Early intervention efforts in psychiatry emphasise the detection of a disorder at the earliest possible stage, followed by interventions proportional to the stage of illness (1). In psychosis, where the evidence is most established, early intervention has globally transformed the understanding of psychosis and has influenced service reform (2). Whilst research on early intervention for eating disorders (EDs) is still in its infancy, the First Episode Rapid Early Intervention for Eating Disorders (FREED) pathway has shown promising outcomes for patients in the United Kingdom (3–5). The case for early intervention for EDs stems from neurobiological, clinical, and socioeconomic evidence that early intervention may improve outcomes and sustain full recovery from an ED (6, 7).

However, individuals often do not receive treatment for an ED until many years after first experiencing symptoms (8). There are many barriers to accessing treatment for an ED, including important patient factors such as delays in help-seeking (9, 10). There are also widescale service-related restrictions to accessing care in the United Kingdom. In some areas, only low-weight patients are prioritised for treatment, with higher-weight patients being regarded as “not sick enough” to access treatment (11), although recent guidance on early intervention clearly states that this is not acceptable practice (12). Early intervention for EDs has recently been rated as a priority for investment in the United Kingdom, with best practice recommendations advocating for sufficient funding for frontline services and early intervention initiatives (13).

FREED is one approach to reducing service-related delays to specialist ED treatment. FREED is a trans-diagnostic service model and care package for young people aged 16–25 years with a recent onset ED that is 3 years or less in duration (14). FREED encourages rapid access to treatment that is well-coordinated, evidence-based, and adapted to young people's needs (9, 11). It operates as a “service within a service.” Within each site, a FREED Champion is appointed to manage the provision of the service and coordinate a “mini team” of clinicians who deliver treatments adapted to the FREED model and ethos. Early access to evidence-based treatment is encouraged through wait-time targets of 2 weeks for an assessment to take place and 4 weeks for starting treatment concordant with national [National Institute for Health and Care Excellence (NICE)] guidelines.

Active outreach to patients is encouraged via a phone call within 48 h of referral to praise help-seeking and screen eligibility for the service. All services using FREED become part of the FREED network, overseen by a core national team. This national team provides training, implementation support, and ongoing evaluation to the growing FREED network and comprises clinicians and academics from the South London and Maudsley NHS Foundation Trust (SLaM) and King's College London (KCL) (9, 15). Online training is freely available via the FREED website (freedfromed.co.uk), and live training supplements this. Due to growing treatment waiting lists, a more recent addition to FREED involves recording other forms of support/intervention offered to patients that do not come under NICE guidelines for treatment, for example, psychoeducational groups or peer support sessions. The FREED pathway and processes are depicted in Figure 1.

The FREED model has been evaluated in a single-centre pilot study and a multi-site (FREED-Upscaled) study, demonstrating reductions in the duration of untreated eating disorder (DUED), significant improvements in clinical outcomes, and notable cost savings (3, 4, 11, 14). Patients treated through FREED valued timely access to treatment and developmentally tailored adaptations to treatment (16).

Since these earlier evaluations, significant investment has been provided by the National Health Service (NHS) and Academic Health Science Networks (AHSNs) to scale FREED nationally (6). AHSNs were established by NHS England to address national and local healthcare priorities by improving the spread of innovation and improvement, with regional implementation specialists managing the adoption of initiatives within their local area (17). In April 2020, FREED became part of the AHSNs' national adoption and spread programme. This supported its scaling across England, with 15 localised “Early Intervention Eating Disorder” programme leads managing the implementation of FREED in local ED services (5, 18). By the end of the AHSN national programme in March 2023, 46 NHS trusts had adopted or started implementing FREED out of 54 eligible NHS Trusts across England. A summary of the national roll-out is depicted in Figure 2.

Data from the evaluation of this national scaling (the “FREED-4-All” dataset) suggest that findings from the earlier studies were replicated here (5). Most of this scaling took place after the onset of the COVID-19 pandemic when ED services internationally faced challenging conditions including increasing numbers of referrals (19). Even before the onset of the pandemic, UK adult community ED services experienced a growing demand and were underfunded/understaffed (20). There is currently no qualitative exploration of the experiences of those implementing FREED during this time. This is important if the challenges and facilitators associated with implementing FREED are to be understood, to inform future planning for the delivery of authentic early intervention, and to ensure sustainability of the FREED model. Qualitative evidence from those directly involved in the implementation of FREED will provide a rich understanding of these facilitators and challenges, and will also be useful for other services looking to adopt an early intervention service.

The scaling, implementation, and evaluation of FREED to date have been guided by the RE-AIM implementation framework (Reach, Effectiveness/Efficacy, Adoption, Implementation, Maintenance; www.RE-AIM.org) (9, 15). Another implementation framework, which is well-suited for evaluating the next stage of scaling and implementation of FREED, is the Non-adoption, Abandonment, and Challenges to Scale-Up, Spread and Sustainability (NASSS) framework (21). The NASSS framework seeks to map areas of possible complexity in innovation projects by considering seven domains where complexity may arise and how these may lead to the abandonment or non-adoption of healthcare technology/innovations (22). This framework is suited for the next stage of evaluating the scaling and implementation as we look for ways to reduce and manage areas of complexity, maintain/sustain FREED, and improve local implementation fidelity.

