It would be really useful if public-health researchers knew how often trans youth got pregnant. That’s because this is a population that is very likely to slip through the cracks. If you are a young trans person with a uterus, but who identifies as and presents as male or genderqueer in day-to-day life, you might not feel comfortable seeking out the reproductive health care you need — one of the many reasons trans activists are hoping to chip away at the stigma surrounding trans health care.
Late last month, a study on this question was published in the International Journal of Transgenderism, and if you read the subsequent headlines, you’d think that the authors had answered it: the pregnancy rate for trans youth is no different than the pregnancy rate for cisgender youth. Vice, for example, announced that “Trans Youth Get Pregnant at the Same Rate as Cisgender Peers,” and a bunch of other outlets covered the study in a similar manner. (The Washington Blade, a major LGBT paper, picked up the study’s press release.)
But if you dig into the study, there’s reason to be very careful about the numbers being presented. This is a useful example — once again — of how overhyped press releases can spread misinformation far and wide.
For the survey, a team led by Jaimie Veale of the University of Waikato in New Zealand looked at data from 923 responses to the Canadian Trans Youth Health Survey, “a large online survey open to people identifying as trans or genderqueer, currently living in Canada, aged 14 to 25 years.” That survey’s questions were drawn mostly from “existing population-based adolescent health surveys,” and different versions of it were administered to the 14-to-18-year-olds and the 19-to-25-year-olds in the sample.
To generate their estimates, the researchers looked at responses to the question, “How many times have you been pregnant or gotten someone pregnant?” among those in the sample who reported that they had previously had sex (111 participants hadn’t and skipped the question). Of the 540 youth who answered the question, 26 reported pregnancy involvement at least once, for an overall percentage of 4.8 percent, which can be safely rounded up to the 5 percent figure reported in the paper and in media accounts. The total percentage of youth in the sample who had been involved in a pregnancy — re-including those who had never had sex — was 1 percent, and the researchers point out that these 1 percent and 5 percent figures are similar to those published in the 2013 British Columbia Adolescent Health Survey, which is a bigger, more representative survey of a slice of all Canadian youth, not just trans ones.
But this paper’s figures should be interpreted with a lot of caution, for two reasons: The first is that youth were asked, in effect, if they had ever been involved in a pregnancy. That means the pregnancy involvement in question could have occurred before or after they transitioned (that’s a slight oversimplificaiton, of course, since plenty of youth go through in-between phases, and transition isn’t always a straightforwardly discrete process). This isn’t nitpicking: If one of the key questions here is to what extent reproductive health resources should specifically target trans youth with uteri, that’s exactly the number we need: how likely it is that a trans youth who can get pregnant will get pregnant. If someone gets pregnant and then transitions, that’s a very good thing for researchers to know in terms of better understanding the trajectories of trans youth, but it doesn’t bear on the most important question at hand.
Since the numbers we have in this study are 26 out of 540, if even just a small percentage of those pregnancies occurred during a period that could safely be called pre-transition, it would cause the rate to decrease significantly. Let’s say just 8 — less than a quarter — of those pregnancies were pre-transition, changing the calculation to 18 out of 540: Suddenly we’re down to 3.3 percent, or 3 percent, of trans youth in the study being involved with pregnancy at the time they were actually members of this group. The researchers, to their credit, note in the Limitations section of the study that it would be interesting to ask “this group … at what stage of transgender identity development did this pregnancy occur[.]”
The other big issue here is natal sex — that is, whether the respondents were born with penises or vaginas. The researchers combine youth who got someone pregnant with youth who themselves became pregnant in a “pregnancy involvement” category, but don’t break those 26 pregnancies out so as to explain how many of them involved getting someone else pregnant, and how many of them involved the trans youth in question getting pregnant. Without that data, we can’t really generate an estimate from this sample of how frequently trans youth in the same became pregnant.
