Healthcare avoidance due to anticipated discrimination among transgender people: A call to create trans-affirmative environments.

Transgender people encounter interpersonal and structural barriers to healthcare access that contribute to their postponement or avoidance of healthcare, which can lead to poor physical and mental health outcomes. Using the 2015 U.S. Transgender Survey, this study examined avoidance of healthcare due to anticipated discrimination among transgender adults aged 25 to 64 (N = 19,157). Multivariable logistic regression analysis was conducted to test whether gender identity/expression, socio-demographic, and transgender-specific factors were associated with healthcare avoidance. Almost one-quarter of the sample (22.8%) avoided healthcare due to anticipated discrimination. Transgender men had increased odds of healthcare avoidance (AOR = 1.32, 95% CI = 1.21-1.45) relative to transgender women. Living in poverty (AOR = 1.52, 95% CI = 1.40-1.65) and visual non-conformity (AOR = 1.48, 95% CI = 1.33-1.66) were significant risk factors. Having health insurance (AOR = 0.87, 95% CI = 0.79-0.96) and disclosure of transgender identity (AOR = 0.77, 95% CI = 0.68-0.87) were protective against healthcare avoidance. A significant interaction of gender identity/expression with health insurance was found; having health insurance moderated the association between gender identity/expression and healthcare avoidance. Providers should consider gender differences, socio-demographic, and transgender-specific factors to improve accessibility of services to transgender communities. A multi-level and multi-faceted approach should be used to create safe, trans-affirmative environments in health systems.

Introduction

Compared to gay, lesbian, and bisexual individuals, transgender people experience a higher prevalence of discrimination, which adversely affects their health and well-being throughout their life course (Kcomt, 2019; Lambda Legal, 2010; Macapagal, Bhatia, & Greene, 2016). Such discriminatory behavior ranges from being denied healthcare to being physically, verbally, or sexually abused in healthcare settings (Grant et al., 2011; James et al., 2016). Evidence suggests that many transgender people either avoid/delay receiving healthcare or engage in selective disclosure about their transgender identity to health providers in order to avoid discrimination (Grant et al., 2011; Institute of Medicine, 2011; James et al., 2016; Reisner et al., 2014; Stotzer, Kàopua, & Diaz, 2014). However, avoiding or delaying care can lead to poorer physical and mental health outcomes (Seelman, Colón-Diaz, LeCroix, Xavier-Brier, & Kattari, 2017), and selective disclosure can result in receiving inappropriate care or missed opportunities for preventive care (Bauer et al., 2009; White Hughto, Reisner, & Pachankis, 2015). Although the discrimination experienced by transgender people when seeking healthcare services has been well documented, transgender populations are not monolithic. There is limited understanding about their differential experiences to healthcare access based on gender identity/expression and how this may be moderated by other factors.

The term transgender encompasses a diverse range of identities including transgender women (those assigned male at birth but who identify predominantly as women), transgender men (those assigned female at birth but who identify predominantly as men), non-binary/genderqueer (individuals whose gender identity may be fluid or ambiguous, or who may perceive themselves as being both male and female), and cross-dressers (individuals who wear clothing typically associated with another gender, either in private or in public) (Institute of Medicine, 2011). Because of their history of stigmatization, the recruitment of transgender respondents remains challenging (Meier & Labuski, 2013). Extant studies on transgender people consist predominantly of transgender women, with transgender men and non-binary/genderqueer identities under-represented (Levin, 2014) and cross-dressers often excluded (Teich, 2012).

Access to healthcare is defined as the “timely use of personal health services to achieve the best personal outcomes” (Institute of Medicine, 1993, p. 4) and is characterized as the ability and ease of the consumer to seek and obtain needed services from providers or institutions as well as the cost of healthcare (Levesque, Harris, & Russell, 2013). Yet, the interpersonal and structural barriers that transgender people encounter compromise their access to healthcare. White Hughto et al. (2015) explained how stigma—the process of labeling, stereotyping, and marginalizing as a form of social control—can impact transgender people's well-being. Stigma occurs at the structural, interpersonal, and individual levels and is a fundamental cause of adverse health outcomes among transgender populations. It can function directly (by inducing stress responses, making individuals more vulnerable to physical and mental health problems) and indirectly (by restricting access to resources). Structural stigma can manifest as providers' knowledge deficit about transgender people's health needs or organizational policies that are not affirming of transgender identities. Interpersonal stigma includes the harassment, abuse, and discrimination that transgender people may receive from healthcare providers. In turn, these experiences can make them vulnerable to individual stigma, inducing feelings of shame, anticipation of rejection, and a desire to conceal one's identity (White Hughto et al., 2015). The prevalence of these forms of stigma in healthcare encounters and their detrimental consequences for transgender people have been well-documented in a growing body of qualitative literature with transgender patients (Santos, Mann, & Pfeffer, 2019; Vermeir, Jackson, & Marshall, 2018).

