Redefining prevention and early intervention to include longstanding and severe eating disorders

Abstract

There is a lack of agreement over the usefulness and coherence of the term 'severe and enduring’ to denote a sub-category of illness amongst patients with eating disorders. Nonetheless, a proportion of patients experience longstanding and severe illnesses, often as a result of a lack of treatment altogether, or poor responses to currently available treatments as they are delivered within existing specialist services. This paper is written from the perspective of lived experience and published research. It calls for a re-imagination of existing principles within the field of eating disorders which have underpinned recent demonstrable improvements in clinical practice and outcomes for patients, namely prevention and early intervention. These concepts can also be applied in the development of more effective and inclusive care pathways for patients with longstanding and severe eating disorders. The author reminds us of the importance of language in the description and categorisation of illness, and the essential requirement that those with lived/living experience of illness share equitable decision-making power in defining their condition and designing the healthcare provision they receive.

Eating disorders: describing and defining diverse experiences

It is widely acknowledged that understandings of eating disorders (EDs) have long been dominated by particular, restrictive stereotypes, which can be experienced as excluding by patients who do not fit them [1,2,3,4,5]. In reality, the range of ED diagnoses represents widely heterogeneous conditions, experienced by diverse groups of people, in ways that can be fluid over time in relation to a wide range of contextual factors [6]. Appreciating that EDs are also rooted in biological processes, and often implicate multiple bodily systems [7,8,9], it becomes clear that holding an understanding of the array of presentations, experiences, courses of illness, and possible responses to treatment that patients with eating disorders may have is a mammoth task.

Nosology, then, has the challenge of determining a shared language for describing and categorising illness in a way that has utility—providing access to understanding and care for what are often highly individually subjective and varied experiences [10]. This process has an inherent tension (or dialectic) which is also intrinsic to the process of creating policies, services, and treatment pathways for EDs—namely between the requirement to provide general (or nomothetic) systems, structures and taxonomy, and the need to allow for individual (or idiographic) features to exist and not be erased or excluded therein. The harder the edges of a definition for any diagnosis, the more the systems we design to facilitate treatment may suit and be effective for patients who fit the definition (or conversely, the more exclusive or ineffective they can be for those who don’t). Yet anyone who works with people as subjects, not objects, will recognise the messiness of human experience, and that few people with mental illness ever really fit neatly into a diagnostic box.

This may be especially the case in EDs, where existing understandings are based on a paucity of well-funded and transparent research [11, 12] which has over-privileged some demographics and diagnoses and neglected others (such as men [13], gender and sexual minorities [14], and those from non-white backgrounds) [15, 16]. The accuracy and efficacy of current diagnostic categories are therefore questionable and will benefit from a more representative evidence base which corresponds to the range of presentations and demographics EDs encompass [17]. This may in turn lead to a refining of existing diagnoses, or the creation of new ones, as seen over time with the introduction of binge eating disorder (BED) [18] and avoidant restrictive food intake disorder (ARFID) [19] to diagnostic manuals. Research with previously under-studied groups gives rise to suggested new categories such as muscularity-oriented disordered eating (MODE) [20], orthorexia nervosa [21], and others.

“SEED”—a well-meaning concept with constraints built in

Within healthcare, there is a well-recognised need to provide care pathways for cohorts of patients who may not yet have had adequate research to justify a diagnostic category/sub-category for their condition, but who nevertheless are perceived to require some kind of specialised treatment not currently made explicit by the evidence base. In the case of EDs, one such group is those patients labelled as having a ‘severe and enduring eating disorder’ (SEED) based on factors including greater illness duration and severity, decreased motivation to change, maintenance of low body mass index (BMI), and the persistence of greater psychological and physiological risks [22].

As a patient who has been described or labelled as ‘SEED’, I think it is important to understand the creation of this concept as arising from a recognition of unmet needs, and a desire on the part of clinicians for better outcomes for patients. It is at least more helpful for me as a patient to engage with my care when I take the view that people are generally doing the best they can for others, with the resources available to them, and that clinicians work in the field of eating disorders to alleviate suffering and help patients recover.

