First impressions online: The inclusion of transgender and gender nonconforming identities and services in mental healthcare providers' online materials in the USA

ABSTRACT

Background: When accessing mental healthcare services, transgender and gender nonconforming (TGNC) individuals face systemic barriers to gender-affirmative care. Initial points of contact, like intake forms, may show limited consideration for the heterogeneity of TGNC identities and can lead to negative consequences prior to face-to-face interaction with providers. Aims: The first aim was to mimic a likely pathway a TGNC individual may follow to seek mental healthcare services in the USA and to describe the extent to which they may encounter enacted stigma or affirmative messages that may impede or facilitate access to care. The second aim was to determine if a positive State legal climate for TGNC people was associated with more affirmative provider materials. Methods: Content analysis was used to examine a national sample of websites and intake forms of mental healthcare providers who advertise online as working with TGNC clients. Intake forms were coded for usage of affirmative language in gender/sex questions and including questions for a client's pronouns and preferred name. Websites were coded for mentioning a variety of services or resources for TGNC clients. Results: While provider websites were found through Google searches for a “gender therapist,” only 56.6% of websites stated a provider specialty to work with TGNC clients and 32.1% of websites had no mention of services or resources for TGNC people. Additionally, a significantly larger proportion of intake forms from States with legal protections for TGNC people used affirmative language in gender/sex questions and asked for a client's pronouns than intake forms from States without legal protections. Discussion: Barriers to affirmative healthcare for TGNC people within patient and provider interactions have been identified in previous research and these data show TGNC individuals may face enacted stigma even in their search for a provider, particularly those TGNC people living in States without legal protections.

As many as three out of four individuals who identify as transgender or gender-nonconforming (TGNC) seek mental health services (James et al., 2016) compared to 14.8% of adults in the general population (Park-Lee, Lipari, Hedden, Copello, & Kroutil, 2016). This high utilization is likely driven by evaluation requirements for receiving transition-related medical care (Coleman et al., 2012) and high rates of discrimination, stigma, and violence due to their gender identity (Bockting, Miner, Romine, Hamilton, & Coleman, 2013; Seelman, 2014). Exposure to these minority-related external stressors and internal stressors, including internalized transphobia and expectations of discrimination and violence, escalates TGNC individuals’ risk for mental health problems (Brennan et al., 2017; Hendriks & Testa, 2012; Meyer, 2003) including depression, anxiety, and suicide risk (Bockting et al., 2013; Bouman et al., 2017; Clements-Nolle, Marx, Guzman, & Katz, 2001; Dawson, Wymbs, Gidycz, Pride, & Figueroa, 2017; Millet, Longworth, & Arcelus, 2017).

Herek (2009) offered a framework for sexual stigma and discrimination that includes enacted, felt, and internalized stigma that can be extended to gender minorities (Herek, 2016). Enacted stigma includes violence, discrimination, or victimization targeted at minority individuals. Microaggressions, the erasure of minority identities, and hate crimes are all types of enacted stigma. The expectation of experiencing enacted stigma is felt stigma. Felt stigma can influence behavior such as concealing one's identity or avoiding a situation where the possibility of enacted stigma is high. Internalized stigma occurs when a minority individual believes the stigma associated with their identity and adopts it into their own belief system. Internalized stigma can be seen in TGNC people who experience internalized transphobia or feel shame about their gender minority identity.

TGNC individuals' healthcare experiences

Despite calls for gender identity affirmative care, (e.g., American Psychological Association, 2015), enacted and felt stigma are well-documented in health care systems. Enacted stigma includes violence in health care settings, refusal of care, erasure of TGNC identities in medical records, misgendering, and providers who lack knowledge and expertise for TGNC-affirmative care (Bauer et al., 2009; James et al., 2016; Xavier et al., 2013). Anticipation of poor or inappropriate care (felt stigma), lack of insurance, and financial instability all create barriers to health care for many TGNC people. In fact, 33% of TGNC participants indicated they had delayed or avoided seeking care due to finances and 23% reported avoided care due to fear of mistreatment by a provider in a large national survey (James et al., 2016).

Barriers to care may arise even before face-to-face interactions with health care providers or their office staff. One key initial point of contact in health care systems is the patient intake paperwork which may show little consideration for the heterogeneity of TGNC identities. Forms that only allow for a selection of “male” or “female” when asking about a client's gender or only ask for a client's sex, create a microaggression by reinforcing the sex/gender binary (Smith, Shin, & Officer, 2012). These limited options can lead to felt stigma prior to face-to-face interaction with providers, which may keep TGNC patients from disclosing their gender identity or fearing refusal of care (Goins & Pye, 2013).