The term cisgender refers to individuals whose gender identity aligns with the social expectations associated with their assigned sex at birth. Transgender stigma is rooted in a pervasive culture of cisnormativity (the systemic marginalization of transgender identities by representing cisgender identities as normal and assuming that all people are cisgender) (Bauer et al., 2009) and cisgenderism (an ideology that privileges cisgender identities by denigrating or pathologizing gender identities that do not align with assigned sex at birth) (Lennon & Mistler, 2014). Cisnormative assumptions are so pervasive that health systems and providers often do not question the experience of gender, do not anticipate the possibility of a transgender existence, and thus, may be unprepared when a transgender person seeks their services. Cisnormativity perpetuates the erasure of transgender identities, bodies, and experiences, reinforcing the lack of knowledge production and information dissemination about transgender people and their needs (Bauer et al., 2009). Qualitative research with healthcare providers has identified how such erasure contributed to their knowledge deficit about transgender people, resulting in culturally and medically incompetent care and systematic deficiencies in health policy (Logie et al., 2019; Snelgrove, Jasudavisius, Rowe, Head, & Bauer, 2012).

Even if they do not align with cisgenderist ideology on an individual basis, healthcare providers often work in broader social contexts that produce and perpetuate cisgenderism at the interpersonal and structural levels (Ansara, 2015). Subtle micro-aggressions may occur through unintentional practices, such as misgendering (where transgender patients are referred to in a manner that is inconsistent with their gender identity) or marginalizing (regarding a person's gender identity as weird) (Ansara, 2015). By relegating transgender individuals as the Other, cisnormativity and cisgenderism create an architecture of social exclusion, breeding prejudice and oppression against transgender identities (Ansara, 2015). Transgender stigma, cisnormativity, and cisgenderism foster disrespect and mistreatment of transgender patients.

Emerging findings on transgender people's differential healthcare experiences have revealed important disparities among gender identity/expression subgroups. Transgender men were twice as likely as transgender women to postpone needed healthcare due to anticipated discrimination (Jaffee, Shires, & Stroumsa, 2016; Kattari, Atteberry-Ash, Kinney, Walls, & Kattari, 2019). Although non-binary individuals were less likely to delay receiving healthcare relative to transgender women (Kattari et al., 2019), they were also significantly less likely to report being treated with respect by healthcare providers after disclosure of their transgender identity compared to binary transgender individuals (Kattari, Bakko, Hecht, & Kattari, 2020). To our knowledge, no studies have examined the healthcare experiences of cross-dressers.

There are also racialized differences to transgender people's healthcare access. Transgender people of color experience significantly higher levels of transphobic discrimination compared to their White counterparts in accessing health services (Kattari, Walls, Whitfield, & Langenderfer-Magruder, 2015). This underscores the need to consider how gender identity/expression may intersect with other minority identities. Individuals who hold multiple marginalized identities may experience disproportionate risk of exposure to discrimination and health inequities (Bowleg, 2012).

Having health insurance enables timely access to health services; being uninsured can lead to adverse health consequences and poorer quality of life (Blackwell, Martinez, Gentleman, Sanmartin, & Berthelot, 2009; Hoffman & Paradise, 2008). Transgender people are more likely to be uninsured compared to cisgender people (dickey, Budge, Katz-Wise, & Garza, 2016). There are also income-related disparities to healthcare access in the U.S. Many low-income people do not qualify for Medicaid and private insurance is often unaffordable even when it is available. Those living in poverty are more likely to lack health insurance, as much of their household budgets are used to cover basic needs (Hoffman & Paradise, 2008; Institute of Medicine, 2009). Transgender people experience higher levels of unemployment and poverty relative to cisgender people (Crissman, Berger, Graham, & Dalton, 2017), making their access to health insurance more challenging (Institute of Medicine, 2011).

Evidence suggests that visual non-conformity (recognizability as transgender by others) (White Hughto et al., 2015) and identity disclosure (“coming out” about their transgender identity to others) (Ducheny, Hardacker, Claybren, & Parker, 2019) are important factors associated with experiencing discrimination and healthcare access. Visual non-conformity is a form of visible stigma and increases the likelihood of experiencing transphobic discrimination in healthcare settings (Kattari & Hasche, 2016; Rodriguez, Agardh, & Asamoah, 2018). Individuals who are visually conforming may choose to conceal their identity and pass as cisgender as a way of managing transgender stigma (White Hughto et al., 2015). Being out when seeking medical care was positively associated with delayed medical care due to anticipated discrimination (Cruz, 2014). Disclosure may increase the risk of the transgender individual's exposure to stigma, but concealment can lead to restricted access to appropriate preventive care or transition-related healthcare (White Hughto et al., 2015).