It is, however, difficult to separate the kind of care professionals desire to provide from the systems within which that care is delivered. The care required for a patient with a longstanding and severe illness can be resource-intensive, and the constraints placed on service providers make highly individualised and patient-centred care a challenge. Attempts to streamline care pathways with the invention of subcategories of patients risk being more about the convenience of services, and resolving a problem faced by professionals, than the clinical needs and preferences of patients themselves [23].

I come across an illustration of this persistently in my work as an expert by experience. In multiple arenas, clinicians and policymakers insist on using the term ‘SEED’ in discussions where I am present, even after requesting that we discuss the terminology we can agree upon when working together and share how I do not want to participate in discussions that use the term ‘SEED’ uncritically. This represents an imbalance of who gets to decide, where even the terms used can be imposed upon those with less power to define experiences which are, importantly, their own.

I am not alone in having concerns about the coherence and impact of the term ‘SEED’. Research has highlighted how other patients may also perceive the label as negatively-focused, lacking in hope, and interpreted by others in a way that results in less support from healthcare, not more [24]. The term has conceptual limitations in its foundations, highlighted by a recent narrative review comparing attempts to define ‘SEED’ from 11 studies, of which none provided a definition for severity [25]. ‘SEED’ does not meet the evidence threshold for a diagnosis or sub-diagnosis within diagnostic manuals [17], and its utility is called further into question by the lack of research into treatment targets and adaptations to existing treatments that patients with longstanding and severe illness may need [26]. More worryingly still, recent research documenting 60 cases of assisted dying in ED and the clinical rationales used to justify it found the use of ‘SEED’ as though it were a diagnosis—or as the authors described, a ‘pseudo-diagnosis’ [27]. As one psychiatrist put it to me during a recent assessment: “there is no such thing as ‘SEED’. They are not severe and enduring eating disorders, they are unresolved eating disorders. ‘SEED’ does not exist.”

Another potentially harmful consequence of the use of the term ‘SEED’ is rooted in semantic qualities of the word ‘endure’, which is etymologically derived from its Latin root, ‘to harden.’ [28] By evoking notions of ‘hardening’, as an inherent characteristic, ‘SEED’ portrays EDs as progressive or staged illnesses by nature. This is not yet substantiated by evidence and is contradicted by research from real-life clinical settings which shows that duration and severity of illness are not good predictors of outcomes of treatment [29].

The semantic connotations I have outlined reflect my own experience of being told I had a ‘diagnosis of SEED’, which bestowed me a sense of futility in the face of the inevitable decline I was told to by clinicians to expect if I didn’t change my ways. Yet because I was termed ‘SEED’, assistance to change was not offered, resulting in me feeling stigmatised and to blame via my marginalisation from treatment and my assumption that it must be my own fault if it is assumed to be within my power to change things without help. The only support I was offered was not change-focussed and focussed instead on harm reduction, which may best serve patients as only a part of broader care pathways ultimately oriented towards recovery [30].

Avoiding dichotomies in understanding illness, treatment, and recovery

The potential creation of a false binary between the quality of life gained in recovery, and gaining quality of life whilst maintaining illness, is an example of the many dialectics at play in considering the subject of longstanding eating disorders. Polarities include those between (1) longstanding and severe illness and early intervention; (2) the prevention of illness, and intractable illness potentially involving inevitable or ‘terminal’ decline [29]; (3) nomothetic taxonomy, conceptualisations of illness, and care pathways, and idiographic presentations, situations and motivations; and so on. The central dialectic described in dialectical behaviour therapy of acceptance and change [31] is also perhaps a central binary seen in clinical practice—between being wholly accepting of illness, inevitability and even terminality on the one hand, and full recovery on the other.

It is naturally unhelpful to characterise illness and treatment in such binary terms, but the risk of doing so is very much present when categorising patients and dividing treatment approaches based on conceptualisations of different presentations or ‘stages’ of illness. As such, having an awareness of this dichotic tendency and its possible role in clinical practice can be important, especially when the characteristics and attitudes of clinicians are known to shape patients’ conceptualisations of their illnesses, treatment, and their attitudes towards recovery (for instance in establishing a shared nihilism, hope, tolerance of uncertainty and complexity, etc.) [23, 32, 33].

