The impact of a needs-based model of care on accessibility and quality of care within children's mental health services: A qualitative investigation of the UK i-THRIVE Programme

Funding information The Health Foundation, Grant/Award Number: 7671 Abstract Background: The i-THRIVE Programme is a needs-based model of care, based on the THRIVE Framework, that is being implemented across the United Kingdom with the aim of improving outcomes for children and young people's mental health and wellbeing. This study aimed to investigate the impact that this programme has on accessibility and quality of care, as viewed by key stakeholders. Methods: Interviews with professionals and service users were conducted during the implementation of the THRIVE Framework in four sites of one mental health and community service provider. Results: Three themes are identified: ‘impact of needs-based groupings on referral’, ‘impact of collaborative and interagency approach’ and ‘impact of i-THRIVE on clinical practice’. Findings suggest that accessibility was seen to be promoted through the integration of a needs-based approach, flexible re-referral, signposting and information sharing, the use of goal-orientated interventions and collaboration over risk and treatment endings. Shared decision making was perceived to improve the experience of care for young people, as was interagency working. Goal-focused interventions and upfront discussion of treatment endings were seen to help clinicians manage expectations and discharge but could also compromise effectiveness and engagement. Obstacles to impact were resistance to interagency working and a shortage of resources across the system. Conclusions: i-THRIVE is a promising approach with the potential to facilitate the accessibility and quality of mental health care. However, a tension exists between enhancing accessibility and quality of care, which points towards the importance of

outcome and satisfaction monitoring. Obstacles to impact point to the importance of a whole-system approach supported by sufficient resources across the locality.

K E Y W O R D S
CAMHS, i-THRIVE Programme, needs-based approach, service innovation

| INTRODUCTION
There is an urgent need to improve the accessibility and quality of mental health care for children and young people (CYP; NHS England, 2015). This has led to an unprecedented level of interest in addressing problems with the structure and delivery of services (Glassgow et al., 2018). In the United Kingdom alone, the government committed £1.25 billion to support the transformation of Child and Adolescent Mental Health Services (CAMHS) over 5 years. This came in the wake of a taskforce review of CAMHS that identified increased demand and found that access and timeliness were key problems despite improvements delivered by large-scale transformations, such as Children and Young People's Improving Access to Psychological Therapies Services (CYP-IAPT;NHS England, 2015;Edbrooke-Childs et al., 2015).
In 2013, the average wait time for routine appointments and urgent help in England was 15 weeks and three weeks, respectively, with only 31% of CYP who required intervention accessing services (NHS Benchmarking Network, 2013). For those who received help, poor engagement was a key issue with the most usual number of contacts CYP had with services being one . To tackle this, user involvement was identified as a priority, as was the need for CYP to be more involved in decision making about their care (Vohra et al., 2014;Young Minds, 2014). Issues such as this led to the development of Future In Mind (NHS England, 2015), which recognized the need to improve quality through better access and timeliness; making service delivery needs-based, equitable and more effective; and the importance of enabling CYP to be involved in making decisions about their care.
The THRIVE Framework for system change (Wolpert et al., 2019) provides a set of principles for establishing a person-centred and needs-based approach that puts shared decision making at the heart of all choices. It is designed to complement existing transformation programmes but recognizes the limitations of evidence-based interventions and the shortage of existing resources. The THRIVE Framework distinguishes between Getting Help (or 'treatment') and Getting Risk Support, emphasizing the need for young people and communities to build on their strengths. Need is conceptualized according to five categories: Getting Advice and Signposting, Getting Help, Getting More Help, Getting Risk Support and Thriving. The first four represent the needs and choices of CYP experiencing mental health problems, whereas 'Thriving' represents proactive prevention and promotion of mental health and wellbeing for CYP with mental health problems and for all CYP in the wider community. Rather than categorizing CYP based on diagnosis or type of problem, their needs are agreed through a shared decision between patient and service providers. This collaborative style supports another of the THRIVE Framework's key tenets: the requirement for a whole-system approach to the provision of CYP mental health and well-being support. The premise is that by grouping needs by similarity, services from all sectors can be tailored and commissioned to better meet those needs and facilitate an integrated and effective delivery of care.
To enable implementation, the THRIVE Framework has been translated into the model of care, i-THRIVE (i-THRIVE: implementing the THRIVE Framework, n.d.). i-THRIVE was initially rolled out nationally as part of the NHS Innovation Accelerator Programme, and programme support is now commissioned individually by sites. The National i-THRIVE Programme aims to provide a mechanism to deliver a whole-system approach to improving accessibility and outcomes, offering tools, training, consultation and support for services to organize their provision of care. It builds on the approach taken by CYP-IAPT that focuses on partnerships within and between NHS, health and social care providers and aims to move towards a population health model that works across entire communities encompassing education, health, voluntary and local authority sectors (NHS England, 2014). The i-THRIVE Programme aims to establish a 'THRIVE-like' system by tackling transformation at three levels: commissioning and population health at the macrolevel; organizing

