What Lesbian, Gay, Bisexual, Transgender, Queer and Intersex Patients Say Doctors Should Know and Do: A Qualitative Study

Abstract

This qualitative study explored the experiences of lesbian, gay, bisexual, transgender, queer and intersex (LGBTQI) people in health care and their recommendations regarding what physicians should know and do to be able to take care of them. Six focus groups were conducted with LGBTQI people (N=48) in four U.S. cities between October 2013 and April 2014. Five overarching themes emerged regarding patients’ suggestions for providers: be comfortable with LGBTQI patients; share medical decision-making; avoid assumptions; apply LGBTQI-related knowledge; and address the social context of health disparities. These core competencies differed in meaningful ways from competencies created by national organizations such as the Association of American Medical Colleges. Community-derived competencies¹ stressed the importance of collaborative patient-physician partnerships, particularly in the setting of hormone prescription for transgender patients; prioritized addressing social determinants of health and focusing on marginalized subpopulations² and stigmatized needs of the community. Limitations, particular of sampling, were considered. Community input could improve medical education interventions to reduce health disparities in marginalized communities.

The medical community has long acknowledged that people who are not heterosexual or do not identify with the gender they were assigned at birth have significant disparities in health outcomes compared with cisgender³ and heterosexual people (American Medical Association, 1996; Institute of Medicine [IOM], 2011; Coker, Austin, & Schuster, 2010; Lambda Legal, 2010; Makadon, Mayer, Potter, & Goldhammer, 2015). Lesbian, gay, bisexual, and transgender (LGBT)⁴ people experience disproportionate levels of mental illness; tobacco, alcohol and other drug use; suicidality; discrimination; and violence (IOM, 2011). Another group at risk is intersex⁵ people whose chromosomes or anatomy do not match medical definitions of male or female.

Disparities in health care received by LGBTQI people were also recognized by the American Medical Association as early as 1996. Women who have sex with women are less likely than other women to have insurance, a primary care provider, or recent cancer screening (Kerker, Mostashari, & Thorpe, 2006). In a recent survey of New York State residents, 32.9% of transgender people lacked insurance, compared to 14.5% of non-transgender people (Frazer, 2009). Denials of care are as common as 19% for transgender people (Grant et al., 2011). Disparities are amplified in communities of color⁶. For example, 9% of white LGBT people reported refusals of care while 22% of Latino, 18% of black, 32% of indigenous, and 14% of mixed race LGBT people reported refusals, suggesting further marginalization among these groups (Frazer, 2009_). These disparities may contribute to mistrust of health care systems. In a survey of LGBT cancer survivors, many expressed fear of substandard care if they came out to their providers (Margolies & Scout, 2013).

Improving LGBTQI health will require physicians competent in providing care for these communities. This is important as 40% of LGBT New York State residents believed there were not enough health professionals who were adequately trained and competent to deliver healthcare to them (Frazer, 2009). Medical education spans many years of training, beginning with medical school, which provides foundational courses and more general knowledge, and continuing with residency and continuing medical education, which hones more specialized clinical skills. Select medical schools have developed LGBTQI health curricula, which have demonstrated positive results, such as improvements in knowledge and attitudes based on assessments created by researchers (Chen, Rodriguez, & Dreger, 2012; Kelley, Chou, Dibble, & Robertson, 2008; Safer & Pearce, 2013; Sequeira, Chakraborti, & Panunti, 2012). However, as recently as 2011, medical schools only taught a median of 5 hours of LGBTQI–related medical curriculum (Obedin-Maliver et al., 2011). Given that context, it is not surprising that students generally felt comfortable, but not fully prepared to care for LGBTQI patients, particularly transgender patients (White et al., 2015).

Medical education theory has shifted focus from recommending instructional processes to determining desired outcomes or competencies for physician trainees (Frank et al., 2012). However, most academic medical centers do not measure the LGBTQI-related competence of their physicians (Khalili, Leung, & Diamant, 2015). Recently, the Association of American Medical Colleges (AAMC) released recommendations for specific training needed to become proficient in providing care to LGBTQI patients (2014). Yet, we were unable to find validated tools for assessing LGBTQI competence within the field of medicine. The mental health fields have created several validated instruments that assess practitioners’ LGB competence and self-efficacy. Research regarding these scales has identified key factors in provider competence. However the research does not draw from patient input nor does it use scales that assess competence in working with transgender clients. As Bidell and Whitman write, “Minority sexual orientation and gender identity present distinctive differences, [therefore] counselors need to develop distinctive competencies regarding transgender clients (2013, p. 120).” We were interested in further exploring competence for working with various members of the LGBTQI community with the input of LGBTQI people.