Understanding the differential experiences of transgender people across gender identity/expression categories and examining the potential ways that sociodemographic and transgender-specific factors may moderate healthcare avoidance can reveal whether certain gender identity/expression subgroups may be in need of targeted outreach efforts to improve their healthcare access. Using a large, national transgender sample, this study addresses the following research questions:

Is there an association between gender identity/expression and healthcare avoidance due to anticipated discrimination?

What are the sociodemographic and transgender-specific factors associated with healthcare avoidance due to anticipated discrimination?

Do sociodemographic and transgender-specific factors moderate the effect of gender identity/expression in transgender people's avoidance of healthcare due to anticipated discrimination?

Methods

Data source

This study was a secondary analysis of the 2015 U.S. Transgender Survey implemented by the National Center for Transgender Equality (NCTE). Conducted online, this survey examined the experiences of transgender adults living in the U.S. The NCTE's multi-pronged approach to sample recruitment received an unparalleled response and resulted in the largest database of transgender people in the U.S. to date (N = 27,715). Non-probability sampling methods consisting of direct outreach, modified venue sampling, and snowball sampling were used. Eligibility requirements for survey completion were: ≥18 years of age; residing in a U.S. state, territory, or an American military base; at any stage in the gender transition process; and self-identify as transgender or another identity on the transgender spectrum. The term “transgender” was defined broadly to be inclusive of all forms of gender non-conformity and to encourage participation from individuals with diverse experiences regarding their gender identity. Further details about the survey methodology are available elsewhere (James et al., 2016). All procedures, including informed consent, received full human subjects review and IRB approval.

Study sample

The USTS contained a disproportionate number of white, young, and highly educated respondents compared to the U.S. general population. To enable the sample to be more representative of the population from which it was drawn, survey weights (based on the American Community Survey) accounting for educational attainment, race, and age provided with the original dataset were applied (National Center for Transgender Equality, 2017).

Health insurance inadequacy is a well-known barrier to healthcare access in the U.S. There are greatest needs for—and variability in—health insurance status among people between the ages of 25 and 64 (DeNavas-Walt & Proctor, 2015). Those aged ≥25 are unlikely to be covered by their parents' health insurance plans and those aged ≥65 are eligible for Medicare. Thus, this study's sample was restricted to persons aged 25 to 64. This age restriction enabled a more accurate analysis of the respondents' socio-economic status and minimized the possibility of confounding their socio-economic status with that of their parents'. Lastly, to acknowledge the spectrum of identities under the transgender umbrella, cross-dressers and non-binary/genderqueer respondents were included in this study, enabling us to explore healthcare access in these under-represented subgroups. The final weighted sample was 19,157 transgender adults.

Measures

Healthcare avoidance due to anticipated discrimination

Respondents were asked: “Was there a time in the past 12 months when you needed to see a doctor but did not because you thought you would be disrespected or mistreated as a trans person?” (yes/no).

Gender identity/expression

Respondents were asked to identify the term that best described their gender identity: cross-dresser, woman, man, trans woman (MTF), trans man (FTM), or non-binary/genderqueer. This was recoded into a four-category variable (transgender women, transgender men, non-binary/genderqueer, and cross-dressers).

Trans-specific variables

These consisted of visual conformity and disclosure of transgender identity. Visual conformity was assessed by the following item: “People can tell I am trans even if I don't tell them.” Response options were: always, most of the time, sometimes, rarely, and never. Responses were recoded to tertiles: conforming (other people could not tell that the individual was trans), somewhat conforming, and non-conforming (others could tell that the individual was trans). The respondents' degree of outness was assessed by the item: “How many people in each group below currently know you are trans?” The response set consisted of five categories (all know that I am trans; most know that I am trans; some know that I am trans; none know that I am trans; and I currently have no people like this in my life) for eight categories of people (immediate family; extended family; lesbian, gay, bisexual, or trans friends; straight, non-trans friends; current boss/manager/supervisor; current coworkers; current classmates; current healthcare providers). Respondents were scored based on their disclosure to each category of people in their lives (α = 0.74). If a respondent selected, “I currently have no person like this in my life,” then those categories were excluded in the respondent's overall score. The disclosure of transgender identity variable was a summary measure computed to reflect four categorical degrees of outness, ranging from disclosure to no one to disclosure to everyone. Due to sample size in each category, this variable was recoded into three categories: none to some, most, and all.

Sociodemographic variables

These consisted of race/ethnicity (non-Hispanic White, Hispanic/Latino/a, Black/African American, Biracial/Multiracial, and others); has health insurance (yes/no); and living in poverty (yes/no). The poverty measure was a recode based on the U.S. Census Bureau's official poverty measure in 2015.

