Healthcare provider communication and current contraceptive use among transgender men and gender-diverse people: Results from an online, cross-sectional survey in the United States.

Over five months of data collection, 5,005 people initiated the study: 4,207 (84%) from among existing participants of The PRIDE Study (35.3% of likely eligible participants of The PRIDE Study at the time), and 798 (16%) from the study-specific website (an unknown proportion of those exposed to study information.) Among the 5,005 initiators, 1,694 (34%) expressed a gender identity that aligned with the broad umbrella of transgender or gender diverse and were female or intersex assigned at birth; 1281 (76%) were existing participants of The PRIDE Study who were invited to opt-in to this cross-sectional, stand-alone survey, and the remainder (n=413, 24%) were recruited from the general population. Among these 1,694 TMGD respondents, over half (n=916, 54%) were between the ages of 20–29, had health insurance coverage (n=1512, 89%), and had completed a college or graduate degree (n=1054, 62%) ( Table 1 ). A minority of respondents (n=322, 19%) identified as American Indian/Alaska Native, Asian, Black/African American, Hispanic/Latinx, Middle Eastern/North African, and/or Native Hawaiian/Pacific Islander, versus White only (n=1293, 76%). Most respondents endorsed multiple gender identities (n=1036, 61%), and multiple sexual orientations (n=1010, 57%). Most respondents (n = 973, 57%) reported a conversation with a provider about contraception use in their lifetime: 397 (23%) had a provider-initiated conversation, 357 (21%) had a patient-initiated conversation, and 219 (13%) had both a provider-initiated and patient-initiated conversation ( Table 1 – 2 ). Among those who had a conversation with a provider, participants reported low comfort asking about contraception: 233 (24%) reported being not at all or only a little comfortable asking their provider all the questions they had about contraception use ( Table 2 ). Among respondents who reported a history of gender-affirming hormone therapy, puberty blockers, and/or gender-affirming surgeries (n=879), 357 (41%) had a conversation about contraception with their provider before starting gender-affirming care, while 196 (22%) had a conversation after initiating gender-affirming care; 335 (38%) of respondents with a history of gender-affirming care reported no conversation about contraception with their provider ( Table 2 ). Overall, 809 (48%) respondents were currently using any method of contraception for any reason, most commonly barrier (n=282, 17%) and LARC (n=279, 17%) methods ( Table 3 ). Respondents who reported any prior conversation with a provider about contraception were currently using contraception at a higher proportion than were respondents who did not have a conversation: 78% among those with a patient-initiated conversation, 42% among those with a provider-initiated conversation, and 26% among those with no conversation (p<0.001). This pattern was most pronounced among LARC users. Overall, the most common reasons for ever having used contraception were to prevent pregnancy (n=837, 70%), avoid period symptoms (n=662, 55%), stop menstruation (n=537, 45%), and protect against STIs (n=439, 36%) ( Table 4 ). Respondents who self-initiated a conversation with a provider about contraception were more likely to use contraception to prevent pregnancy or to minimize and prevent menstruation and its symptoms than were respondents who only had provider-initiated conversations, or no conversation ( Table 4 ). The most frequently reported barriers to contraceptive use included difficulty finding affirming healthcare providers (n=535, 44%), the cost of contraception (n=379, 32%), and the cost of healthcare visits (n=357, 30%) ( Table 4 ). In adjusted logistic regression analyses, TMGD respondents who had any conversation with a provider were more likely to currently use a contraceptive method than were respondents who had never had a conversation ( Table 5 ). Across all models, this association was strongest for those who reported a patient-initiated conversation – these respondents had 3.8–11.9 times the odds of current contraception use compared to those who never had a conversation with their provider. However, the association between provider-initiated conversations and contraceptive use was strong as well. Adjusting for reasons for contraceptive use attenuated the association substantially across models. Communication with a healthcare provider was most strongly associated with LARC use.