Understanding the perspectives of potential health care consumers as they relate to physician competency is important. Improving the patient experience is part of the Institute for Healthcare Improvement’s Triple Aim (Berwick, Nolan, & Whittington, 2008) and the Agency for Healthcare Research and Quality’s National Quality Strategy (2016). Specific community health needs should be considered when developing medical education curricula (Habbick & Leeder, 1996; Strasser, et al., 2015). Community-identified competencies could improve health care experiences by ensuring that the priorities of community members are addressed. In addition, this input could improve relationships between patients and medical institutions. For example, the Northern Ontario School of Medicine gathered the perspectives and suggestions of local aboriginal communities to inform curriculum interventions (Strasser & Marsh, 2014). Other medical schools have used patient narratives to inform specific curricular content (Rogers, Morris, Hook, & Havyer, 2016; Foster, Robb, Cordar, Chaudhary, Noseworthy, & Lok, 2015; DeFries, Rodrigues, Ghorob, & Handley, 2015), but we could not find other examples of medical schools formally soliciting the input of community members regarding curricular objectives.

The experiences of LGBTQI people in health care and other settings have been explored through qualitative research. One such study used community-identified competencies to make practice recommendations specific to LGBTQI people with cancer. However, this community input has not typically led to recommendations for physician competency (Margolies & Scout, 2006; Sevelius, 2012; Woody, 2016; Poteat, German, & Kerrigan, 2013). Psychologists have created cultural competency recommendations based on the formal solicitation of LGBTQI patients (Borroughs, Bedoya, O’Cleirigh, & Safren, 2015), which could be an example for the field of medicine.

Community perspectives can aid in determining expertise needed to care for people who disproportionately experience discrimination and violence. LGBTQI people have individual and collective histories of trauma in medical settings, including: (a) therapies aimed at changing the sexual orientation or gender identity of LGBTQI patients; (b) psychiatric diagnoses, such as homosexuality, ego-dystonic homosexuality, gender identity disorder, and gender dysphoria; (c) non-consensual medical interventions such as surgeries or hormone treatment for intersex people; and, (d) biased or prejudicial treatment by healthcare providers (IOM, 2011; Roen, 2004). Given this history, community-identified competencies may be important to guide medical providers’ behavior and speech, which could build trust and avoid trauma. Therefore, our objective was to determine the competencies LGBTQI community members perceived physicians would need to deliver effective and accessible care.

Methods

Research Design

Underpinning our research are interpretivist paradigms, which suggest that reality is constructed between people and findings emerge through dialogue and negotiations of meanings among community members and researchers (Cohen & Crabtree, 2008). We held focus groups to create opportunities for LGBTQI participants to theorize together. We used a thematic analysis to develop a list of ideas for medical competencies necessary to provide care for LGBTQI patients. The University of Vermont Institutional Review Board (IRB) determined that this study was exempt.

Researcher-as-instrument

The first author (AA) facilitated the six focus groups. She is a LGBTQI-identified white woman who conducted the study as part of her medical school scholarly project. For three years prior to medical school, she had provided HIV prevention interventions for LGBT young people in New York City. A number of the participants were people she had known prior to the study and all participants knew she was a medical student at the time of the study. She was trained in basic qualitative methods at Oberlin College with Anna Agathangelou whose expertise is in international feminist political economy and postcolonial and decolonial thought. The second author (EC) is a white, straight, cisgender woman who was the Director of Educational Instruction and Scholarship and Associate Professor of Family Medicine at the University of Vermont at the time of the study. She received formal training in qualitative methods of analysis in her graduate and doctoral programs in education at the University of Albany where she studied with Dr. Lynn Gelzheiser and Dr. Joel Meyers who served as the president of the American Psychological Association Division of School Psychology. She is the parent of a child who is gender non-conforming. The third author (AF) is a white, straight, cisgender man who was an Assistant Professor of Medicine and Clinician Investigator in the Division of General Internal Medicine at Montefiore Medical Center during the study. He is a practicing general internist at a community health center in a low-income urban setting.

As an LGBTQI person, the first author has an investment in the creation of interventions to decrease health disparities for LGBTQI people as well as her own preconceptions about the nature of obstacles to receiving LGBTQI-competent care. In order to clarify her beliefs in relation to the subject matter, she and EC independently reviewed the data to clarify their findings. AF reviewed the manuscript and provided feedback on discrepancies between interpretations of findings and supporting participant quotations. In addition, the participants were given the option of reaching out to the researchers to review preliminary findings. Approximately eight of the participants as well as note-takers and the translator reviewed the manuscript and provided feedback, which was used for revisions.

