Correlation of polycystic ovary syndrome with lesbianism

Factual Questions
1

I was just chatting with a lesbian friend who explained that she grew very thick body hair as a result of a polycystic ovary. Then she told me something that amazed me: she said “80% of the lesbians have it, and 15% of straight women.” I went and googled up information, and found such numbers reported in a medical journal–Fertility & Sterility reported:

“Eighty percent of lesbian women, compared with 32% of the heterosexual women, had PCO on pelvic ultrasound examination. Thirty-eight percent of lesbian women, compared with 14% of heterosexual women, had PCOS. There were no significant differences in the androgen concentrations between lesbian and heterosexual women with normal ovaries. However, lesbian women with PCO and PCOS had significantly higher androgen concentrations compared with heterosexual women with PCO and PCOS.”

But still - what astonished me was a positive correlation between an ovarian condition and lesbianism. I’d seen the news about research suggesting that certain brain/neural structures might correlate with lesbian, gay, and transgender - sexuality and gender identity being located in the brain makes intuitive sense - but homosexuality can be associated with gonads too?

**“Conclusion(s): There is a significantly higher prevalence of PCO and PCOS in lesbian compared with heterosexual women. Lesbian women with either PCO or PCOS had more pronounced hyperandrogenism than did heterosexual women with either PCO or PCOS.” **

How to explain the correlation? Maybe if higher levels of androgen were present during brain development, and tended to produce neural structures associated with lesbian brains?

2

Perhaps it has something to do with taking hormonal birth control (I know some lesbians take it for period-related problems, but certainly the number must be way lower).

3

Isn’t obesity a risk factor for PCOS? There was a recent study that found lesbians are much more likely to be overweight or obese than heterosexual women. It’s not clear from the abstract of the study in the OP that adjustments were made for the weight of the participants, so that could explain some of the difference.

As an aside, typing “obese lesbians” into google is not a good way to find scientific publications.

4

It’s interesting, but I’d want to read about the sample used. It’s well-documented that lesbians are less likely to use health services, expecially GYN services, so there may be differences in severity of presenting issues.

5

I just finished reading the full text of the paper (unfortunately, it’s not free for everyone). I admit that I was expecting to find a lot of crap - poor design, poor statistics, overcalling conclusions, etc. In fact, it was not too bad a paper. Still, I have a number of concerns.

As ultrafilter notes, obesity is more common in lesbian women and is, itself, a potent risk factor for PCOS. And, indeed - the lesbian women in this study were significantly fatter than the heterosexual women (BMI 25.1 vs 23.6, p < 0.01). I was disappointed to discover the authors did NOT attempt to separate out the effect of this disparity on the prevalence of PCOS (e.g. by using multivariate analysis). IMHO, failure to dissect out the obesity from the lesbianism is a near-fatal flaw.

The paper did not mention physical fitness at all. One can postulate that the more obese, lesbian women are also less fit. One should then recall that being physically unfit leads to insulin resistance and thus higher insulin levels. High levels of insulin promote PCOS. Neither insulin levels nor physical fitness (eg. MVO2 measurement) were done in this study.

I am very concerned that the authors measured total testosterone levels and not free testosterone levels. The former are notoriously misleading and uninformative. Frankly, any paper on PCOS that uses total testosterone and not levels of free testosterone should be highly suspect.

Finally, the hypothesis here is rather obvious and simplistic. It must have been studied earlier. But, if the results were negative (i.e. “not exciting”), they may not have been published. OTOH, if one study fluked out a positive (“exciting”) result, it will be published. In such circumstances, then, the positive results will be mistakenly assumed to be the only available data, when in fact there is lots of negative, albeit unpublished, conflicting data.

Even if I’m wrong about ‘publication bias’ (as described above), it is highly unusual that such “interesting” results have never been duplicated. The paper was published three years ago - more than enough time to do so. In fact, over the last 50 or so years, out of hundreds of thousands of research papers, only two have even looked at the notion of increased PCOS in lesbian women or transexual women. This paper makes three. The paucity of such studies may be telling us that others have looked into it, come up empty-handed, and not bothered to publish their (negative) results. Or, perhaps it means that others have rejected the hypothesis outright due to its obvious over-simplicity. Just my WAG.

6

Apparently, it’s not known if obesity is the cause of PCOS or if PCOS is the cause of obesity. My own (hetero) step-sister has PCOS. She got pregnant only when she lost a significant amount of weight.

7

Very true. To add to this, the insulin resistance that leads to the obesity is either a cause or a symptom of polycystic ovaries. I’m being treated for PCOS at the moment. I wasn’t overweight, but taking up regular exercise and taking a medication normally prescribed for diabetes seems to be helping. I agree with KarlGauss, it seems that leaving out obesity and physical fitness (not to mention diabetes/insulin data) is quite the oversight.

Many thanks to KarlGauss for the summary, BTW.

Oh, and I consider myself heterosexual. I have felt some attraction to women in the past, but I haven’t gone and earned anyone a toaster oven, yet. :wink:

8

Am I missing something here that says it’s not due to pregnancy differences?
I can’t imagine the % of birth mothers is as high as the % of heterosexual birth mothers.

9

I think the most you can say is that weight is correlated with PCOS.

It may just as well be that higher average BMI in lesbian populations are caused by higher rates of PCOS, or that both have are caused by a third unknown factor. By the way I’m not convinced of the quality of the epidemiological data on weight in lesbians - I think the jury is out on this until more rigorous studies are done.

10

I did a quick google, but couldn’t find the answer to this:

How big were the sample sizes in the original study?

11

There were 618 women (undergoing ovarian stimulation) in the study. Of thee, 254 were "self-identified as lesbian, and 364 were heterosexual.

12

Thanks. Enough to make it interesting.

I also wonder if there was a location effect: it seems a little odd to me that the samples were so close to being equal–41 percent lesbian. That fits with the premise of a vast overrepresentation of the condition in lesbians, but still, one has to wonder.