The seven domains of the NASSS framework comprise the health condition(s) that FREED addresses (EDs); the technology/service (FREED) and its functionality; the value proposition of FREED to the whole healthcare system; the useability and acceptability of FREED for adopters (NHS ED service staff); organisational (NHS ED services) factors; wider context issues (e.g., policy-related drivers); and an overarching consideration of the emergence and adaptation of FREED over time. Considering ways to reduce innovation complexity within real-world and “rough ground” implementation of innovations is vital for futureproofing an intervention, and this rough-ground implementation is where attention in innovation projects should be directed (23).

The current study therefore seeks to use the NASSS framework to investigate the views and experiences of AHSN Early Intervention Eating Disorder (EIED) programme leads in supporting ED services to implement FREED, as well as to seek their views and opinions on the sustainability of FREED. The study aims are to understand (1) the perceived challenges and facilitators to implementing FREED in ED services across England and adaptations made to the model during implementation and (2) explore views on the sustainability of FREED.

An exploratory, qualitative study design involving individual semi-structured interviews was used for this study. A critical realist perspective was adopted, which assumes the existence of an objective world that can be known through scientific endeavour, but our understanding is gained through subjective human experiences and interpretation (24). Critical realism emphasises a theory-based (not determined) approach to gaining knowledge; it posits that our understanding of reality is mediated through theoretical frameworks and interpretations of researchers (25). As such, we seek to understand the unique and subjective experience of those involved in implementing FREED as guided by the NASSS framework. Ethical approval for this study was granted by King's College London Minimal Risk Ethics (MRA-21/22-26307).

Thirteen participants representing 12 of 15 different AHSN regions and the national AHSN programme were included in the sample. Each participant supported multiple ED services to adopt FREED, and most had experience managing other mental healthcare projects alongside a portfolio of physical health improvement projects. All participants had an active role in facilitating the adoption and implementation of FREED, for example, by managing regional funding, engaging with relevant stakeholders, sourcing implementation materials for teams, and facilitating shared learning opportunities.

Table 1 shows all themes and sub-themes and whether they were facilitators, barriers, or both. Data analysis revealed 28 sub-themes (10 facilitators, 13 barriers, and five considered as both). Two domains were classed as simple, three were classed as complicated, and two were classed as complex (Figure 3). Descriptions of the sub-themes felt to be most prominent are included in this manuscript. The additional sub-themes are described in Supplementary Data S1.

This study aimed to explore the perceived facilitators and challenges associated with implementing FREED in ED services across England and to seek views on the sustainability of the model from the unique perspective of AHSN implementation specialists. The application of the NASSS framework supported the categorisation of implementation facilitators and barriers and areas where complexity is present across each interconnecting domain. Prominent facilitators were a high-value proposition of FREED and engaged, passionate adopters. AHSN participants were largely supportive of early intervention and could see the potential for positive patient outcomes. The complexity analysis suggests the greatest challenges to implementation are illness-related factors and organisational barriers. As the national AHSN programme has ended, complexity may endure and/or increase, and strategies to manage this in the sustainability phase of FREED will need to be monitored. When thinking towards the future, participants expressed concerns about the stability of implementation and networking across services using FREED and stressed the importance of flexibility and creativity in local adaptations of FREED.

Even when coping with circumstances such as the COVID-19 pandemic and an increase in patient referrals, AHSN leads felt that there was a will among clinicians in ED services to adopt FREED. Further, FREED Champions worked hard to implement FREED and increase enthusiasm in their wider teams. How adopters perceive a healthcare innovation and their beliefs and values about patient care are essential to innovation engagement and adherence (22, 32). Innovations that go against the intended users' values or that they cannot see advantages in are less likely to be supported (21, 33). The role of FREED Champions appeared to be an integral part of the perceived success of FREED, and AHSN participants discussed the importance of sustaining the funding for these posts to ensure that early intervention is continually championed across ED services. Ensuring protected spaces for clinicians to share learning and best practices was also important to participants. As some services are still early in their implementation journey, ensuring they have these forums to adopt FREED with optimal fidelity will be important. Careful planning of how the FREED network will operate after the AHSN EIED programme ends is therefore required.

The perceived value of an innovation is arguably one of the most important enabling factors for scaling, spreading, and sustaining a healthcare innovation (34). AHSN leads believed that FREED could demonstrate high value for patients and that FREED's evidence base was sufficient and compelling. For example, FREED has demonstrated reductions in day/inpatient admissions, which translates to considerable cost savings for the NHS (9). Therefore, we considered the value proposition for FREED to be simple. However, participants desired to see additional research on treatment accessibility for underrepresented groups and further evaluation of the national scaling of FREED. Ultimately, more robust evidence, such as that provided via randomised controlled trials evaluating FREED against credible control interventions, is needed. Early intervention initiatives for those over 25 years are lacking (6), and whilst ethnic minority and LGBTQIA+ people may be more likely to experience EDs (35–37), they may be underrepresented in ED services (38). Research addressing some of these gaps is currently in progress (www.EDIFYresearch.co.uk) (39).

