Abstract
Objective:
Treatments in anorexia nervosa (AN) have not been wholly effective and, accordingly, practices need to be reviewed. The recovery model is an approach to treatment that has become a guiding principle for mental health policy worldwide that might provide promise for AN treatment. The model has received much attention in recent years; however, there is a dearth of literature exploring how useful this model is for AN. The aim of the current article was to consider the relevance of this model in AN.
Methods:
This article provides a summary of the recovery model and reviews the literature to establish whether it is compatible with AN. The possible utility of the approach in AN is explored and suggestions are made as to how the model might be implemented in treatment.
Results:
Qualitative studies examining the patient’s perspective of AN support the recovery model. Many evidenced-based treatments currently used in AN have elements that are consistent with a recovery model approach. Treatments that are most consistent with recovery approaches have been effective for those with chronic AN.
Conclusion:
It is proposed that the model might offer a way in which to add to current practice and might have particular relevance for those with chronic AN. Future research is required to better understand how the model can best be utilised in AN.
Introduction
Anorexia nervosa (AN) is a severe and chronic mental illness associated with a vast array of physical complications as well as harmful effects on psychological and social functioning (Berkman et al., 2007). It has the highest mortality rate of any mental illness (Beumont and Touyz, 2003). Among surviving patients, less than half of those diagnosed with AN do well over time (46–50%), approximately 30% improve but continue to experience symptoms, and approximately 20% remain chronically ill (Steinhausen, 2002).
AN is further complicated by the lack of definitively successful treatments and prevention strategies. Despite many models and proposed approaches there remains no conclusively effective treatment for adults (Hay et al., 2009). Additionally, 74 of the 75 AN treatments recommended by the UK’s National Institute for Health and Care and Excellence (NICE) have only received a grade of ‘C’. One the largest randomised controlled trials conducted in AN, recently published in
While the difficulties in conducting research in the area of AN are well known, reviews of the evidence-base for AN treatment paint a clear picture that more needs to be done to better support patients and their families. Not only is it important to determine more clearly the effectiveness of currently available treatments, but established treatment approaches also need to be improved. Many questions still remain concerning the optimum treatment model, the correct dosages of treatments, and the illness stage at which treatments should be administered. At the same time, questions remain, as to whether the correct approach to treatment is generally being taken. While treatments remain not wholly effective, it is critical, not only to continue to improve recognised treatments, but also to develop novel and innovative therapies and question established treatment approaches in AN. We suggest looking to our colleagues in other fields of psychiatry and psychology to see if we can learn from other ways of managing chronic, debilitating, and so-called ‘untreatable’ mental illnesses.
In the areas of psychiatry and psychology, two key paradigms have emerged as a means to conceptualise treatment and recovery from mental illness: the medical model and the recovery model (Mountain and Shah, 2008; Roberts and Wolfson, 2004). Historically, the medical model, emerging from professional-led research and practice, is the primary way in which recovery has been conceptualised (Andresen et al., 2011). This model positions recovery as an objective ‘cure’, a condition defined by the absence of symptoms and a return to normal, pre-morbid functioning (Roberts and Wolfson, 2004). This is the model under which most research in AN has been conducted, with dominant thinking in AN treatment tending to support symptom-centric treatments. Accordingly, the AN treatment research to date has similarities: traditional clinician led therapy conducted from a medical model perspective where ‘good’ outcome is conceptualised as symptom abatement. However, in other areas of psychiatry the recovery model has emerged as an alternative way to conceptualise treatment and outcome. The recovery model, which emerged from the consumer/survivor movement, emphasises the personal experience of recovery, involving hope, connection, and establishing a personally fulfilling life. This model stands in contrast to the medical model and traditional understanding of good outcome, which is conceptualised as symptom reduction alone (Anthony, 1993; Jacobson and Greenley, 2001).
There is a body of literature examining the recovery model and its significance for people with mental illness, their treatment, and the development of professional practice and policy (e.g. Andresen et al., 2011; Anthony, 1993; Jacobson and Greenley, 2001). However, there is a relative absence of research explicating how this model applies to AN. The aim of the current paper was to consider the relevance of the recovery model in AN and how recovery-oriented practice might be implemented in the field. Following an outline of the recovery model this article will address two questions: (i) Is the model compatible with the conceptualisation of AN recovery and AN treatment? and (ii) How might a recovery approach be implemented in the treatment of AN?
