from the East of England – Cambridgeshire and Hertfordshire Research Ethics Committee (ref: 20/EE/0047) on 30 June 2020.

Information governance statement

Under the Data Protection legislation, the University of Hertfordshire and Norwich Clinical Trial Unit are the Data Processor; The University of Hertfordshire is the Data Controller and we process personal data in accordance with their instructions. The University is committed to the Data Protection Principles set out in the Data Protection Act 2018 (‘Data Protection Legislation’) and the UK General Data Protection Regulation (UK GDPR). You can find out more about how we handle personal data, including how to exercise your individual rights and the contact details for the University of Hertfordshire’s Data Protection Officer via the following link: www.herts.ac.uk/about-us/legal/freedom-of-information-data-protection/data-protection.

Department of Health and Social Care disclaimer

This publication presents independent research commissioned by the National Institute for Health and Care Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, MRC, NIHR Coordinating Centre, the Health Technology Assessment programme or the Department of Health and Social Care.

This article was published based on current knowledge at the time and date of publication. NIHR is committed to being inclusive and will continually monitor best practice and guidance in relation to terminology and language to ensure that we remain relevant to our stakeholders.

Trial registration

This trial is registered as ISRCTN66452702 (registered 9 April 2020, www.isrctn.com/ISRCTN66452702).

Funding

This article presents independent research funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme as award number 17/78/10.

About this article

The contractual start date for this research was in October 2019. This article began editorial review in June 2023 and was accepted for publication in August 2024. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The Health Technology Assessment editors and publisher have tried to ensure the accuracy of the authors’ article and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this article.

This article was published based on current knowledge at the time and date of publication. NIHR is committed to being inclusive and will continually monitor best practice and guidance in relation to terminology and language to ensure that we remain relevant to our stakeholders.

Copyright

Copyright © 2024 Smith et al. This work was produced by Smith et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, distribution, reproduction and adaption in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.