Between May and September 2019, we recruited individuals for an online survey from (1) among enrolled participants of an existing panel of lesbian, gay, bisexual, transgender, queer, questioning, intersex, asexual and more (LGBTQIA+) adults, The Population Research in Identity and Disparities for Equality (PRIDE) Study, as well as (2) an anonymous study-specific website distributed to the public via community organizations, listservs, and in-person LGBTQIA+ health events. Eligible respondents identified as transgender men, nonbinary, and/or gender-diverse, some also identified as intersex, selected “Female” or “Not listed” for their sex assigned at birth, and were aged 18 years or older, resided in the US or its territories, and could read and write in English. We excluded individuals who were assigned male at birth. We programmed the survey on the Qualtrics platform (Qualtrics LLC; Provo, UT), and implemented several custom and automated steps to prevent duplicate submissions from the same respondent. Respondents who completed the survey were entered in a raffle to win one of 100 $50 electronic gift cards. Further details on recruitment and survey content, design, and format have been described previously[ 14 , 15 ]. Respondents provided electronic informed consent prior to starting the survey. The Institutional Review Boards of Stanford University School of Medicine and the University of California, San Francisco reviewed and approved this study and continuing review is now maintained additionally via WCG (WIRB-Copernicus Group, Inc). The primary outcomes for this analysis included current use of contraception for any purpose and communication with healthcare providers about contraceptive use. We measured current contraceptive use with the following question: “Of the birth control methods you have ever used for ANY reason, please select below the method(s) you are currently using.” Respondents could select one or more of the 21 contraception options (described in detail) that were listed or could select that they were not currently using any method. Based on the descriptions of each contraceptive method, we categorized individual methods as: (1) long-acting reversible contraception (LARC) (hormonal intrauterine device (IUD), copper IUD, implant); (2) combined hormonal contraception (ring, patch, combined pill); (3) barrier methods (external condom, internal condom, diaphragm, cervical cap, sponge); (4) progestin-only methods (progestin pill, shot); (5) permanent contraception (partner permanent contraception, respondent permanent contraception); or (6) other (abstinence, spermicide, emergency contraception, fertility awareness, withdrawal). To assess patient-provider communication about contraception, respondents could select all that applied from: “Has a provider ever discussed birth control methods with you for the purposes of pregnancy prevention?”; “Yes, I brought it up”, “Yes, my provider brought it up”, “No”. We categorized respondents as having had provider communication that was patient-initiated, provider-initiated, both patient- and provider-initiated, or never had a conversation. We determined the timing of conversation about contraception relative to receipt of gender-affirming care with the question: “When did you and your provider discuss these birth control methods?”, for which respondents could select before and/or after pursuing gender-affirming hormone therapy and/or surgery. We also assessed respondent comfort: “How comfortable did you feel asking your provider all of the questions you had about birth control?” with answer choices that ranged from very comfortable to not at all comfortable. To evaluate experiences trying to access contraception care, we asked: “Have any of the below difficulties ever made it HARDER for you to get birth control?” with a list of possible barriers. Respondents could also indicate their reasons for not using contraception: cost of contraception, cost of healthcare visit, time required, travel required, and difficulty finding affirming healthcare providers. Sociodemographic characteristics measured included: age in years; race or ethnicity; gender identity (agender, cisgender man, cisgender woman, genderqueer, man, nonbinary, transgender man, transgender woman, Two-Spirit, woman, another gender (specify if desired), and prefer not to say); intersex identification; sexual orientation (asexual, bisexual, gay, lesbian, pansexual, queer, questioning, same-gender-loving, straight/heterosexual, or another sexual orientation); relationship status; educational level; and health insurance coverage (yes, no). For racial or ethnic identity, sexual orientation, and gender identity, respondents could select multiple options. We also assessed the proportion of healthcare providers who were aware of the respondent’s gender identity on a scale of 1 (out to 0% of providers) to 11 (out to 100% of providers). Respondents self-reported ZIP code, which we used to convert to US census region. We described sociodemographic characteristics, reproductive history, and contraceptive use data with frequencies and ranges, overall and by history of provider communication. To evaluate differences across respondents who reported varying levels of provider communication about contraception, we conducted