I thought about writing a post about the factor structure of popular eating disorder scales to celebrate my completion of an advanced statistics course in structural equation modelling. When I sat down to read some articles about that, though, I found myself side-tracked– and thoroughly uninterested in deconstructing scale psychometrics. So with a promise to do that at some point, I return to a favourite topic of mine: culture and eating disorders.
When I was writing about culture and eating disorders for the blog last year, I received quite a few requests for articles about eating disorders in developing countries. I suspect that the desire for this kind of article stems from a need to highlight (for the doubters) that eating disorders are serious mental health issues that can impact anyone who is predisposed, regardless of whether they live in a media-saturated landscape or not. As I noted in the series, eating disorders and culture live in complex relation – saying that eating disorders are solely caused by either biology or culture is to miss the point of the complex interrelationship between bodies and worlds.
What I have also tried to highlight in this work, though, is how culture doesn’t mean media. Culture is a whole lot more than the popular magazines that litter our doctors’ offices and coffee tables. Culture includes all manner of beliefs and customs, arts and intellectual endeavours. Thinking about eating disorders requires that we expand what we mean by culture, and also to situate eating disorders not only within “culture” as such but also social context including political and economic milieu.
None of these things are static; as we are a part of our sociocultural and political-economic milieu, we are continually and actively involved in creating and re-creating these fields ourselves. Though we might not feel powerful or even able to make any change at all (particularly in the face of capitalism, which has a history and metamorphosis as rich as it is troubled), the things that we do are continually holding up, changing, or breaking down social contexts (often all three at once).
So, you might ask, why is this important to eating disorders? In my view, it is important because when we engage in debate about “whether eating disorders are culture bound” or talk about the sociocultural aspects of eating disorders, we need to look beyond the TV set, computer screen, and magazine page. Media literacy, as I’ve said many times, is no guarantee against a) feeling bad about yourself or b) developing an eating disorder, if you’re so predisposed. Neither are genes alone enough to cause an eating disorder. There is more going on in this picture, so let’s unpack it a little more.
The Study
Pilecki, Salapa & Józefik published an article in the Journal of Eating Disorders this year, looking at eating disorders in Poland. From the outset, the authors align themselves with those who consider eating disorders “culture-bound” – that is, tied to a particular kind of culture. The authors do focus primarily on the trappings of culture that I just noted were not the only aspects of culture relevant to our understanding of eating disorders: mass media, thinness ideals, and a hyper-focus on body scrutiny. They also consider how others in the lives of those with eating disorders may perpetuate problematic body relationships through comments and teasing, or by exhibiting disordered eating patterns themselves.
We’ve heard all of this before – we know that body ideals aren’t helpful, we know that they are not enough to cause eating disorders and that a body positive environment is not enough to protect against all eating disorders. Where things start to get interesting, in my opinion, is when the authors begin to explore Poland as a former Eastern Bloc country.
For those who didn’t pay attention in history class – or who are younger than I am (I was born in 1989, the year of the revolutions that led to the dissolution of communist rule in these nation-states) – the Eastern Bloc was comprised of a number of countries that were under Communist leadership from the late 1940s until the late 1980s/early 1990s. In 1989, a series of revolutions brought about massive change in the social, political and economic structure of many Eastern Bloc states, including what is often referred to as Westernization (a thinly veiled term for capitalization, really, meaning a huge intensification of trade with nations outside of the Eastern bloc, restructuring of social services, work, and class structures, etc.).
Pilecki et al. note that prior to 1989, the prevalence of eating disorders in Poland was quite low; they attribute an uptick in cases to media sources primarily, but were interested in continuing to explore other Westernization-related socio-cultural correlates. To me, it is much more interesting to think about how people relate to their bodies as individualistic, productivity-machines in Western political-economic context, but that’s just me. In this article, they explored socio-cultural risk factors for eating disorders in the Polish context, using data collected between 2002 and 2004, some 20 years post-Eastern Bloc.
The authors surveyed:
- 47 women with anorexia-nervosa restrictive subtype (AN-R)
- 16 women with anorexia-nervosa binge-purge subtype (AN-BP)
- 34 women with bulimia nervosa (BN)
- 19 women with eating disorder not otherwise specified (ED-NOS)
- A control group of 85 girls and their parents
They asked these individuals about their beliefs about issues that have been considered risk factors for eating disorders. I’ll flag this methodological issue right away as one that irks me and yet seems somewhat unavoidable in social research: using pre-determined “risk factors” without leaving open the possibility of factors not often considered a part of “socio-cultural” (like oh I don’t know… political economic factors).