Moving away from unhelpful extremes of ‘either/or’ towards a middle ground where ‘both/and’ can coexist is achieved through thinking more dialectically in order to integrate rather than segregate such ideas. For instance, it is not the case that patients either recover or inevitably progress in illness towards death. It is not the case that recovery is entirely clinically measured or entirely based on individual definitions. It is not the case that treatment is entirely change-focussed and absent of acceptance of things that cannot be changed, at least in that particular moment. It is not the case that diagnostic categories and descriptions of illness need to be entirely perfect or are therefore entirely inadequate—and so on. By integrating opposing features, a middle ground is created that makes available a greater range of ideas, resources, knowledge, and approaches upon which to draw in the understanding and treatment of patients with longstanding and severe illnesses. We turn now to two of these ideas—prevention and early intervention.

Prevention and early intervention in eating disorders: current theory and practice

The role of early intervention and prevention of illness is widely recognised as important in the field of EDs with a recent review of 130 early intervention and prevention initiatives [34] highlighting the effectiveness of programs targeted at reducing risk factors such as body dissatisfaction, particularly for adolescents and young adults. This broad area of research and public health practice also includes efforts to early detect the development of EDs and to smooth access to treatment, for instance by increasing help-seeking behaviours using population-targeted co-designed materials [35]. Underpinning these measures is a drive to reduce the duration of untreated eating disorder (DUED) [36]. Reducing DUEDs not only decreases the length of suffering for individuals but is suggested to increase the likelihood of a positive treatment response compared to those with a longer duration of illness before treatment [37].

A key example of early intervention in the UK is the development of the First Episode and Rapid Early Intervention for Eating Disorder (FREED) pathway for 16–25-year-olds with an illness duration of under 3 years. [38, 39] FREED aims to reduce the length of time between treatment-seeking and receipt of specialist evidence-based treatment—with most patients starting treatment within 4 weeks of referral. FREED also adapts treatment planning to the specific developmental needs of emerging adults, f taking a flexible approach to appointment scheduling, increasing involvement of family/friends, and focusing on managing transitions (such as to university) [40]. This is a laudable example of utilising knowledge about at-risk groups and mobilising resources towards providing rapid and tailored responses early in the course of illness. Most importantly, FREED has proven effective, reducing the need for more intensive treatment and producing better recovery outcomes after 24 months than treatment as usual [41].

There is a need to further develop the FREED model and other examples not detailed here of initiatives to prevent, detect and intervene early in the treatment of EDs. For instance, FREED only includes patients with anorexia nervosa (AN), and the majority of patients do not have AN. FREED also raises questions about where else effective intervention may prove rewarding both in terms of the prevention of suffering and reducing costs downstream for healthcare providers.

The need to widen the net

The lessons of FREED can be applied for equal benefit for those of any age, with any duration of untreated or partially treated illness, however severe that illness may be. This requires a re-imagination of the principles of prevention, detection, and early intervention, to include other groups traditionally seen as outside of the narrow window in which these concepts are thought to apply. Whilst a recent meta-analysis of 192 epidemiological studies showed 82.4% of feeding and eating disorders were present by age 25 worldwide [42], studies into the trajectories of illness have been found to have limited and mixed quality [43], and the possibility for relapse later across the age range is well established [44].

Despite some indication that increased DUED is associated with a more refractory illness trajectory [35], there is evidence to show that both length and severity of illness are not adequate predictors of recovery [29]. Recent data contradicts the idea of there being a time-limited window to instigate recovery, showing that the longer duration of illness increases the likelihood of recovery. Solmi et al.’s recent comprehensive analysis [45] showed recovery rates across EDs of 42% at < 2 years, 43% at 2 to < 4 years, 54% at 4 to < 6 years, 59% at 6 to < 8 years, 64% at 8 to < 10 years, and 67% at ≥ 10 years. This suggestion of increased likelihood of remission over time may not be universally established or applicable across ED diagnoses, yet it offers a very different narrative from the nihilistic ones I was told as a teenage patient. Further cause for optimism comes from examples in research of recovery being made more likely even for the most unwell patients by the development of integrated models of adequately intensive and individualised treatment [46]. The goal of recovery therefore needs to be at the forefront of designing care pathways for all presentations of illness [27].

