Interpreting genetics in the context of eating disorders: Evidence of disease, not diversity

Abstract

How is genetic involvement interpreted for disorders whose medicalisation is contested? Framing psychiatric and behavioral disorders in terms of genetics is expected to make them seem “more medical.” Yet, genetic etiology can also be used to frame behavior as acceptable human variation, rather than a medical problem (e.g., sexual orientation). I analyze responses to the idea of a genetic component in anorexia and bulimia nervosa (AN/BN) via semi-structured interviews with a sample of 50 women diagnosed with an eating disorder (25 recovered). All but three volunteered that genetics would medicalise AN/BN, by (1) making eating disorders seem more like “real” diseases; (2) implying a need for professional treatment; or (3) for biologically-based treatment. Results also indicate several “counter-logics,” by which genetic framing could support non-medical definitions of AN/BN. I argue that genetic framing reduces perceived individual responsibility, which can support definitions of behavior as either a reflection of disease (which entails intervention) or a reflection of normal human diversity (which does not). In the context of public skepticism as to the “reality” of AN/BN, genetic involvement was taken as evidence of disease in ongoing negotiations about the medical and moral status of people with eating disorders.

INTRODUCTION

Genetic involvement in health conditions is expected to draw attention away from alternative explanations, thereby changing how a physical or mental problem is conceived and acted upon (Conrad 2000; Lippman 1991; Nelkin and Lindee 1995). Eating disorders are frequently interpreted as something one does, rather than something one has; more like behavioral choices than psychiatric medical disorders (e.g., Crisp 2005). Does the idea of genetic involvement encourage conceptual reclassification, i.e., from choice to disease? If so, then by what logic, and with what consequences?

After a condition has been officially adopted as a disorder by professional organisations--as have anorexia nervosa (AN) and bulimia nervosa (BN) (APA 2000)--a “reservoir” of public skepticism may remain (Conrad 1992: 271). To solidify the medicalisation of psychiatric disorders, advocates have emphasised biological aspects (Pescosolido et al. 2010, Phelan 2005). The underlying expectation is that genetic framing will reinforce the medicalisation of a psychiatric diagnosis (e.g., Conrad and Schneider 1992:281). Yet, genetic framing can also serve a different, anti-medical purpose, as when genetic research on autism is interpreted to support “neurodiversity” (Ortega 2009) and genetic research on homosexuality to support equal rights (Shostak et al. 2008, Brookey 2001). Empirical examination is needed to elucidate how genetic influence is interpreted, particularly because genetic reframing can increase stigma (e.g., Phelan 2005, Kvaale et al. 2013) and elicit other, possibly unwanted meanings. Little research exists on interpretation of genetics by people with psychiatric diagnoses, and none on those with eating disorders.

This article examines how genetic influence is interpreted in the context of public ambiguity about whether eating disorders are medical phenomena. Despite being official psychiatric disorders, the extent of their medicalisation is arguably limited by the widespread perception that they are volitional--not only by skeptical outsiders (Crisp 2005) but also some insiders (Fox et al. 2005)—and by feminist critique of simple medical models (e.g., Bordo 1993). Because AN and BN are not always understood as diseases, or even as problems, they provide a potentially interesting source of data as to the effect of genetic framing.

BACKGROUND

“Medicalisation occurs when a medical frame or definition has been applied to understand or manage a problem…” (Conrad 1992:211). This definition leaves open who does the medicalising and for what purpose; non-professionals can apply medical frames for their own reasons (Barker 2008, Conrad 2007). Furthermore, medicalisation is not a simple dichotomous process, but can also occur in “degrees” (Conrad 2007). Some aspects of a phenomenon may (not) be medicalised—etiology, diagnostic category, treatment—and medicalisation may not be accepted by all social actors (also see Halfmann 2011). An example is the “contested disease,” in which patient-advocates strive for official medical acceptance of conditions as somatically “real,” rather than imagined or psychological (e.g., Barker 2008).

Within a traditional sociological “medicalisation” framework, professional adoption of a disorder is the endpoint of a process of medicalisation (e.g., Conrad and Schneider 1992); such adoption represents its entry into “medical jurisdiction” (Conrad 1992). Similar processes may continue to shape understandings well after official adoption (e.g., Shostak et al. 2008), though not always theorised as “medicalisation” per se (see Williams et al. 2011: 236). In the present article, professional adoption of eating disorders is a point of departure, such that the medicalisation of eating disorders continues after their official acceptance as disorders, via changes in perceived etiology. As noted by Pickersgill, medicalisation can be understood as “a set of processes enabled by and co-produced through the interactions between a heterogeneous assemblage of standards, clinical practice, scientific research and patient activism” (2012: 336). My focus is the ongoing interaction between professional and personal understandings of a condition, which can be conceived in terms of ideology and control, Foucaultian discipline, or constructionist reflexivity.¹ Moreover, personal understandings may be consonant, dissonant, or autonomous with regard to professional knowledge (Broër and Heerings 2012).

