Contraceptive use among transgender men and gender diverse individuals in the United States: Reasons for use, non-use, and methods used for pregnancy prevention.

There were 1694 eligible participants for this sub-study. Details on exclusions have been previously described [ 19 ]. The median age was 27 years (IQR 23–33), and participants were mostly White, insured, college-educated, and had a range of gender identities and sexual orientations. Almost all participants were assigned female at birth (99.4%), and 69 (4.1%) identified as intersex ( Table 1 ). Most participants 1225 (72.3%) had never used testosterone; 8.2% had used it prior, and 19.5% reported current use. Approximately half (50.9%) of the transgender men had ever used testosterone, versus 20.5% among nonbinary individuals, and 18.3% among genderqueer individuals. The distribution of gender identities differed between those who had never versus ever used testosterone. Among ever users of testosterone, 71.9% identified as transgender men, 62.9% indicated multiple gender identities, and 38% identified as nonbinary individuals. There was no difference in testosterone use by sex assigned at birth or comparing those who did and did not identify as intersex. Among participants with a pregnancy history, 21.9% had ever used testosterone. Approximately half of participants with a hysterectomy (48%) and/or bilateral oophorectomy (54.6%) had ever used testosterone ( Table 1 ). Most participants 1203 (71.0%) had used some contraceptive method in their lives and pregnancy prevention was the most common reason (49.4%) for use cited. Reasons for contraceptive use were similar across ever versus never users of testosterone, except for the desire to avoid period symptoms (39.1% overall; never (40.7%) vs. ever (34.8%), p = 0.02) and preventing hair growth (never (3.3%) vs. ever (1.3%), p = 0.02) – both of which were more prominent among never users of testosterone than ever users ( Table 2 ). Overall, 27% of respondents reported never having used contraceptives. The most common reasons cited for non-use were: not engaging in penis-in-vagina sex (5.7%) or not having sex with people who produce sperm (4.8%). A smaller proportion of the sample reported being incapable of pregnancy 46 (2.7%). Among those reporting being pregnancy incapable, many had a hysterectomy ( n = 30/46, 65.2%) and some also reported a bilateral salpingoophorectomy ( n = 26/46, 56.5%; none had bilateral salpingoophorectomy without concurrent hysterectomy), and fewer without a hysterectomy reported testosterone use ( n = 16/46, 34.8%). Being incapable of pregnancy was more commonly reported among ever versus never testosterone users (4.5% vs. 2.0%, p = 0.01), but when we examined only those without a hysterectomy, there was no difference between individuals currently using testosterone ( n = 7/16) and those who never used testosterone ( n = 9/16). Few individuals who never used contraceptives were open to pregnancy (0.4%). Among non-pregnancy related reasons, the most commonly cited reasons were “I do not want to use estrogen or feminizing hormones/am concerned birth control will interfere with my gender affirmation process” (5.0%), which was more commonly reported by ever vs. never testosterone users (7.5% vs. 4.0%, p = 0.006) and concern about contraceptive side effects (3.7%), which did not differ by testosterone exposure. The answer “I do not have any symptoms such as chest tenderness, bloating, acne, pain from cramping, heavy bleeding, sometimes referred to as pre-menstrual syndrome or PMS that birth control is sometimes used to help treat” differed between ever (4.9%) vs. never (2.4%) testosterone users ( p = 0.012) ( Table 2 ). Approximately half of respondents reported ever using contraception to prevent pregnancy 837 (49.4%). Among this group, the three most reported contraceptive methods were external condoms (91.8%), combined hormonal contraceptive pills (63.0%), and withdrawal (45.9%). Responses differed among ever and never testosterone users for the internal condom (14.2% vs. 8.3%, p = 0.02) ( Table 3 ). Among those currently using contraception for pregnancy prevention, the most common methods included external condoms (35.4%), hormonal IUDs (24.8%), and abstinence (19.2%). Among current contraceptive users for pregnancy prevention, responses differed significantly between ever and never testosterone use in whether individuals reported use of an internal condom (5.0% vs. 1.0%, p = 0.01), use of the combined hormonal contraceptive pill (2.1% vs. 14.1%, p < 0.001), and fertility awareness methods (0.7% vs. 4.1%, p = 0.03) ( Table 3 ). In multivariable logistic regression analyses among participants who reported no desire for future pregnancy and were not currently surgically sterilized, we found no association between ever vs. never testosterone and current use of LARC, barrier, or progestin-only methods. However, participants who reported ever using testosterone had lower odds of currently using combined hormonal contraceptive pills, patch, or ring for pregnancy prevention than did those who had never used testosterone (aOR, 0.11; 95% CI 0.03–0.45) ( Table 4 ). Some participants reported relying on testosterone for pregnancy prevention either currently ( n = 30, 3.6%) or in the past ( n = 55, 6.6%). Among those currently relying on testosterone for birth control, most ( n = 26/30, 86.7%) were not using other contraceptive methods to prevent pregnancy and did not report surgical sterility. Contraceptives used in addition to testosterone were abstinence (6.7%), external condoms (3.3%), and withdrawal (3.3%) ( Table 5 ).