Participants

Inclusion criteria were LGBTQI or other related identities. Attempts were made to exclude people under the age of 18 given the complications in obtaining IRB approval for minors. However, after reviewing the data, the researchers discovered that one participant in an early group was 17. This was reported to the IRB. Thereafter, ages were verbally verified to avoid further instances. Additionally, people who did not identify as either lesbian, gay, bisexual, transgender, queer, or intersex (or in a related identity group not previously named) were excluded.

We used a convenience sample and recruited study participants in the following ways: facilitating a focus group at a statewide transgender conference; sending emails to personal contacts; posting Facebook events; announcing focus groups on LGBTQI community center listservs; and placing flyers in LGBTQI community centers, LGBTQI health centers, coffee shops, and at local universities. A participant in an early focus group also assisted with community entry, offering to organize focus groups in his home city, one with individually-recruited LGBTQI health policy experts and one at a community center run by and for Latina transgender women. A colleague of his served as a liaison with the director of the community center who helped to establish rapport with participants. The participants for that focus group were recruited on site the day of the group.

Settings

The focus groups took place in Burlington, Vermont; New York City; Oakland, California; and Washington, D.C. between October 2013 and April 2014. At each focus group, AA was present along with a note-taker and, in one instance, one additional facilitator. In each group, all non-participants identified as LGBTQI. Focus groups were held at the following locations: a transgender conference at a state university; a LGBTQI community center in a small city; a transgender community center run by and for Latina transgender women; a private home in a large city; the office of an LGBTQI service organization; and an LGBTQI community center in a large city. The cities sampled varied significantly. For example, Burlington is a small city with a mostly white population and without a LGBTQI-specific health center, while New York City has one of the largest LGBTQI populations in the country.

Demographics Questionnaire

We developed a self-administered questionnaire, which contains five questions regarding age, race / ethnicity, sexual orientation, gender identity, and intersex identity. For the first four questions, participants were asked to place checks in the boxes next to identities that applied to them and could check as many as applied within each category and/or fill in identities under the category other. Options for age included ranges starting at the age of eighteen, as we had planned to exclude participants under 18. See above. Checkboxes for race / ethnicity included Latino/a, African American/Carribean/Black, Pacific Islander / Native Hawaiian, Asian American, Indigenous/Native American/American Indian, Middle Eastern/Arab American, European/white, multiracial, and other. Checkboxes for sexual orientation included lesbian, gay, bisexual, queer, pansexual, and other. Checkboxes for gender identity included transgender, transwoman, transman, genderqueer, cisgender woman, cisgender man, and other. The form also included the question, “Do you identify as intersex?” See Table 1 for a glossary of terms. Of note, we used more nuanced identity terms in the questionnaire than in the acronym LGBTQI, which is used throughout the paper, as we wanted to understand more specifically how participants identified.

Table 1.

Glossary of Terms

Term	Definition
Bisexual	A person who is attracted to and has romantic relationships with people who identify as men and women.
Cisgender	A person who identifies with the gender they were assigned at birth.
Gay man	A person who identifies as a man and is attracted to and/or has romantic relationships with people who identify as men.
Genderqueer	A person whose gender identity is not simply male or female.
Gender nonconforming	A person whose gender identity does not conform to gender stereotypes.
Intersex	A general term used for a variety of cases in which a person is born with a reproductive or sexual anatomy that does not fit the usual medical definitions of female or male.a
Lesbian	A person who identifies as a woman and is attracted to and/or has romantic relationships with people who identify as women.
Out	A person who is open about their sexual orientation, gender identity, or other aspects of their identity.
Pansexual	A person who is attracted to and/or has romantic relationships with people of all genders.
Queer	An umbrella term for people whose attractions and/or romantic relationships are not heterosexual or whose gender identities are not cisgender.
They	A gender-neutral pronoun used in the place of he or she.
Transman	A person who is transgender and identifies as a man.
Transwoman	A person who is transgender and identifies as a woman.
Transgender	An umbrella term for people who identify with a gender other than that which they were assigned at birth.
Transition	The process of undergoing physical changes, including those accompanying hormone blockade, hormone therapy, silicone injection, and surgery; changes in dress; or other changes in order to make one’s physical presentation more in line with one’s gender identity.

Note. The above definitions come from the authors with the exception of the definition for intersex.

