Eating Disorders & Socioeconomic Status in the Community

The assumption that eating disorders only impact young, white, affluent women seems so out dated as to be laughable – and yet somehow this image persists, one of the most prominent stereotypes about eating disorders. It’s a damaging stereotype on so many levels; as we know, stereotypes about who might suffer from an eating disorder can lead people to feel that they don’t actually have an eating disorder and de-legitimizing their distress. The stigma that stems from having a body not expected to have an eating disorder can lead people to avoid seeking treatment out of fear of being dismissed by doctors, not thinking the type of treatment on offer will be appropriate or helpful, and more. Somehow, in the face of this, the image of the privileged and vain young woman who chooses to not eat marches on. And it is a shame.

Researchers are exploring stereotypes such as these versus the actual presentation of eating disorders in communities. While studies that look at demographics in treatment centres are unlikely to disprove these stereotypes – after all, if people who don’t fit the expected are not seeking treatment, they won’t be included in such studies – community based studies have a higher likelihood of revealing nuance in who might suffer from distress around food, weight & shape.

The Study

Mitchison, Hay, Slewa-Younan & Mond (2014) conducted a study based around population surveys in 1998 and 2008. By looking at a sample of the general population instead of looking at demographics within a treatment centre, they are likely to catch undiagnosed or untreated eating disorders. If that’s the case, you might wonder, why don’t people always conduct this kind of research? There are a number of drawbacks to this kind of study that make this research logistically challenging, including that population based surveys don’t always include eating disorder-relevant questions and that it is fairly time consuming and costly to conduct quality population based research.

The data isn’t perfect either; though samples are generally representative of the people within the population, one can’t diagnose people within that sample with an eating disorder based on their responses, so we’re capturing eating disorder-like behaviours here, not diagnosed eating disorders (which, of course, could also be seen as a positive thing). You could ask people if they’ve been diagnosed with an eating disorder as well, but then you’d face similar drawbacks to treatment-based studies.

A benefit of population surveys like the one in this study is that they tend to be repeated over time with only minor question changes, which means that researchers can compare data at one point to data at a later point. Importantly, though, there is no guarantee that they are comparing data from the same people at both time points – that would be a longitudinal study, where the same people would be followed over time. Here, researchers took a cross-sectional approach, taking data from the 1998 survey and comparing it to data 10 years later, in 2008. Think about this as a study of trends within a group, not trends within individual people. They were interested in knowing more about the link between eating disorders and:

  • Sex
  • Age
  • Urban vs. rural living status
  • Household income

They were also interested in whether those with eating disorders see a drop in quality of life over time. It would have been amazing to know more about the links to other variables like race, sexual identity, etc., but alas the researchers did not look specifically at those categories.

The researchers conducted this study in South Australia, where there is an annual household survey that includes face-to-face interviews. There were 3010 participants in the 1998 round, and 3034 in the 2008 round. Interviews include questions related to health and demographics, including questions from the Eating Disorder Examination questionnaire such as whether participants:

  • Binged
  • Purged
  • Engaged in extreme dieting

For general health, the interviews included questions from the Medical Outcomes Study Short Form; these specifically focused on quality of life as it related to health, both physical and mental.


General demographics were fairly similar in 1998 and 2008, which makes for good comparison: for instance, there was a nearly equal distribution of men and women in each sample, and participants were primarily married or common law (61.5% in 1998 and 62.7% in 2008). Participants in 2008 were slightly older, of a higher weight, more likely to be employed, slightly more educated and urban-dwelling.

More participants in general binge ate in 2008 versus 1998, but the increase was more marked amongst those of a lower income – compared to those in higher income brackets, there was a more significant, substantial increase in “objective binge eating” (from about 2.5% of the population in 1998 to approximately 5.5% of the population in 2008 in this income bracket). Binge eating also seemed to more significantly impact physical quality of life among participants in rural settings.

Purging appeared to be on the rise amongst older participants, which I find quite interesting in terms of breaking the stereotype that eating disorders only happen to the young and silly. Men and those from lower socioeconomic statuses also seemed to have greater increases in purging. Note, however, that purging was still not terribly common in the sample, increasing from about 0.8% to 1.1% of those at a lower socioeconomic bracket between 1998 and 2008, for example.

There was also more extreme dieting in the 2008 sample, but only in more urban areas; results for increased extreme dieting were not significant, however, nor did they appear to impact quality of life. There was a small relationship between extreme dieting and poorer quality of life related to physical health amongst people from lower income brackets.

So, What Does it Mean?

I thought this study was quite interesting, at least in terms of highlighting population trends. Of course, there are many aspects of eating disorders not captured in these results – notably, restriction and over-exercise, which may or may not be caught in “extreme dieting.” There is a distinction to be made between extreme dieting and an eating disorder in terms of both tenaciousness of the behaviour and orientation toward the behaviour (i.e. the distress underlying the restriction) that would need to be teased out before saying seeing an increase in extreme dieting indicates an increase in eating disorders.

One of the most interesting points, actually, is around how increased binge eating behaviours appear to be tied to steeper declines in quality of life related to physical health for those living in rural areas. As the authors note, these people may not have the same access to treatment for physical ailments and/or may feel stigmatized and unable to seek care. This also stands for the observed link between declines in quality of life linked to extreme dieting for those from lower socioeconomic statuses.

The authors oddly refer only to the stigma around disordered eating amongst men; I would argue that there is almost certainly a specific stigma associated with being poor and having an eating disorder that I think would be worth exploring in more detail. What are the implications of living in a situation where food is more scarce and suffering from an eating disorder?

Mitchison et al. suggest that these results could help us to develop “more gender- and age-neutral, or specialized, health promotion strategies” (p. 8). I actually think that these results are more useful in highlighting the need for a better continuum of care for eating disorders such that those in rural areas or lower socioeconomic statuses could access care; I suspect this need would be even more marked if more demographic variables were included.

Of course, we do need to work on decreasing stigma so that people then feel more comfortable seeking that treatment. The results might also help to highlight, as the authors point out, how we need to broaden the types of samples generally used in eating disorder studies. It’s never just one thing, is it?

Mitchison D, Hay P, Slewa-Younan S, & Mond J (2014). The changing demographic profile of eating disorder behaviors in the community. BMC public health, 14 PMID: 25213544


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.