Eligible participants included TGD individuals AFIAB 18 years old and older, in the US or its territories, who could complete an English survey. We recruited participants through a community-facing anonymous website as well as The Population Research in Identity and Disparities for Equality (PRIDE) Study, a national community-engaged prospective cohort study hosted by Stanford University School of Medicine. PRIDEnet [ 17 ], whose mission is to catalyze health research among sexual and/or gender minority people, facilitated study participation through email, social media, and in-person and online community outreach [ 18 , 19 ]. An online survey was administered from May to September 2019 via the survey platform Qualtrics (Provo, UT, USA) after an extensive previously described community-engaged design process [ 19 , 20 ]. Survey variables included self-reported sociodemographic characteristics, medical and surgical history, pregnancy history, current and ever use of contraceptives and/or gender-affirming hormones (including testosterone), and contraception use specifically for pregnancy prevention. We used three screener questions to determine eligibility as a TGD person AFIAB: (1) “What is your gender identity?”[free text]; (2) multiple-selection 10-item categorical gender identity question wherein participants could “select all that apply”, and (3) “What sex were you assigned at birth, for example on your original birth certificate?” ( Table 1 ). The 10 gender categories listed were: agender, cisgender man, cisgender woman, man, nonbinary, transgender man, transgender woman, Two-Spirit, woman, additional gender identity [free text], and “prefer not to say”. Participants who selected “male” or “prefer not to say” for sex assigned at birth screened ineligible. We categorized respondents as TGD if responses included any gender besides “woman” or “cisgender woman” exclusively. A fourth question “Do you identify as intersex?” in the full survey categorized participants as intersex-identifying even if they were assigned female sex at birth. Additionally, we collected sexual orientation, race/ethnicity (as a social characteristic), education, geography, and parental status. For sexual orientation, there were 10 response options: asexual, bisexual, gay, lesbian, pansexual, queer, questioning, same gender loving, straight/heterosexual, and/or another sexual orientation. For race/ethnicity, there were also 10 response options ( Table 1 ). Sexual orientation and ethnoracial identity options were “select all that apply” and participants could write in additional answers. US Census Bureau regions was assigned from self-reported zip-codes. Participants reported any past use of gender-affirming hormones (including testosterone), which (if any) gender-affirming surgeries they had undergone (specifically, hysterectomy and oophorectomy), and separate questions queried ever and current contraceptive methods used and reasons for use and non-use. For questions analyzed see Appendix Table 1 . We classified contraceptive method use and testosterone use into three categories: “never”, “ever”, or “current”. The “ever” classification combined any past and current use. “Never” and “ever” categories were mutually exclusive, while “current” and “ever” categories were not mutually exclusive. Some “ever” users were classified as “current” users if use was ongoing. To characterize the sample, we tabulated frequencies and percentages for sociodemographic variables and reproductive history-related variables overall, and by a history of testosterone use. To describe motivations behind ever use of contraception, we tabulated frequencies and percentages of the reasons individuals reported having used contraception as well as the reasons for never contraceptive use, overall and by a history of testosterone use. To explore contraceptive use patterns for pregnancy prevention, we tabulated frequencies and percentages of each method by ever use or current use, overall, and by history of testosterone use. In all these analyses, we evaluated outcomes differences between ever versus never users of testosterone using Chi-square, Fisher’s exact tests, and Kruskal-Wallis tests. Additional analyses tabulated the frequency and percentage of other contraceptive methods concurrently used among the sub-sample of respondents who reported