Control variables

The covariates controlling for healthcare need included general health status and disability. Respondents were asked to rate their general health as: excellent, very good, good, fair, and poor. Because of category sizes, fair and poor health were collapsed into one category. Disability was assessed by asking respondents if they had any difficulties with hearing, seeing, concentrating/remembering/decision making, walking/climbing stairs, dressing/bathing, and completing errands, with the response options of yes or no. A summary measure was computed to identify respondents who reported having any one or more of the individual disabilities.

Interaction terms

To examine whether the effect of gender identity/expression was moderated by transgender-specific and sociodemographic variables, we created several 2-way interaction terms consisting of gender identity/expression and (a) race, (b) poverty, (c) having insurance, (d) visual conformity, and (e) disclosure.

Statistical analysis

Univariate frequency distributions were used to describe the study sample. Bivariate analyses consisted of Pearson's chi-square (χ2) tests to examine the associations between gender identity/expression with the covariates, and all covariates with the outcome measure. Multivariate logistic regression analyses were conducted to estimate the adjusted odds and 95% confidence intervals (CI) of avoiding healthcare due to anticipated discrimination. The logistic regression models were constructed in the following manner: First, gender identity/expression was entered as the sole main predictor into Model 1. Second, the covariates that provided some control for variabilities on healthcare need were entered into Model 2 (general health status and disability). Third, remaining predictors were entered into Model 3 (race, poverty, insurance, visual conformity, and disclosure), thereby enabling the estimation of the association of gender identity/expression with the outcome measure while controlling for other factors. Lastly, to determine whether the relationship between healthcare avoidance and gender identity/expression was moderated by race, living in poverty, having health insurance, visual conformity, and disclosure, we tested 2-way interactions between gender identity/expression and each of these predictors individually in the model. If the interaction was significant, subsequent analyses were conducted separately by gender identity. All analyses were conducted using SPSS, version 26.

Results

Overall, 55.1% identified as transgender women, 23.9% as transgender men, 16.1% as non-binary/gender queer, and 4.9% as cross-dressers. Table 1 displays the estimated distributions of the key study measures in the overall sample and across gender identity/expression subgroups. Almost one-quarter (22.8%) in the overall sample avoided healthcare due to anticipated discrimination, with the highest prevalence among transgender men (27.0%). In bivariate analyses, statistically significant associations were found between gender identity/expression and the control variables, and between all covariates with the outcome measure (results not shown).

Table 1

Estimated distributions of key study measures.

Variables	Overall		Transgender Women		Transgender Men		Non-binary/Genderqueer		Cross-dressers
Categories	N = 19,157		N = 10,561		N = 4,576		N = 3,087		N = 933
	n	(%)	n	(%)	n	(%)	n	(%)	n	(%)
Healthcare Avoidance	4,364	(22.8)	2,408	(22.8)	1233	(27.0)	596	(19.4)	128	(13.8)
Race
Non-Hispanic White	11,994	(62.6)	6,717	(63.6)	2,505	(54.7)	2,103	(68.1)	668	(71.6)
Hispanic/Latino/a	3,168	(16.5)	1,701	(16.1)	895	(19.6)	409	(13.3)	162	(17.4)
Black/African American	2,784	(14.5)	1,479	(14.0)	905	(19.8)	341	(11.0)	59	(6.3)
Biracial/Multiracial	398	(2.1)	176	(1.7)	118	(2.6)	90	(2.9)	13	(1.4)
Others	814	(4.2)	487	(4.6)	152	(3.3)	144	(4.7)	31	(3.3)
Lives in Poverty	5,498	(29.8)	3,238	(31.8)	1,201	(27.3)	917	(31.0)	142	(15.9)
Has Health Insurance	15,851	(82.9)	8,685	(82.5)	3,866	(84.5)	2,467	(80.0)	832	(89.2)
Visual Conformity
Conformers	9,778	(51.2)	4,515	(42.8)	3,025	(66.2)	1,644	(53.5)	594	(63.7)
Somewhat conforming	6,651	(34.8)	4,132	(39.2)	1,155	(25.3)	1,095	(35.6)	269	(28.9)
Non-conformers	2,679	(14.0)	1,891	(17.9)	387	(8.5)	333	(10.8)	69	(7.4)
Disclosure of Trans Identity
None/Some	6,251	(34.9)	2,609	(26.8)	1,023	(23.5)	1,842	(62.7)	776	(88.5)
Most	8,924	(49.8)	4,968	(51.0)	2,875	(65.9)	994	(33.8)	87	(9.9)
All	2,752	(15.3)	2,171	(22.3)	462	(10.6)	104	(3.5)	15	(1.7)
Health Status
Excellent	2,578	(13.5)	1,692	(16.0)	421	(9.2)	326	(10.6)	139	(14.9)
Very Good	6,082	(31.8)	3,209	(30.4)	1,682	(36.8)	865	(28.1)	326	(34.9)
Good	6,378	(33.3)	3,323	(31.5)	1,614	(35.3)	1,110	(36.0)	331	(35.5)
Fair/Poor	4,109	(21.4)	2,333	(22.1)	857	(18.7)	783	(25.4)	137	(14.7)
Any Disability	6,395	(34.2)	3,269	(31.7)	1,589	(35.5)	1,324	(44.3)	213	(23.3)

