Living in a Large City: A Risk Factor for Bulimia Nervosa?

The link between urban living and mental disorders such as schizophrenia and depression has been known for quite some time (Sundquist et al., 2004). In one study, Sundquist et al found that individuals living in a densely populated area had a 68-77% higher risk of developing psychosis and 12-22% higher risk of developing depression.

The question then arises, do eating disorders follow a similar pattern? And if yes, what are some possible explanations? Certainly we know that both genetic and environmental factors are involved in the development of eating disorders, but what specific factors and to what extent remains unclear.

In this study, Gabrielle E. van Son and colleagues set out to explore whether increasing urbanization was an environmental risk factor for the development of eating disorders.

In order to answer this question, the researchers had a network general practitioners (GPs) record each newly diagnosed case of anorexia or bulimia nervosa between 1985-1989 and 1995-1999. The total number of patients served by the group of GPs was around 1% of the population in the Netherlands.

The researchers then divided the population under study into three groups: (1) rural areas, defined as 20% or more of the population was involved in agrarian labour; (2) large cities, defined as having more than 100 000 inhabitants; (3) urbanized areas which included all other areas.

Then the calculated the age-adjusted incidence rate ratios among the different areas. I’ve summarized the results below:


Mean age of detection in this study (among women):

  • Anorexia nervosa: 22 years
  • Bulimia nervosa: 27 years

Age-adjusted incidence rate ratio (in other words, the ratio between the rate (adjusted for age) in one area compared to the other, * = indicates significant difference in rates between the two areas):

Anorexia nervosa:

  • Urban vs. rural: 0.90
  • Urban vs. city: 0.88 
  • City vs. rural: 0.98

Bulimia nervosa:

  • Urban vs. rural: 2.60*
  • Urban vs. city: 2.10*
  • City vs. rural: 5.22*

So what do those numbers mean? Essentially, the authors found no differences in incidence rates for anorexia nervosa in rural, urban, or city areas. On the other hand, incidence bulimia nervosa was markedly higher in cities compared to rural areas (in other words, the incidence rate of bulimia nervosa was 5x higher in large cities versus rural areas).

Having said that, the 95% confidence interval was also relatively large (2.08-13.14), which means there is a 95% chance that the true ratio is somewhere between 2.08 and 13.14, so it could very well be that the true rate is only 2x higher in cities compared to rural areas (or 13x!).

The authors conclude that “urban life is a potential environmental risk factor for bulimia nervosa but not for anorexia nervosa.”

But the interesting question is why???

Does living in a city somehow contribute to causing bulimia nervosa? Would moving back to rural towns be a good (but seemingly extreme) preventive measure for those at risk of developing bulimia nervosa?

van Son et al. cite other factors that seem to suggest a strong environmental influence in causing bulimia nervosa:

Another indication of the strong environmental influence on the incidence rate of bulimia nervosa is the unstable pattern of the incidence rate over time, as evidenced by the sudden and sharp rise in the incidence of bulimia nervosa since the 1980s (Soundy et al, 1995), the relative rarity of bulimia nervosa before 1970 (Kendler et al, 1991) and the failure to find conclusive evidence of the existence of bulimia nervosa in history (Keel & Klump, 2003). Such fluctuations in the incidence rate pattern cannot be caused by changing genetic factors because the time scale is too limited.

Uh-oh, please don’t tell me you think bulimia nervosa is a “Western” disease caused by the prevalence of thin models, I thought to myself. After all, it is normal to expect a sharp rise in incidence after a disorder is identified and there’s increased surveillance (bulimia nervosa first came onto the “scene” in the 1970s).

van Son et al provide two plausible (and not exclusive) hypotheses to explain these results: (1) migration hypothesis and (2) opportunity hypothesis.

Basically, the migration hypothesis relies on the idea that because there are more treatment and medical facilities in larger cities, individuals with bulimia nervosa are more likely to be diagnosed/identified , and/or seek help. In such cases, the large city had no effect on the development of bulimia nervosa, just its identification. 

The opportunity hypothesis basically suggests that it is easier to engage in bulimic behaviours (obtaining large amounts of food, eating large amounts of food, and purging) in large cities. It is easier to hide bulimia in a large city than in a small town.

I think both of these factors play a role.

I have no doubt in my mind that opportunity played a big role in my bulimic symptoms. Living on my own definitely increased the frequency and severity of my bingeing and purging behaviours (unfortunately), and living in a large city helped, too. Routinely buying large amounts of “typical binge food”, going through large amounts of food, and/or spending inordinate amounts of time in the bathroom, is much easier when you live alone (or with strangers who don’t care/don’t notice) and in a large city where you don’t run into the same grocery store clerks.