There is conceptual “room” for additional medicalisation if some social actors do not behave as though eating disorders are legitimate medical disorders. The extent of medicalisation, including acceptance of diagnostic categories and perceived legitimacy of treatment, is continually under negotiation as actors interact with institutional definitions and their consequences. An official definition is powerful because people are more likely to be influenced by it (see Schudson 1989), but it is not coextensive with a cultural object (e.g., a diagnosis) as used in practice. After an official diagnostic category has defined a group of symptoms as a disease entity and thereby stabilised the uncertainty surrounding them (Jutel 2009), some instability may yet persist. In this study, I understand genetic etiology as a framing device for the phenomenon of eating disorders. This framing device may lead people to see eating disorders as less or more “medical.” People with eating disorders may or may not frame their own behavior in such terms, but empirical examination can show what people think these “framing tools” would do in their social contexts, i.e., highlight certain characteristics of eating disorders and/or elicit certain social responses. This study examines the reasoning by which genetic involvement shores up the perception of AN/BN as natural, and therefore real and legitimate medical entities (Hacking 2001: 7, Jutel 2009).

There has been little direct empirical examination of the hypothesis that genetic framing may “reaffirm” (Conrad and Schneider 1992: 281) or “further” (Conrad 2000:326) the medicalisation of psychiatric disorders (but see Shostak et al. 2008). In U.S. surveys about people with mental illness, genetic etiology is associated with endorsement of hospitalisation or pharmaceutical treatment (Phelan 2005; Schnittker 2008). But genetics can also support non-medical interpretations of conditions or behaviors, such as homosexuality (Shostak et al. 2008, Brookey 2001) and autism (Ortega 2009). Shostak and colleagues (2008) argued that genetic information is interpreted according to prevailing understandings of a phenotype at particular historical junctures; perhaps if research on genetics of sexual orientation had been popularised when homosexuality was listed in the DSM, it would have been interpreted to support the medical definition.

Eating disorders have been officially medicalised by virtue of their inclusion in the DSM, but some, including people with eating disorders themselves, continue to see them as bad behavioral choices that could be controlled through willpower. Over a third of the public (U.K.) interpret eating disorders as bad behavior—e.g., people “could pull themselves together” (Crisp et al. 2005: 148; also see Stewart et al., 2006)—and blame those with eating disorders for their conditions more than those with depression, schizophrenia, or panic attacks (Crisp et al. 2000; Roehrig and McLean 2011). A person who could voluntarily stop his or her illness but chooses not to is in violation of the “sick role” (Parsons 1951, Williams 2005). Moreover, some with AN/BN actively resist medical understandings (Fox et al. 2005, Rich 2006, Weaver et al. 2005). Stigma-reduction experiments with nursing students found that genetic descriptions of AN made it seem less volitional than social descriptions (e.g., Crisafulli et al. 2008). Public contestation about whether eating disorders are to be considered genuine medical problems shapes the context in which social actors interpret the idea that genes could play a role.

AN and BN involve bingeing, purging, and/or food restriction, with high mortality rates and serious health consequences compared to other psychiatric disorders (Arcelus et al., 2011; Sullivan, 1995). These disorders run in families, and twin studies estimate high heritability (48–76% for AN; 50–83% for BN, Striegel-Moore and Bulik 2007), though interpretation is contested (see Horwitz et al. 2003 exchange). Molecular approaches build on these findings, and despite recent disappointments research continues using larger samples and different approaches (Pinheiro et al. 2010; Striegel-Moore and Bulik 2007). People with eating disorders are likely to have heard about genetic involvement in AN/BN and psychiatric conditions generally (Pescosolido et al. 2010). Genetic contributions to AN have been publicised (e.g., Tyre 2005) and popular health websites refer to genetic risk factors for both AN and BN (e.g., Mayo Clinic 2009, 2010). Psychiatric researchers and clinicians generally portray genetic factors as part of a probabilistic multifactorial model that involves behavioral and environmental factors (e.g., Striegel-Moore and Bulik 2007), not as a simple, deterministic one-gene-one-disease model (Kendler 2005; Rose 2007; Pickersgill 2009, 2012). Nevertheless, talking about genetics even in relation to complex disorders can cue determinism (see Condit et al. 2009). Interpretation may also be shaped by perceived social or political impact of (not) endorsing genetic causation (e.g., Easter 2012, Brookey 2001). Interpretations of genetics are thus likely to be formulated in relation to a public discourse, not just an individual’s evaluation of its relevance to the condition in question.

