The hardest part of science blogging is picking an article to blog about. In times when I’m indecisive–when I spend hours sifting through the literature, inevitably creating several draft posts before deciding each article isn’t interesting enough–I turn to the list of topics that have been suggested by readers. The last suggestion I received was “eating disorders in the lesbian community.” It is a great suggestion, but I thought my search wouldn’t turn up much. But, to my surprise, it did turn up some studies.
But please, don’t expect too much: it is not a well-studied area, and most of the data comes from self-reported questionnaires, which are not particularly reliable:
- First, there’s selection bias: the 50% or so of people who return the surveys could be different in significant ways from the 50% that don’t. For example, in a survey about mental health, perhaps individuals that have had personal experiences with mental health issues are more likely to respond. This skews the data in a such a way that it might appear that a specific subgroup that’s being studied has higher rates of mental health issues than is actually true.
- Second, if one or two people are evaluating hundreds of individuals, we can be more confident that their criteria are the same. But, if every individual fills out a questionnaire, with limited instructions, how do we know that “bisexual” for one person is the same as “bisexual” for another? How can we be sure that someone’s definition of “binge eating” is the same as the clinicians and researchers studying it? Short answer: we don’t. The hope is that sufficiently large sample sizes and properly matched (same age, for example) control groups will help offset any biases here.
My advice is: look for large trends in the data. Small differences or just barely statistically significant findings are often just noise.
One can hypothesize that discrimination and stigma of homosexuality may lead women to experience more mental health issues, including eating disorders. But is that borne out by the data?
A study by Krieger and Sidney (1997) found that among 25 to 37 year-olds who had at least one same-sex sexual partner, 33% of African-American women and 56% of Caucasian women reported experiencing discrimination based on sexual orientation… a study by Mays and Cochran (2001) [that] lesbian and bisexual women–as compared to heterosexual women–reported higher frequencies of discrimination, which in turn, was associated with psychiatric morbidity.
But when it comes to really establishing whether homosexual women exhibit higher rates of mental health issues, the data is mixed. A lot of studies suggest that rates of anxiety and depression are equally common among homosexual and heterosexual women. Others suggest that homosexual women are more likely to seek out help for mental health concerns (psychotherapy, counseling) than their hetereosexual counterparts. Two studies (Brand, Rothblum and Solomon, 1992; Siever, 1994) seem to suggest that eating disorders are less common among lesbians than heterosexual women. But these studies were done more than 20 years ago… are they still relevant today?
In addition to the limitations I mentioned above, many of these studies lack a proper (age-matched, etc..) control group, and some don’t have a control group at all. In others, the sample sizes are too small, and all too often, bisexual women are not included or not counted as a separate group.
The study by Koh and Ross had methodological improvements over the previous research by including more appropriate controls and generally being more comprehensive.
The sample of 1,304 women was composed of 524 lesbians, 143 bisexuals, and 637 heterosexual women.
So, what did they find? Again, please keep in mind: this is self-reported data.
In a nutshell, with regard to eating disorders (square brackets are mine):
An eating disorder history also demonstrated a relationship to being out. Out bisexuals were more likely to have [reported to have] an eating disorder than heterosexual women. It is unclear why this association would be found. Regardless of degree of outness, bisexual women were twice as likely to have [reported to have] had an eating disorder than lesbians. There is no other existing data on bisexual women and eating disorders. Despite the differing rates of eating disorder histories, the B.M.I.s and current self-perceptions of weight were equal among the lesbian, bisexual and heterosexual women subgroups.
Why am I so suspicious of the reported data?
In the general population, the prevalence of anorexia nervosa is around ~.09%, bulimia nervosa is around 1-2%, and binge eating probably around ~3%. In total (even though there is a lot of crossover), that’s about 5%. Say we relax the criteria and prevalence rates double as a result, that’s still 10%.
In this study, 13% of heterosexual and 14.5% of lesbian women reported having had an eating disorder. For bisexuals: over 25% reported having had an eating disorder.
Now, here’s my take: maybe the trend is legitimate, maybe there’s some unique pressures that self-identified bisexual women face that results in an increased likelihood to develop EDs. I think that’s a completely legitimate hypothesis, and could very well be true. But eating disorder rates 5x as what’s reported for the general population?
Also keep in mind: there were only 143 self-identified bisexual women in this study, that’s a very, very small amount.
If we were to combine the lesbian and bisexual women: 85% of self-identified bisexual women had sex with both men and women as adults and 55% of self-identified lesbians did), the number drops to almost 15%. Compare that to almost 13% of heterosexual women in this study (which is the proper control, not previously reported numbers from other studies), and it seems like there aren’t really any major differences between the two groups in this regard.
What about other studies?