Notes. Data obtained from a weighted sample of the 2015 U.S. Transgender Survey. Missing data ranged from 0.1% (health status) to 6.4% (disclosure of trans identity) on individual items.

The results of the logistic regression model predicting the transgender respondents’ avoidance of healthcare due to anticipated discrimination are displayed in Table 2. The difference in odds among the gender identity/expression subgroups in their healthcare avoidance can be seen across the top of the table. Model 1 depicts the unadjusted odds ratios (OR) identifying significant differences in healthcare avoidance by gender identity/expression using transgender women as the reference. The odds ratios among the subgroups remained relatively similar in Model 2, after adjusting for covariates that controlled for healthcare need. Model 3 depicts the results of the fully adjusted model with covariates that controlled for sociodemographic and transgender-specific factors. Significant differences were found among the gender identity/expression subgroups in their healthcare avoidance, which answers the first research question. Compared to transgender women, transgender men demonstrated increased odds of avoiding healthcare due to possible mistreatment (adjusted odds ratio [AOR] = 1.32, 95% CI = 1.21–1.45), whereas non-binary/genderqueer individuals and cross-dressers had lower odds of avoiding healthcare due to possible mistreatment (AOR = 0.71, 95% CI = 0.63–0.80 and AOR = 0.66, 95% CI = 0.53–0.82, respectively). Race, poverty, health insurance, visual conformity, and disclosure were significant covariates, which addresses the second research question. Hispanic/Latino/a, biracial/multiracial, and other racial/ethnic groups had greater odds of healthcare avoidance compared to non-Hispanic White people. Living in poverty was robustly associated with healthcare avoidance (AOR = 1.52, 95% CI = 1.40–1.65) compared to those who were not impoverished. Individuals with health insurance had lesser odds of healthcare avoidance (AOR = 0.87, 95% CI = 0.79–0.96) compared to those who did not have health insurance. Visual non-conformity proved to be a significant risk factor, with non-conforming individuals having the greatest odds of avoiding healthcare (AOR = 1.48, 95% CI = 1.33–1.66) compared to those who were visually conforming. Lastly, those who disclosed their transgender identity to everyone in their network had decreased odds of avoiding healthcare due to possible mistreatment (AOR = 0.77, 95% CI = 0.68–0.87) compared to individuals who had disclosed to none to some people in their social network.

Table 2

Logistic regression models of healthcare avoidance during the past year due to anticipated discrimination.

Variables	Model 1		Model 2		Model 3
Categories	OR	95% CI	AOR	95% CI	AOR	95% CI
Gender Identity/Expression
Transgender women	1.00	…	1.00	…	1.00	…
Transgender men	1.25***	1.15, 1.35	1.26***	1.16, 1.37	1.32***	1.21, 1.45
Non-binary/Genderqueer	0.81***	0.73, 0.90	0.72***	0.65, 0.80	0.71***	0.63, 0.80
Cross-dressers	0.54***	0.45, 0.66	0.57***	0.47, 0.69	0.66***	0.53, 0.82
Health Status
Excellent			1.00	…	1.00	…
Very Good			1.30***	1.13, 1.49	1.57***	1.34, 1.84
Good			2.34***	2.04, 2.67	2.83***	2.43, 3.29
Fair/Poor			3.74***	3.25, 4.31	4.26***	3.63, 5.01
Disability			1.35***	1.25, 1.45	1.32***	1.21, 1.43
Race
Non-Hispanic White					1.00	…
Hispanic/Latino/a					1.49***	1.34, 1.65
Black/African American					1.11	1.00, 1.24
Biracial/Multiracial					1.32*	1.03, 1.69
Others					1.49***	1.24, 1.78
Lives in Poverty					1.52***	1.40, 1.65
Has Health Insurance					0.87**	0.79, 0.96
Visual Conformity
Conformers					1.00	…
Somewhat conforming					1.21***	1.11, 1.31
Non-conformers					1.48***	1.33, 1.66
Disclosure of Trans Identity
None to Some					1.00	…
Most					1.07	0.98, 1.17
All					0.77***	0.68, 0.87
Model Results
Nagelkerke R2	0.9%		7.9%		11.7%

Notes. AOR, adjusted odds ratio; CI, confidence interval; OR, odds ratio. An odds ratio of 1.00 is the baseline. Participants with valid data on all variables were included (91.4%). Asterisk(s) denote significant results (*p < .05; **p < .01; ***p < .001).