This study begins with the idea that medicalisation is a continuous and negotiated process that can extend even after official adoption of a diagnosis. Compared to the general public, people with eating disorders are well-positioned to assess conceptual trade-offs of genetic reframing for themselves and others in their social contexts. Their detailed knowledge of behaviors, treatments, stigma, and skepticism surrounding eating disorders qualify them to interpret potential impacts of genetic ideas “on the ground.” They may also have an investment in one or another conception of eating disorders, which will shape their interpretations of genetic framing. The present study is the first to examine perceptions of genetics by people diagnosed with an eating disorder, and contributes to an emerging literature on perception of genetics by those diagnosed with psychiatric disorders (Laegsgaard et al. 2010, Meiser et al. 2005, Rusch et al. 2010; Turney & Turner 2000).

METHODS

Recruitment and Sample

I conducted semi-structured interviews with 50 U.S. women² with a history of either AN or BN, half in treatment and half recovered (see Table 1). Those in treatment included inpatients, day program patients, and outpatients (including participants in a BN treatment study) at a multidisciplinary University hospital-based program emphasising cognitive-behavioral therapy. They were recruited via posted flyers and word of mouth by clinic personnel. In recruitment and consent materials, the study topic was described as “your ideas and opinions about eating disorders and what causes them.” Prospective respondents contacted me by phone or email, or gave permission to clinic personnel to give me their contact information. I informed them I was not part of a treatment team, did not have access to their medical records, and would not share information with others.

Table 1.

Respondents’ Diagnosis, Recovery and Treatment Status (N=50)

	Primarily AN	Primarily BN	Total
Recovered	14 (56%)	11 (44%)	25 (100%)
In treatment	13 (52%)	12 (48%)	25 (100%)
Inpatient	9	0	9
Day program	2	2	4
Outpatient	2	0	2
Treatment study	0	10	12
Total	27 (54%)	23 (46%)	50 (100%)

AN= Anorexia nervosa, BN=bulimia nervosa

Recovered women were recruited via mass email to female faculty, staff, and students of the same university. The email subject line read, “Have you recovered from an eating disorder?” and the message, “If so, I would like to hear your ideas about eating disorders and what causes them.” The email described eligibility requirements, study features, my contact information and provided a URL to view the consent form. Recovery was defined as three years without significant restricting, bingeing, or purging, based on responses to five questions. Diagnosis was based on self-report of professional diagnosis of AN or BN. An eating disorders psychologist helped develop eligibility questions and provided guidance in cases of borderline eligibility for either recovery status or diagnosis.

The study was designed with input from treatment providers to guard against risk of distress. Four clinicians reviewed the interview guide, all respondents were given information about local and national treatment resources, and a safety plan included access to counselors in case of serious distress during the interview (there was none). I excluded those with current AN/BN who were not in treatment because if genetic ideas were disturbing I wanted such respondents to be in contact with a treatment provider. The University IRB approved the study.

The majority of respondents was white and had at least some college education (see Table 2). They ranged in age from 18 to 64, with average age 33. Treatment experience ranged from no treatment to multiple hospitalisations.

Table 2.

Demographic Characteristics of Sample. Frequency (%) or Mean (S.D.) (Range) N=50

	All (N=50)	Recovered (n=25)	In Treatment (n=25)
Treatment experience
Little or no treatment	10 (20%)	7 (28%)	3 (12%)
Outpatient treatment only	15 (30%)	11 (44%)	4 (16%)
One structured program*	11 (22%)	4 (16%)	7 (28%)
Two or more structured prog.	14 (28%)	3 (12%)	11 (44%)
Age: Mean (S.D.)	32.7 (12.8)	36.0 (12.6)	29.4 (12.4)
Range	(18–64)	(20–58)	(18–64)
Race/ancestry/ethnicity
White/European desc.	42 (84%)	22 (88%)	20 (80%)
Black/African descent	4 (8%)	2 (8%)	2 (8%)
Asian descent	2 (4%)	1 (4%)	1 (4%)
Hispanic/Latino	2 (4%)	0 (0%)	2 (8%)
Education
H.S. dipl or less	2 (4%)	0 (0%)	2 (8%)
Some college/Assoc.	19 (38%)	8 (32%)	11 (44%)
Bachelor’s degree	8 (16%)	3 (12%)	5 (20%)
Some grad./Master’s	18 (36%)	13 (52%)	5 (20%)
PhD or MD	3 (6%)	1 (4%)	2 (8%)

^*“Structured program” refers to psychiatric hospitalisation for an eating disorder (not for symptoms like dehydration), residential program, or structured day program

Data Collection and Analysis

Interviews lasted 1½ −2 hours, were conducted by me, digitally recorded (with respondents’ permission) and professionally transcribed with identifiers removed and pseudonyms given. Recovered people were interviewed at a location of their choosing, usually one of two private offices on the medical campus; outpatients in these offices; and inpatients and day program respondents in private clinic locations at times that did not conflict with program activities. Participants were paid $40 cash.