The recovery model
The recovery model began in the 1930s as a grass-roots consumer-advocacy movement but did not gain prominence until the 1980s and 1990s. It emerged during this time as a response to the medical model’s historical pessimism regarding outcome in mental illness, particularly schizophrenia (Andresen et al., 2011). Historically, and according to a traditional medical model approach, people with schizophrenia were perceived as having a life-long sentence with little hope of symptom remission and limited future prospects in terms of overall functioning (Andresen et al., 2003); that is, they were not expected to recover. During the 1980s, evidence emerged that suggested people with schizophrenia were living full and meaningful lives despite the presence of symptoms of mental illness (Andresen et al., 2011). In particular, personal accounts of recovery, that emphasised hope and empowerment, began to be published by sufferers themselves (e.g. Deegan, 1988). Alongside these accounts of a more positive outlook for schizophrenia were a number of quantitative outcome studies that suggested recovery from schizophrenia was more prevalent than had previously been recognised (e.g. Harding et al., 1987). The increasing evidence that people with schizophrenia could live a meaningful life led to a new conceptualisation of recovery and the emergence of the recovery movement. These developments led to a reconsideration of the notion of ‘outcome’ in psychiatry and psychology and how ‘recovery’ should be defined.
The most commonly accepted definition of recovery within this paradigm was developed by Anthony (1993: 527) who describes recovery as:
A deeply personal, unique process of changing one’s attitudes, values, feelings, goals, skills, and/or roles. It is a way of living a satisfying, hopeful, and contributing life even with limitations caused by illness. Recovery involves the development of new meaning and purpose in one’s life as one grows beyond the catastrophic effects of mental illness.
According to this definition a person may recover without necessarily experiencing symptom remission or a return to pre-morbid functioning (Anthony, 1993; Davidson et al., 2005). The main tenets of the recovery model are: hope, spirituality, personal responsibility and control, empowerment, connection, purpose, self-identity, symptom management, and overcoming stigma (Schrank and Slade, 2007). Quality of life, general functioning, employment, and access to housing are also emphasised. In this way, the model is more holistic, encompassing the restoration of the whole person beyond symptom reduction. There is an understanding that consumers need more than just symptom relief. The model also takes a different position regarding notions of expertise. While medical knowledge is respected, those who have experienced mental illness are considered experts on their lives and experiences and, accordingly, peer support that draws from the lived experience of others is valued.
Internationally, the recovery model has emerged as a principle aim of mental health services in the last two decades. The recovery approach has been integrated into public mental health policy in many English-speaking countries, including Australia, Canada, Ireland, Israel, the United Kingdom, New Zealand, and the United States (Andresen et al., 2011; Leamy et al., 2011). Both the American Psychiatric Association and American Psychological Association have made significant moves to implement recovery-oriented practice for mental illnesses (American Psychiatric Association and American Association of Community Psychiatrists, 2011).
Many elements of the model are long-accepted, fundamental principles of therapy, such as respect for consumers, the emphasis of hope, and the need for partnerships between mental health professionals, consumers, and families (Reisner, 2005). The model is not an empirical treatment in and of itself but, rather, an overarching philosophy of treatment, conceptually close to a common factors model (Reisner, 2005). Nonetheless, elements of the recovery model have received empirical support (Resnick et al., 2005) and evidence-based treatments can be used within this approach (Peebles et al., 2007).
The body of literature on the recovery paradigm has focused on definitions of recovery (Anthony, 1993), the relevance of the model to specific diagnoses (e.g. schizophrenia) (Bellack, 2006), training programs for practitioners to deliver recovery-oriented practice (Peebles et al., 2009) and consumer perceptions of recovery-oriented practice (Marshall et al., 2009). However, the literature has not specifically examined the model in relation to AN. Questions therefore remain as to the usefulness of this paradigm for AN and if the lessons learnt from other areas of psychiatry can be applied to AN treatment approaches.
Is the recovery model compatible with AN?