Lastly, a significant interaction of gender identity/expression by health insurance was observed. Among transgender men, the insured had decreased odds of healthcare avoidance compared to the uninsured (AOR = 0.70, 95% CI = 0.58–0.85), whilst among other gender identity/expression subgroups, there was a trend towards decreased healthcare avoidance (AOR = 0.93, 95% CI = 0.83–1.04), though the results were not statistically significant. Insurance status moderates the association between gender identity/expression and healthcare avoidance.

Discussion

Our results show that transgender stigma is a barrier to healthcare access for transgender people, with a disproportionate prevalence of healthcare avoidance among transgender men relative to the other gender identity/expression subgroups. After controlling for covariates, transgender men had increased odds of healthcare avoidance due to anticipated discrimination compared to transgender women. Furthermore, having health insurance is a moderating factor in transgender people's healthcare avoidance due to anticipated discrimination. Importantly, these results support the call to create trans-inclusive environments in healthcare systems and the need for targeted intervention efforts to improve healthcare access for transgender men. Barriers that produce missed opportunities for preventive, general, or trans-related care may magnify the physical and mental health disparities among transgender men relative to other subgroups.

Although non-binary/genderqueer individuals experience interpersonal and structural discrimination like other transgender subgroups, their experiences are unique from transgender individuals with a binary gender identity. This study found that non-binary/genderqueer individuals were less likely to avoid healthcare relative to transgender women. It is possible that non-binary individuals with an androgynous gender expression may be less susceptible to discrimination from health providers and thus, they may be less concerned about being mistreated. However, our findings also show that non-binary/genderqueer individuals were less likely to reveal their identity to most or all people in their network compared to transgender women and transgender men. The needs of non-binary/genderqueer people are often misunderstood by inexperienced healthcare providers, with the assumption that non-binary/genderqueer individuals do not want medical interventions for their transition (Kattari et al., 2019). Although they appear to be less disadvantaged relative to binary transgender people in healthcare avoidance due to possible mistreatment, it is important for providers to embrace gender diversity so that non-binary/genderqueer individuals may feel safe to disclose their identity while accessing needed healthcare. Our findings showing differential access to healthcare among gender identity subgroups are consistent with previous research (Cruz, 2014; Jaffee et al., 2016; Kattari et al., 2019).

Our study revealed that cross-dressers had decreased odds of healthcare avoidance due to anticipated discrimination relative to transgender women. They were also the least likely to have disclosed their identity to most or all people in their social network. The inclusion of cross-dressers in this study brings visibility to a subgroup that is under-represented in transgender health research. The recruitment of cross-dressers in transgender studies is challenging because these individuals may not necessarily experience a dissonance between their gender identity and biological sex, nor subscribe to the transgender identity label (Miner, Bockting, Romine, & Raman, 2012). Because cross-dressers are often excluded in studies, more research is needed to understand their unique experience of access to healthcare.

The poverty rate in this study sample is notably higher compared to the poverty rate among the general U.S. population in 2015 (29.8% vs. 13.5%, respectively) (Proctor, Semega, & Kollar, 2016). This is consistent with previous findings that transgender people experience higher levels of unemployment and poverty than cisgender people, which is indicative of structural inequities such as employment discrimination (Crissman et al., 2017). The finding that impoverished transgender individuals were more likely to avoid healthcare due to anticipated discrimination underscores how poverty can be an additional source of stigmatization in healthcare—that people with low socio-economic status often perceive receiving differential treatment and lower quality of care from health providers (Martinez-Hume et al., 2017). Poverty can produce multiplicative disadvantages by influencing other facets of the transgender experience and contributing to an increased risk of mistreatment from providers.

Poverty and visual non-conformity are interrelated (Begun & Kattari, 2016). Within a culture that adheres to a binary conceptualization of gender, individuals who do not fit prescribed gender stereotypes are more likely to experience transphobic discrimination (Miller & Grollman, 2015), which can result in greater poverty and increased vulnerability for homelessness (Begun & Kattari, 2016). Furthermore, though not all transgender individuals may wish to receive medical transition procedures, those who do are generally considered more privileged because they have the financial means and access to do so (Serano, 2007). Visually non-conforming individuals are more likely to be economically disadvantaged (Meier & Labuski, 2013) and are at heightened risk of extreme poverty in the form of housing instability (Begun & Kattari, 2016). This study found that transgender individuals who were somewhat visually conforming or visually non-conforming had increased odds of avoiding healthcare due to anticipated discrimination. Within cisnormative and cisgenderist healthcare contexts, individuals who are recognizable as transgender are more likely to encounter discrimination and less likely to receive acceptance from healthcare providers (Rodriguez et al., 2018). Thus, visually conforming transgender individuals experience greater privilege relative to their non-conforming counterparts. Health providers should be cognizant of the challenges experienced by visually non-conforming individuals and guard against magnifying the inequities that these individuals encounter.