^aAdapted from “What is intersex?” by the Intersex Society of North America, no date. Retrieved from: http://www.isna.org/faq/what_is_intersex

Each set of responses were given identification numbers in order to anonymously link demographics to quotes. Because the facilitators had not known in advance that one group would be conducted primarily in Spanish, the demographics form was not translated but was instead interpreted aloud. We considered excluding this data; however we believed that excluding these demographics would provide a less accurate representation of the participants than including them. Therefore, we included them.

Data Collection

Focus groups were facilitated using a focus group script, which was not pilot tested but was reviewed between co-authors for face validity and is included as an appendix. The questions were structured based on a recommendation from the IRB at the University of Vermont to ask participants to speak in the third-person to further ensure anonymity. Additionally, focus group participants were instructed to use the third person. Focus groups were audio-recorded and note-takers anonymously connected verbiage with participant demographics as above. Given limited funding and time, data saturation was not reached as we were unable to repeat a group conducted in Spanish or further explore ideas elicited from specific subpopulations.

Data analysis

All focus group recordings were transcribed by AA who used note-takers’ documentation to contextualize quotes with demographics. Focus group conversations that were conducted in Spanish were translated into English. After the first three focus groups, transcripts were independently reviewed by two authors (AA and EC) who identified key concepts, initially broken out into attitudes, skills, and knowledge. AA and EC met to compare these initial codes and, through further discussion, grouped them into a set of mutual categories. As further focus group transcripts were added to the data set, EC and AA reviewed them independently, expanding the categories and creating new ones. EC and AA then met again to compare and build a final set of categories. After further conversation with AF, these were revised again into a final set of competency themes, which were directive and each of which included knowledge, attitudes, and skills. Afterward, AA ensured that all quotes were accounted for in resultant themes. The data presented here includes the final set of thematic competencies along with illustrative quotes.

As described above, in order to deepen the analysis, the manuscript was distributed to participants, note-takers, and the translator who provided feedback, which was incorporated into revisions. The manuscript and transcripts were also sent to members of the AAMC Advisory Committee on Sexual Orientation, Gender Identity, and Sex Development who also provided feedback and suggestions, which were incorporated into the manuscript. The feedback of a qualitative research expert, along with peer reviewer feedback, was also used to further revise the manuscript.

Results

Participants

48 participants took part in the focus groups with distribution as follows in the six groups: 12, 5, 7, 6, 4 and 15. Of the people who provided data in each category, the majority of participants were between the ages of 26 and 35, although one was below the age of 18 and two were older than 55. Approximately half of the participants were people of color and half were white. Approximately one-third identified as transgender, one-third as genderqueer, and one-third as cisgender. Four participants identified as intersex. See Table 2 for the complete demographic breakdown of the sample.

Table 2.

Characteristics of Focus Groups Participants

Characteristic	Participantsa
Age
0–17	1
18–25	11
26–35	20
36–45	7
46–55	3
56–65	1
>65	1
Missing	4
Gender
Transgender	15
Transwoman	5
Transman	5
Cisgender	12
Genderqueer	14
Other	4
Missing	7
Intersex identity	4
Sexual Orientation
Queer	20
Pansexual	0
Lesbian	7
Gay	7
Bisexual	2
Heterosexual	5
Other	1
Missing	8
Race/Ethnicity
Latino/a	15
African American/Caribbean/Black	2
Asian American/Asian	2
Indigenous / Native American	2
European/White	24
Missing	5

Note. Participants were given the option to check more than one box for gender, sexual orientation and race/ethnicity. 3 participants did not fill out demographic surveys.

^aN=48

Community-Identified Provider Competencies

The following data is organized by themes we labeled community-identified provider competencies. Direct quotes are provided from participants with minimal editing to highlight key findings. At least one-fourth of participants made comments that related to each of the following themes. A unique identifier based on the focus group attended is given for each participant along with demographic information to provide greater context. See Table 3 for specific participant recommendations by theme.

Table 3.