The interview progressed from broad, relatively easy questions to more specific and difficult questions (i.e., the role of genetics). First, I elicited personal narratives, conceptions of eating disorders, and how/why they develop. Then I presented specific terms describing eating disorders--including “mental illness,” “brain disease,” “physical illness,” “choice,” and “lifestyle”--to elicit discussion of medical and other conceptions of eating disorders. Finally, I introduced the idea of genetics (“some say genes could play a role in eating disorders…”), and asked for reactions (e.g., “What is/was your reaction to the idea of a genetic causal factor?”) and speculations about what genetics would plausibly entail (e.g., “Would everyone with those genes develop [AN/BN]?”). I presented two hypothetical scenarios--a “media campaign” featuring genetics³ and a “genetic test”⁴–to prompt imagination of potential consequences in both personal and public arenas. My requests for speculative “best guesses” about new information and hypothetical scenarios elicited talk about genetics in a non-intimidating way, for the purpose of capturing how and why genetic discourse might make a difference for the respondent personally and in her social world. All of my references to genetic contributions were general enough to be open to multiple interpretations, including probabilistic polygenic and multifactorial models, but nevertheless may have cued determinism (Condit et al. 2009) and/or “genes for” AN/BN. At the end, I debriefed about the focus on genetics and explained that it did not imply a gene “for” eating disorders.

I analyzed transcripts in several stages. The first stage was note-taking during review of transcripts for accuracy and identifier-removal. When transcripts were entered into N6 (QSR 2002), I began “open” coding of a broad range of material relevant to research questions and created theoretical memos to record surprising observations, new hypotheses, and ideas for future coding (Lofland et al. 2005). I then carried out focused coding to capture the ways that genetics elicited medical concepts (described below). Finally, to check my analysis, I searched for interpretations of genetics that implied non-medicalising logic about AN/BN (“counter-logics”).

Focused coding of medicalisation themes followed two rules, narrow and broad. The narrow definition included only the following two themes: unambiguous medical language and concepts (e.g., words like “medical” and “disease,” and comparisons to uncontested disease like diabetes or cancer) and the perceived importance of treatment by a healthcare provider of any kind. The broad definition included spontaneous mentions of a third theme: expectation of biologically-based treatments if genes are known to play a role. Although biologically-based treatment could simply represent a change in the type of treatment, such treatment would only be available from physicians. I understood these themes as implicit “logics” connecting genetic etiology to medical conceptualisations. This analytic strategy set a high bar by requiring that respondents independently and spontaneously mention medical concepts; I never directly asked respondents whether genetics would encourage medicalised conceptions of eating disorders. Quotations are attributed using pseudonyms, recovery status (R=recovered, T=in treatment) and primary diagnosis (AN or BN).

RESULTS

Before the idea of genetics was introduced, most respondents (37 out of 50) had already brought it up themselves. Some described well-developed theories; others only mentioned it in passing, sometimes to refute the idea. Respondents held a variety of theories about how genes might play a role--sometimes multiple theories--including genes “for” eating disorders, other psychopathology (e.g., depression, anxiety, and addiction), personality traits, and large body size. All respondents thought eating disorders were problems, but not necessarily medical problems. Twenty-four (48%) disagreed, expressed uncertainty, or distanced themselves from the term “mental illness.” Yet more resisted “brain disease” (n=36, 72%). Endorsing or resisting biomedical terms could shape interpretation of genetic framing to support medical or non-medical understandings.

Respondents’ mixed endorsement of medical terms suggests there was “room” for additional medicalisation. Would genetic ideas be interpreted to make AN and BN seem more disease-like? Most respondents (n=37, or 74%) spontaneously brought up at least one of two medicalising themes, and nearly all (n=47, or 94%) brought up one of three themes to describe potential shifts in their own thinking about eating disorders, or that of other people.