For some transgender people, the decision to conceal their transgender identity has been key to their survival, as disclosure can increase the risk of experiencing transphobic violence and discrimination. Yet, identity concealment can also result in a constant worry about being discovered or accidently outed, creating a sense of hypervigilance and self-consciousness in public interactions (Ducheny et al., 2019). In the healthcare context, passing as one's affirmed gender identity and concealment of transgender identity can result in missed opportunities to receive preventive care, which can lead to delayed diagnosis and treatment of disease (e.g., sex-specific disorders) and increased morbidity and mortality (Lombardi & Banik, 2015; Unger, 2014). In this study, those who were out to everyone about their transgender identity were less likely to avoid seeking care. Disclosure can potentiate the ability to access social support and to experience affirmation of gender identity (Hoffkling, Obedin-Maliver, & Sevelius, 2017) as well as to receive appropriate healthcare (Bauer et al., 2009). However, healthcare providers and organizations must offer welcoming environments for transgender individuals to feel safe to disclose their transition status.

This study also reinforces findings from a growing body of scholarship highlighting the need for intersectional analysis of transgender experience. Consistent with a large body of research highlighting racial disparities among healthcare consumers in the general population (Hausmann, Jeong, Bost, & Ibrahim, 2008; Institute of Medicine, 2009), we found that racialized transgender people experienced a higher likelihood of avoiding healthcare utilization due to anticipated healthcare discrimination. Importantly, this study assessed anticipated discrimination specifically on the basis of transgender identity (not anticipated discrimination on the basis of racial identity); however, transgender people of color still reported higher rates of anticipated discrimination. As transgender people of color live at the intersection of racial and gender identities that are historically oppressed, their fears regarding discrimination on the basis of gender identity cannot be divorced from their fears of experiencing discrimination as racialized people (Kattari et al., 2015). Intersectional analysis directs our attention to the ways in which multiple forms of oppression are not additive but are intersectional (Bowleg, 2012). Our findings highlight that the concerns experienced by transgender people of color regarding potential discrimination are compounded by their experiences as healthcare consumers of color. Further research that continues to document and complicate our understanding of the diverse experiences of transgender people experiencing multiple forms of oppression is critical to the creation of affirming and welcoming services for transgender communities.

Implications for practice and policy

The findings of this study raise awareness about how transgender stigma, underpinned by cisnormativity and cisgenderism, can affect healthcare access for transgender people. We found differential experiences of healthcare access across gender identity/expression subgroups. Health systems should consider these differences in their efforts to improve accessibility of services to transgender communities. Providers must also be cognizant of the multiple marginalized identities that transgender individuals may hold and how these individuals may be exposed to multiple forms of discrimination (such as racism, ableism, etc.) in addition to cisgenderism. Outreach efforts and intervention should be sensitive to the intersection of these oppressions (Kattari et al., 2020).

In their positions of power, healthcare providers hold responsibility for advancing equity in healthcare access (Miller & Grollman, 2015). Given the pervasive nature of cisnormativity and cisgenderism, combating stigma and creating a safe, trans-affirmative environment requires a multi-level and multi-faceted approach (Hatzenbuehler & Link, 2014; White Hughto et al., 2015). To lessen healthcare avoidance, providers need to develop self-awareness by examining their own values and biases that may perpetuate cisgenderist practices. They should avoid assumptions about a person's gender identity and use affirming and inclusive language that is respectful of their patients' identities (e.g., using names and pronouns consistent with a patient's identity). Providers are encouraged to build their knowledge base about transgender people and their needs, including an understanding of the socio-political and historical context of transgender populations and how stigma may impact their health outcomes, morbidity, and mortality (ANA Ethics Advisory Board, 2018; National Association of Social Workers, 2016). Ongoing education is necessary because the terminology about transgender identities, distinctions between various transgender communities, and the cultural norms among transgender populations often evolve (Collazo, Austin, & Craig, 2013). For example, not all non-binary/genderqueer individuals and cross-dressers subscribe to the transgender identity label (Miner et al., 2012). Awareness of community resources (e.g., support groups, service organizations, referral networks, etc.) and other trans-friendly providers is also important, as these resources can provide social support, promote resilience, and enhance transgender individuals' self-affirmation (National LGBT Health Education Center, 2016).