Providers who wish to make their initial points of contract trans-affirmative have several options. First, providers who recognize systemic discrimination related to having a TGNC identity could invite patients to make improvements and suggestions to materials, such as intake forms (Bauer et al., 2009; Goins & Pye, 2013). However, this places the burden on the community being served and some surveys of the TGNC communities have suggested that providers conduct their own research on specific TGNC issues in order to be culturally responsive and avoid making TGNC clients responsible for their provider's education (Bauer et al., 2009; Elder, 2015). Second, some attempts at practical guidelines to limit stigma, such as demonstrations of an affirmative client intake form and intake questions are available (Coren, Coren, Pagliaro, & Weiss, 2011; Donatone & Rachlin, 2013; Makadon, Mayer, Potter, & Goldhammer, 2015; Makadon, Mayer, Potter, Goldhammer, 2007). However, some of these products show limited consideration for the full spectrum of trans identities, use outdated, stigmatizing language like ‘transgendered,’ or focus on a small population within the TGNC communities (Bouman et al., 2017; McIntosh, 2016). The most updated sample intake form from The Fenway Institute, a leading information source on LGBT health, shows increased attention to the variety of TGNC identities (Makadon, Mayer, Potter, & Goldhammer, 2015) but the level of detail may not be appropriate for services in a general practice where TGNC patients are more infrequent than in a specialty LGBT clinic. Despite the importance of intake paperwork as an initial indicator of whether services will be TGNC-affirmative or not, there are no data on how or whether this potential barrier to care is being addressed in general clinical practice.

Online health information seeking

A Pew survey of over 3,000 adults in 2010 revealed that 80% of adults who use the Internet had searched online for health information and 44% of that group specifically looked for information about doctors or other health professionals (Fox, 2011), suggesting the initial point of contact for health care may be online for many people. Understanding how people are finding and using that information can help reduce barriers to care. When using a search engine to find health information, most people stay on the first or second page of results (Morahan-Martin, 2004). Once on a webpage, people use several criteria to evaluate the credibility of the health information including the authority of the source, how understandable and professional the writing is, how clean and professional the layout is, and the inclusion of external links (Morahan-Martin, 2004). When appropriately designed, provider websites save both providers and patients time (Alpert, 2015). Providers and their staff are able to avoid unnecessary phone calls if their website answers frequently asked questions, and patients are given more information and tools to identify a provider who meets their needs. Given that patients frequently visit a healthcare provider's website before their first visit, it is recommended that the website design should facilitate a trusting relationship between the provider and patient (Erdem & Harrison-Walker, 2006). Erdem and Harrison-Walker recommend topic-specific pages, online forms, and information about how to obtain a referral.

There are populations for whom online health information seeking is particularly important. For example, individuals who have experienced low provider-patient communication and less patient-centered healthcare in the past are more likely to engage in online health behaviors, including searching for a healthcare provider (Hou & Shim, 2010). Additionally, LGB youth use the Internet as a primary source of health information. Like TGNC people, LGB youth face barriers in health care systems such as lack of trust between patient and provider (Rose & Friedman, 2013). Online health information seeking is useful when there is a lack of culturally-appropriate content in traditional health resources, such as sexual health resources. One study that included transgender individuals found LGBT young people use the Internet to search for sexual health information that is inclusive of sexual and gender minority identities. Unfortunately, barriers to care for LGBT youth also occur when accessing online health information, as they may perceive the content as stigmatizing or untrustworthy (Magee, Bigelow, DeHaan, & Mustanski, 2012).

Little is known specifically about the online-health seeking behaviors of TGNC people. However, the Internet has been a vital tool for advocacy, organizing, and sharing information amongst TGNC communities, particularly in resource-limited areas (Shapiro, 2004). As the Internet has become a place of empowerment for TGNC people, individuals may look for content on provider websites that affirm TGNC identities. In a qualitative study of the experiences of TGNC individuals who had sought psychotherapy, participants expressed that they would only select a therapist who is visibly supportive of TGNC people and active in the community (Benson, 2013). Thus, mental healthcare providers’ online materials are an opportunity to signal support of TGNC people and earn potential clients’ trust.

Content analyses of healthcare providers’ products

A few studies have explored what providers’ materials communicate to LGBT clients, but none are specific to TGNC clients. A thematic content analysis of sexual health intake forms revealed limited inclusion of LGBT identities in questions about sexual orientation, relationship status, and gender identity (Goins & Pye, 2013). Wright and McKinley (2011), conducted content analyses of college counseling center websites for inclusion of mental health resources specifically for LGBT students. Categories of mental health resources included individual and group counseling for LGBT students, a campus peer group, and counselor specialty in LGBT issues. Inclusion of LGBT mental health resources on the websites ranged from 5.4% to 31.5%. A follow-up study (McKinley, Luo, Wright, & Kraus, 2015) showed minimal increases, and some decreases, in the proportion of websites that mentioned specific LGBT mental health resources. It should be noted that these studies consider the LGBT community as a whole despite specific differences and needs of TGNC individuals in comparison to sexual minority individuals (Su et al., 2016). Nevertheless, these content analyses revealed that even providers, such as a university services, who wish to serve sexual and gender minorities, may not reach these communities. The online content often appears to omit the signals TGNC individuals may seek to identify an affirmative provider and/or may create barriers to care by excluding TGNC identities in the materials.

Legal climate and LGBT mental health

TGNC individuals’ experiences with health care occur in a cultural context that is formed, in part, by the legal climate. Just as stigma is enacted and felt in health care systems, stigma permeates the larger social and legal climate where those health care interactions take place. A handful of studies have demonstrated that a favorable legal climate, such as employment protections and hate crime statues that include sexual orientation, is associated with improved mental health in LGBT populations (Goldberg & Smith, 2011; Hatzenbuehler et al., 2012; Hatzenbuehler, Keyes, & Hasin, 2009; Kail, Acosta, & Wright, 2015). Other research has indicated that municipal policies banning employment discrimination based on sexual orientation are associated with more fair treatment in a potential hiring situation (Barron & Hebl, 2013) suggesting that such policies impact the behavior of non-LGBT people in a way that ultimately benefits the LGBT communities.