Here’s a sample of some of the questions, so you can get a sense of what they asked (the article is open access so you can see all of the questions on page 4 here):
- A person should control his various weaknesses
- It is important to the members of my family that we be “up to date”
- I know many people who have problems with food”
- In my social circles, appearance is crucial to a successful social life
- When I see a thin model, I feel fat and unattractive
- My greatest desire is to know how to control my life
- What my family considers delicious food is unhealthy in my opinion
An important note: this is not meant to be a scale that assesses for eating disorders or to be used as a diagnostic instrument. Nor was this study a comparative prevalence study, though the authors write a lot about prevalence.
Results
Most of the people they surveyed who had BN and AN-R were situated in rural villages (about 30%) and those with AN-BP lived in small towns (about 56%) whereas the control group lived in Krakow (about 77%). I think this is a fascinating finding that runs counter to what we might expect if we assume that Westernization is directly tied to the prevalence of eating disorders. Assuming that big cities like Krakow are more “Westernized,” if we assume that Westernization increases the prevalence of eating disorders we would expect to see fewer eating disorders in rural settings. While this is not a study specifically exploring prevalence, and so we can’t and shouldn’t read too much into this finding, it is still interesting.
The authors suggest that this difference might be tied to how living in rural areas may actually reflect social privilege. I’m not sure I get behind this argument, as it seems to reflect an assumption that eating disorders are more common in middle to upper class groups, which contradicts their other key finding that “social affiliation” was not actually different between ED and non-ED groups (i.e., there was not a strong association between being lower or middle or upper class and being in the ED group or control group).
Other markers of social class were somewhat conflicting. Parents of those with eating disorders in this study were more highly educated than national average. Those in the BN group were more likely to come from single parent families. It is hard, from this kind of conflicting information, to draw solid conclusions about “what kind” of person in Poland has an eating disorder – which in my view reflects the heterogeneity of those with EDs.
It’s also interesting to look at how the families of those with EDs related to food. In general, the parents of those with EDs were not more appearance or food focused than those without. The authors note that in the 1980s, Poland was in the midst of a serious economic crisis. During this time, having more access to food signalled economic success. In the movement out of the economic shift, the meaning of food consumption and bodies shifts – but slowly. The authors suggest that parents and children might orient differently to bodies and the meanings associated with different bodies – possibly, they argue, children consider thinness desirable where their parents might continue to see thinness as indicative of economic loss.
Implications
I would love to see these results analyzed more thoroughly in the context of what it means to be an adolescent in a country still recovering, years later, from serious economic crisis. Beyond valuing or not valuing thinness, what kinds of pressures exist to be a productive citizen? What is the specific meaning of social class in a formerly communist nation? This kind of thing fascinates me far more than whether people buy into thinness ideals.
Overall, this study provides an interesting look at eating disorders in a context I had not read much about. It reminded me how much of a nerd I can be about the political economy and history, so that’s also fun. What’s your take?
References
Pilecki MW, Sałapa K, & Józefik B (2016). Socio-cultural context of eating disorders in Poland. Journal of eating disorders, 4 PMID: 26998306
Rathner et al. published “Prevalence of eating disorders and minor psychiatric morbidity in Central Europe before the political changes in 1989: a cross-cultural study” in 1995. Using the EDI and the GHQ, the authors determined that Hungary (behind the iron curtain at the time) actually had twice the prevalence rates of bulimia compared to Austria and GDR (German Democratic Republic (confusing name no?) otherwise known as East Germany at the time — also behind the iron curtain).
So the interesting thing about the study (access to complete paper here: https://www.researchgate.net/profile/Michael_Geyer2/publication/14612549_Prevalence_of_eating_disorders_and_minor_psychiatric_morbidity_in_central_Europe_before_the_political_changes_in_1989_a_cross-cultural_study/links/549c77ea0cf2fedbc30fdd46.pdf) is the use of the General Health Questionnaire (GHQ) which is commonly used to uncover non-psychotic psychiatric problems in the population. The authors used the GHQ to uncover “psychatric caseness” — basically looking to uncover predisposition to eating disorders.