Specific Recommendations for Community-identified LGBTQI Competencies

Competency	Specific Recommendations
Be comfortable with LGBTQI patients	Avoid behaviors and language, which could be perceived as disrespectful or stigmatizing.
	Avoid body language or tone that conveys disgust
	Use the preferred name and pronoun for patients.
	Use words for anatomy that patients prefer. Avoid using words for anatomy that may be perceived as assigning the incorrect gender to patients.
	Do not avoid touching patients.
	Avoid denials of care.
	Do not demonstrate a comfort that is disingenuous
Share medical decision-making with patients	Use shared decision-making for all medical decisions
	Avoid so-called gatekeeping or presenting obstacles to hormone therapy and surgeries for transgender patients.
Avoid assumptions about sexual orientation, gender identity, behavior, or anatomy	Avoid assumptions of cisgender and heterosexuality
	Avoid assumptions connecting gay men and transgender people with HIV
	Avoid assumptions about anatomy
	Avoid assumptions connecting identity and behavior
	Avoid assumptions about lack of sexually transmitted infection and other disease risk factors for women who have sex with women.
Apply knowledge of LGBTQI identities, anatomy, and common sexual practices	Apply knowledge of transgender health care, including need for routine health maintenance
	Apply knowledge of common sexual practices of LGBTQI people.
Acknowledge and address the social context of health disparities	Advocate for patients
	Create resources to fill gaps in care
	Improve language access by learning languages
	Provide health care to undocumented people
	Understand the histories of stigma that transgender and other LGBTQI patients experience and do not stigmatize patients.
	Avoid assumptions that health concerns of LGBTQI people are the result of psychiatric pathology
	Understand and address the sequelae of lack of health care access.
	Provide health care to address stigmatized experiences including sex work, use of hormones obtained without a prescription, intravenous drug use, immigration, incarceration, poverty, and silicone injection.

Be comfortable with LGBTQI patients

The first key theme related to providers feeling comfortable with patients, their bodies, and their sexual lives. Approximately a quarter of participants perceived they were being treated differently because of their LGBTQI identities or reported that their providers shifted to a negative attitude when they disclosed their sexual orientation or gender identity. A white cisgender participant who identifies as queer and a lesbian (RU1) reported, “It’s so uncomfortable to be talking to a doctor and then they treat you differently once they find out.”

Participants reported that discomfort could also manifest as outright denial of care. Four participants reported instances of this. A lesbian, indigenous participant (RU5) with multiple sclerosis described an instance in which a provider’s knowledge of her sexual orientation led to unprofessional conduct. “He never sent me for any scans. He never did any tests. He just came in, looked at that front page of my paperwork [where I’d written ‘lesbian’ and] called me a ‘fucking dyke’ to my face.”

Other instances of discomfort with LGBTQI patients were also perceived to negatively affect care. Six participants reported instances of providers using the wrong pronoun or name or otherwise not respecting the gender identity of patients. Four participants also reported that providers avoided touching transgender patients. One heterosexual, Latina transgender woman (CR6) explained, “[Providers] keep that distance like you’re not supposed to touch the body. You’re taking care of my body, but you really don’t want to look at my body.” RU1 expressed her feelings about provider discomfort this way. “I shouldn’t ever have to feel like my partners or sex life makes you feel icky.” These negative experiences were perceived to exacerbate mistrust of the medical system. One cisgender African American gay man (HP 5) explained, “Especially for queer people of color there’s already this ingrained mistrust of the medical system. If they have one bad experience … I see it with my trans friends like, ‘I’m just not going back.’”

Examples of positive experiences were also reported. One transgender, indigenous participant (OA4) said,

As soon as [a transman I know] talked about his gender experience with his gynecologist, they were very careful to not use gendered language during the exam. It was all very matter of fact and they actively took steps to minimize any chest exposure, any referring to the chest tissue as “breasts” and things of that nature.

Participants also clarified that what they wanted was for providers to be rather than simply seem comfortable. OA4 said, “It is more useful to teach the skills in how to build that comfort then it is to teach someone to demonstrate a comfort that they may not feel.” A biracial lesbian participant (RU2) put it this way, “Back in the day when people were less self-conscious about their opinions, I had a gynecologist say he would not accept me as a patient. Now he probably would, but would I want him?”

Share medical decision-making with patients

Another common theme was that medical decisions should include patients’ perspectives and preferences. Approximately one-fifth of participants reported instances where their preferences were excluded in medical decision-making. They described situations in which, instead of a patient advocate, providers acted as a gatekeeper, implicitly or explicitly asking transgender or genderqueer people to follow the recommendations of the provider before medical care such as surgery or hormones would be provided. A white, straight transgender man (TI6) gave this example,

I know someone [who feels] they have to meet certain standards or jump through certain hoops to get access to the healthcare that they need. For example, this person likes to have long hair and typically the gender stereotype is that guys have short hair and this person felt that their counselor was pressuring them to into getting a haircut because that would mean that they were more serious about their transition.

This approach to medical decision making could also exacerbate mistrust. One Latino, Caribbean, gay transgender man (TI2) explained, “I have friends who are counselors and they try to steer away from gatekeeping because they feel it breaks any trust that has built in the relationship.”

