Prenatal Diethylstilbestrol Exposure and Likelihood of Being Transgender

By Aly W. | First published July 19, 2020 | Last modified August 7, 2021

Abstract / TL;DR

A new high-quality study of prenatal diethylstilbestrol (DES) exposure and transgenderism in assigned male at birth (AMAB) people has been published and provides insight on the incidence of transgenderism in individuals AMAB and exposed to DES in utero. A very low incidence of transgenderism in people AMAB was found, which contradicts a low-quality unpublished Internet survey from the mid-2000s that reported a high rate of transgenderism in people AMAB and prenatally exposed to DES. While it still remains a hypothetical possibility that DES could increase the likelihood of transgenderism in individuals AMAB, there is no quality clinical evidence at this time to support this being the case. Moreover, the new data indicates that such an effect would be small at most.


Diethylstilbestrol (DES) is a potent synthetic nonsteroidal estrogen which was previously used clinically for a variety of estrogen-type indications including pregnancy support (to help reduce the risk of miscarriage). Following the discovery of various estrogen-related toxicities of DES in the 1960s and 1970s, DES and other nonsteroidal estrogens were discontinued. People exposed to DES in utero have been found to have increased risk of various subtle birth defects and have been referred to as “DES daughters” and “DES sons”. It has been theorized that prenatal DES exposure may also increase the likelihood of transgenderism in individuals assigned male at birth (AMAB). However, no quality clinical data has been available to assess this question. That is, until now.

Troisi et al. (2020)

A new study on sexual orientation and gender identity in individuals assigned female at birth (AFAB) and AMAB and prenatally exposed to DES has recently been published:

  • Troisi, R., Palmer, J. R., Hatch, E. E., Strohsnitter, W. C., Huo, D., Hyer, M., … & Titus, L. (2020). Gender Identity and Sexual Orientation Identity in Women and Men Prenatally Exposed to Diethylstilbestrol. Archives of Sexual Behavior, 49(2), 447–454. [DOI:10.1007/s10508-020-01637-7]

Here is the abstract of the paper:

We assessed the associations of prenatal diethylstilbestrol (DES) exposure, a potent estrogen, with sexual orientation and gender identity in 3306 women and 1848 men who participated in a study of prenatal DES exposure. Odds ratios (OR) and 95% confidence intervals (CI) were derived from logistic regression models adjusted for birth year, study cohort, and education. Among women, the OR for DES in relation to reporting sexual orientation identity as nonheterosexual was 0.61 (95% CI 0.40–0.92) primarily due to a strong inverse association with a lesbian identity (OR 0.44, 95% CI 0.25–0.76). Among men, the OR for DES in relation to reporting a nonheterosexual sexual orientation identity was 1.4 (95% CI 0.82–2.4), and ORs were similar for having a gay identity (1.4, 95% CI 0.72–2.85) and bisexual identity (1.4, 95% CI 0.57–3.5). Only five individuals reported a gender identity not conforming to that assigned at birth, preventing meaningful analysis. Women who were prenatally exposed to DES were less likely to have a lesbian or bisexual orientation, while DES-exposed men were somewhat more likely to report being gay or bisexual, but estimates were imprecise.

And here are some additional notable excerpts regarding DES and transgenderism in people AMAB:


The present study used data from a longitudinal cohort study. In 1994, the National Cancer Institute (NCI) combined cohorts that were followed previously (1970s–1980s) and assembled new cohorts (comprising mothers or offspring of existing cohort members) to establish the US NCI DES combined cohort follow-up study. The resulting combined cohort study is comprised of prenatally exposed and unexposed women and men whose mothers attended a large, private infertility practice (Horne Cohort), or participated in the Women’s Health Study of women who were and were not given DES during pregnancy (WHS Cohort; Greenberg et al., 1984), or enrolled in a clinical trial of DES (or placebo) use in pregnancy (Dieckmann Cohort; Bibbo et al., 1977). […] The men’s combined cohort additionally included prenatally DES-exposed and unexposed men who participated in a study at the Mayo Clinic (Leary et al., 1984).