Key messages
• A needs-based and integrated approach is reported to facilitate the accessibility and quality of mental health care and support for children, young people and their families.
• Shared decision making and multi-agency working are seen by clinicians and service users to improve quality of care provision.
• A needs-based and integrated service model, supported by information sharing and collaboration over ending treatment, is seen by a range of professionals to facilitate accessibility of mental health care.
• Barriers to programme impact identified by professionals included insufficient resources and resistance from sectors outside CAMHS to multi-agency collaboration • Further evaluation of the National i-THRIVE Programme is needed to fully assess impact on clinical practice services around needs-based groupings at the mesolevel and interactions between CYP, families and professionals at micro level.
To date, the National i-THRIVE Programme has worked with over 70 cross sector organizations, and over half (63%) of CYP in England live in localities implementing the framework (National i-Thrive Programme, 2019). However, despite this scope, no published evaluation has focused on i-THRIVE, nor examined how it has been received by professionals and service users. This study aims to explore how changes introduced during the implementation of the programme are viewed by key stakeholders, with a focus on their perception of innovations in service structure and clinical practice and how this is seen to impact the accessibility and quality of care for CYP.

| Setting
The study was conducted in four London boroughs. Each consisted of a 'site' (CAMHS and services from Clinical Commissioning Groups) and providers from the 'wider system' (education, third sector and local authority). Although the i-THRIVE implementation was led from within the NHS, quality improvement (QI) projects had a focus on education and integration with the wider care system. The THRIVE Framework was implemented across a four phased approach: • Phase 1: engagement, understanding the system, re-design and planning.
• Phase 2: building capacity within the system, workforce planning and training.
• Phase 3: implementation of QI changes, technical assistance and coaching.
• Phase 4: learning, embedding and sustaining Practical support was provided by the i-THRIVE Academy practice and development modules in five areas: 'Shared Decision Making', 'Using i-THRIVE Grids to improve shared decision making', 'Getting Advice, Assessment and Signposting', 'When to Stop Treatment: building confidence in letting go', and 'Risk Support'. Transformation efforts were facilitated by an implementation support team and the i-THRIVE Toolkit.
During Phase 1, transformation priorities were identified with each site and an implementation plan created. This included training to build capacity, as well as QI projects targeted at developing services related to two needs-based groupings: 'Getting Advice and Signposting' and 'Getting More Help'. This was seen as a mechanism to improve the priorities of access to services, waiting times, efficiency, engagement with services, experience and participation.

| Participants
The sampling strategy aimed for maximal variation in terms of staff roles and stakeholder groups, and participants were identified through ethnographic fieldwork. The implementation team and service leads also identified informants. The resulting sample was heterogeneous, reflecting the main stakeholders. It included CAMHS clinicians, commissioners, service leads, service users and their parents or carers.
Participants from the wider referral pathway and implementation team were also included. The eligibility criteria is indicated in Table 1.
Of 80 participants, 13 contributed during both first and second data collection periods ( Table 2). The first period (Time 1) consisted of 16 interviews and 4 focus groups (n = 15); the second (Time 2) consisted of 32 interviews and 9 focus groups (n = 30).

| Analysis
All data were transcribed verbatim, anonymized, imported into NVivo software and analysed thematically (Braun & Clarke, 2006). The analysis used an inductive-deductive approach (Fereday & Muir-Cochrane, 2006) where initial codes were generated prior to analyses that were framed by the research questions. These codes provided a guide but, during the process of analysis, inductive codes were assigned that either expanded existing codes or introduced unanticipated topics. This was consistently applied to all data and the resulting codes were grouped into themes that reflected the substantive experiences and perceptions of participants. Data from both time points were analysed independently and compared before being merged to create a table of themes.
Coding was conducted by two members of the research team and thereafter all themes and codes were reviewed by a third member.

| RESULTS
Three superordinate themes, each with associated sub-themes, were identified (Table 3).
3.1 | Impact of needs-based groupings on referral 3. At some sites, initial telephone contact and discussion with referrers was prioritized, engendering a more collaborative approach that could increase efficiency. CAMHS STAR (Support Time and Resilience) workers were employed to focus on outreach in schools and signposting. One commented: 'All the schools, they can pick up the phone and they can call us before the referrals. That saves them time and that saves us time.' (CS25, Time 2).