Medical Theme 1: Genetics legitimise a disease

Respondents thought genetic ideas could help convince others – and sometimes themselves -- that eating disorders were “real,” “legitimate” and “valid” disorders. Against a background of skepticism about the reality of eating disorders qua disorders, genetics helped defend against a competing claim: that the eating disorder was one’s own choice and fault. Genes helped BN (and other mental illnesses) to be “legitimised as an actual disease,” and others would be “more likely to see it as a mental disorder,” a “mental defect or disease,” “more of a brain illness [or] mental illness” and “sick.” Respondents volunteered that genetic involvement would increase the resemblance of AN/BN to a variety of physical illnesses (diabetes, heart disease, gangrene, and cancer), neurological diseases (Alzheimer’s, autism, mental retardation) and schizophrenia.

Diseases were contrasted with volitional decisions; having a disease meant you were not to blame. According to respondents, genetics would help make the eating disorder something one has, rather than something one does or is: something external to the person, rather than emanating from the self. Vanessa likened it to cancer and explored how a genetic component would be perceived by others.

If it’s genetic, then I think that that would imply that the disease is not one that you choose. It would be viewed more like cancer. It’s just you have no control over it. And yes – and I think people would be more apt to feel sorry for you and have more empathy for you or sympathy. Because you can’t empathise unless you’ve been through it… Because they feel like, “You chose to be this way. You can correct it. This is just you.” But if you say – if you prove it’s a genetic thing, then it’s not. It takes that away.

(Vanessa, R-BN)

Rather than the eating disorder being “just you,” genetics reframes it as a disease for which sympathy is warranted. Respondents relied on a common sense meaning of a disease as something not chosen; genetic origins undermined the idea of eating disorders as “chosen” and thereby increased the resemblance to disease. Genetics could also help “validate it in the eyes of medicine” (Isabelle, T-BN), suggesting that even healthcare providers did not view eating disorders in medical terms.

Some laughed as they spoke about the conceptual transformation of AN and BN into “diseases,” suggesting the disease analogy was at odds with common understandings. Paula playfully speculated that with genetics, BN could one day be perceived as an un-stigmatised childhood disease, like chicken pox.

I can see it becoming something that is spoken of more in the terms of other just sort of things that you get (laughs). Like, diseases. Like, “Oh. I had the chicken pox when I was a kid.” “Oh. I had bulimia when I was a kid.” That societally might not be then associated with some kind of character defect.

(Paula, R-BN)

Similarly, Yvette’s laughter and vocal performance suggested that while a disease analogy could “help,” it might not be plausible.

I think it [genetics] might help society’s view of it. Like, that it is actually, like, a disorder and a disease and not just somebody’s choice to not eat. So I think then people might start, like, worrying about it like any other disease (slight laughter). Like, [said in a dramatic, serious tone of voice] “Oh my gosh. I could catch that.” Like, “I could get that from my mom.”

(Yvette, R-AN)

With genetics involved, AN seemed to Yvette like a disease one “catches” or “gets,” and thus humorously in conflict with choice.

Some respondents cautioned that genetic re-framing could go too far and oversimplify eating disorders, leading to false expectations of a simple cure. For example, Sydney said that genetics would make it more a “medical thing” and was concerned that this would wrongly ignore the complexity of the disorder, because “it is not just a disease.”

I think a biological link or sort of discovery in that field would make it more a medical thing. Which hopefully could lead to a drug or something that might be able to help it. So I’m totally in favor of that discovery (laughing). But at the same time I understand that it’s very hard to separate out. Because it is not just a disease. It’s NOT a disease. It’s a very complex cycle of mental and physical collection of symptoms.

(Sydney, T-BN)

Sydney contrasts the “complex cycle” of eating disorders with a simple “drug”, suggesting the two are incompatible.

From the opposite direction, if one’s own AN/BN were not genetic, it could seem less real compared to “genetic” cases. Gillian was concerned that her own AN, self-understood as non-genetically influenced, would seem like something she invented.

I: … what was your reaction to the idea of genetics as making – as playing a role somehow when you first heard about it?

R: Well, initially I thought “Well, no one in my family has it. So, does that mean I don’t really have it? And, it’s all in my head…. I would be very distressed. Because, I would think that meant on some level I was making it up. Which bothered me.

(Gillian, T-AN)

Victoria raised a similar concern:

R:… And then if I found out that I didn’t have the genetic make-up, then it would be like, “Well why do I have an eating disorder?”

I: What would it make you think?… Or feel.

R: That it was, like, all my fault. And this [BN] wasn’t even something.

(Victoria, T-BN)

If it were genetic, the disorder would seem to have more reality: it was “something”, it existed, rather than having been invented or chosen by someone.

Medical Theme 2: Genetics necessitates treatment by health professionals

In addition to genetic influence making eating disorders seem more like “legitimate, real diseases” (the first theme), it also made treatment for eating disorders seem more appropriate and necessary. If genes played a role, professional treatment was warranted because individuals could not be expected to overcome eating disorders by themselves.