I actually think opportunity is a really crucial factor in why there’s little evidence of bulimia nervosa through-out the centuries, prior to significant levels of urbanization and increased access to food. Not having access or ability to engage in behaviours will certainly limit the prevalence of those behaviours in the population and the frequency among those that do engage in them. Quite obviously, you can’t develop an addiction to substance if you don’t have access to that substance. (Not to say bulimia nervosa is exactly like an addiction, I don’t think it is.)

Of course, I would never suggest that opportunity is the only factor–that’s clearly not the case, by far. But it might explain, in part, why rates seem to be higher in urban versus rural areas.

Readers, what do you think? What are some other factors that you think might contribute to this effect? Why do you think this isn’t true for anorexia nervosa (or do you think it is)? 


van Son, G., van Hoeken, D., Bartelds, A., van Furth, E., & Hoek, H.Epidemiology"> (2006). Urbanisation and the incidence of eating disorders The British Journal of Psychiatry, 189 (6), 562-563 DOI: 10.1192/bjp.bp.106.021378

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


Tetyana is the creator and manager of the blog. She has an Honours BSc in Neuroscience and an MSc in Medical Science. She can be reached at tetyana[at]scienceofeds[dot]org.


  1. I find the opportunity hypothesis totally sound, and my intuition is that it is much more of a factor in the disparity than migration. Nothing enables binging and purging behavior quite like presence of temptation and lack of short-term consequences (e.g. someone finding out).

    • I agree, especially when you’ve already started… the increased opportunity can definitely quickly exacerbate it. (At least that’s true in my case.)

  2. I agree that opportunity is huge. I also wonder if, on a more micro level, this is why so many people migrate from AN to BN or AN-BP over time–not just that starvation is difficult to sustain even for anorexics, but because purging can be easier to hide in certain social environments than straight restricting. Adaptation due to opportunity, or something like that.

    • Eva, I completely agree on all counts! Bulimia, in particular, was nice because no one really knew I was sick but it gave me a way to deal with anxiety (super maladaptive and super shitty, but a way). I remember someone in a support group, a long time ago, saying that as she was recovering from bulimia, she noticed she was much less social because the as the opportunity to purge decreased (at least she didn’t want to purge), she was more anxious about going out and eating.

      • Yeah–I developed AN as a child, and could pass it off as being a “picky eater” and “shy,” but after I became seriously ill and needed emergency hospitalization, my family was no longer so dumb as to let me use excuses to restrict. So I figured out that purging is much easier to deceive people with. (Finding: people with EDs are smart?)

        Anyways, I think that’s especially true for people with “adult responsibilities”–I’d always much rather restrict but it’s harder to lie to your partner, save face at work luncheons, etc that way. So I actually wonder if you see crossover in long-term sufferers or people in relapse as a function of life stage changes–that would be an interesting study to read. Like, do people who live alone have different crossover diagnoses than people who are married/have children/are in certain professions? (…I feel like I should make a pun on “regression” here but I’ll stop writing;)

  3. Hi Tetyana,

    I think it’s absolutely right that opportunity is a major factor — it’s great how intelligently you break down the ‘rise’ of bulimia in the later part of the 20th century.

    For me, it’s both opportunity *and* temptation (as specifically differentiated from pure or ‘objective’ opportunity).

    In the city I used to live in, it was much easier to b/p for all the reasons you cited — that’s opportunity playing a role.

    At the same time, in that particular city (as in most cities) there were just so so so many temptations: the plethora of indulgent, over-the-top and highly varied foods. The amount of choice was a major precipitating factor.

    I’m now spending my summer in a small, remote town where there’s still plenty of opportunity to b/p, but where the temptation is quite low. It’s almost like, now that I’ve binged in the city… it’s far less exciting here where there’s literally nothing good to eat. The cravings still have a life of their own, but I can say for sure that I’m able to distance myself from them much better than when the cravings had a more tempting outlet.

    As a result, my progress in recovery has really astonished me… it’s really promising to think of how much our environment can influence the ED, because then we can start to make choices to engineer a more supportive environment, little by little.

    Anyway, thanks for this post! 🙂

    • Thank you so much for your comment, Susan!

      Sorry it took my a while to respond, been busy.

      I agree with everything you wrote, particularly the point on differentiating between opportunity vs. temptation. That’s definitely important. It is both. If I’m in a good mood, temptation plays a bigger role, but in a bad mood where I feel I need to b/p to deal with my mental state, almost anything goes (seriously, stuff I don’t find tasty at all).

      “The cravings still have a life of their own, but I can say for sure that I’m able to distance myself from them much better than when the cravings had a more tempting outlet.”

      Yup! What an important observation, too — not only for your recovery but perhaps for treatment practices in the future.

      I’m glad you are making good progress in recovery!

Leave a Reply

Your email address will not be published.

7 + 9 =