The patient’s perspective
The recovery model is consumer driven; therefore, in order to explore the relevance of the model in AN, research examining the consumer’s perspective of AN recovery was reviewed. Understanding the patient’s experience of recovery from AN is an area of research that has grown in the last decade, with a number of qualitative studies published (e.g. D’Abundo and Chally, 2004; Dawson et al., 2014; Hay and Cho, 2013; Keski-Rahkonen and Tozzi, 2005), including a meta-synthesis of 16 such studies (Espindola and Blay, 2009). Espindola and Blay reviewed the literature in scientific databases using the keywords eating disorder, anorexia nervosa, AND qualitative research. Using quality inclusion criteria, they selected 16 studies that explored the patient’s perspective of AN for analysis. The basis of analysis for the current paper included Espindola and Blay’s dataset, along with more recent literature sourced using the same keywords. These studies have identified common themes that are associated with recovery, according to consumers.
When consumers cite the factors that were helpful in leading to change, the endorsement of recovery principles is evident. The importance of connection and satisfactory relationships with others, for example, is repeatedly highlighted by consumers (e.g. Cockell et al., 2004; D’Abundo and Chally, 2004; Hsu et al., 1992; Redenbach and Lawler, 2003; Tozzi et al., 2003) and was the most commonly endorsed factor favouring recovery in the review of qualitative studies conducted by Espindola and Blay. On the other hand, lack of social support and being misunderstood by others have been identified as barriers to change (Cockell et al., 2004; Dawson et al., 2014; Espindola and Blay, 2009). Similarly, in a review of personal accounts of recovery from AN, Hay and Cho found that relationships with others who shared the lived experience of AN, such as those in a support group, were considered important to people’s recovery journeys (Hay and Cho, 2013).
The importance of hope is emphasised under the model and, similarly, hope has been identified as an important psychological construct underpinning successful outcome in AN by patients (Hay and Cho, 2013). AN research additionally suggests that patient and clinician beliefs and attitudes about prognosis have important implications for outcome (Dawson et al., 2014). Likewise, autonomy and empowerment, that have been identified in the recovery model literature as important variables (Andresen et al., 2011), have been found to support improvement in AN. Establishing an internal locus of control and making active decisions to recover have been identified by consumers as components in initiating change in AN (Dawson et al., 2014). Determination to recover has also consistently been identified as helpful by AN sufferers (Dawson et al., 2014; Hay and Cho, 2013; Keski-Rahkonen and Tozzi, 2005; Redenbach and Lawler, 2003; Tozzi et al., 2003). In contrast, low self-esteem and ineffectiveness have been found to predict poorer outcome in AN in quantitative intervention studies (Wade et al., 2011).
Qualitative studies have also identified that patient attitudes towards the status of AN treatment are consistent with the recovery model. The model emphasises that change can occur without professional intervention and places importance on the external or ecological factors associated with recovery as well as the interaction between the individual and their environment (Onken et al., 2007). AN patients have cited multiple variables associated with successful outcome, suggesting that there is no one single pathway to recovery (Hay and Cho, 2013). While therapy might be important for change, patients suggest that successful outcome goes beyond treatment (Espindola and Blay, 2009). Positive life experiences, spirituality, satisfactory relationships, new interests, and meaningful life activities have been identified by consumers as helpful in leading to change (Espindola and Blay, 2009; Hay and Cho, 2013). Examples of such experiences include starting a hobby or activity (e.g. yoga), starting a new relationship, or even getting a pet (Hay and Cho, 2013). These findings suggest that there are multiple ‘ingredients’ that lead to change in AN and some of these ingredients are achieved outside of treatment. This is consistent with the recovery model, which posits that change can be impacted by a range of environmental factors and not necessarily limited to professional intervention.
The qualitative literature exploring the patient’s perspective of AN recovery is consistent with the recovery model, providing preliminary support of its relevance in AN. There are common themes across the studies suggesting that these findings might be generalisable to the AN population. However, further confirmation via systematic reviews (e.g. Noyes et al., 2008) is needed.