currently relying on testosterone for contraception. We modeled the association between a binary (ever/never) measure of testosterone use and current contraceptive use for pregnancy prevention via four separate multivariable logistic regression models. Each model evaluated the adjusted and unadjusted association between ever versus never testosterone use and use of (1) Long-acting reversible contraceptive (LARC) methods (i.e ., hormonal intrauterine device [IUD], copper IUD, or implant); (2) combined hormonal contraceptives (pill/patch/ring); (3) barrier/external methods (external or internal condom, diaphragm); and (4) progestin-only methods (pill or depo-provera (medroxyprogesterone acetate) shot). Based on a directed acyclic graph (DAG [ 21 ]), we identified covariates for adjusted models to minimize confounding bias ( Appendix Fig. 1 ). We included the following covariates in adjusted models: formal education (binary: college degree (reference) vs. less than a college degree), health insurance coverage (binary: no insurance (reference) vs. insurance coverage), and race/ethnicity (reference: non-Hispanic White, vs. all other ethnoracial identities combined (American Indian and Alaska Native, Asian and Pacific Islander, Black/African American, Hispanic/Latinx, Middle Eastern and North African, Other, and unknown (discrete answer choice)). We used Stata 15.1 (StataCorp; College Station, TX, USA) for quantitative analysis and Microsoft Excel for qualitative analysis. The Institutional Review Boards of Stanford University School of Medicine, The University of California, San Francisco, and WIRB-Copernicus Group approved this study. All participants provided consent electronically before participation.

To our knowledge, this is the first paper to disaggregate contraception use for pregnancy prevention from other uses among TGD individuals AFIAB. Among the 1694 TGD AFIAB individuals in the sample, contraception was used across all genders – including users and non-users of testosterone – and 49.4% of respondents had used contraception for pregnancy prevention. The three most reported types of contraception ever used for pregnancy prevention were external condoms (91.8%), combined hormonal contraceptive pills (63.0%), and withdrawal (45.9%). Taken together, these methods can be characterized as having high personal autonomy, are noninvasive, and two (condoms and withdrawal) do not require any interaction with health care providers. Types of contraception used and reasons for using contraception sometimes differed between ever and never users of testosterone – with implications for contraceptive counseling given the unknown efficacy of testosterone for pregnancy prevention [ 9 ], and poor experiences of managing unintended pregnancies for TGD AFIAB individuals [ 22 , 23 ]. The contraceptive use prevalence reported here – specifically condoms and contraceptive pills - is similar to or slightly higher than previous reports [ 2 , 10 ], which may be attributable to the lower proportion of testosterone users in our sample. Although we cannot determine whether method use differences stem from participant preference, contraceptive need, and/or sampling strategy and sample characteristics, our results suggest that assumptions that TGD individuals will not want estrogen-containing contraceptives may be unfounded, and patient-centered counseling across method types should be employed [ 24 ]. This should be considered on balance with our finding that some participants did not want to use contraception because of possible interference with gender-affirmation, echoing prior research [ 25 ] noting concerns regarding gender dysphoria from estrogen-containing contraceptives. These findings are limited by a convenience sample wherein TGD individuals who are White with high socioeconomic status are over-represented, and thus, results cannot be generalized broadly. Although we assessed intersex assignment at birth and intersex identity separately, the resulting sample size was too small for us to analyze contraceptive use and needs. This begs for research focused on the contraceptive needs and experiences of intersex individuals. Finally, contraceptive use was not clinically verified. Study strengths include comprehensive sexual and reproductive health data collected via an affirming survey from a large sample of