Healthcare organizations should conduct a self-assessment of their policies and practices, and engage their staff in designing and implementing change to build a trans-affirmative environment. Examples of such change efforts include: the development of patient and employee non-discrimination policies (to ban discrimination against sexual orientation, gender identity, and gender expression); creating a process for reporting and redressing discrimination if it occurs; creating trans-affirmative policies (such as patient room assignments within sex segregated systems); incorporating inclusive language on intake forms and assessment tools; promoting a welcoming environment within the physical space of the organization (such as the availability of universal bathrooms or placing trans-friendly visual cues in the waiting area); adding gender affirmative imagery and content on patient education and marketing materials; and providing education to staff to promote medical and cultural competence in serving transgender patients. These efforts help to create a gender-affirming organizational culture (Moone, Croghan, & Olson, 2016; National LGBT Health Education Center, 2016).

As policymakers continue the heated discourse on healthcare reform in the U.S., consideration must be given to underserved and vulnerable populations. Policymakers must consider the cascading effects of employment discrimination and its potential to magnify health and economic disparities among marginalized populations. Though advocacy efforts supporting the rights of sexual and gender minorities have increased public awareness of their marginalization, transgender people experience higher rates of discrimination relative to sexual minority populations (Kcomt, 2019; Lambda Legal, 2010; Macapagal et al., 2016). Based on the deleterious health consequences associated with discrimination (Albuquerque et al., 2016; McCann & Brown, 2017; Winter et al., 2016), transgender people need explicit legal protection against discrimination in employment and other facets of life. Presently, at the writing of this article, federal laws explicitly banning discrimination based on sexual orientation and gender identity/expression do not yet exist and only some states offer full non-discrimination protections for sexual and gender minority populations. Thus, eliminating structural barriers to healthcare access requires lawmakers to consider the impact that the lack of explicit legal protections may ensue.

Strengths and limitations

The NCTE is a respected organization that advocates on behalf of, and is led by, transgender people. Their visibility and active engagement with transgender communities enabled them to access these populations for sample recruitment. The primary researchers used a multi-pronged approach in their sample recruitment efforts, thereby producing an extraordinarily robust national sample. This large dataset enabled the construction of models using multiple variables to examine factors experienced by transgender people in their access to healthcare.

Although survey weights were applied to this study sample, the non-probability sampling methods used in the primary study limit the generalizability of our findings. The sole use of an online platform for data collection may have contributed to online survey bias. Respondents to online surveys are generally more likely to be white, young, and with higher socio-economic status (Hash & Spencer, 2007; Miner et al., 2012; Rachlin, 2007). Disadvantaged transgender individuals may not have responded to the survey because of their lack of access to the internet or a web-enabled device, lack of familiarity with the use of the internet, or inability to travel to a community organization to complete the survey. Little is known about the individuals who chose not to complete the survey. Therefore, these findings may underestimate the true prevalence of vulnerabilities experienced by transgender people. Lastly, the study used a cross-sectional design and thus, causality cannot be inferred.

Future research needs

Further research on transgender men and non-binary identities are needed to gain a deeper understanding of their experiences. In particular, qualitative or mixed methods approaches may facilitate theory building to understand the patient experience of transgender men, and to elucidate the challenges encountered by non-binary identities within a predominantly dimorphic gender healthcare system. The present study showed how the respondents' visual nonconformity and disclosure of their transgender identity impacted their healthcare access. Future research can explore the complexity of these factors in transgender patients' interactions with their healthcare providers and include prospective data to examine the temporal relationship between visual conformity and healthcare avoidance in a more nuanced manner. Lastly, more information is needed regarding the impact of healthcare avoidance on transgender people's morbidity and mortality rates.

Conclusion

In conclusion, we found that transgender men, those living in poverty, and visually non-conforming individuals had increased odds of avoiding healthcare due to fear of mistreatment. Healthcare is a basic human need and thus, access to healthcare is a basic human right. Compromised access to healthcare can lead to untreated disease and increased morbidity and mortality, magnifying the health disparities experienced by vulnerable populations. Providers should be mindful of the increased risks that may be experienced by certain transgender subgroups, and to consider them when designing services or conducting outreach. This study underscores the importance of cultural competence among healthcare providers in caring for transgender patients. It is a call to action to create trans-affirming environments, so that transgender people may feel safe and supported when seeking care.

Funding information

The development of this article was supported in part by the Ontario Graduate Scholarship.

Ethics approval

This study was approved by the Research Ethics Board at the University of Windsor in Windsor, Ontario, Canada (REB #18–193).

CRediT authorship contribution statement

Luisa Kcomt: Conceptualization, Formal analysis, Funding acquisition, Writing - original draft. Kevin M. Gorey: Methodology, Writing - review & editing, Supervision. Betty Jo Barrett: Conceptualization, Writing - review & editing. Sean Esteban McCabe: Methodology, Writing - review & editing.

Declaration of competing interest

None.