The condition domain was determined to be complex (unpredictable, multiple interacting components) when taken together with what we know about EDs. EDs are highly varied and comorbid with other psychiatric and physical health conditions and carry long-term morbidity and mortality burdens for the individual (40, 41). Furthermore, EDs are often stigmatised, under-researched (7), and disproportionately affect certain groups of people, as has been discussed. Participants frequently expressed concerns about services' abilities to cope with an increase in the number and acuity of ED referrals observed since the onset of the COVID-19 pandemic. Our own data show that FREED services experienced a ∼1.4-fold increase in referrals after the onset of the pandemic, which calls for careful continued monitoring of the long-term impacts of the pandemic (19). Participants also considered issues around help-seeking in EDs as a potential barrier to early intervention. The FREED team has developed a mobile application (FREED-M), which is currently being evaluated in a feasibility trial assessing whether FREED-M can improve help-seeking in young people with an ED (42). Other ways to manage complexities related to EDs must be considered.

One of the main factors presenting as a facilitator or barrier under the technology theme was related to funding options; participants desired more investment opportunities for ED services and described a lack of continued funding for FREED as a source of concern. When discussing sustainability and “futureproofing” FREED, most participants highlighted the need for continued investment, which aligns with recommendations to invest in ED services and early intervention initiatives to cope with the increased demand (6, 20). Complexity emerged as the implementation of FREED requires clinicians to deliver NICE-recommended treatments, and participants stated that more training opportunities to upskill clinicians to deliver these are needed. There was also a desire to see training on managing age-related transitions of care in a FREED context. Such a training package is currently being developed.

The organisation was rated as complex, where barriers included variable leadership/team support for FREED. Sufficient clinical leadership and support are central to implementation projects and innovation in healthcare, and leaders are essential in cultivating an environment where staff feel confident to take appropriate risks and engage in innovation (43). However, adult ED services have historically been under-resourced, and the NHS has an ethos of gatekeeping to optimise scarce resources (11, 20), which, as described by some participants, has led to anxiety about change and risk aversion. The AHSNs were instrumental in helping address risk aversion by ensuring sites could implement gradually and by encouraging teams to decide launch dates. The support of the AHSN has therefore undoubtedly facilitated and encouraged services to launch FREED. However, other crucial challenges across the NHS include significant staff recruitment difficulties and unfilled vacancies (44). Participants identified struggles to recruit and keep FREED Champions, which could pose a threat to the sustainability of FREED. Also, as FREED currently operates on a “train the trainer” approach, a high turnover of staff could lead to a lack of consistent messaging about the principles of the FREED pathway. These accounts of the impact of recruitment challenges highlight the need for assessments of implementation fidelity to ensure consistent messaging and practice. Adopting a team-based approach to FREED may also help guard against deteriorating fidelity as one person does not have to solely uphold the pathway and values.

Important issues concerning the sustainability of FREED over time included, first, implementation fragility, for example, a “partial” adoption of FREED, or “pausing” of FREED. Participants highlighted a need for flexibility when implementing the pathway, allowing at least to start with a partial adoption (e.g., within part of a service's catchment area or covering only certain diagnostic groups). Greenhalgh and Papoutsi (43) stress that individuals create adaptive solutions according to their local contexts, creating an inevitable gap between the evidence-based ideal and the real-world context, but that these workarounds and amendments should be studied carefully. These workarounds appeared to help stretched services adopt FREED at a slower pace. Nonetheless, as FREED has now been adopted across an increasing number of ED services and organisations, it is important to investigate whether FREED is being delivered as intended, i.e., as a transdiagnostic service model covering all ED diagnoses and severities. Assessments of implementation fidelity should allow for some flexibility in local adaptations, and it appears these have been an important factor for the realistic adoption of FREED, but it should also be determined how much and which adaptations are acceptable and which are not (45). The latter is important to avoid a delivery of “watered down” early intervention and FREED “in name only.” For example, if FREED is only being rolled out for certain diagnoses (i.e., anorexia nervosa and other low-weight presentations), this will disadvantage young people with bulimic presentations at higher weight, a group that is often underserved and experiences multiple systemic and health disadvantages (46).

This study revealed several sources of possible complexity relevant to implementing the FREED pathway across ED services in England. However, central to most of the interviews were descriptions of services, clinicians, and stakeholders being enthusiastic about early intervention and FREED. In recent years, clinical, research, and campaign groups have strongly advocated for investment in early intervention for EDs, setting a strong policy consensus for early intervention (6, 12, 13, 47). The experiences of FREED patients and other individuals with lived experience also confirm that early access to treatment is a priority (6, 16). Despite operational complexities, we believe that this momentum and support will help drive FREED forward. Collaborating with the AHSNs has been of great benefit for scaling FREED across the NHS, and these organisations have been identified as a key component to creating an innovative work culture in the NHS (48). Work must focus on sustaining investment, improving fidelity, and futureproofing FREED beyond the national programme.