Participants who perceived that they lacked control over their medical care also feared that it would be taken away. One participant (demographics missing) described a transgender friend’s fear that asking his provider questions about hormone injection techniques might lead them to think, “Oh, you don’t know how to inject. We should just take this away.” Given this context, participants highlighted the importance of shared decision-making during which the provider discusses risks and benefits, and the patient has an opportunity to decide how to proceed. A straight, intersex, white woman (TI1) said,

A lot of times doctors want to reserve the role of decision-maker to themselves. Rather than giving the patient all the information, their idea is to pick the patient’s brain, but then make a decision themselves … but I think doctor’s should think in terms of genuine informed consent.

Avoid assumptions about sexual orientation, gender identity, behavior, or anatomy

Another common theme was that medical providers often make assumptions that negatively impact care. Approximately one-quarter of participants discussed providers’ assumptions of heterosexual or cisgender identities. For example, one Latina transgender woman (CR5) explained her difficulty with intake forms.

Starting with how to identify – you don’t have options [during registration]. It is easy for me to sign in as a woman but then [the provider] ends up asking me inappropriate questions, for example when was my last period or if I might be pregnant.

Participants also perceived that providers made inaccurate assumptions based on their LGBTQI identities. For example, three participants (a gay cisgender man, a transgender woman, and a genderqueer person) explained that they were assumed to have sexual transmitted infections (STIs) including the human immunodeficiency virus (HIV). In contrast, a white cisgender queer lesbian (RU1) explained, “When [the provider] finally figured out that I was sleeping with women he was like, ‘Well you don’t really need to worry about [STIs],’and recommended that I didn’t get tested for anything.” Similarly, a white genderqueer, queer participant (HP2) warned against making assumptions about sexual behavior and risk factors for women who have sex with women.

Doctors will often not believe the person’s past … [and] in a community that has such high rates of drug use and sex work, the doctors are not providing the right care because in their mind they’re like, “I’m providing the right care for a woman who has sex with a woman.”

Apply knowledge of LGBTQI identities, anatomy, and common sexual practices

A smaller number of participants (five) discussed instances in which providers’ lack of knowledge of LGBTQI sexual practices created distance between patients and providers or meant missed opportunities for providing care. One white, queer transgender man (TI4) said,

My friend had carpel tunnel, [and] the advice given was, “Oh, just stop doing what you’re doing.” [For a] hetero-normative person who is not necessarily using their hands as much for sex [that] may be okay, but if you’re someone who’s relying on that as primary sex organ, no.

A white queer cisgender participant (OA10) said, “It could be helpful to have more specifics about what is normal, [sex] with dildos, fisting … so doctors are not freaking out and shaming patients.” Another white cisgender queer and lesbian participant (OA1) discussed instances in which providers did not perform routine screening, e.g. prostate exams for transgender women or pap smears for transgender men. Approximately one-fourth of participants discussed the need for greater knowledge regarding care for transgender patients.

Acknowledge and address the social context of health disparities

A final theme was that LGBTQI care should address forms of social marginalization that may affect health status. An Asian American, heterosexual transgender woman (CR7) said, “We have to think not just [about] providing good quality, friendly health services but about human rights. Being transgender myself, my identity is associated with stigma, pathologization. The doctor should know these things and not reinforce [them].”

Approximately one-third of participants highlighted that social marginalization was common in LGBTQI communities and could not be ignored. A Latina, genderqueer, gay participant (CR12) discussed the needs of immigrant patients, “Being monolingual in this globalized world doesn’t cut it and so it’s not just for the immigrant to learn English. It’s also for the American to learn other languages.” That participant went on to address health issues for undocumented people saying, “I am undocumented and I’m a human … Physicians need to understand that when they make a pledge, they make a pledge to humanity …”

Addressing stigmatized experiences, including silicone injection, use of hormones accessed without a prescription, intravenous drug use, sex work, immigration, and incarceration was also perceived by participants to be an important component of LGBTQI care. Approximately one-fifth of participants discussed these themes. CR7 quoted above also said,

A lot of us come from the other country and start transitions very young [with silicone injections] before we moved to the U.S. When we move here, we go to the doctor and ask about silicone injection and they’re not willing to help us. It’s difficult to meet the doctor that just say, “Okay, we can talk about it.”