“General vulnerability factors and major elements of ED were present in Hungary and the GDR even before the political changes and are not dramatically different from Western countries where these problems have received more attention. Given this fact, perhaps the only additional requirement to translate this pre-disposition into ED is dieting which may be sufficient to precipitate ED in those who are vulnerable, as the rate of dieters in a population may determine the prevalence of ED. It is likely that Western culturally mediated dieting behaviour was perceived e.g. via electronic mass media in the CCEE [Central and Eastern Europe], at least in students [they were measuring these things in students obviously]; personal and cultural contacts, either direct or indirect, have been hampered in some way by the ‘Iron Curtain’, but not totally excluded…”
Rathner and colleagues reference 3 studies from Czechoslovakia, Poland and Russia on eating disorders that pre-date 1989 but I can’t seem to locate them online sadly. In the Szabó and Túry (1991) “The prevalence of bulimia nervosa in a Hungarian college and secondary school population” paper, the authors list the prevalence of BN at 1.3% in the college population which seems in line with the prevalence of BN at the time across Western countries in 1991 [Stein, David M. “The prevalence of bulimia: A review of the empirical research.” Journal of Nutrition Education 23, no. 5 (1991): 205-213.]
The problem with all of this stuff is the “guessiness” of how authors decide to frame reasons for prevalence one way or the other. But there seems to be plenty of evidence in your review and the original smattering of papers on the topic of eating disorders minus Western culture that the predisposition for EDs is universal in human populations *and* the sociocultural/environmental frameworks in which that presdisposition might be activated are infinitely complex as is the way in which sociocultural/frameworks further shape the progression of the condition (once activated) as well.
Sorry for blog post response on your blog!
Haha you’re welcome to leave lengthy comments any time. That’s really interesting re: Hungary having twice the prevalence rates of bulimia compared to Austria and GDR at that time. With all of these things what I really wonder is about diagnostic sensitivity, and how much people (read: doctors, psychs) just catching EDs has to do with what the prevalence rate is – and beyond that, actually, cultural aspects around the ability or desire to even seek out help and care. These things are not insignificant factors in what prevalence rates end up being! Agreed re: the “guess-i-ness” of it all. There’s a whoooole lot of complexity and nuance that no single measure, scale, or prevalence rate is going to capture..
Another study trying to prove that EDs are caused by Western culture and the Thin Ideal. But what struck me most was their reaction when the results appeared to disprove their hypothesis.
“What may be significant here are the relationships identified by items 12, 16 and 18, which indicate the irrelevance of thin pressure in the ANR group”.
They later add…
“This brings the credibility of any self-assessment test performed by girls suffering from ANR into question, including self-assessment tests of cultural issues”.
If girls with ANR disprove the researchers’ hypothesis they are dismissed as being unreliable.
One problem with this study, and so many others, must surely be the use of a single word “thin” to describe both desirable slimness and rank emaciation, when they are completely different things. It’s perfectly possible (and many of us know this from personal experience) that someone who becomes emaciated never at any time intended to become slim – they may in fact have been totally slim to begin with. Trying to become slim and becoming emaciated are completely different things. I find it really frustrating that so many researchers seem completely unable to grasp that. Both the research they choose to carry out and the way they present the results are completely blinded by their preconceptions around the thin ideal. I for one wish they would stop. IMHO it’s no better than claiming “refrigerator mothers” cause autism, or any of those other harmful claims of the past.
Re: the discussion around prevalence studies. Check out this recent study which calculated that 42% of teenage girls in Pakistan have anorexia nervosa.
http://www.ncbi.nlm.nih.gov/pubmed/26870084
The Pakistan study I cited basically categorised every respondent who indicated they were dieting, or believed themselves to be too fat as “having AN”. Weight was not considered to be a factor, and there was no difference in average weight between the girls “having AN” and those “not having AN”. Both groups had an average weight well within the healthy range, given their age. This study is an example of reductio ad absurdum in the conflation of AN with dieting and body image, but it is only an overtly extreme example of what happens in research all the time. Researchers take it for granted that AN is nothing more than an exaggerated attempt to look good through dieting: so these are the factors they keep examining.
I keep coming across studies that show prevalence is remarkably similar across the world, regardless of culture.
