Patient Perspectives on Anorexia, Treatment, and Therapeutic Alliance

Dear Science of Eating Disorders readers, please welcome Andrea, our newest contributor! Below is her introduction and first post.

Hello SEDs readers, my name is Andrea and I’m excited to be contributing to the blog. I have an undergraduate degree in sociology and I am currently a Masters student studying family relations and human development. My research is looking at the experiences of young women in recovery from eating disorders, and uses qualitative methods including narrative interviews and digital stories to explore stories of eating disorders and recovery. I am particularly interested in stories that fall outside of the “norm,” as I feel that we sometimes hear a limited, scripted story of what it means to be someone who has had and recovered from an eating disorder.

I myself am recovered from ED-NOS, and I am happy to be making meaning from my experiences by exploring eating disorders in an academic way. I hope to be able to add my voice to the conversation–I’ll be looking mainly at the qualitative literature on eating disorders, their treatment, and recovery. You can reach me at


We often hear about treatment programs and outcomes from the perspective of the service providers and those in positions of power. Particularly where eating disorders are concerned, there has been much emphasis on exploring which forms of treatment most effectively address complex symptoms and presentations. While hearing from clinicians and researchers is undoubtedly important, and quality randomized controlled trials (which are actually few and far between, in the eating disorder realm) give us important guidelines in terms of shaping the future of eating disorder treatment, focusing solely on these perspectives may neglect to incorporate patient voices into the equation.

A previous post by Gina explored how former patients define recovery but how do individuals diagnosed with anorexia nervosa define the illness itself and the therapeutic alliance? A study by Boughtwood & Halse (2010) explores how teenage girls admitted to Australian hospitals for AN construct their illnesses, and whether these constructions align with the perspectives of their treatment programs and teams.

The authors use poststructuralist feminism as a framework for their study.


Post-structuralism is based on the writings and philosophies of several key thinkers, including Foucault, Barthes, and Derrida. These philosophers explore discourses, or how social and historical forces work to frame individuals’ thoughts and actions including language, attitudes, values and behaviours.

A major interest of post-structuralism is power, and the dynamics of power that serve to favour certain points of view or groups of individuals over others, and how this comes to impact the creation of subjectivity.

So, poststructuralist feminism is largely interested in factors that serve to position individuals at different levels of power, with a particular interest in how gender comes into the picture. These power dynamics make an exploration of clinical settings such as eating disorder units particularly interesting, as the context is layered with power differentials, for example between patients and carers, between different members of treatment teams, and between patients.

Straightforward? Not so much, but hopefully some of the intricacies of poststructuralist feminism will become more clear as I describe this study and what the authors found.

(For a more nuanced explanation of poststructuralist feminism and how it comes into play with eating disorders, I recommend reading articles by Helen Malson, including Malson, H. (1998) The thin woman: feminism, post-structuralism and the social psychology of anorexia nervosa (London, Routledge), or works by Maragaret Wetherall, including Wetherell, M. (1998). Positioning and interpretative repertoires: Conversation analysis and post-structuralism in dialogue. Discourse & Society, 9(3), 387–412.)


For their research, Boughtwood & Halse interviewed 25 teenage girls, all of whom met the DSM-IV criteria for AN. Demographically, most of these girls were Anglo-Australians, with 2 of Chinese descent and one of Italian descent. All had participated in inpatient treatment at large urban hospitals, and 20 of the girls were engaged in this treatment at the time of the study. This treatment was focused on weight restoration and the normalizing eating through timed, nurse-supervised, structured meals and snacks.

Interviews took the form of semi-structured, life history interviews conducted in a recursive (conversational) way; the girls were able to introduce ideas that were not addressed directly through the interview guide if they thought they were important to their stories. The authors focused their analysis on the descriptions the girls provided about interactions with doctors and other aspects of treatment. Within these descriptions, they located power, subjectivity, self, and relationships.

These aspects of the data were then related to notable patterns in the literature dealing with power and subjectivity in hospital settings more broadly and those constructing the “anorexic patient” more specifically. Field notes were also collected to make note of the context in which the girls were engaged in treatment.