Reflections from lived experience

Whilst those early in the course of illness have benefitted from focussed interventions such as FREED, it could be argued that the need to prevent the prolongation or progression of illness could be even more pertinent amongst those with longstanding and severe illness like me. Patients like me suffer high personal costs and require a greater intensity of care from resource-poor services [47]. However, I do not think that positioning the healthcare needs and suffering of one group of patients against those of another is a helpful way of looking at this subject.

Yes, it has been painful as a patient who has often been unable to access any treatment at all to see new pathways spring up for patient groups who I consider to be far less unwell than myself. This can cause natural feelings of resentment, which likely are a response to a deeper emotional pain arising from feeling overlooked or neglected—like your suffering matters less, or even by extension, you are worth-less. However, I have learnt that a more mature, compassionate and effective attitude is to recognise that we do not have the resources and energy to help all groups all at once, and that we have to start somewhere.

A dialectical attitude has helped me in overcoming a splitting mindset: bringing an appreciation that it is both ‘good’ that more people are being more effectively helped, and at the same time ‘bad’—or more accurately, painful—to feel cast outside of this group when you wish you could benefit from similar help. Initiatives like FREED shouldn’t be shouted down simply because they rightly can’t include everyone as they are, or else we have a race to the bottom where nobody receives care because not everyone can. Instead, building upon and expanding the work that has already been done could help share the principles of prevention and early intervention for the benefit of a greater number of patients including myself.

My own experience as a patient with a longstanding and severe ED has given me an appreciation of how the opportunity to prevent suffering and to intervene early is present in every interaction with healthcare, irrespective of how long or severely you have been unwell. For example, when presenting to the GP/primary care in a new area, I have always had to wait a long time before being seen by the specialist ED service despite meeting a description of ‘severe’ illness (- waiting a minimum of 4 months and a maximum of 4 years). I cannot comprehend how these instances, when I have been severely unwell with significant medical instability, could represent less of an opportunity for the prevention of suffering than preventing someone from developing illness in the first place.

I’ve felt ineligible for swift responses to my illness—except in certain medical emergencies, of which there have been many—as though the concept of early intervention does not apply to people like me on the grounds that my ED was diagnosed many years ago. I have struggled to understand how early intervention is not viewed as something that isn’t just about early intervention in the course of illness, but also about early intervention in the given moment of presentation to any healthcare setting. A useful concept I learnt in therapy described how the best time to plant a tree may have been 100 years ago, but the second-best time is today. We deal in the here and now, and someone’s past treatment history and illness trajectory resulting in the denial or withdrawal of care feels punitive and is a widespread theme in my own experiences.

In reserving concepts like prevention and early intervention for certain groups of patients, others, in this case those with longstanding and severe illness, are excluded with the result being a reinforced sense of nihilism and un-treatability [23]. The prioritisation of people early in the course of illness over patients like myself has left me feeling as though it’s been decided, somewhere, that there is less point in providing me with timely treatment. This can even feel like being penalised for being unwell, or like the fact that you have demonstrated that you can just-about-survive without any support or treatment proves that you don’t really need any. This is ultimately stigmatising, in my experience.

The seeming inapplicability of the ideas of prevention and early intervention I have described has not however been the case in my experience of other areas of healthcare, such as dentistry, where the contrast has been stark. On presenting to the dentist, they have immediately taken a preventative approach to stop further wear to my teeth and have initiated restorative interventions at the earliest opportunity. Instead of a care pathway of managed decline, or strategies to live alongside severe dental pain and dysfunction with a better quality of life, preventing poor oral health and recovering from my individually unique presenting problems has been seen as just as appropriate in approaching my care as it is for anyone else.