I always have to wonder if AN is indeed present many places but just presents differently in different cultural contexts. Again, the lack of sensitivity in instruments used to assess evidently limit our detection capacity; also, I feel that a lot of the treatments available are not cross-culturally effective, which makes me slightly concerned about just slapping the label of AN on anyone…
I’m not sure they were really trying to prove that AN is CAUSED BY the thin ideal; I think they just fall into the usual trap of finding what they were looking for. I think what we don’t talk enough about it that most (all?) of the instruments we use to assess were developed in a very specific time and place, and in a context wherein there are big assumptions about what an eating disorder is and who it impacts. I think what they do miss, in stating that self-assessment isn’t always super credible (which is for sure a part of it, people in general are not the very best at self-reporting), is that the assessment instrument is itself socioculturally located.
It’s a good point that “thin” is often used interchangeably with emaciation or malnutrition; there are shades of thin and there’s “ideal thin” and “crossing the line thin” – i.e., also illustrative of the trouble with how bodies are socially surveilled. There’s privilege associated with thinness, but once you become “too thin,” this is no longer seen as desirable.
There is much this study that consists of setting a hypothesis, finding the hypothesis is not proven, then questioning the results in order to uphold the hypothesis.
I want to examine one particular example in more detail.
The researchers write:
“What may be significant here are the relationships identified by items 12, 16 and 18, which indicate the irrelevance of thin pressure in the ANR group”.
12. When I see a thin model, I feel fat and unattractive.
16. To be more attractive, I should be skinnier.
18. I am jealous of a model’s slim appearance.
(Girls with AN-R scored lower than the healthy control group for all of these questions.)
So the researchers conclude: “It is possible that thin pressure was relevant before the manifestation of ED, but then lost its relevance following significant weight loss”. And they also then go on to suggest that girls with ANR are unreliable at filling-in self-report questionnaires.
They simply cannot let go of their preconceptions, even when they have conducted research and the data is staring them in the face: girls with ANR are LESS influenced by the thin ideal then a control group, not more.
Speaking as someone with the condition, I am not merely frustrated by this, I actually feel personally attacked – and silenced. These girls have taken part in the survey in good faith, stated clearly that they are not motivated by the thin ideal, and the researchers just won’t listen. They conclude that the girls are either lying, don’t know what they are saying, or must have been motivated by this ideal at an earlier stage.
In my experience AN is in no way about looking good – if anything, it is more about looking ill or broken, in order to properly reflect what is within. It never once crossed my mind than going from a healthy weight to severely underweight was an improvement in appearance. It was an inner compulsion, triggered by trauma. I believe that that sort of thing happens to all sorts of people, all across the world, and that it is utterly irrelevant whether the surrounding society has some belief around “slim is beautiful”. I don’t even begin to understand why researchers cannot see the difference between thin as beautifully slim and thin as desperately broken. I am totally in accord with the ANR girls in Poland. I don’t agree at all with statements 2, 16, or 18 – either now or before I got ill. Such ideas have absolutely nothing to do with the development of anorexia, which is fundamentally a withdrawal from the world, not an engagement in its values.
All research of this type is troubling, but this study really seems to cross some lines in continuing to espouse the harmful dated prejudices, even when their results seem to disprove them. The questions they posed were 100% loaded in favor of the thin ideal hypothesis – and even then they didn’t get the results they hoped for.
I can definitely see how it would be upsetting. I have to say that the approach is not uncommon, which is certainly frustrating – but I think illustrative of broader problematic trends in research, as opposed to a particular fault of the authors. Not that I’m coming to their defence necessarily, but situating the critique within a broader research enterprise that calls for research that replicates instead of challenging. It’s something I certainly take issue with. Being critical in research is not as welcome, in my opinion, as it should be. An important note, I think, is that their questions were not drawn from thin air; they were using a questionnaire established by other researchers. I do agree that they read into the response what they were looking for; as I wrote in the post, I think there was more to unpack in terms of things other than the thin ideal.
I’m so sorry to hear that the article made you feel silenced – hopefully my post did not do that? I think there is/at least there should be room for multiple ways of experiencing EDs, whether thinness ideal-related distress is salient or a factor for the person. It’s certainly unfortunate, as I wrote here & continue to write everywhere that will listen (and sometimes where they won’t), that “sociocultural” gets taken as shorthand for “Western media.” It’s so much more complex than that.
Great research find and analysis. Related: I am listening to your podcast interview on Finding Our Hunger. Fantastic stuff.
Thanks JD!