Kruskal-Wallis tests, 𝒳 2 tests of independence and Fisher’s exact tests. Secondly, we also implemented a series of nested multivariate logistic regression models to examine the association between communication with a healthcare provider about contraception use and current contraception use. In these models, we categorized the exposure as (1) never had a conversation with a healthcare provider about contraception use (reference group), versus (2) had a patient-initiated conversation, (3) had a provider-initiated conversation, or (4) had both patient- and provider-initiated conversations with a healthcare provider about contraception use. We treated the outcome variable as binary: (1) any current use of a contraceptive method versus (2) no current use of a contraceptive method. In the nested models, we adjusted for potential confounders of the communication and contraceptive use relationship ( Supplemental Figure 1 ). In Model 1, we modeled the unadjusted association between having spoken to a healthcare provider about contraception and current use. Model 2 adjusted for the following covariates: age (continuous, centered on the mean), education (categorical: less than a college degree (reference), college degree/some graduate or professional study, or grad school degree), gender identity (transgender man/man only (reference), both transgender man + gender-diverse identities, gender-diverse identities only), gender identity outness (discrete: 0% (reference) up to 100%), health insurance coverage (categorical: yes, no (reference)), and due to limited racial diversity in the sample a crude binary measure of race/ethnicity as an imperfect proxy for experiences of racism in patient-provider interactions (binary; anyone who indicated any of the following identities (reference): American Indian/Alaska Native, Asian, Black/African American, Hispanic/Latinx, Middle Eastern/North African, or Native Hawaiian/Pacific Islander, versus people who selected White only). Model 3 adjusted additionally for a binary indicator of reasons for contraceptive use (for pregnancy prevention, versus any other reason (reference)). Given low levels of missingness across included covariates, we modeled associations among respondents with complete data. Additionally, we modeled the association between communication with a provider as described above in model 3 and the current use of specific method types (barrier, LARC, combined hormonal methods, progestin-only methods, permanent contraception, and other methods). All analyses were conducted in Stata version 18.0 (StataCorp, College Station, TX).

In a cross-sectional survey of 1,694 TMGD people in the United States, we found notable gaps in communication with healthcare providers about contraception: one third had never had a conversation about contraception with their provider, and among those who had, nearly one in four did not feel comfortable asking all their questions. These findings highlight a need for provider training on contraceptive counseling for TMGD people, and the importance of patient-initiation and advocacy during these conversations. Indeed, previous research has noted that providers who encourage patient involvement and tailor their counseling to the patients’ needs and preferences – patient centered and shared decision-making – are more likely to improve patient outcomes and long-term contraceptive adherence[ 11 , 24 – 28 ]. Additionally, considering half of our respondents disclosed their gender identity to their provider, provider knowledge and respect for gender identity may also influence comfort with self-advocacy[ 29 , 30 ]. This study is limited by several factors. One limitation is the potential for recall bias in the exposure, such that those who had a conversation with a provider that led to starting a contraceptive may be more likely to recall that conversation than those who did not go on to use a method. This could lead to misclassification of the exposure differentially by outcome, and thereby inflate the observed association. Similarly, it could be that those already using contraception are more likely to initiate a conversation with their provider to ask questions; due to the limitations of survey design, we cannot tease apart the direction of the association. Further, given that the exposure requires lifetime recall, it is possible that respondents might fail to recall conversations further in the past. An additional limitation is the difference in scope between our primary exposure of interest (a conversation about contraception use for pregnancy prevention ) and our outcome (contraceptive use for any reason) . Respondents may have had conversations with their providers about contraceptive use for other reasons, which could have resulted in some respondents reporting “no conversation” here, even if they had discussed contraception for OTHER reasons with a provider. This would result in misclassification of respondents by exposure and could conservatively bias our estimates toward an underestimation of the relationship. Another limitation is that the survey did not ascertain if there were intervals beyond the past 12 months in which the respondent engaged in sexual activity that could lead to pregnancy. Thus, it is likely that some respondents were incapable