Of course, healthcare providers need adequate resources to provide this kind of care, and a lot of the problems I have described in my experiences of ED treatment have been the product of a lack of resources—not a wilful withholding of care, or a punishing, nihilistic approach by design. It has just felt this way, at times, and I can understand this with compassion and without seeking to blame. The competition for scarce resources means that only some groups can win at any given time. I have attempted to argue that the answer to providing healthcare under such resource constraints is not to give up on achieving the universal principles and values of care for all patients. Even if we do need to segregate and prioritise certain groups of patients in order to have effective pathways through treatment for different groups, this doesn’t necessitate a siloing of healthcare principles for some groups and not others. We need the resources to design systems that can afford to offer such dignity to all patients, and to consider how to achieve this together.

Recommendations for integrating prevention and early intervention as universal principles of care

Resourcing a more compassionate approach

The problems articulated in this paper have to be considered in the context of the resource-starved ecosystem in which EDs are understood and responded to. A recent survey demonstrated services in England are 84% under-resourced in relation to population need, even before the COVID-19 pandemic [48]. Resources for research are also sparse, in particular in relation to physical health conditions [26]. Outside of specialist services, human resources are inadequately equipped to identify and respond effectively to EDs, with an acute lack of training in undergraduate medical education, for example [49]. The resulting attitudes towards EDs as a niche, specialised area [50] are perhaps in part responsible for the siloing of care experienced by patients in navigating care for the often multifaceted components of their condition [51].

An increase in resources in the broadest sense is an essential first step towards being able to more effectively prevent adverse impacts of illnesses of any duration and severity. Services that can afford to intervene in a timely and responsive manner when any patient with an ED presents for help will afford such patients with greater opportunities for recovery and the alleviation of suffering. Within an adequately resourced system, there is no need for justifications for providing treatment in a timelier manner for some than for others. The resources required include but are not limited to: (1) human resources via adequate staffing levels; (2) knowledge resource via sufficient high-quality training; (3) a robust evidence base via research across all relevant disciplines; and (4) resources to fund the meaningful inclusion of patients and carers/supporters in the prioritisation, design, implementation, and evaluation of improvements in healthcare.

Co-creating shared language and principles of inclusion

The examples given in this paper illustrate the vital role of language in shaping the understandings we have about health and illness, including how this may inadvertently result in experiences of exclusion and marginalisation from care. This is a perspective centralised in medical anthropology and narrative approaches which attend to the role of conceptualisations and stories in shaping the possibilities conceived of by (and made available to) patients [52, 53].

If we are to avoid perpetuating the experiences discussed here, more expansive ways of talking about prevention and early intervention in EDs are required than those which reinforce current narrow conceptualisations. Some solutions may include:

• To avoid conflation with early intervention as conventionally understood (i.e. in terms of illness duration), it may be preferable to adopt a language of ‘timely’, ‘prompt’, or ‘rapid’ intervention.

• In order to be explicitly inclusive of all ED patients and communicate the universally desirable nature of the principle of timeliness, talking about ‘responsive’ intervention may help reinforce the need for treatment to accommodate rather than exclude those with less normative presentations.

• Similarly, reiterating how ‘equitable’ intervention can help prevent or ameliorate health inequalities is centrally important.

These terms could benefit from being examined via further dialogue across stakeholder perspectives (i.e. researchers, clinicians, and those with lived experience), with equal voice given to each in the process of deciding.

Operationalising an enhanced understanding

Work done to date to provide early intervention in ED (such as FREED) needs to be applauded and expanded, with an urgent next step being to include patients with long-standing and severe EDs in research efforts that translate rapidly to implementable clinical models. Making use of existing structures (e.g. replicating or expanding the FREED model) may be the most feasible way to start, alongside an increased quantity of research specifically focussing on longstanding and severe presentations. This paper has raised concerns about an imbalance in research efforts, where attempts to create or concretise new categories of illness such as’SEED’ have been a primary focus alongside identifying the patient characteristics explaining unresolved and unresponsive illness. This may have been at the expense of attending to systemic factors that maintain EDs and features of treatment delivery itself that result not in patients being ultimately ‘untreatable’, but show a system that is ‘unable to treat’ [23].