Only one known study has explored the association between legal climate and mental health in a TGNC population. Blosnich et al. (2016) found TGNC veterans who lived in States in the USA with employment protection had a 26% decreased odds rate of having a mood disorder and 43% decreased odds rate of self-directed violence compared to TGNC veterans who lived in States without this protection. No significant differences were found in mental health variables based on hate crimes legislation. Using the Human Rights Campaign's Municipal Equality Index (MEI), which reports municipal-wide policies, to look for differences in mental health, Blosnich et al. (2016) found TGNC veterans who lived in locales with higher MEI scores, or more LGBT protections, had increased odds of having alcohol abuse disorders and substance use disorders. This finding was unexpected, but the effect was relatively small as a 1 point increase on MEI (maximum score of 100) led to a 1% increase odds of having a substance use disorder. These findings suggest legal climate is an impactful dimension of TGNC mental health though more research is needed to further understand the underlying mechanisms.

Current study

Given the high rates of utilization of mental health care, calls to provide TGNC-affirmative care (Hope, Mocarski, Bautista, & Holt, 2016), and the known barriers to health care, the current study used content analysis to examine intake forms and websites of mental health care providers identified online by searching for a “gender therapist.” The first aim was to mimic a likely pathway a TGNC individual may follow to seek mental health services and to describe the extent to which they may encounter enacted stigma or affirmative messages that may impede or facilitate access to care. The second aim of the study was to test whether more LGB and TGNC affirmative laws and public policies would be associated with more affirmative providers, as Barron and Hebl (2013) found for employment non-discrimination laws. Hypothesis 1 was that websites from States with TGNC-affirmative policies would be more likely to include specific services and resources for TGNC clients than websites from States with fewer affirmative polices. Hypothesis 2 predicted that intake forms from States with TGNC-affirmative policies would be more inclusive of TGNC identities by using affirmative language in gender/sex questions and asking for clients' pronouns and chosen name.

Method

Context of the study: Trans Collaborations

The current study was part of a larger project, Trans Collaborations, which has established an ongoing research relationship with TGNC communities in the Central Great Plains. Trans Collaborations uses a community based participatory research model, which considers academic researchers and community members as equal partners and formulates research questions and topics from community-identified needs (Travers et al., 2013). As part of this community partnership, Trans Collaborations is advised by both a Local Community Board of TGNC leaders in the Central Great Plains region and a National Advisory Board of interdisciplinary TGNC researchers.

Content analysis

Sample selection

Systematic Google Chrome Incognito searches were used to identify websites of mental healthcare providers who categorize themselves as working with TGNC clients. A search engine was chosen because this approach has been used in past research to collect online materials for a content analysis and it mimics the research method TGNC clients would use to identify a provider (Deutsch, 2016; Goins & Pye, 2013). The search phrase ‘gender therapist <state>’ was conducted for 25 U. S. States (every other State in an alphabetical list) to obtain a geographically representative sample. The first 10 hits from each search were explored for potential inclusion. Often search result hits were links to a consolidated list of referrals for providers in a certain geographic location, such as a Psychology Today list of providers in one city who checked that they work with transgender clients. In this case, the providers on the first page of the consolidated list were considered for inclusion if there was a direct link to the provider's website from their Psychology Today (or similar website) profile. This means one search result hit could yield multiple providers. There is no verification process on sites such as Psychology Today to confirm a provider who marks working with TGNC clients is competent to provide these services or ascertain if the provider is a gender specialist such as someone who has been trained by a recognized entity, such as the World Professional Associations for Transgender Health's (WPATH) Global Education Initiative. However, clients searching for a provider may trust these sites as valid and reliable.

To be included for data collection, a provider's website must have displayed an intake form or client information form available for download. Google searches were completed from May 2016 – September 2016. Duplicate results, such as a provider appearing on two consolidated lists, were removed. Of an estimated 1500 providers who appeared in the searches, about 500 included a link to their personal website. From that group, 249 providers included an intake form or client information form on their website and thus were included in the sample.

Of the 249 providers in the sample, 245 appeared on a consolidated list while 4 appeared directly on the Google search result page. Providers were coded into the sample as they appeared. Some providers who first appeared on a consolidated list were also directly on the Google search result page further down the list, but had already been selected for inclusion from the consolidated list. Two-hundred and five providers appeared on consolidated lists from therapist search engines, such as Psychology Today, 26 providers were on consolidated lists from TGNC websites, such as local TGNC organizations list of providers, and 14 providers were on lists from LGBT group websites.