Building flexibility into the referral process
At some sites, clinics were introduced to allow flexible care that enabled self-referral and top-up support. One service lead described the way this could be used to support changing need over time: If the child is not ready to engage or they've had enough sessions and they're not reacting, we give the clinician the ability to say, 'OK, guys, have a conversa- New multi-agency forums were established to facilitate collaboration between sectors (social care, education and health) and services.
Some CAMHS staff emphasized the benefits of these forums in enabling regular time and space to build relationships, share knowledge and develop a joined-up approach.

| The challenges of instigating interagency collaboration
Although participants recognized the value of improving interagency collaboration, it was impeded by specific barriers.

Resistance to interagency working
A complaint made by some CAMHS staff was that there was resistance from outside agencies to building relationships and sharing knowledge. A service lead described her experience: 'I've got some of my leads going in to spend time with the local authority team, the social services team. But I'm not getting that kind of support back. […] It cannot be a "one-way street"' (CS05, Time 1).
During Time 2 interviews, one school mental health officer commented on the continuing lack of collaboration between social care and CAMHS: 'My experience of mental health and social services working together with students … is not good. The communication between them seems pretty awful at times. There doesn't seem to be collaboration. There's almost competition' (WR07, Time 2).

The THRIVE framework not embedded within the wider system
A small proportion of CAMHS staff involved in outreach activities felt they had limited scope to influence the wider system towards a group were asked about implementing the framework, they often viewed it as a CAMHS-specific transformation.

Lack of resources in the system
Participants across the sectors highlighted the lack of resource in schools to offer support for CYP outside CAMHS. One school mental health lead explained: 'I'm not sure there's enough out there, certainly in Borough C, and I'm not sure many schools feel equipped to deal with lots of these things' (WR06, Time 2).
A major challenge to system change was a shortage of resources in the community to facilitate signposting. Three of six commissioners recognized the priority to build support in the system.

| Focusing on episodes of care and treatment endings
The i-THRIVE Programme offered a framework to develop a more boundaried role for CAMHS, with a change in clinical focus towards episodes of care supported by goal-oriented and evidence-based interventions.

Freeing capacity by making CAMHS a boundaried service
For several CAMHS staff, changes were positively received because, not only did they respond to need, but they had the potential to free up capacity. Some CAMHS staff emphasized a positive shift. One psy-

Positive and negative impacts of goal-focused interventions
CAMHS staff and service users described contrasting impacts of the upfront discussion of treatment endings. One psychologist described an improvement in engagement: 'There is less drift. They tend to come more. I think they tend to do the goal work a little bit more, because they know it is very time-limited' (CS33, Time 2).
However, one service user described the demotivating effect of being presented with a limited number of sessions: 'At the beginning, on my first assessment thing, it was said that we'd only need three sessions.
That didn't make me feel good, because it was like I'd just waited and then they were like, 'Oh it's still not serious enough' (SU03, Time 2).
A psychotherapist echoed the problems of boundaried care for some service users: When they have got problems with trust, have problems with making a good relationship and use of the

| Improving clinical capacity by focusing on discharge
One of the i-THRIVE Programme's key objectives was to improve CYP's access to treatment by freeing up clinical capacity. This could be achieved by increased discharge rates but that required changing clinicians' attitudes towards risk and ending treatment. CAMHS staff described being inspired by i-THRIVE Academy training days that introduced a new approach to ending treatment. In practice, however, this was either facilitated or hampered by two factors.

Collaboration over risk and ending treatment
The introduction of multi-disciplinary meetings designed to facilitate team collaboration over high-risk cases was well received by CAMHS staff who felt that approaching risk and discharge through group discussion helped build their confidence in decision-making. As one psychotherapist explained: The idea of listening to everyone's viewpoints and coming to a coherent synthesis of that to take back into your practice where you might be able to discharge cases more easily and with less conflict because you've really thought about it. (CS37, Time 2) Lack of support throughout the whole system Outreach into schools was seen to promote the appropriate use of CAMHS and, consistent with other studies, to foster schools' knowledge of suitable referrals and enable them to signpost (Cane & Oland, 2015;Wolpert et al., 2013). Improved awareness, efficiency and early identification, however, can lead to an increase in not just appropriate referrals but the overall number of referrals (Pettitt, 2003), highlighting the importance of increasing knowledge about alternative services in the community, as well as targeting waiting times.
Long waiting times are the most commonly identified barrier to accessing mental health support in the UK and elsewhere (Anderson et al., 2017). Goals-focused systems that aim to improve accessibility through increasing patient flow have been found to reduce waiting times (Clark et al., 2018;Naughton et al., 2015). In the current study, CAMHS staff viewed the i-THRIVE Programme's boundaried, flexible and goal-orientated approach as advantageous and highlighted the benefits of a multi-disciplinary approach to risk in increasing their confidence in ending treatment.