Respondents presumed the need for some kind of treatment if genetics played a role, whether by a “doctor,” “therapist,” or other “professional.” For example, Joelle expected it would make people with BN see it as a “mental disorder” that needs to be addressed by a “professional” or “someone who knew more.” Thinking of it as genetic would encourage someone to seek help, just as they would for another medical condition.

…it might be good in some contexts for people to be able to know there was a root cause to this. And, that maybe they would actually be more likely to see it as a mental disorder and something they should see a therapist for…. I feel like when people think of things that are genetically pre-disposed, they think heart disease and cancers. And, those are the kinds of things that you put on forms. And, that are talked about. And, that you discuss with doctors. And, recognising it as a problem that you have that is genetic I feel like would make it – I don’t know. Somehow easier I feel, like, to discuss with a professional if you wouldn’t have initially.

(Joelle, R-BN)

Wendy (T-BN) also thought genetics would place BN under medical jurisdiction; “if it was just this genetic part of me, I would feel like only doctors could help me. As opposed to being thought patterns where… friends could help me.” A few respondents also envisioned professional long-term monitoring, to be “kept in check” as for a chronic condition. Colleagues, friends, and family members would also support professional treatment if genetically influenced; genetics might “make it a little bit easier to understand that I might need some help” (Gena R-BN).

Several respondents reasoned that professional help was justified because genes were too powerful to overcome on one’s own. An individual could not be expected simply to stop the behaviors if genes were involved. Betty said a genetic causal factor would help overcome her concern about the expense and trouble of her inpatient treatment for AN:

I mean I would be more accepting that I did have an eating disorder if I knew it was genetic. Or, could be. And, I would go get some professional help or the in-hospitalisation program. Because, I would… not expect myself to be able to do it with no help…. Because, my coming here I thought “Oh, my goodness. I don’t want anybody to pay that amount of money. Why can’t I do this myself?”

(Betty, T-AN)

Obtaining inpatient treatment was easier for Betty to justify if the condition were “genetic” because recovery would not be something she could “do” by herself. Mireya (T-BN) also thought genetics encouraged a medical conception for people with bulimia (though she herself was not convinced); people with BN “might think it’s not in their willpower or in their hands. That it’s outside of their ability to change anything. That it needs to be treated.” Fran, who reported recovering decades ago by stopping bingeing and purging just as she had quit smoking, speculated that if she knew genes were involved, she would have sought professional help “instead of just diving into it and taking care of it myself” (Fran R-BN). Yvette made explicit how reduced responsibility might translate into a more passive patient role; “instead of like, ‘You need to look and mentally understand why you don’t want to eat, ’… it would be ’You just need to go to the doctor. And they’ll handle it for you.’”

To summarise, genetic involvement in eating disorders prompted three-quarters of respondents (n=37) to bring up one or both themes illustrated above from my narrow definition: (1) medical concepts, terminology or comparisons to uncontested medical diseases, or (2) perception that treatment by health professionals would be appropriate.

Medical Theme 3: Genetics imply biologically-based treatments

Under an expanded definition of medicalisation with three qualifying themes (the two described, plus biologically-based treatments), ten more respondents were added, for a total of 47 respondents (94%). They imagined that if genetics played a role, there might be new biologically-based treatments: medications, injections, gene therapy, surgery and the like. When Amy (T-AN) was asked about good or bad effects from the hypothetical media campaign, she immediately imagined new biologically-based treatments: “They would, like, probably try to produce more medications. Like, there’s a whole line of meds for depression. A whole line of meds for bipolar and schizophrenia. They may try to develop one for anorexia.” Some would welcome such a possibility: “Well, if they found a genetic link, then hopefully they could find some kind of treatment early on… Gene therapy or something.” (Deena, R-AN) Others were more cautious. For example, Mireya objected to a reductive medicalisation because, “in order to really, like, heal somebody with an eating disorder you really need to think of it holistically and not like as a medical disorder that’s genetic, that’s purely genetic.” Although none of my questions or hypothetical scenarios suggested a “purely genetic” explanation, some respondents interpreted them as such; genetics connoted biological interventions.

Three respondents did not mention even one of the three medical themes (Jackie, T-BN; Liana, T-AN; and Emma, R-BN). They stated that genetic involvement would lessen perceptions of volition, but did not link this to medical concepts or treatment in their answers. Initial endorsement of medical terms such as “mental illness” did not appear to affect later interpretation of genetics; virtually all respondents made a similar interpretation; the lack of variability precluded statistical comparison.