Throughout their analysis, the authors engaged in group analysis to check for trustworthiness. This kind of check to quality of analysis is particularly important in qualitative studies such as this, as unlike quantitative research, there are few established parameters for easily determining “validity” or “reliability”- researchers must instead rely on solid theoretical backing, strong grounding of analysis in the data, and other checks for quality, including soliciting the perspectives of experienced peers and “experts” (sometimes, though not in this case, including the participants themselves).


The “Perfect Patient”

Perhaps unsurprisingly for those of us who have participated in hospital-based treatment, one of the key themes that emerged was the idea of being or performing as the “perfect patient.” Treatment systems, according to the authors’, present guidelines for how patients should behave. These expectations are transmitted through policies and procedures, including the provision of incentives for “good behaviour” (e.g. gaining weight) and sanctions for “bad behaviour” (e.g. refusing to finish a meal). These guidelines are not always accepted without resistance, and the authors found that they may lead to adversarial relationships between patients and treatment teams.

Nonetheless, their findings suggest that girls may be willing to comply for a number of reasons; for example, they may recognize that the doctor occupies a position of power and therefore must necessarily “know what is good for them” or as “means to an end,” with the end being the ability to leave treatment having completed the program.

Despite recognizing the power and knowledge held by doctors, girls may still feel ambivalent about whether they are really sick or not, and thus feel unsure about how much power to surrender in order to “get better.” Doctors may interpret patient compliance (i.e. through gaining weight, etc.) as a demonstration of their commitment to treatment. However, the girls in this study described their compliance as a way to escape the hospital environment, which many described as a stressful. The “perfect patient” identity is one of uncertainty, the authors note, as it may conflict quite markedly with the individual’s own perception of “perfection.”

Therapeutic Alliance

Boughtwood & Halse note several key findings in terms of the therapeutic alliance, or the relationship between carers and patients.

Girls engaged in inpatient treatment may possess a great deal of knowledge about their disorders and about the different roles played by members of the treatment team. This may lead to the patients making evaluations about clinician competency: new doctors or members of the treatment team deemed less knowledgeable than others (or than the patients themselves) may hold relatively less power, and patients may not see them as credible “experts.”

Similarly, young women engaged in treatment may not buy into common biomedical understandings of “health,” describing differences between physical sensations and medical measures (e.g., “feeling normal” while having extremely low blood pressure). This may lead to biomedical descriptions of the physical dangers associated with anorexia holding little sway over behaviour; the study revealed that in this group, even the diagnosis of AN may itself be rejected.

Subversive Behaviours & Therapeutic Alliance

The authors also describe girls’ accounts of engaging in practices that subvert the doctors’ authority. (I will not include here as they may be potentially instructive for current sufferers.) These behaviours may be learned from other patients engaged in treatment, and in performing them the girls are described as constructing doctors as a malicious enemy.

These subversive behaviours may hurt the therapeutic alliance in a number of ways: the patients are mistrustful of their doctors, who in turn perceive the patients as dishonest, exacerbating damaging stereotypes of individuals with eating disorders as deceitful and in need of strict, unbending treatment. What is particularly interesting about this study’s description of this process is a recognition the interactions between patients and treatment teams. Rather than seeing patients solely as passive recipients of care, the dynamics of doctor-patient interaction are explored in a way that recognizes that many patients are themselves very aware of the power dynamics at play, as well as the “ins and outs” of treatment.

Societal Gender Dynamics

The study also attends to the ways in which the dynamics of treatment map onto societal gender dynamics.

Girls are described as “colonizing” the hospital setting by bringing in items from home, which the authors suggest re-shapes the context of treatment. Further, girls on the unit may cry or give doctors the silent treatment, behaviours which are often socially coded as “feminine.” Interestingly, what the authors describe as exercises of femininity may serve to impact the therapeutic alliance by humanizing and individualizing the patients in the eyes of their treatment teams.


Overall, this study presents an interesting take on the therapeutic alliance, adding the patient perspective to the picture.

I do feel that the conclusions related to feminization of the hospital space may be overstated–I am not sure that boys and men would not also bring items from home to make their spaces feel more comfortable, and I think this would be worth exploring.

Also, as is true of most qualitative research, these findings relate specifically to the eating disorder units at which the study was conducted, and to the group of girls interviewed. The findings are interesting, but they are also context-specific; while these constructions of anorexia and the therapeutic alliance might be true in other settings, more research would need to be conducted to check the findings against other, similar studies. Finally, it would be interesting to bring back the findings to the girls themselves.