Nine participants also discussed the ways a lack of understanding of patients’ social context could lead to harm. A white genderqueer queer (OA7) person gave this example

An incarcerated trans woman attempted self-orchiectomy because she was unable to access appropriate transgender care. She was brought to the Emergency Department [where urology was consulted and] “salvaged” the testicles and then she was admitted to the jail psychiatric unit and discharged back to the county jail where a couple months the exact same thing happened…

A Discrepant Case: In Defense of Doctors

One participant defended medical providers against the critiques made by other focus group participants. On the topic of gatekeeping TI1 said, “It’s part of [providers’] role to challenge, make sure you’re serious about [transition] because it’s a big thing. You don’t want to do something you’re not sure about …”

Discussion

The competencies identified by community members included a focus on patient autonomy and shared decision making. In particular, participants described the importance of avoiding gatekeeping or presenting obstacles to transition-related health care. Community priorities also included being comfortable with patients, avoiding assumptions and behaviors that reinforce stigma, increasing knowledge of sexual practices and transgender health, and working to decrease the effects of social determinants of health. The latter was of particular concern to marginalized subpopulations, including transgender and genderqueer participants of color, who described multiple obstacles to health and wellbeing.

Participants’ concerns that physicians and other medical providers do not feel comfortable caring for them is striking, confirming what has been found in past studies. Physicians’ attitudes toward gay patients have shown a decrease, but not disappearance, in negative attitudes (Smith & Mathews, 2007). Additionally, on a standardized test of implicit bias, health care providers showed a preference for heterosexual people over gay men or lesbians. (Sabin, Riskind, and Nosek, 2015; Anselmi, Vianello, Voci, & Robusto, 2013). Stigma in the health encounters of transgender patients has also been documented (Poteat, German, & Kerrigan, 2013). This is in contrast to self-reported comfort of medical students in working with LGBT patients (White et al., 2015). One participant did defend providers, which could genuinely reflect high-quality care or could be due to a need to defend providers given a lack of available options. Educational interventions will be necessary to increase comfort and proficiency in caring for LGBTQI patients. Based on participants’ feedback, providers should be comfortable in taking a sexual history, using the correct pronouns and name for transgender patients, and providing a physical exam that respects gender identity.

Other organizations such as the AAMC and the American Psychological Association (APA) have created competencies for providing care for LGBTQI individuals. Many similarities exist, including the importance of medical knowledge related to caring for LGBTQI people and the importance of acknowledging and navigating power imbalances. The differences between community-identified competencies and national organizations’ competencies are important, especially regarding patient autonomy. Community-identified competencies focus on collaborative patient-physician partnerships, which could disrupt historic power imbalances that may perpetuate social determinants of health. This is especially salient regarding hormones and surgeries for transgender patients, as participants expressed concerns about health care providers limiting access to care that they perceived as essential. Medical paternalism, a term which may be applicable here, has been defined as “the interference with a person’s liberty of action, justified by reasons referring exclusively to the welfare, good, happiness, needs, interest or values of the person being coerced (Dworkin, 1972, p. 65).”

Although these were mentioned in the AAMC competencies in a parenthetical statement, “… identifying various harmful practices (e.g. … withholding hormone therapy from transgender individuals) that perpetuate the health disparities … (2014, p. 58),” patient autonomy may require more explicit focus by national organizations in order for change to occur. The APA competencies addressed these concerns directly and may be an important model for national medical organizations (2012, p. 840, p. 847). Similarly, while the AAMC competencies did not mention the stigmatized experiences of LGBTQI patients, including intravenous drug use, poverty, sex work, silicone injection, immigration, incarceration, and use of hormones acquired without a prescription, which were important to focus group participants, a number of these concerns including sex work, incarceration, and poverty were discussed in detail by the APA.

A unique community-identified competency elucidated in our focus groups was addressing common stereotypes of LGBTQI people. These stereotypes included the connections between gay men, transgender people, and HIV; assumptions that women who have sex with women do not have risk factors for STIs; and assumptions regarding sexual practices based on patients’ gender presentations or sexual orientations. Focus group participants suggested that these assumptions create barriers to effective care as they cause providers to make irrelevant or inappropriate recommendations and miss screening and treatment opportunities. Furthermore, these stereotypes may become barriers to building trusting relationships with patients. In addition, the layers of oppression for low-income and immigrant LGBTQI people, including documentation, language access, and financial access to quality care, were of importance to focus group participants, but not included in either the AAMC or APA competencies. Balancing the needs of marginalized subpopulations with more broadly generalizable principles may be challenging for national organization developing competencies; however, because of the multiple barriers to health and medical care that these patients face, we believe that their needs are particularly important to consider when creating medical education competencies.

Participants recognized several areas, such as the impact of paternalism or the needs of marginalized subpopulations, which were not included in the AAMC competencies. Including community stakeholders in competency development may better ensure that gaps in physicians’ knowledge are addressed and that physicians-in-training obtain necessary skills to build collaborative relationships with patients. Given this, we recommend formally including the perspectives of marginalized populations in competency development. Including community perspectives in recommendations for medical training may be one way to rebuild trust with marginalized groups. In every focus group, participants provided informal feedback expressing gratitude for this work.