Provider demographics

Intake forms and websites were coded for the provider's demographic information including State, individual or group practice, provider field of study, and education, if this information was discernable from the provider's materials. The number of providers from a given State ranged from one to twenty-two. Table 1 details how many providers from the sample of 249 are from each included State. Most providers (158 out of 249 or 63.5%) worked in an individual practice. Level of education could be determined for 238 (95.6%) providers: 70.2% were Master's level clinicians and 29.8% held a doctorate degree. The field of study was clearly identifiable for 195 (78.3%) providers: 28.2 % had a degree in Counseling Psychology; 26.7% in Social Work; 17.4% in Clinical Psychology; 16.9% in Marriage and Family Therapy; 9.2% in Counseling/Clinical Mental Health Counseling; 1% in Psychiatry; and .5% in Substance Abuse Counseling.

Table 1.

Number of providers from each state.

State	Number of providers
Alabama	12
Arizona	8
California	22
Connecticut	5
Florida	14
Hawaii	8
Illinois	9
Iowa	5
Kentucky	11
Maine	6
Massachusetts	2
Minnesota	10
Missouri	13
Nebraska	15
New Hampshire	5
New Mexico	6
North Carolina	20
Ohio	10
Oregon	17
Rhode Island	1
South Dakota	10
Texas	16
Vermont	4
Washington	16
Wyoming	4
Total	249

Coding

The trans-affirmative coding categories were developed from recommendations for intake forms and trans-affirmative practices identified in previous literature and with advisement by the Trans Collaborations Local Community Board. The coding categories for websites included and expanded the coding system used by Wright and McKinley (2011) and McKinley, Luo, Wright, and Kraus (2015), to be most appropriate for the current sample of websites beyond university counseling centers. The coding categories for intake forms and websites are detailed in Table 2 along with their theoretical rationales.

Table 2.

Coding categories for content analysis.

Material	Coding Category	Options	Theoretical Rationale
Intake form	Gender and Sex	Gender Asked? Y/N	Carabez, Pellegrini, Mankovitz, Eliason, and Scott (2015) and Goins and Pye (2013), recommend asking about sex and gender separately; Bauer et al. (2009) and Carabez et al. (2015), suggest avoiding binary gender options; Goins and Pye (2013) recommend open-ended forms
		Sex Asked? Y/N
		If yes to either, complete following code for affirmative language
Intake form	Affirmative Language in Gender/Sex Questions	Used? Y/N	Carabez et al. (2015) and Goins and Pye (2013), recommend asking about sex and gender separately; Bauer et al. (2009) and Carabez et al. (2015), avoid binary gender options; Goins and Pye (2013) recommend open-ended forms
		Language considered affirmative if options beyond “male” and “female” were included and/or only asked for client's sex if additionally asked for client's gender
Intake form	Pronouns	Asked? Y/N	Carabez et al. (2015) and Goins and Pye (2013), suggest asking for a client's preferred pronouns
Intake form	Chosen/Preferred name	Asked? Y/N	Beagan et al. (2013) recommend a space for the client to list their preferred name in case it differs from their legal name
Website	Group counseling for TGNC individuals	Y/N	Included in Wright and McKinley (2011) and McKinley et al. (2015), coding system
Website	Additional information/pamphlet on TGNC issues	Y/N	Included in Wright and McKinley (2011) and McKinley et al. (2015), coding system
Website	Website link to TGNC or LGBT resource	Y/N	Included in Wright and McKinley (2011) and McKinley et al. (2015), coding system with added code of trans specific resource given the documented differences between LGB individuals and trans individuals
		If yes, is resource TGNC specific or broadly LGBT
Website	Provider specialty in TGNC issues/individual counseling with TGNC clients	Y/N	Combined two categories included in Wright and McKinley (2011) and McKinley et al. (2015), coding system
Website	Affiliation with professional organization related to TGNC issues/expertise	Y/N	Adaptation of Wright and McKinley (2011) and McKinley et al. (2015), Website link to LGBT resource and Provider Specialty categories
		If yes, is organization TGNC specific or broadly LGBT

Note: Binary coding categories were used for intake forms and website analyses.

The first author and a second coder completed coding for intake forms and websites. A subset of the sample was double coded to calculate interrater reliability. Interrater reliability for intake form coding categories was excellent (Cohen's К ranged from .94-1.00). Interrater reliability was good for website coding categories of Website Link to TGNC/LGBT Resource (Cohen's К = .79), Provider Specialty (Cohen's К = .85), and Professional Organization (Cohen's К = .78). For website coding categories Group Counseling (Cohen's К = .57) and Additional Information (Cohen's К = .54), interrater reliability was low. The two coders came to consensus on the final coding decisions for these two categories when there was disagreement.

State legal climate

State legal climate for TGNC people was assessed in the following categories: Employment Laws and Policies, Gender Marker Change on Identification Documents, Transgender Healthcare, Housing Laws and Policies, Public Accommodations Laws and Policies, Hate Crimes, School Anti-Bullying Laws & Policies, and School Non-Discrimination Laws & Policies. The specific details for State policies and legislation were collected from the Human Rights Campaign's (HRC, 2016) State Maps (http://www.hrc.org/state_maps) in September 2016. These maps classify the type of protection enacted in each State and demonstrate which type of legislation is considered the highest level down to the lowest level by a color gradient. Due to small sample sizes, the legal climate coding options within each policy category were collapsed into bivariate options of “No protections for TGNC people” or “Protections for TGNC people.” Table 3 details the type of protection captured for each legislation coding category. Classifying State legal climate according to HRC State Maps has been used in other studies that demonstrated the impact of legal climate on LGBT mental health (Blosnich et al., 2016; Goldberg & Smith, 2011).