Though the authors have stayed true (as is demonstrated through their use of excerpts throughout the paper) to the descriptions offered by the participants, they have also interpreted the data in light of their own knowledge and those of others in the field. It would be interesting to see how well the analysis fits with the participants’ understandings of their acts of resistance and compliance in treatment.


Despite some limitations, this study has opens up a number of important implications and possibilities.

Primarily, recognizing that individuals with eating disorders are not simply passive recipients of care is extremely important. While this might seem like an obvious conclusion, the patient perspective is notably missing from many studies looking at therapeutic alliances and the establishment of treatment systems.

The authors did a good job of pointing out the reality that individuals with eating disorders may in fact be extremely knowledgeable about their disorders, about treatment systems, and have strong views about particular programs and doctors. To discount their knowledge and presence in treatment interactions will likely result in ineffective or, at best, temporarily effective treatment.

By recognizing the roles played by all involved in treatment, including the patients themselves, it might be possible to move toward more solid therapeutic alliances. Specific ways of integrating patient perspectives warrants further investigation and is obviously not unproblematic; again, we come up against the balancing act of “who knows best” when it comes to treating eating disorders. While I don’t think that this study answers the question entirely, I am hopeful that it will help to open a dialogue that includes patient voices.


Boughtwood, D., & Halse, C. (2009). Other than obedient: Girls’ constructions of doctors and treatment regimes for anorexia nervosa Journal of Community & Applied Social Psychology DOI: 10.1002/casp.1016

Liked it? Take a second to support Andrea on Patreon!


Andrea is a PhD candidate focusing on individual, familial, and health care definitions and experiences of eating disorder recovery. She has an MSc in Family Relations and Human Development and a BA in Sociology. In her Masters research, she used qualitative and arts-based approaches (digital storytelling) to explore the experiences of young women in recovery from eating disorders. Andrea has recovered from EDNOS. She can be reached at andrea[at]scienceofeds[dot]org.


  1. Awesome post Andrea! I loved reading it. I’m really happy to have you contribute because your expertise is so different from mine and I think it will be really useful for both myself and the blog readers. I especially value your background in qualitative research, because all of that is new to me! I only started reading it after I began the blog… So yes, welcome!

    This is a very interesting topic. In your own opinion, how could clinicians integrate these findings into practice? I mean, on a one-on-one basis, it is probably not that difficult, because the clinician can gauge the patient’s level of understanding and awareness. But in a group setting? I can’t imagine what that kind of integration would look like–some individuals having more say than others in their treatment?

    Like, what are the dynamics between patients who are aware they are sick and want to get better, and those who are more ambivalent or don’t think they need to get better at all?

    When I did treatment (outpatient), I didn’t go to their groups because I was surrounded by patients whose parents forced them to be there; they didn’t want to get better. I was 15 at the time. I, on the other hand, literally fought to be there, and I didn’t want to be around patients who were ambivalent and denying their illness (which I why I preferred support groups at Sheena’s Place, where participants were aware of their disorder and the facilitator wasn’t infantilizing). So, I imagine in an inpatient or residential setting, the dynamics get even worse.

    This kind of relates to my previous post, actually, in that it highlights the pros and cons of inpatient treatment versus something like FBT. And also the double-edged sword of being around patients with the same disorder.

    Your post didn’t explore gender as much as your intro to poststructuralist feminism led me to believe it might. Do you think it matters, in these settings, the gender of the clinicians and treatment staff?

    • Thanks Tetyana, I enjoyed writing the post and look forward to writing more. As you suggest, I think it could be quite difficult for clinicians to integrate findings like these into their practice. Especially in group settings, I imagine that the focus, for clinicians, is on finding the approach that works best for the majority of people; evidently, clinicians are dealing with many patients at a time, and perhaps the individual therapeutic alliance can fall by the wayside a little. I think it is important to note that I think that clinicians are often doing the best that they can with relatively limited resources (especially time!). I think one of the biggest problems here is (what I consider to be) the highly individual nature of eating disorders. I doubt that any two people would experience their disorders in the same way- there are just way too many contextual/individual factors that play into things like readiness for and resistance to treatment. While there might be broad trends, I think for the most part everyone is coming into treatment with different histories, different expectations, and different levels of commitment. Clinicians, too, may have different histories and experiences, even with particular (sometimes repeat) patients that colour the therapeutic alliance experience.