During an email exchange with Science of Us, Elizabeth Saewyc, a youth-health researcher at the University of British Columbia and a co-author on the study, said that she was about to head off on a trip and wasn’t able to break down the pregnancy numbers by natal sex for me in time for this post, but pointed out that in the broader survey from which these numbers were drawn, about 75 percent of the sample was female at birth, and that she thought it likely the percentage would be about the same in her subsample (Veale, the paper’s lead author, echoed this in an email). But she also said she didn’t think this was particularly important: “I’ve been a public health professional and researcher involved in sexual and reproductive health issues both clinically and from a public health perspective for 20 years, but I’m not sure what the public health issue is that would require a focus only on those who become pregnant, as opposed to any of those involved in pregnancy, either becoming pregnant or causing someone else to become pregnant.” More broadly, Saewyc had “always been curious as to why people put so much emphasis on pregnancy-related research and monitoring that is solely focused on the person with the uterus, when, at least for this age group, there are always two people involved in creating a pregnancy, whose lives and decisions are often affected by this reproductive moment, and who may be further involved in parenting, if the pregnancy leads to a birth.”
I can’t speak for public-health researchers since I’m not one, but as a layperson, it doesn’t seem difficult to come up with situations where you’d want to know specific things about one group or the other. People who can become pregnant, but not those who cannot, benefit from access to birth-control pills, for example, and might need to get an abortion — and there’s no way of knowing whether the father or potential father will be in the picture. That’s why obtaining better estimates of the trans pregnancy rate is so important in the first place: because trans youth might not feel comfortable seeking out those services if they don’t look like a “typical” young woman, and are afraid they won’t be treated as such by reproductive-health providers. To lump the experiences of getting someone else pregnant and getting pregnant yourself together into a“pregnancy involvement” measure makes it harder to gain a clear sense of the public-health picture. Plus, even the document Veale, Saewyc, and their colleagues point to in their comparison between their numbers and the broader numbers for adolescents, that survey of British Columbian youth, mentions that “The rate [of pregnancy involvement] was 5 percent among male and female students who had ever had sexual intercourse,” indicating researchers thought it useful to break things out by sex. And in the U.S., of course, this is regular government practice — people closely follow the teen pregnancy rate, not the teen “pregnancy involvement” rate.
Plus, with trans youth there are two other issues that render this distinction relevant in different ways: the hormones some trans people take as part of their transition process, and gender dysphoria, the often intense and visceral feeling of discomfort many trans people feel with the bodies they were born with. As the researchers point out, “It is known that physical transitions that transgender people undertake through hormonal and surgical interventions generally disrupt their reproductive functioning.” So to better understand to what extent hormones (and to a lesser extent surgical interventions, which tend to occur later) affect fertility rates among trans youth who can carry children, it’s helpful to break things out by natal sex when reporting pregnancy-prevalence data. Then there’s dysphoria: As the Human Rights Campaign recently noted in a trans safe sex guide, it can cause trans people to be less likely to engage in “traditional” sex, for lack of a better term. There could be natal-sex differences there as well that would be worth knowing about, and the way to find out, again, is by looking at and reporting the data. While it’s true that trans people often want to de-emphasize what they view as a sex assigned to them at birth without their consent, that doesn’t mean researchers can’t respectfully collect and report zoomed-out statistics that will, in the long run, improve access to health care in important ways.
Anyway, the point is that between the timing issue and the natal-sex issue in this survey, it is very easy to imagine the “true” pregnancy rate (not the rate of pregnancy “involvement”) in this sample either being significantly lower — if just a small chunk of those pregnancies occurred pre-transition — or significantly higher — if natal females are, in fact, 75 percent of the subsample — than what was reported. Or maybe those two factors cancel each other out; without the numbers, there’s just no way to know. This is a really noisy estimate, and both Veale and Saewyc cautioned in their emails against extrapolating from such small numbers, at least on on the likelihood-of-getting-pregnant question — and included many caveats in their paper itself.
Of course, none of those caveats made it into the headline of UBC’s press release: “Transgender youth are as likely to become pregnant as other adolescents.” And then, because this is how science journalism often works, a bunch of outlets ran with that severely overblown headline, and now a lot of people think we have a good estimate of the rate of trans youth pregnancy. We don’t yet, and that’s unfortunate.