Counter-logics

To further check my conclusions, I searched for “counter-logics” by which genetic framing elicited non-medical conceptions of AN or BN. A few who articulated the medical logic also thought genetic etiology could connote futility of treatment and/or imply that genetic AN/BN is “part of who I am, so I’ll just accept it” (Yolanda, T-BN). No respondent argued that genetic etiology connoted natural and acceptable human variation (i.e., diversity rather than disease). However, such a view was raised--and quickly rejected--by one respondent, well before I introduced genetics. While discussing AN as “lifestyle,” Gillian mentioned that AN could be seen as a genetic capacity to live on very few calories. In response to probing, she offered that different genetic theories could support different understandings of eating disorders:

I think if you look at it as you’re pre-dispositioned to have an eating disorder, it sounds more like mental illness. If you say you’re pre-dispositioned to be capable of surviving –of subsisting on 400 calories and being a very low weight, that sounds more like lifestyle.

(Gillian T-AN)

Gillian clearly endorsed the “mental illness” genetic theory, not the “lifestyle” one. In addition, a few initially understood genetics to imply conserved traits that are evolutionarily adaptive, but doubted the application to AN/BN (i.e., AN/BN are not adaptive, and therefore not “genetic”).

DISCUSSION

Results show that in the context of eating disorders, genetic framing tends to support medical definitions of eating disorders, even in a sample where half resisted medicalised terms such as “mental illness.” I identified three “logics” underlying such medicalisation: it makes an eating disorder seem more like a “real disease”; more like something a professional should handle; and more likely to require biologically based treatments (that only medical professionals can provide). These themes were highly salient to respondents; 47 out of 50 raised at least one despite never being asked directly about them. I also identified several counter-logics by which genetic etiology could discourage medicalisation. This study provides empirical support and individual-level detail for the hypothesis that genetic framing of an officially medicalised psychiatric condition will support medicalised understandings (Conrad 2007; Shostak et al. 2008).

Respondents took for granted widespread skepticism about the legitimacy of eating disorders, and a binary opposition between valid somatic disease and invalid moral or psychological volition. In the context of public ambiguity about whether eating disorders are to be considered “real” psychiatric or medical disorders or personal choices, genetic framing strengthened the disease definition and medical treatment-seeking. Respondents conceptually lumped genetically influenced disorders with “real” diseases and split them from ongoing choices (Zerubavel 1996). Some respondents imagined using genetic framing in their social contexts to draw or reinforce a desirable conceptual boundary (Pachucki et al. 2007) between eating disorder as a medical problem and eating disorder as a moral problem; i.e., the conceptual boundary between “sickness” and “badness” (Conrad and Schneider 1992).

If genetic framing draws, redraws, or reinforces the boundary between eating-disorder-as-choice and eating-disorder-as-real-disease, it can change expectations about socially appropriate action. Respondents interpreted genetics to imply social acceptability of the need for treatment, biological or otherwise. Interestingly, genetic involvement legitimised treatment itself (Theme 2), not just biologically based treatment (Theme 3); if “genetic,” then it is for a professional to handle. It should be noted, however, that while genetic etiology can alleviate responsibility for having a condition, it may not alleviate responsibility for managing it and staying well (Condit et al. 2006; Rose 2007; Weiner 2009); the “sick role” entails an obligation to cooperate with healthcare providers (Parsons 1951, Williams 2005, also see Rose 2007 on genetic responsibility). The idea that “genetic” problems entailed pharmaceutical treatment (Marteau et al. 2004) was resisted as implausible, inappropriate, or undesirable by some (also see Easter 2012), even as others hoped for it. The molecularisation (Rose 2007) of AN/BN treatment had reductionist biopolitical implications for some respondents: if others presumed the availability of a pill, those with AN/BN might be blamed all the more for “choosing” not to take advantage of it. Similarly, genetic framing could discourage interventions emphasising familial and societal factors, even though clinicians and scientists resist such conclusions (Striegel-Moore and Bulik 2007; also see Pickersgill 2009, 2012).

One strength of this study is its sample of individuals not already participating in genetic studies, who may be more familiar with and approving of genetic explanations (Laegsgaard et al. 2010; Meiser et al. 2005). However, my sample may hold more medicalised conceptions than the wider population of people with eating disorders, because all defined themselves to have had an eating disorder; none had “active”, untreated AN/BN; and half were recruited through a hospital-based program. Nevertheless, half did not endorse the term “mental illness.” These data are arguably limited because the conceptual reclassification described is due to hypothetical genetic information provided in the context of an interview, not a real-world effect of knowing one’s individual genetic risk. With regard to measurement, the interview question about the hypothetical genetic test may have been interpreted to suggest doctor-administered tests and simplistic theories of genes “for” AN/BN (Kendler 2005), despite intentional vagueness on many counts: When describing the test I mentioned environmental factors, used probabilistic rather than deterministic language, referred to “genes” not “a gene” (in keeping with polygenic perspective), and did not refer to doctors (to accommodate direct-to-consumer testing). This question was useful for eliciting discussion about whether it is desirable to think of and/or present oneself as having a genetic predisposition, given that one has already developed an eating disorder. Respondents may have shifted between deterministic and other understandings of genetic influence throughout the interview; my approach did not track such fluctuations. Nor could I track shifts between interpretations of genetics that anticipated a critical or sympathetic audience. I suspect many of my findings were generated in relation to a skeptical public that knew little about eating disorders and held simplistic understandings of genetics.