      Like you said, the other people involved in group therapy settings in particular can play a huge role in the dynamics between patients and between patients and clinicians. This is likely exacerbated in the adolescent context, because often adolescents are placed into treatment by parents, but I did observe similar trends in my own experiences of treatment, which was in a hospital setting (intensive outpatient/day-hospital treatment) with other over-18-year-olds. There were times when I was frustrated by the fact that other patients seemed to be “playing the system” in a way- subverting clinician authority etc., and other times when I felt like the rules and regulations imposed by the system were ridiculous and rigid, despite being extremely invested in attaining recovery. At the same time, I never had much luck with individual, outpatient therapy- in a way, seeing some patients deny the severity of the illness or be unwilling to let go of the eating disordered behaviours made me re-evaluate where I stood in relation to my own illness, and almost be more committed to recovery as a result. As I read this article in particular I thought about how the experiences described related to my own, but differed in some significant ways. For example, I could relate to the idea of (at first) relinquishing the behaviours in the name of appearing to be the “perfect patient.” I think that at first, engaging in treatment and following the rules was probably more about not letting anyone, including the treatment team, down. But then, I noticed that I began to actually buy in to the treatment, including all of the little rules and regulations I found grating. It reminds me of the studies indicating that in order for cognitions relating to eating disorders to be challenged, the behaviours need to be changed first (the idea that the psychological “work” of recovery occurs once eating is normalized and weight restored).

      Anyway, those are my own little anecdotal comments, and I don’t know that I really answered your question- basically, I think it would be quite difficult for clinicians to integrate these findings into (particularly group) practice. One thing that I would say would be that studies like these might help to remind clinicians of the importance of individual therapeutic alliances in recognition of the patient’s understandings, even in the group setting.

      In terms of gender, you’re right; it didn’t come up in my analysis as much as I thought it might- little strands of gender dynamics emerged throughout the article itself, and I could have examined these in more detail. Part of my reason for not doing so was a fear that the article was becoming absurdly long (kind of like this comment!). Gender dynamics likely do play into the therapeutic alliance: besides the mention of the “colonization” of hospital rooms with items coded as feminine, the authors did mention that the male clinicians may be perceived as being more powerful/authoritative, and also that patients often cried or gave the silent treatment to doctors to express their dislike of certain decisions etc. in a “feminine” way. To be honest, I wasn’t 100% sold on this part (i.e. the crying as representing a female patient’s exertion of “feminine wiles” to get her way- not that the authors put it that way but something about the framing irked me); however I do agree with the potential for gender to compound power differences between patients and clinicians. Something that I think would be extremely interesting to explore would be power differences and the therapeutic alliance between male patients and male and female clinicians. I think it’s more common to hear about male clinicians exerting power over female patients, but less common to really delve into the dynamic between male patients and male clinicians or female clinicians. Looking at it in this more unexpected way might help us to draw out more novel findings!

      Anyway, I hope this epic comment helped to answer some of your questions!

      • You wrote:
        “At the same time, I never had much luck with individual, outpatient therapy- in a way, seeing some patients deny the severity of the illness or be unwilling to let go of the eating disordered behaviours made me re-evaluate where I stood in relation to my own illness, and almost be more committed to recovery as a result.”

        I actually had this same experience but through Sheena’s Place and what I experience when I come across ‘pro-ED’ things (not really pro-ED, as much as really ambivalent about recovery). I realize what I have and don’t want to lose it.

        I found treatment rather useless, except for weight restoration because I didn’t own a scale, so it was useful to know what I weighed and how I was gaining. I found, in my experience, focusing too much on recovery was usually, except when I was much sicker, counterproductive. I think IP and residential can be lifesaving and really important, for sure, but I think it is almost important to ‘recovery in-real time’, because it is so easy to get into a state where the ED feels like such a huge part of you that letting go is nearly impossible because life revolves around being sick.

        Yeah, the “feminine” crying thing, yeah, that also irks me.

        I haven’t found differences in my experiences with clinicians of different gender–then again, my experiences are limited. Very limited. I did find that I preferred clinicians who didn’t have an ED history, though.

Leave a Reply

Your email address will not be published.

9 − five =