Table 3.

Coding categories for State legal climate.

	Specific Laws and Policies included in Legislation Coding
Legislation Category	No Protections for TGNC People	Protections for TGNC People
Gender Marker on ID	- No relevant legislation	- Facilitate change in gender marker on driver's licenses
		- Facilitates change in gender marker on birth certificates
		- Facilitates change in gender marker on both driver's licenses and birth certificates
Transgender Healthcare	- No relevant legislation	- Bans on insurance exclusions for transgender healthcare
		- Trans-inclusive health benefits for state employees
		- Both bans on insurance exclusions and trans-inclusive health benefits for state employees
Employment	- No relevant legislation	- Prohibit discrimination based on sexual orientation and gender identity only for public employees- Prohibit discrimination based on sexual orientation and
	- Prohibit discrimination based on sexual orientation only for public employees
	- Prohibit discrimination based on sexual orientation only	gender identity
Public Accommodations	- No relevant legislation	- Prohibit discrimination based on sexual orientation and gender identity
	- Prohibit discrimination based on sexual orientation only
Housing	- No relevant legislation- Prohibit discrimination based on sexual orientation only	- Prohibit discrimination based on sexual orientation and gender identity
Hate Crimes	- No relevant legislation- Sexual orientation included in hate crime statue	- Sexual orientation and gender identity included in hate crime statue
School Anti-Bullying	- Prevent school districts from specifically protecting LGBT students	- Address harassment and/or bullying of students based on sexual orientation and gender identity
	- Restrict inclusion of LGBT topics in schools
	- No relevant legislation
School Non-Discrimination	- No relevant legislation- Prohibit discrimination for sexual orientation only	- Prohibit discrimination for sexual orientation and gender identity

Note: Binary coding categories for State legal climate.

Data analysis

Frequencies were used to examine the prevalence of gender-affirmative language in intake forms and mention of resources on websites. Chi-square analyses were conducted to identify any pattern of mean differences in the intake and website coding among the provider demographics and different levels of legal protections for TGNC people.

Results

Frequency

Websites

Nearly three-fourths (74.3%) of the websites referred to at least one resource or service for TGNC clients and 25.7% of the websites mentioned more than one resource or service. A remaining 32.1% of websites did not include any mention of services or resources for TGNC clients.

The most common reference to services for TGNC clients was a provider stating a specialty with TGNC clients or that they provide individual therapy for TGNC clients, with 56.6% of websites including this mention. The percentage of websites that mentioned other resources or services were considerably lower. Only 6% of websites mentioned that the provider conducts group therapy with TGNC clients. Similarly, just 10.4% of websites provided additional information on TGNC related issues, such as definitions of different gender identities or a description of gender dysphoria. A minority of websites, 27.7% of the sample, linked to a LGBT or TGNC resource, such as local support groups. Of the 68 websites that linked to an outside resource, 69.1% of the links were broadly LGBT in focus while 30.9% were TGNC specific. Finally, 12.4% of websites mentioned a provider's involvement with a professional organization dedicated to LGBT or TGNC issues, such as the World Professional Association for Transgender Health (WPATH). Among the websites that mentioned a professional organization, 86.2% were TGNC specific.

Intake forms

In the sample of 249 intake forms, 185 or 74.3% included questions about gender and/or sex. Of these 185 intake forms, 56.8% were rated as using TGNC-affirmative language, while 43.2% were not. Only 6.0% of intake forms included space for potential clients to mark their pronouns and 18.1% of intake forms included an option to designate their chosen name if it differed from their legal name.

Provider demographics and gender-affirmative language and resources

Websites

Chi-square analyses were used to look for patterns in the mention of services and resources for TGNC clients on websites based on the size of the provider's practice, education level, and field. There were no significant relationships between these variables (all p > .05).

Intake forms

Chi-square analyses were used to look for differences in the use of gender-affirmative language and questions about pronouns and chosen names based on the size of the provider's practice, education level, and field. There was a significant relationship between gender-affirmative language and size of practice, such that intake forms from providers in individual practice (N = 81, 66.9% affirmative; N = 40, 33.1% non-affirmative) were more likely to use affirmative language in gender/sex questions than intake forms from providers in group practice (N = 24, 37.5% affirmative; N = 40, 62.5% non-affirmative), χ²(1) = 14.784, p < .001. There were no other significant relationships between provider demographics and intake form codes (all p > .05).

State legal climate and gender-affirmative language and resources

Websites

Chi-square analyses were performed to examine differences in mention of TGNC resources and services on websites between States with legal protections for TGNC people and States without legal protections. The only significant relationship existed between employment policies and a provider mentioning membership in a professional organization, χ²(1) = 3.97, p < .05. However, this relationship is in a direction contrary to Hypothesis 1, such that providers in States with no employment protections for TGNC people noted membership in a professional organization more often (N = 21, 16.5% belong; N = 106, 83.5% do not belong) compared with providers in States that have employment protections (N = 10, 8.2% belong; N = 112, 91.8% do not belong). There were no significant relationships between the remaining variables (all p > .05).^1.