Over the course of the NCI combined cohort study, 2924 men had responded to at least one questionnaire and were considered participants; 256 of these had died since 1994. Of the remaining 2668 (1322 exposed, 1346 unexposed) men, 1885 (947 exposed, 938 unexposed) agreed to complete the full 2016 questionnaire. Of these, 1848 (933 [98.5%] exposed and 915 [97.5%] unexposed) answered the relevant questions on sexual orientation identity and gender identity.


Only two women, both DES exposed, and 3 men (2 exposed and 1 unexposed) reported gender identity that did not conform with the sex they were assigned at birth. Five DES-exposed persons and five unexposed preferred not to answer the question on gender identity.


The NCI combined cohorts of women and men represent one of the only human studies allowing evaluation of the influence of a potent synthetic estrogen on early fetal development in a population with documented DES exposure status.

There were too few women or men reporting a gender identity different from that assigned at birth to analyze potential effects of prenatal DES exposure, but this suggests that any effect would be small.

In summary, the findings showed that women who were prenatally exposed to DES were significantly less likely to report being lesbian or bisexual. In contrast, while men who were prenatally exposed to DES were somewhat more likely to report being gay or bisexual, the estimate was very imprecise and compatible with chance. Finally, very few individuals reported currently thinking of themselves as a gender different from that assigned at birth in either the exposed or unexposed group.

This is the very first quantitative study on prenatal DES exposure and transgenderism in people AMAB to be conducted (DES Action, 2020a). The incidence of transgenderism in individuals AMAB was too low to allow for analysis of relative likelihood. Hence, the study couldn’t determine whether the incidence of transgenderism in people AMAB was actually increased with prenatal DES exposure or not. In any case, based on the very few cases encountered, it can be assumed that the influence of prenatal DES exposure on gender identity in people AMAB, if any, is small at most. Only 2 out of around 930 (or about 0.2% of) DES-exposed people AMAB were transgender. Put another way, 99.8% of people AMAB prenatally exposed to DES ended up being cisgender rather than transgender.

Scott Kerlin, Ph.D. is a DES son and clinical psychologist who runs (or ran) an online support forum for other DES sons (DES Action, 2020b). In the 2000s, he conducted an Internet survey on gender/sexual identification in members from his forum and shared his results (Kerlin & Beyer, 2003; Kerlin, 2004; Kerlin, 2005). He found a strikingly high rate of transgenderism in the AMAB survey respondents (more than 100 out of 500 or about 31%) (Kerlin & Beyer, 2003; Kerlin, 2004; Kerlin, 2005). Kerlin presented the results of his survey as papers at a couple of scientific conferences (DES Action, 2020b). However, they were never peer-reviewed or formally published in an actual scientific journal (DES Action, 2020b). Kerlin’s study is among the most important contributions to the popular notion that prenatal DES causes transgenderism in people AMAB. Unfortunately however, the findings are very low-quality data and should be interpreted with caution (DES Action, 2020b). The most important limitation of the study is the nature of how its sample was obtained and the very high likelihood of sampling bias (i.e., a biased and non-representative sample). As evidenced by the much higher quality data provided by Troisi et al. (2020), it’s clear that Kerlin’s findings aren’t reliable evidence.

It’s certainly still possible that prenatal DES exposure could increase the probability of transgenderism in people AMAB. But we don’t have the data to say whether this is actually the case or not. Put another way, there is no quality evidence at this time that prenatal DES increases the likelihood of transgenderism in individuals AMAB. In any case, the Troisi et al. (2020) data indicates that at best any such effect would be small—far smaller than that implied by Kerlin (2005). This is of course not to say that prenatal DES exposure couldn’t cause other problems (e.g., reproductive tract abnormalities like later vaginal/cervical clear cell cancer in individuals AFAB) however.


Taken together, the support for the notion that prenatal DES exposure increases the likelihood of transgenderism in people AMAB is very weak and this idea remains hypothetical rather than a confirmed finding. Moreover, the data we do have indicates that any such effect could at most be small and that the incidence of transgenderism in individuals AMAB with prenatal exposure to DES is very low.


Some additional coverage of the study and this topic by the advocacy group DES Action can be found here and here. Scott Kerlin does not seem to be happy about the study at all and discusses it here.