Given time constraints and competing priorities in medical education, including curricula that addresses the needs of marginalized community members may be met with resistance. Prioritizing feedback from community members may also require a shift in theoretical perspective among the medical education community. As Mogedal points out, “Learning from the community confirms that the community actually has something to contribute … [and] implies the willingness to share power (1993, p. 128).” Community advisory boards have become commonplace in research and clinical settings, and community involvement could also play an important role in competency development. Incorporating community members into medical education curricular committees could be one way to better ensure the needs of patients are represented.

A strength of our study is that we sampled diverse groups of LGBTQI patients, which is important as LGBTQI community members have different needs and experiences. A group made up mostly of Latina transgender women was concerned with substandard care, immigration status, language access and stigmatized experiences, including silicone injection by unlicensed medical providers; a group made up mostly of white transgender men discussed difficulties accessing hormone therapy; and a group of mostly white cisgender women was concerned with doctors’ assumptions and negative attitudes after the women disclosed their sexual orientations. We may have failed to capture the perspectives of other subgroups, but our findings emphasize the importance of being broadly inclusive.

Limitations

As the reader interprets and examines our findings, we would like to highlight several key limitations of our study. Though we tried to be broadly inclusive, important subgroups may not have been represented. Few participants were African American or Asian American. We did not collect data on socioeconomic status or physical ability so important subgroups may have been missed. Nearly all participants were between 17 and 40, and older or younger individuals may have different health care experiences and needs. Geographic locations were limited to large urban centers on the east and west coasts and a small city in Vermont; LGBTQI people in other parts of the country, including more rural locations or locations in the south; and outside the United States may have very different experiences. An extremely small number of participants identified as intersex. Therefore our findings are not necessarily relevant to these populations. The demographic form was not translated into Spanish, limiting this data. Aside from one focus group conducted in Spanish, all the groups were conducted in English, limiting accessibility to people who primarily speak other languages. Most of our recruitment was conducted through email or social media, limiting accessibility to people without access to computers. Moreover, LGBQTI people who are not open about their identities may not have felt comfortable attending the focus groups. In addition, given limitations of time and funding, data saturation was not reached. Another potential limitation is that because of an IRB request, we asked participants to speak in the third person, limiting first-person narratives.

Conclusions

Our study uncovered multiple LGBTQI community-identified physician competencies including being comfortable with patients, sharing medical decisions, avoiding assumptions, applying LGBTQI-specific knowledge, and acknowledging and addressing the social context of health disparities. Overall, LGBTQI participants overwhelmingly desired patient-centered medical care that acknowledged the experiences of marginalized communities. We learned that community-identified competencies may include the needs of marginalized subpopulations, which may be missed by general guidelines developed for health and mental health providers. As such, we recommend the following next steps for both medical and mental health professional training programs: (1) incorporate the five community-identified competencies into curricula; (2) create positions for local community members to sit on curricula committees; (3) formally query local marginalized populations regarding their health needs and incorporate those into curricula development; (4) conduct further research regarding the recommendations of LGBTQI populations, particularly marginalized subpopulations, to improve their health care experiences and decrease health disparities. In addition, we would suggest that health and mental health professional organizations incorporate the recommendations of LGBTQI community members into guidelines for working with LGBTQI patients.

Appendix

Focus Group Script

I am going to give you a few minutes to read this information sheet. I am looking for information about what you think doctors should know and be able to do to take good care of lesbian, gay, bisexual, transgender, queer, and intersex patients. When responding to the questions I ask, please do not share any identifying information about the individuals involved in the experiences, even if one of those individuals is you. We ask this because this is the only way we can protect the confidentiality of the people who are in the group. For example, if you are reporting an experience you have had, please frame this experience as “I know a person who experienced the following…” None of what you say should include anything that might help someone identify that person even if that person is you.

1. What are some positive experiences you’ve heard LGBTQI people describe with doctors?

2. What are some negative experiences you’ve heard LGBTQI people describe with doctors?

3. What knowledge do you think providers should have in order to work effectively with LGBTQI populations?

4. What attitudes do you think providers should have in order to work effectively with LGBTQI populations?

5. What skills do you think providers should have in order to work effectively with LGBTQI populations?

Note. This script was modified in settings with a more specific subset of LGBTQI people. For example, the acronym LGBTQI was replaced with transgender, genderqueer, and gender non-conforming in an early focus group held at a transgender conference.