Intake forms

Chi-square analyses were performed to examine differences in affirmative practices on intake forms between States with legal protections for TGNC people and States without legal protections. There was a significant relationship between use of affirmative language in gender/sex questions and the following legal climate variables: Gender Marker on ID, Transgender Healthcare, Housing, Public Accommodations, Hate Crimes, School Anti-Bullying, and School Non-Discrimination (all p < .05). These relationships were in the direction consistent with Hypothesis 2. Intake forms from States with greater legal protections for TGNC people were more likely to use affirmative language than intake forms from States with fewer or no legal protections for TGNC people. These results are detailed in Table 4.

Table 4.

Significant chi-square analyses for use of affirmative language in gender/sex questions by State legal climate.

	Affirmative Language?
	Yes	No
State Legislation or Policy	N (%)	N (%)	p
Gender Marker on ID			0.017
No protections	38 (46.9)	43 (53.1)
Protections	67 (64.4)	37 (35.6)
Transgender Healthcare			0.002
No protections	61 (48.8)	64 (51.2)
Protections	44 (73.3)	16 (26.7)
Employment			0.007
No protections	46 (47.4)	51 (52.6)
Protections	59 (67)	29 (33)
Public Accommodations			0.014
No protections	52 (49.1)	54 (50.9)
Protections	53 (67.1)	26 (32.9)
Housing			0.007
No protections	50 (48.1)	54 (51.9)
Protections	55 (67.9)	26 (32.1)
Hate Crimes			0.003
No protections	48 (47.1)	54 (52.9)
Protections	57 (68.7)	26 (31.3)
School Anti-Bullying			0.009
No protections	48 (48.0)	52 (52.0)
Protections	57 (67.1)	28 (32.9)
School Non-discrimination			0.003
No protections	58 (48.7)	61 (51.3)
Protections	47 (71.2)	19 (28.8)

There were also significant relationships in the hypothesized direction between intake forms asking for a client's pronouns and the legal climate variables of Transgender Healthcare, Housing, Hate Crimes, School Anti-Bullying, and School Non-Discrimination (all p < .05). A larger proportion of intake forms from States with these protections provided space for clients to write in their pronouns, than compared to intake forms from States without these protections. There was no significant relationship between asking for pronouns and State legislations for Gender Marker on ID, Employment, or Public Accommodations (all p > .05).

Additionally, there was a significant relationship between State legislation for Hate Crimes and intake forms having a chosen or preferred name question, χ²(1) = 5.0, p < .05. Specifically, 24.1% of intake forms from States that include gender identity in their hate crime statutes (N = 112) asked for a chosen name while only 13.1% of intake forms from States without hate crime protections (N = 137) included a similar question. The remaining analyses for intake legal climate variables and including a chosen/preferred name question on an intake form were not significant (all p > .05). Table 5 documents the results of Chi-square analyses for State legal climate variables and intake form variables of Pronouns and Chosen/Preferred Name.

Table 5.

Chi-square analyses for intake coding categories by State legal climate.

	Pronouns?		Chosen/Preferred Name?
	Yes	No	Yes	No
State Legislation or Policy	N	N	N	N
Gender Marker on ID
No protections	4 (3.8)	102 (96.2)	21 (19.8)	85 (80.2)
Protections	11 (7.7)	132 (92.3)	24 (16.8)	119 (83.2)
Transgender Healthcare
No protections	4 (2.4)*	160 (97.6)*	25 (15.2)	139 (84.8)
Protections	11 (12.9)*	74 (87.1)*	20 (23.5)	65 (76.5)
Employment
No protections	4 (3.1)	123 (96.9)	22 (17.3)	105 (82.7)
Protections	11 (9)	111 (91)	23 (18.9)	99 (81.1)
Public Accommodations
No protections	5 (3.6)	135 (96.4)	24 (17.1)	116 (82.9)
Protections	10 (9.2)	99 (90.8)	21 (19.3)	88 (80.7)
Housing
No protections	4 (2.9)*	134 (97.1)*	23 (16.7)	115 (83.3)
Protections	11 (9.9)*	100 (90.1)*	22 (19.8)	89 (80.2)
Hate Crimes
No protections	4 (2.9)*	133 (97.1)*	18 (13.1)*	119 (86.9)*
Protections	11 (9.8)*	101 (90.2)*	27 (24.1)*	85 (75.9)*
School Anti-Bullying
No protections	3 (2.4)*	124 (97.6)*	20 (15.7)	107 (84.3)
Protections	12 (9.8)*	110 (90.2)*	25 (20.5)	97 (79.5)
School Non-discrimination
No protections	4 (2.6)*	149 (97.4)*	25(16.3)	128 (83.7)
Protections	11 (11.5)*	85 (88.5)*	20 (20.8)	76 (79.2)

*p < .05.

Discussion

The current study used content analysis to examine gender-affirmative practices in websites and intake forms of mental health care providers who advertise online as working with TGNC clients. The findings extend previous research that looked at the use of affirmative language in sexual health intake forms (Goins & Pye, 2013) and content analyses of college counseling center websites (McKinley, Luo, Wright, & Kraus, 2015; Wright & McKinley, 2011). This study focused on mental health care providers and accessed materials in a way that mimics how potential TGNC clients may find providers in their area. Additionally, this study extended previous research on the association between legal climate and professional business practices to mental health providers offering their services to the TGNC communities.