TGD individual AFIAB - created, deployed, and analyzed using community engagement. Since data were collected in 2019, several shifts have made accessing contraception and gender-affirming care more challenging, including the 2022 Supreme Court ruling in Dobbs v Jackson Women’s Health Organization [ 26 – 28 ] and legislation limiting gender-affirming care in over 23 states [ 29 ]. Gender-affirming care sites are often also sites of contraceptive care and vice versa, so restrictions on either may limit access to both [ 30 , 31 ]. However, shifts have expanded care access as well – including emboldened protections for gender-affirming care and reproductive health care in some states, rising civic activism, medical society position papers [ 32 ], and expanded medical guidelines, such as the new World Professional Association for Transgender Health (WPATH) Standards of Care (2022) that calls for gender-affirming contraceptive care [ 33 ] and the 2024 U.S. Selected Practice Recommendations for Contraceptive Use, which consider contraceptive care for transgender people [ 34 ]. Our data illustrate that TGD individuals AFIAB use contraception for many purposes, including pregnancy prevention. These data serve as both a proof of principle and a date-stamped backdrop against which more contemporary data from a post- Dobbs landscape with severely restricted care and education will be compared. Future research is needed to explore how the desire for or receipt of gender-affirming processes influences contraceptive decision-making, contraceptive outcomes, and gender-affirming goals [ 9 ].

Transgender and gender diverse (TGD) individuals assigned female or intersex at birth (AFIAB) experience unintended and undesired pregnancies [ 1 ], yet research is limited on contraceptive use for pregnancy prevention in this population. Research indicates that TGD individuals AFIAB use a wide range of contraceptive methods – often condoms [ 2 – 4 ] and depot medroxyprogesterone [ 5 ] and use patterns differ from cisgender women [ 6 ]. Reasons for use include menstrual suppression [ 7 , 8 ] balanced against concerns about hormonal methods interacting with testosterone therapy or undermining gender-affirming goals [ 2 , 9 , 10 ]. Prior research, however, has not differentiated contraceptive use for pregnancy prevention versus other reasons, nor differences in contraceptive use between testosterone users and non-users. Approximately 49% of TGD individuals AFIAB use testosterone [ 11 ], and some who use it for gender affirmation also use it for pregnancy prevention [ 2 , 11 , 12 ]. Many clinicians and TGD individuals mistakenly believe testosterone reliably prevents pregnancy and can be used as a contraceptive [ 2 , 10 , 13 ]. In one study of 197 transgender men, 16% used testosterone as contraception, and 6% reported being advised to do so by a healthcare provider [ 2 ]. However, pregnancy can occur during testosterone use, and ovulation patterns while using testosterone have not been well-described [ 1 , 3 , 14 , 15 ]. However, contraceptive efficacy for pregnancy prevention may not be the most important factor for TGD individuals [ 9 ]. Other considerations may include contact with medical providers, needing examinations/procedures, impact on menstrual bleeding, and actual/perceived interference with affirmed secondary sex characteristics [ 9 ]. Note that throughout this manuscript, we use the term sex, as in “assigned sex at birth” and “secondary sex characteristics” to connote anatomic and physiological features. We use the term “penis-in-vagina sex” to connote the sexual act that could results in pregnancy if sperm are released near the cervix. We use the term gender to describe the socially constructed roles, behaviors, attributes, and activities that pertain to categories of humans and which are usually ascribed to women, men, and people of diverse genders [ 16 ]. To inform equitable, evidence-based contraceptive counseling and shared decision-making with TGD individuals, data are needed on contraceptive use, experiences, and rationale for use among TGD individuals who do and do not use testosterone. We sought to close these gaps by surveying TGD individuals in the US about current and past contraceptive use, reasons for use and non-use, and then contextualizing these findings by gender-affirming testosterone use.