AN treatments
Some AN treatments demonstrate compatibility with a recovery model approach, particularly in terms of common factors. There are many parallels between the common factors and recovery models. Both encompass many basic long-accepted principles of therapy including a very collaborative relationship with the client, encouraging self-efficacy (including empowerment and individual responsibility), promoting hope and expected improvement, and a strong therapeutic relationship (Peebles et al., 2007). Research suggests that specialised manualised treatments are not the only way to treat AN (Bulik, 2014). Indeed, non-specific therapies delivered by eating disorder specialists have been found to be as (Zipfel et al., 2014), or more (McIntosh et al., 2005), effective. In psychotherapy, common factors theory holds that it is the shared components of psychotherapy (such as thinking and feeling processing and the therapeutic relationship) that accounts more for outcome than the factors that are unique to individual treatment models (Tracey et al., 2003). Common factors is an established paradigm for exploring treatment efficacy (Hubble et al., 1999). McIntosh and colleagues (2005) have argued that the important non-specific factors of AN psychotherapy might allow for an increased sense of patient control and autonomy and support for a patient group in need of these. The exploration of non-specific therapies and common factors is therefore a potential area that could be explored further in AN treatment approaches.
To a certain degree, many current evidenced-based AN treatments tap into recovery model elements. What differs between treatments is the relative importance that is placed on these principles, with some treatments placing greater importance on such principles. Treatments that are more patient driven, such as Motivational Interviewing (MI) and Specialist Supportive Clinical Management (SSCM; McIntosh et al., 2006), for example, are consistent with a recovery model approach. MI, for instance, focuses on generating a patient’s own reasons for change, as opposed to imposing paternalistic values upon a person (Treasure and Ward, 1997). Here, the therapist is encouraged to avoid the traditional medical model stance of being the instigator of change (Treasure and Ward, 1997). Similarly, in Maudsley Anorexia Nervosa Treatment for Adults (MANTRA; Schmidt and Treasure, 2006), rooted in the traditions of MI, therapists do not seek to ‘re-educate’ patients. AN treatments that emphasise strengths, agency, repositioning of expert power and employ a non-pathologising culture, such as family therapy, also support a recovery approach (Gehart, 2012). In Family-Based Treatment for AN (FBT; Lock et al., 2001), the therapist avoids taking an ‘expert’ stance, instead treating the family as expert and encouraging them to take ownership and responsibility for recovery.
Treatments that emphasise peer support, involvement of carers and wider community also support a recovery model approach. Building on FBT, multiple family therapy has been introduced as a means for families to learn from each other and receive peer support. Multiple family therapy also provides a way to overcome the isolation and stigmatisation associated with AN and is a way of injecting hope (Scholz and Asen, 2001). Narrative therapy approaches in AN have emphasised the use of role models and consumer-driven accounts of recovery. Developed by narrative therapists, the ‘Archives of resistance’ is a collection of ‘anti-anorexia’ narratives from sufferers themselves. This encourages the use of counter-narratives that make ‘acts of resistance’ to AN possible (Lock et al., 2004). MANTRA has a focus on more than symptom abatement, including creating a life outside the illness, an emphasis on individual values, and helpful relationships with others (Wade et al., 2011). In MANTRA, the therapist works closely with the patient and the patient’s family or a support network is involved.
Principles of the recovery model are identifiable in currently used AN treatments. This demonstrates that, while the recovery model approaches treatment from an alternative standpoint to current treatments, aspects of the model are consistent with empirically supported concepts. The model complements existing professional standards. Questions remain as to whether we can better capitalise on recovery model principles in the establishment of new treatments and/or the augmentation of current treatments.
At present, elements of the model tend to be scattered within various treatments for AN. Further integration could potentially be more efficacious. In reviewing the literature on therapy approaches, two recent examples of AN treatment were identified that we believe are most closely aligned with a recovery model stance as a whole. Both show significant promise as alternative approaches to treatment. The first example is a recent randomised controlled trial that examined the efficacy of two established therapies (CBT and SSCM) for the treatment of severe and enduring AN (SE-AN) (Touyz et al., 2013). SE-AN is very difficult to treat and is associated with high levels of disability, low motivation to change, and high drop-out from treatment (Wonderlich et al., 2012). Accordingly, a new paradigm for treating such patients has been advocated, in which symptom reduction is not the primary goal. The authors of this study adapted the existing treatments and reprioritised how the goals of treatment were presented to patients (Touyz et al., 2013). Participants were told that the focus of treatment was not on weight gain per se but on quality of life, and treatment goals were collaboratively articulated. Using this approach, both treatment groups experienced gains and, importantly, this study had a very low drop-out rate. These findings contradict conventional beliefs that those with SE-AN cannot benefit from treatment and raise questions as to whether traditional treatment targets need to be reviewed.