CONCLUSION

These findings arguably represent one of two broad genetic behavior-framing strategies: human behavior as a reflection of abnormal disease or normal diversity. If a diagnostic category expresses “what society is prepared to accept as normal [vs.] what it feels should be treated,” (Jutel 2009: p. 279) genetics help stabilise it with evidence that it is an unconstructed natural category of pathology (Hacking 2001, Jutel 2009). A similar phenomenon is observed with neuroimaging (Buchman et al. 2012). In general, I expect that the cultural logic by which genetic involvement implies medicalisation will be relevant for other behavioral disorders such as conduct disorder and addictions (e.g., gambling, internet, as well as substances). Genetics provides a way of conveying to others that behaviors are not mere choices or habits, with the result that they may seem more incontrovertibly medical and require professional help.

However, a second, entirely different genetic framing strategy could support definitions of behavior as acceptable, normal human variation not requiring intervention, that is, as a “human specificity” like race, sex, or sexual orientation (Ortega 2009: 425). Genes can be used to strengthen claims that a behavior or trait is natural and immutable and ought not to be corrected. Such reasoning obtains in genetic “just-so” stories asserting an evolutionary reason for observed behavior (e.g., domestic women; unfaithful, violent men; see Fausto-Sterling 1992). Such genetic behavior-framing has been used to resist medical intervention and discrimination not only related to sexual orientation (e.g., Brookey 2001, Shostak et al. 2008) but also to “neurodiversity” on the autism spectrum (Ortega 2009, Bumiller 2008). There is an interesting parallel to the shifting interpretation of genetic “variants” (i.e., differences from the current “reference genome”) as deleterious mutations or benign individual differences (see Rose 2009). The line between disease and acceptable diversity increasingly may be blurred as “mental abnormality has practically become normalised” (Rose 2009: 79). Yet maintaining a distinction between a “condition to be treated” and other atypical behavior is economically necessary for individuals whose behavioral phenomenon must be defined as a disorder in order to be treated (and covered by insurance).

Genes need not provide a warrant for medically “fixing” a trait, though clearly they do so in this sample of respondents. The polysemic capacity of genetics might be exploited differently by subgroups with different motivations or understandings of a trait as either ability/capacity or symptom/disorder. Within the neurodiversity movement there are “pro-cure” and “anti-cure” contingents (Ortega 2009), which employ biogenetic discourses for mutually incompatible ends--i.e., to argue for or against treatment. A similar divide exists in eating disorders; the “pro-anorexia” contingent (Fox and O’Rourke 2005) treats it more like an identity than an illness, though even some who seek treatment see it as both (Rich 2006). A “pro-anorexia” sample might have interpreted genetics differently than my sample (as hinted by respondent Gillian, above). Perhaps “disease” and “diversity” contingents could unify around a hybrid genetic framing strategy that supports both definitions: helpful adaptation under one set of environmental conditions, but problem in another (e.g., the “adapted-to-flee famine hypothesis” of restricting behavior in AN, Guisinger 2008; also see Belsky and Pluess 2009, Homberg and Lesch 2011).

Biological processes are part of all thought and feeling, disordered or not: the biological underpinnings of psychiatric disorders do not actually distinguish them from normal healthy behavior. Yet, whether interpreted to connote “disease” or “diversity,” genes support a claim of innocence (Luhrmann 2001)—it’s my body, not something I chose or can directly control. This claim of innocence can lead either to a request for forgiveness and assistance (disease), or for toleration, if not respect (diversity). Under both interpretations, “genetics” is arguably mapped onto the “body” side of body-mind dualism. But it may signify otherwise in the future, particularly if genetic effects are understood not only as ubiquitous (e.g., Turkheimer 2000), but also highly contingent on other genes, behaviors, and environments. In the future, body-mind dualism may be reinscribed in new ways, with finer-grained genetic distinctions replacing the current implicit dualism of “genetic vs. not genetic.” Such reinscriptions are likely to be developed and negotiated in relation to social understandings of the disorder, just as with eating disorders.