Despite using search parameters to seek TGNC services, only three out of four websites included reference to a service or resource for TGNC clients. Most of the references to TGNC clients in the current study's sample consisted of providers’ indication of expertise with TGNC (or LGBT, if the ‘T’ for ‘transgender’ was included) clients or that they offer individual therapy for TGNC clients. Mentions of affirmative content such as group therapy, additional information, or links to LGBT/TGNC websites occurred infrequently, similar to the findings of college counseling websites. It appears that even providers who advertise as working with TGNC clients are unlikely to offer additional information that might signal TGNC-affirmative care (e.g., Benson, 2013).

On intake forms, just over half of providers who included questions about gender and/or sex used gender-affirmative language that allows clients to note a response outside of a traditional male-female binary. This was more common for solo practitioners than those in group practices. Nearly half of the intake forms that had gender and/or sex questions used language that limited response options and did not consider the range of gender identities outside “male” or “female.” This means a substantial number of mental health care providers online who explicitly advertise working with TGNC people are using intake forms that enact stigma by erasing TGNC identities. Additionally, the percentage of intake forms that asked for a client's pronouns or a name other than a client's legal name was very low. The most common approach on websites and intake forms was that most providers mentioned they work with TGNC clients or perhaps asked for a client's gender in a non-stigmatizing manner, but then did little to further demonstrate that they offer gender-affirmative services to potential TGNC clients seeking health information online.

Surprisingly, there was little support for Hypothesis 1 that providers in States with progressive laws and policies such as workplace non-discrimination laws would have more TGNC affirmative material on their websites. The only difference in the mention of resources of services for TGNC clients on websites between States with and without legal protections was a larger proportion of websites from States without legal protections for TGNC people in employment referenced a professional organization than websites from States with employment protections. Though this relationship occurred in the opposite direction of Hypothesis 1, the percentage of websites that referenced a professional organization was low regardless of legal climate (16.5% of websites from States without protections versus 8.2% of websites from States with protections). This trend of a small percentage of websites including resources or services for TGNC people occurred for all website categories, other than provider specialty. The lack of discrepancies in resources based on legal climate speaks to a broader issue. Most websites did not include any additional affirmative markers for potential TGNC clients other than the 56.6% of websites that noted a provider specialty. A lack of specific and culturally-responsive information on provider websites is contrary to best practices for establishing trust with patients in online spaces (Erdem & Harrison-Walker, 2006). Providers who wish to reach TGNC clients may want to consider the high likelihood that TGNC people will seek health information online and examine if their websites are inclusive of TGNC identities and explicitly describe what gender-affirmative services and resources they can provide.

There was substantial support for Hypothesis 2 that a more progressive State legal climate would be associated with more affirmative practices on intake forms. Based on all legal climate variables, the proportion of intake forms from States with legal protections used gender-affirmative language was significantly larger than the proportion of intake forms from States without protections. Additionally, a significantly larger proportion of intake forms from States with protections for TGNC healthcare, housing, hate crimes, school anti-bullying and school non-discrimination asked for pronouns than intake forms from States without protections. There was also a relationship between school non-discrimination policies for TGNC students and asking for chosen or preferred name on intake forms. Given the documented impact of legal climate on LGBT mental health, it is important to consider how and if legal climate interacts with providers demonstrating gender-affirming practices. TGNC people who live in States without legal protections experience increased rates of mental health problems (Blosnich et al., 2016). When these people seek mental health services, they also are a greater risk of encountering intake forms that do not include their identities. Intake forms that limit gender and/or sex options to “male” and “female” may create a negative first point of contact with the provider and felt stigma. These forms reinforce binary notions of gender and constitute a unique form of microaggresion that can be stigmatizing to TGNC clients (Smith, Shin, & Officer, 2012).

This examination of websites and intake materials offered an important, although modest, assessment of TGNC affirmative-practices in mental health across a variety of disciplines. Providers who appeared in the search lists indicated a willingness to serve TGNC clients in some fashion, often by checking a box to that effect in a provider listing service. The extent to which a given provider had training, supervision, and experience for working with TGNC clients is unknown, as it is unknown for a perspective client. Appearing in a Google search for a “gender therapist” does not guarantee the provider will be a gender specialist or even be competent to work with TGNC clients. The materials reviewed in this study provide clues that many of them may not be well-prepared, given that one in four made no further mention of such services on their website and nearly half failed to ask basic questions about gender in a non-stigmatizing manner.