Similar to the patient population targeted by Touyz and colleagues, the Community Outreach Partnership Program (COPP), proposed by Williams and colleagues (2010), is a treatment designed for people who are not responsive to traditional treatment approaches that focus on symptom reduction. Rather than continuing to offer similar treatments, the authors developed a new model of care for these patients which centres on improving quality of life, minimising distress, and increasing hope for the future. This treatment approach recognises the role of environmental factors in promoting recovery. There is a focus, for instance, on integrating people within their community; establishing meaningful work or activity is valued. The program uses an MI approach placing responsibility for change with the client and encouraging client autonomy. However, there are treatment non-negotiables, however, which are mutually agreed and used to maximise autonomy. Using this approach, individuals in the program experienced significant improvement in eating disorder symptoms, increased values on relationships, and decreased hopelessness and distress.
These two studies provide support for the adaptation of treatment targets to focus more on patient values and holistic goals and not solely on symptom abatement. In both studies the definition of improvement was broadened to include quality of life. These studies demonstrate that recovery-focused treatment can be successful and might be most successful for particular patients, such as those with chronic AN, or those who have not responded well to traditional symptom-focused treatments. They raise questions about the need for a new paradigm in AN treatment and shine some preliminary light on how a recovery-oriented treatment approach might be implemented in AN treatment.
Implementing recovery-oriented practice in AN
Appropriate implementation of recovery-oriented care requires research and discussion. As an initial starting point we propose the model can add to current practice in AN by: (i) providing a philosophical framework and professional language that can complement the scientist–practitioner model and directly inform practice; (ii) providing a rich resource of new ideas that may serve to augment current models of treatment; and (iii) providing a means to integrate consumer voices into practice.
There is evidence to suggest that expanding treatment goals beyond symptoms might be effective. Qualitative findings suggest that there are multiple factors outside treatment which promote successful outcome in AN, such as the encouragement of empowerment, individual responsibility, holistic care, promoting hope, improving quality of life, and peer support (Cockell et al., 2004; Dawson et al., 2014; Hay and Cho, 2013; Tozzi et al., 2003). A traditional medical model approach to treatment de-emphasises these factors, whereas a recovery model approach promotes these issues. These factors, which have largely been on the periphery of clinical practice and research, need to shift to become a central part of research and treatment. Research that continues to improve treatments is vital; however, we also need to broaden our scope and examine factors external to treatment and consider how these can be integrated into holistic, person-centred care. This would include expanding the intentions of therapy where goals shift from symptom abatement to including well-being in a complete sense, which includes relationships, career, and community (Gehart, 2012). Such a treatment focus may promote patient engagement and allow for greater engagement with the person rather than the diagnosis (Davidson et al., 2009).
The recovery model embraces continuity of care, which might be effective in AN treatment. Traditional mental health services have worked on an acute-care model, which conceptualises mental illness as diseases that can be treated and cured. Many people suffer from acute forms of mental illness that are essentially ‘cured’ with one treatment. For others, this is not the case and for these individuals the current system of mental health care is mismatched. The recovery model acknowledges that recovery is often not achieved following a single episode of treatment and that we should not necessarily be discouraged by this (Davidson et al., 2009). Furthermore, there is a tendency for practitioners who treat those with chronic mental illness, such as long-standing AN, to focus on symptoms and impairments (i.e. what is ‘wrong’ with the person), and overlook strengths and competencies that the person continues to have (Davidson et al., 2009). This can lead to helplessness and despair for patients and practitioners alike. The recovery model promotes a strengths-based approach, which might be particularly important for those with chronic AN.