A recent global study commissioned by Lancet Psychiatry (“Gone Too Soon”) [54] which aimed to map the range of factors leading to premature mortality in mental illness and identify actions to prevent this may provide a useful starting point for this research, as it addresses issues of complex aetiology and treatment using a socioecological model. This confers the benefit of incorporating factors pertinent to the timely, responsive and equitable treatment of longstanding EDs such as addressing the role and impact of iatrogenic harm, stigmatisation, and co-occurring psychiatric and physical conditions as well as the wider food environment, for example.

Another model that may be useful to centralise is a staged model of prevention, as used widely in other areas of healthcare and public health [55, 56]. This includes four stages of prevention: (1) primordial prevention, relating to actions that impact systems and environments in which risk factors arise; (2) primary prevention, relating to detecting and responding to risk factors for future illness that may develop; (3) secondary prevention, relating to screening for early illness and initiating treatment; and (4) tertiary prevention, relating to treating patients with established illness and preventing the progression of illness in treatment.

Currently, specialist ED services overwhelmingly operate in the realm of tertiary prevention. On the basis of the research and personal experiences highlighted in this paper, it is evident that current interventions relating to prevention and early intervention are excessively narrow, based on constrained conceptualisations of who these may apply to and what they may involve. We need to create a far broader framework for understanding prevention and early intervention in EDs, from which targeted interventions can be implemented that range from the environments in which the risk for EDs may arise, to interventions to prevent illness progression amongst those with longstanding and severe illness.

An example map of possible interventions for each stage of prevention is proposed in Fig. 1, below.

Fig. 1

A proposed map identifying possible interventions relevant for multiple levels of prevention in relation to EDs

Coproducing healthcare with lived and living experience

Developing a more comprehensive and evidence-based framework for prevention and early intervention in EDs needs to be an endeavour of co-creation, drawing upon expertise from all relevant stakeholders. Similarly, to the “Gone Too Soon” project, this mapping needs to include experts in healthcare policy and practice, such as public health professionals, health economists, scientific researchers, specialist clinicians, and more, in order to ascertain the relevant factors ranging from the broad/ecological to the specific/individual.

Vitally, this knowledge must be co-produced with a diverse range of people with lived and living experience, both directly as patients and indirectly as caregivers. Inclusion in the process of knowledge creation can be a positive experience for those with expertise by experience when it is well-resourced, reciprocal, and values the subjective phenomenology of experiences of illness as providing important insights into that illness [57].

In the case of EDs, there are traditionally held notions that people with EDs are unreliable narrators of their own experience [58], which need to be overcome if the power to make decisions about the knowledge we have—and what to do with it—is to be shared on an equitable and just basis with people with lived/living experience [59, 60]. Whilst examples of co-production exist in the field, these lack a standardised approach [61]. As such, a commitment to sharing best practice is needed, for instance through developing agreed guidelines for how to effectively embed coproduction in decision-making activities that range from knowledge-generation about the nature of EDs to the interventions needed for their prevention and treatment.

In general terms, Richmond et al. (2023) [57] identified six characteristics that can act as important ingredients for positive involvement in mental health research, based on their own experiences, namely: reframing painful memories; recognising value; practising reciprocity; bridging gaps; countering stigma; and challenging established narratives. A good starting point for both research and clinical teams is to coproduce agreed principles for working together collaboratively. Specific to EDs, the Australian National Eating Disorders Collaboration’ National Eating Disorders Strategy 2023–2033 centralises the role of lived experience guidance and leadership—from policy development to service delivery and evaluation [62]. This serves as inspiration for those aiming to collaborate with people with lived/living experience in the development of care pathways and treatment options for all patients, irrespective of their presentation. Measuring the success of these will depend on thorough, routine data collection, and the evaluation of outcomes which are then communicated with transparency to inform quality improvement and the dissemination of good practice.

Rather than siloing patient groups and the approaches they are offered, a conceptual shift is needed towards integrating the principles of prevention and early intervention as universal qualities of the care we wish all patients to benefit from. In working together to achieve this, much suffering can be prevented, and more timely treatment will be more likely to be accessed by all who need it.