of pregnancy and thus unlikely to be interested in the use of contraception to prevent pregnancy. General limitations include the racial and ethnic composition of the study population. Studies have demonstrated that marginalized racial and ethnic groups are more likely to experience barriers to healthcare access, receive unequal treatment, and are less likely to advocate for themselves due to racial bias[ 15 , 31 – 35 ]. Thus, the results from this study are unlikely to completely capture these disparities. Most of the sample also had at least some college education or had health insurance. Collectively, these advantages could bias the findings toward underestimation of barriers to contraceptive care among populations who must additionally navigate structural racism, underinsurance, and lack of education and/or health literacy. Further, we did not ascertain respondents’ gender identity, age, or other socio-demographic characteristics at the time of their conversations with their provider, nor the frequency of these conversations – these characteristics may moderate the relationship between communication and contraceptive use. Further, as most respondents found the survey through broad LGBTQIA+ related health events and organizations, the sample may not represent the full TMGD population. Despite these limitations, this research is one of few studies among a large sample of TMGD people assigned female or intersex at birth from across the United States, conducted in a community-engaged approach, with detailed findings on contraceptive use, motivations, and barriers to care with a particular emphasis on provider communications. Given formidable barriers to sexual and reproductive healthcare faced by TMGD people, training providers on how to initiate affirming and relevant conversations with TMGD patients about contraceptive needs and preferences is essential, as are efforts to educate and empower TMGD individuals to advocate for the information they need in these interactions. Future research should explore the relationship between health care provider communication and contraceptive use among TMGD people in longitudinal studies that allow for the establishment of temporality for new initiators of contraceptives and that capture accurate contemporaneous measures of gender, body parts, and sexual activity. This will enable better characterization of the role and influence of health care provider communication on contraceptive initiation and sustained use among TMGD people in the United States.

Transgender men and gender-diverse (TMGD) people use contraception for many reasons, including pregnancy and sexually transmitted infection (STI) prevention, for menstrual suppression to reduce gender dysphoria, and for other reasons[ 1 , 2 ]. However, research on contraception use and access among TMGD is limited[ 1 – 4 ]. Across studies, 33–70% of TMGD participants report negative healthcare experiences, such as poor and/or abusive communication, explicit discrimination based on their gender identity, denied or delayed healthcare service, and inconsistent levels of care[ 5 – 8 ]. Considering that over 1 million individuals in the United States are transgender and that access to contraception is a core indicator of healthcare quality, research is needed to understand how to facilitate access to contraception for TMGD people in the United States[ 5 , 6 , 8 – 10 ]. Research indicates that contraceptive counseling can positively affect long-term contraceptive use and achievement of reproductive goals[ 11 – 13 ]. Conversely, inadequate counseling and poor provider communication about contraception can lead to the use of misaligned methods given contraceptive goals or dissuade people from contraceptive use entirely[ 14 ]. Factors such as deterrence of patient advocacy ( i.e., disregarding patient preferences and values, passive provider listening, lack of respect for patient bodily autonomy, minimum or no discussion of options, discouraging informed choice, etc .) and lack of provider knowledge about contraceptive options can prevent patients from making informed decisions about their contraceptive needs[ 13 , 15 – 17 ]. These deficits in care are experienced more acutely by TMGD individuals and can be exacerbated by provider misconceptions, discrimination, and misinformation[ 18 , 19 ], as well as exclusion from family planning research. In previous studies, 5.5% to 9% of transgender men believed testosterone had contraceptive benefits based on a healthcare provider’s advice[ 2 , 3 ]. This statistic highlights the role of inaccurate counseling by providers; while testosterone can induce hypothalamic-pituitary-gonadal suppression resulting in anovulation, the degree and duration of this anovulation is not fully understood and thus testosterone is not generally considered a fully reliable or primary form of contraception to prevent pregnancy[ 20 – 23 ]. A lack of national-level data on the experiences of TMGD people with contraceptive use, reasons for use, and if and how TMGD individuals talk with their healthcare providers about contraception limits evidence-based care. To address this gap, we analyzed survey data collected from a cross-sectional study that examined contraception care experiences of TMGD adults in the United States.