Approaches such as peer-support also have the potential to build solidarity and community in a disorder where sufferers often report isolation and feelings of being misunderstood. National eating disorder associations, such as The Butterfly Foundation in Australia and the British Eating Disorder Association (B-EAT) in the UK, have placed a strong emphasis on consumer voices, for example by promoting support groups. In conjunction with the National Health Scheme, B-EAT introduced a ‘buddy system’ using ex-sufferers’ expertise to provide support and guidance to current sufferers.
Considerations in applying the model
While evidence suggests that there are elements of the recovery paradigm that are consistent with AN, there are some unique features of AN that need to be considered. Some caution in translating lessons learned from recovery in other fields, such as emphasising adaption to symptoms, is warranted. AN has been distinguished from other primary psychiatric disorders because of the ego-syntonic nature of the disorder and associated low motivation to change (Vitousek et al., 1998). While for other mental illnesses, like schizophrenia, patients might be encouraged to learn adaptive ways to live with symptoms, such as voice-hearing, adaption to AN symptoms, such as severely restricted food intake, would be indefensible from the perspective of many. While the symptoms of mental illness are key targets for treatment, equally, the consumer movement has made known that hopelessness, loss of control, and loss of sense of self can have a significant effect that can be more distressing than symptoms (Bellack, 2006). It is important to note that the recovery model does not support poor mental health care (such as disregarding symptoms), but rather emphasises that factors outside of symptom abatement, such as quality of life, are equally important.
A recovery model approach advocates treatment decisions to be made in accordance with consumer values. In the case of AN, it can be the values of the illness itself, such as low motivation to change, driving an individual’s choices. Critics of the recovery model approach have warned against the dangers of empowering patients, such as letting incapacitated patients make important choices about their treatment (Peyser, 2001). It should be noted that a recovery perspective does not advocate for seriously unwell patients to take responsibility for their mental health. The model acknowledges that there are times when patients will be unable to make decisions for their care and might need to be treated on an involuntary basis (Munetz and Frese, 2001; Reisner, 2005).
While a medical model approach emphasises the role of treatment, the recovery model, in contrast, places greater emphasis on personal experience. However, moving to recovery-oriented practice does not involve abandoning evidence-based practice. A recovery approach can be delivered within such a framework. Identifying ‘what works’ is the crux of evidence-based practice; elements of the recovery paradigm have empirically been found to be beneficial in mental health (Mountain and Shah, 2008; Resnick et al., 2005). Historically, the medical model and recovery model have been perceived as conflicting paradigms; however, while these models emphasise different aspects of recovery, they are not mutually exclusive (Barber, 2012). In fact, these differing conceptualisations can be viewed as complementary (Bellack, 2006). The combination of these approaches might offer the best way forward in that we promote recovery whilst maintaining good clinical and evidenced-based care. Indeed, it has been recently proposed that the recovery model reflects the evolution of the medical model (Barber, 2012; Flaherty, 2012).
Conclusion and future directions
While much has been written about recovery-oriented practice in mental health, to our knowledge, this paper is the first attempt to examine the model’s potential application to AN specifically. Accordingly, future research is required to examine the ways in which this approach could best be integrated into current AN practice. An evidence base is needed for how best to improve factors for AN patients such as quality of life, hope and self-efficacy, and to connect people to support networks and community. Future research could involve collaborating with patients to develop knowledge about recovery and treatment, examining the utility of ex-sufferers’ help with current patients, and exploring how those affected by AN define and conceptualise recovery. Measures examining functioning, quality of life, and user-defined outcomes could be included in treatment trials in addition to symptom-based outcome measures (McPherson et al, 2003). There is also a need for education of recovery principles in the field of eating disorders. In particular, practical methods and guidelines are needed if we are to promote recovery-oriented practice in AN.
Until current treatments yield better outcomes for those with AN, practices need to be reviewed. The recovery model might offer a means by which to rethink the way in which treatment is delivered and outcome is conceptualised. Given the paucity of successful treatments for adults with AN, the potential role of the recovery paradigm needs to be examined. Whether services integrating this approach produce better results remains to be seen but the model does have relevance for AN and holds promise. As other areas of psychology and psychiatry move forward with the recovery approach it is important that the use of this model in the eating disorders be given attention.
Footnotes
Declaration of interest
The authors report no conflict of interest, the authors alone are responsible for the content and writing of this paper.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.