TGNC individuals seeking an expert gender specialist may have difficulties locating an appropriate provider simply through an online search. Clients seeking a gender specialist should look for additional markers of a providers’ expertise. This may include extensive continuing education in TGNC health, association with interdisciplinary gender clinics, or appearing on a list such as WPATH's “Find a Provider”. Professional certification programs, such as the Global Education Initiative offered by WPATH may help protect the public and improve care. Ongoing efforts to increase awareness of these resources in TGNC communities will offer alternative routes to identify a gender specialist as opposed to Google searches or databases like Psychology Today which have little oversight. While WPATH and similar organizations are makings strides to increase access to high quality mental health care, including in underserved areas, barriers to care still exist for some TGNC people (Rosenkrantz, Black, Abreu, Aleshire, & Fallin-Bennett, 2017; Seelman, Young, Tesene, Alvarez-Hernandez, & Kattari, 2017). In these cases, the best option for these prospective clients is to tap into local expertise and get recommendations from their local TGNC communities for therapists. In fact, research shows that such networks are an important source of information (Benson, 2013).

One important avenue for helping TGNC individuals access culturally-responsive mental health services is to provide patient education and advocacy. Ideally TGNC people should not be responsible for removing barriers to care. However, the urgent need for appropriate services, particularly in underserved locations like the Central Great Plains which includes States in the central portion of the USA such as Nebraska and Kansas, means TGNC clients are taking control of their own health care. The Trans Collaborations Local Community Board has identified patient education and advocacy as a critical need. To this end, we are implementing patient self-advocacy workshops to empower TGNC patients and improve patient-provider communication. It is imperative that patient education programs empower TGNC clients, rather than increase burden and stigma they already face in health care systems.

Below are recommendations for best practices for websites and intake materials for providers who wish to signal they provide culturally competent care for the TGNC communities. These practices were informed by these data, and the literature review, as well as input provided by Trans Collaborations Local Community Board through creation of the coding manual for this study. Although similar recommendations are likely included in basic TGNC-affirmative training for therapists, many providers in underserved areas have little access to such training.

Websites

• ○Mention a provider has expertise in working with TGNC clients, if such expertise exists
• ○Provide links to TGNC resources or support groups
• ○Detail all the services offered for TGNC clients
• ○Identify any membership in relevant professional organizations (i.e., World Professional Association for Transgender Health or Division 44 of the American Psychological Association)

Intake forms

• •
  Ask for “Gender”, not “Sex”

  • ˆ
    If a client must identify their sex for medical records, ask for both “Legal Sex” and “Gender” or “Gender Identity”
  • •
    Include a blank option for clients to write in a response
• •Ask for “Chosen Name,” “Name to call in waiting room,” “I wish to be called…,” or another variation that separates the client's name and their legal name that may be on official documents
• •Ask for which pronouns the provider and office staff should use when referring to the client.

There are a number of limitations to the present study, including the correlational nature of the data that limits conclusions about causality. The findings with intake forms indicate that a positive State legal climate for TGNC people is associated with the use of TGNC-affirmative practices in demographic questions, but the nature or direction of this relationship cannot be determined from these data. Future research should examine whether passing laws protecting LGBT communities changes professional practices, such as the ones assessed in this study or if third variables such as community norms lead to both the passage of laws and affirmative practices. There was limited variance in the website coding categories that affected statistical power of the analyses comparing websites from different legal climates. This may explain why the hypothesis about intake forms and the legal climate were more likely to be supported.

Interrater reliability was good for all coding categories except Group Counseling (Cohen's kappa = .57) and Additional Information (Cohen's kappa = .54). Discrepancies in these categories were discussed and the coders agreed on the final decisions. Discrepancies in these categories may be due to the two coders accessing websites at different times and the possibility of websites being updated with different resources, such as one provider updating their website to be responsive to the night club shooting in Orlando, Florida in the summer of 2016.

The online sampling method does not permit understanding of how or why providers make choices about their intake forms and websites. It is possible electronic health records limit the demographic options on intake forms in some cases and these are not easily changed. Given the significant difference in the use of affirmative language on intake forms between providers in individual practice and providers in group practice, it may be that some providers wish to be more inclusive but do not have decision making power. Providers who operate in a group setting and strive for cultural competency, should consider that advocacy and action are an important component (Sue, 2001) and may want to take further steps to implement TGNC-affirmative changes in their practices.

Finally, not all providers who appeared in the online searches were included in the study because they did not have an accessible website or their intake materials were not online. It is unknown whether these data generalize to those providers. On the other hand, the information used in this study was the information a potential client would be able to access as well, increasing the external validity of the study.

Future research should focus on disseminating best practices, such as validating the proposed recommendations in a sample intake form that is concise and useful for clinicians in different settings. Qualitative studies with providers should seek understanding of how they enact affirmative practices, make decisions and design their materials, and navigate influences like electronic health records. Despite these limitations, the current study found some inclusion of TGNC identities in a national sample of mental health care providers’ intake forms and websites, though demonstrated significant areas for improvement, particularly amongst providers from States with no legal protections for TGNC people. First points of contact with mental health care providers, like websites and intake forms, prime TGNC clients to how affirmative a provider is likely to be. Being culturally-responsive must extend beyond the therapeutic interaction.

Funding Statement

National Institute of Mental Health (MH108897-01A1). National Institute of General Medical Sciences (1U54GM115458-01).

Declaration of conflict of interest

This work was supported in part by grant MH108897-01A1 from the National Institutes of Mental Health and grant 1U54GM115458-01 from the National Institute of General Medical Sciences. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The authors declare that they have no conflict of interest.

Acknowledgments

The authors would like to express their gratitude to Gabrielle Albeck for her assistance with data collection and coding.