Wednesday, April 22, 2009

Counting LGBT People in Health Studies. ACT NOW!

Counting lesbian, gay, and bisexual people in health studies has a relatively short history. And yet, a fairly large number of lesbian, gay, and bisexual people have been identified so far, especially from the California Health Interview Survey (CHIS), and various States have included information on sexual orientation in their Behavioral Risk Factor Surveillance Systems (BRFSS) and Youth Risk Behavior Surveys (YRBS).
But there is no nationally representative dataset that yet captures sexual orientation data, making it very difficult to do the kind of research I am most intereted in - looking at the impact of normative heterosexuality on health. Because most of the studies that do include sexual orientation data happen in States that are relatively friendly (VT, RI, MA, CA, WI, etc.), it is difficult to find a comparison group exposed to higher levels of societal homophobia.

To a small degree, one can use relationship status from the national BRFSS and NHANES datasets to identify sexual minorities, but because these people are by definition in marriages and marriage-like couples, it is unclear what biases play out in becoming a couple, and what impacts this has on health.

So, to do the kind of research I want to do, we need national datasets that ask about sexual orientation identity, not just imputing it from partnership status.

So, is this post totally self-serving? Self-serving, yes, but I wouldn't be the only beneficiary.
Right now, there's no way for people in most of these United States (and her territories, PR, VI, GU, AS, CM) to know whether the trends that have been identified in Massachusetts, Vermont, Wisconsin and California apply to their populations as well.
And because the data would be Federally collected, it would be available to a much broader range of researchers (such as graduate students and people without a huge research funding apparatus behind them), enabling a much broader set of eyes and minds to grapple with LGBT health concerns than has been able to so far.

A question about gender identity has, as far as I know, only been asked in one wave of the Massachusetts BRFSS, and those results have not yet been released. So there's no solid information on the health of transgender people in any State.

Hats off to Senator Sheldon Whitehouse (D-RI) for initiating a "Dear Colleague" letter to ask the Senate to set aside $2M to ask questions about sexual orientation and gender identity on the National Health Interview Survey is a good start. Those questions should be standard questions asked on the National Health and Nutrition Examination Survey (NHANES) and the BRFSS core demographic module as well.

Call your own Senator at (202) 224-3121 and ask them to sign on by calling Jordanna Davis in Senator Whitehouse’s office.

Another place that we should be getting more information is from birth and death records. Vermont does this already, but many of the States with civil union laws do not yet collect this information on standard birth and death certificates. And the National Vital Statistics Sytem should revise its procedures to allow capturing this information from the States that do collect it.


  1. Hear, Hear! As someone who worked with Sen. Whitehouse's office to get him to take the lead on this process, I can't say enough about how clearly he gets LGBT issues and health needs. And he needs some support, so please do get on the horn and call your own Senator today!

  2. Please call or email Senator Whitehouse to thank him. Being great on LGBT stuff is so often a thankless job!

  3. I've run into the same troubles attempting to quantify the number of same-sex couples who would choose to marry if their state legalized same-sex marriage for the purpose of estimating the net economic benefits of legalizing same-sex marriage

  4. Hey T,
    It actually shouldn't be too hard to estimate the number of same-sex couples who would get married, as the number of un-married partners is counted by the Census.
    Not all of them would get married given the chance, of course, I suppose you could use the distribution of unmarried mixed-sex couples : married mixed-sex couples as a guide to estimating that.
    If you want, I'd be happy to run those numbers out for you, it's pretty straightforward.