| Perceived impact of programme changes on quality of care
Quality of care is a key driver of policy and service redesign, and enhancing CYP participation in, and satisfaction with, care is associated with improved engagement and positive outcomes (Department of Health and Social Care, 2010). Though person-centred care is advocated across children's services, evidence indicates that service users are rarely involved in decision-making, leading to reduced satisfaction and engagement (Anderson et al., 2017;Coyne et al., 2015).
In the current study, some clinicians perceived their practice to be sufficiently person-centred, but those that adopted shared decision making reported an increased involvement of CYP in the process.
Service users reported improved experience and sense of control as a result, findings consistent with evidence in this area (Joosten et al., 2011).
Multi-agency working promotes a more comprehensive delivery of care and is perceived by young people, professionals and carers as improving treatment experience (Morgan et al., 2019;O'Reilly et al., 2013). In the present study, multi-agency working was viewed positively and, in accordance with existing literature, multi-disciplinary forums were seen to deliver an integrated approach (Cooper et al., 2016).
The benefits of a goal-orientated approach with clear communication about the number of sessions were recognized by CAMHS staff.
Yet some viewed it as compromising effectiveness and one service user was demotivated by limited sessions. Although some studies suggest that focused interventions are inappropriate for complex cases, others indicate that goal-focused systems can improve efficiency without reducing clinical effectiveness or user satisfaction (Fuggle et al., 2016;Naughton et al., 2018;Robotham et al., 2010). Taken together, this highlights the importance of monitoring clinical outcomes and user satisfaction.

| Perceived constraints to programme changes and their impact on accessibility and quality of care
Though a key priority in improving mental health and wellbeing support for CYP, the move towards integrated delivery of care is contingent on cross-sector collaboration (Nooteboom et al., 2020). In the current study, projects focused on multi-agency working were seen by some CAMHS staff to be met by resistance from other sectors and to inhibit programme impact. This finding is consistent with existing literature that highlights the challenges of promoting cross-collaboration, in particular, if there is an absence of shared protocols, joint leadership and when change is perceived to be forced upon staff by outside agencies (Auschra, 2018;Cooper et al., 2016;Henderson et al., 2019). In the present study, the i-THRIVE programme was perceived as a CAMHS-focused transformation and CAMHS staff felt they had limited ability to influence other sectors towards a THRIVE-like approach. Taken together, findings point to the need for a whole system approach to promote impact (Alderwick et al., 2015), whereby the programme is implemented across sectors in the local system.
Lack of resources is a major challenge to policy and service transformation targeting the integration of delivery of care and support across systems (Raus et al., 2020). In the current study, insufficient third sector resources were seen to constrain the use of signposting and timely discharge, the impact of which relied on adequate support in the community to promote accessibility. Geographical variations in third sector mental health provision have been identified in UK reports with some areas indicating falling funding in tandem with rising need (Newbigging et al., 2020). In the current study, similar challenges were highlighted by commissioners, who recognized the priority to assess and build sufficient resource provision across the system to support programme impact. Consistent with other transformation projects, insufficient capacity within CAMHS was seen as a barrier to implementation, scalability and sustainability of programme change, particularly because signposting, outreach and mapping required focused resources (Cane & Oland, 2015).

| Limitations and future research
The study findings need to be considered within limitations. The results provide an early view of the impact of changes introduced during the programme and some implementation projects were still in their early stages during Time 2 interviews. Participant recruitment was not equal across sites or stakeholder groups, and the data may have provided a bias towards larger groups, such as CAMHS staff, or towards sites that were more engaged.
Future studies should place greater emphasis on the views of CYP and providers across the whole system as this will enable a more comprehensive assessment of the impact and its implications for accessibility and quality of care.

| CONCLUSION
As the first study to evaluate the i-THRIVE Programme, the findings provide an early indication of a promising model of care that was perceived by participants to facilitate the accessibility and quality of care.
Accessibility was seen to be promoted through integration of a needsbased service model, flexible re-referral, sharing of information, the use of goal-orientated interventions, and a new approach to managing risk and treatment endings.
Shared decision making was seen to improve the involvement and experience of CYP, while multi-agency working was viewed positively. The impact of goal-focused interventions and upfront discussion of treatment endings was less clear-cut: although they were seen to improve the quality of accessibility by helping clinicians manage expectations and discharge, some felt that effectiveness and engagement could be compromised. This tension highlights the importance of outcome and satisfaction monitoring.
Impact was constrained by resistance to multi-agency working along with insufficient resources in CAMHS and the system to support the implementation process, as well as to deliver the service changes required for effective signposting and discharge, or to scale up and sustain. Findings point to the potential benefit of a whole-system approach that implements the full scope of the i-THRIVE Programme across CAMHS, education, health, the voluntary sector and local authority.