In looking for more, um, satisfying studies, I suppose, I found this gem from a 2002 paper by Tamara Share and Laurie Mintz (“Differences Between Lesbians and Heterosexual Women in Disordered Eating and Related Attitudes“):
There is much pressure on women in our culture to attain a culturally defined ideal body shape (Root, 2001). Few researchers, counselors, or educators would disagree that such pressure to attain the thin ideal is related to negative body image and disordered eating (Delaney, O’Keefer, & Skerne, 1997; Root, 2001; Striegel-Moore & Cachelin, 2001). Likewise, attaining the thin ideal is important to women because they are socialized to equate self-esteem with body esteem, to see themselves as sexual objects, and to pay vast amounts of attention to their physical appearance in order to attract men (Laidlaw, 1990).
Given these theorized links between attaining the culturally defined ideal body shape and being sexually attractive to men, an emergent question is whether lesbians would suffer from negative body image and associated disordered eating at the same rates as heterosexual women. There are two diverse views on this question. Dworkin (1989) describes one such view.
“Lesbians are a segment of the female population who have undergone the same socialization process as all women but who have rejected traditional female values. By definition a lesbian is a woman whose primary ties are to other women (Ferguson, 1981) and therefore she ought not to be bound by the prescriptions of a patriarchal society. . . . Lesbians do not think of themselves as objects to be defined by male subjects. Therefore, it seems lesbians ought to be able to escape from the negative body image and lack of self-acceptance that other women in our society suffer from. (p. 28)”
Similarly, Striegel-Moore, Tucker, and Hsu (1990) hypothesized that because lesbians tend to “challenge culturally prescribed beauty ideals” (p. 494), they should be less susceptible to the damaging consequences of internalized cultural norms, more accepting of their bodies, and less likely to engage in disordered eating. Brown (1987) echoed this sentiment, noting that lesbians tend to question attitudes about beauty that are rooted in cultural values and therefore, should be less susceptible to the damaging consequences of internalized cultural norms, more accepting of their bodies, and less likely to engage in disordered eating.
Now, let me be clear: I am not questioning lesbian individuals’ attitudes towards societal constructs of beauty, or how they feel about traditional female values.
Is there really enough evidence to support that the extent of internalization of cultural norms plays a significant role in predisposing individuals to develop eating disorders?
I’m not sure.
The assumption that the “thin ideal” plays a huge role in ED etiology is frustrating because, while I don’t deny that it does play a role for some, there’s so much more to it than that (if you are not convinced there’s more to it than that, please refer back to all the previous blog posts). There are non-fat-phobic anorexia nervosa patients, and eating disorders exist in non-Western countries.
Genetics is an important (but not sole) factor in predisposing individuals to eating disorders. That doesn’t mean that sexual orientation and gender identity doesn’t play a role in the development of eating disorders or in how we make sense of our eating disorders (our personal narrative): it can, and it most certainly does for many. But the overemphasis on cultural idealization of thinness as a significant causal factor in the development of eating disorders is erroneous.
If lesbian and bisexual women were to report higher rates of eating disorders, I would first hypothesize that this is due to the increased stress, stigma and discrimination they face. This wouldn’t cause eating disorders, but it may increase the risk of developing them for those who are otherwise already predisposed. (Genetics loads the gun, environment pulls the trigger as Francis Collins said. Pulling the trigger without loading the gun first is not going to do all that much.)
This is a very complicated topic that I’ll definitely blog about it more. Please comment with your thoughts and opinions about this topic! This is something that really interests me (although I haven’t had much time to explore it) because I’ve been open about my non-heterosexuality since I was 11 and so I’ve thought a lot about how my sexual orientation has influenced my eating disorder. I don’t think it played a role in causing or “triggering” my eating disorder, but at some points it did figure into my personal narrative of rationalizing and trying to make sense of my eating disorder behaviours. But that’s a complicated story.
By the way, what did Share & Mintz find?
Differences were found in terms of both internalization of cultural attitudes toward appearance and body esteem related to sexual attractiveness. Specifically, when compared to heterosexual women, lesbians reported significantly lower levels of internalization of cultural attitudes toward appearance and significantly higher levels of body esteem related to sexual attractiveness. Conversely, no significant differences were found in terms of awareness of cultural attitudes toward appearance, eating disorder symptomatology, or body esteem related to either physical condition or weight concern.
Share & Mintz’s findings were supported by Feldman & Meyer, who, in their 2007 paper “Eating Disorders in Diverse Lesbian, Gay, and Bisexual Populations” also reported no significant differences in the prevalence of eating disorders among lesbian and bisexual women and their heterosexual counterparts.
SEDs readers, I’m interested in whether you think your sexual identity/orientation plays a role in your eating disorder, and if yes, how so?
Koh, A.S., & Ross, L.K. (2006). Mental health issues: a comparison of lesbian, bisexual and heterosexual women. Journal of Homosexuality, 51 (1), 33-57 PMID: 16893825
Share, T.L., & Mintz, L.B. (2002). Differences between lesbians and heterosexual women in disordered eating and related attitudes. Journal of Homosexuality, 42 (4), 89-106 PMID: 12243487