eAppendix 1. Interview Questionnaire
eAppendix 2. Participant Recommendations for Clinical ACP Encounters
eTable. Interview Participant vs Noninterview Participant Comparison
eAppendix 3. Interview Guide
eFigure. Joint Display Analysis
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Reich AJ, Perez S, Fleming J, et al. Advance Care Planning Experiences Among Sexual and Gender Minority People. JAMA Netw Open. 2022;5(7):e2222993. doi:10.1001/jamanetworkopen.2022.22993
What are the barriers and facilitators to advance care planning discussions for sexual and gender minority (SGM) people?
In this qualitative study of 201 SGM people and 402 non-SGM people, survey participants reported experiences of discrimination in health care, and interview participants described concerns regarding whether their preferences would be supported.
The findings suggest that discrimination in the health care system is an important barrier to advance care planning for SGM people.
Advance care planning (ACP) can promote patient-centered end-of-life (EOL) care and is intended to ensure that medical treatments are aligned with patient’s values. Sexual and gender minority (SGM) people face greater discrimination in health care settings compared with heterosexual, cisgender people, but it is unknown whether such discrimination occurs in ACP and how it might affect the ACP experiences of SGM people.
To increase understanding of barriers and facilitators of ACP facing SGM individuals.
Design, Setting, and Participants
This mixed-methods national study of ACP included a telephone survey of self-identified SGM and non-SGM participants in a nationally representative sample drawn from a larger omnibus national panel by SSRS. Qualitative interviews were conducted with a subset of survey participants who identified as SGM. Data were collected from October 2020 to March 2021.
Main Outcomes and Measures
The survey included 4 items from the validated ACP Engagement Survey, adapted to capture experiences of discrimination. Interviews asked about participants’ experiences with ACP, including the appointment of medical decision-makers, sharing preferences, and experiences within the health care system more broadly.
A total of 603 adults participated in the survey, with 201 SGM individuals (mean [SD] age, 45.7 [18.7] years; 101 [50.2%] female; 22 [10.9%] Black, 37 [18.4%] Hispanic, and 140 [69.7%] White individuals) and 402 non-SGM individuals (mean [SD] age, 53.7 [19.2] years; 199 [49.5%] female; 35 [8.7%] Black, 41 [10.2%] Hispanic, and 324 [80.6%] White individuals). Regarding reasons for not completing ACP, SGM respondents, compared with non-SGM respondents, were more likely to say “I don’t see the need” (72 [73.5%] vs 131 [57.2%], P = .006) and “I feel discriminated against by others” (12 [12.2%] vs 6 [2.6%], P < .001). Of 25 completed interviews among SGM participants, 3 main themes were identified: how fear and experiences of discrimination affect selection of clinicians and whether to disclose SGM identity; concerns about whether EOL preferences and medical decision-makers would be supported; and a preference to discuss EOL decisions and values outside of clinical settings.
Conclusions and Relevance
This study found that fear of disclosing sexual orientation or gender identity information and discrimination are important barriers to ACP for SGM in clinical settings, but discussions of preferences and values still occur between many SGM people and medical decision-makers. More SGM-specific patient-centered care might better support these discussions within the health care system. Furthermore, health systems can facilitate improved engagement by supporting clinician sensitivity training, including guidance on documentation and requirements.
More than half of sexual and gender minority (SGM) adults experience some form of discrimination in health care; for those who are transgender or gender nonconforming, the percentage increases to 70%.1 SGM communities face multilevel stressors, including family rejection, financial insecurity, and anxiety over concealment of their sexual orientation and/or gender identity; the stress of discrimination toward SGM individuals can reduce life expectancy by up to 12 years.2 SGM people have been left out of the recent conversations regarding the outcomes (or lack thereof) of advance care planning (ACP)3 because there is little evidence, data, or stories from SGM communities about the value (or not) of ACP.4 While ACP may positively affect quality end-of-life (EOL) care and ensure preferences are aligned,5 ACP among SGM is not well understood, and patient-centered approaches for this population are lacking.
Existing studies from Canada6 and Australia7 suggest a need to foster greater engagement in ACP among SGM people, given low ACP rates, historical mistrust, and a lack of training for clinicians to address planning needs unique to this population, including documentation. A review of EOL preparatory behaviors found a low prevalence of ACP discussions between SGM people and primary care clinicians, particularly for transgender people.8
Little is known about the ACP experiences and documentation for this group. This study examines readiness of SGM people to engage in ACP and appoint a medical decision-maker (MDM), concerns about discrimination, and their experiences with EOL discussions with MDMs.
In this study, a national, explanatory sequential mixed methods design was used.9-11 In the first phase, we collected survey data, and in the second phase, we conducted qualitative telephone interviews. Four items related to ACP and documentation were inserted into a nationally representative weekly omnibus phone survey conducted by SSRS research over a period of 7 weeks between October and November 2020. The study protocol was approved by the MassGeneralBrigham institutional review board and reporting adheres to Consolidated Criteria for Reporting Qualitative Research (COREQ) 12 and American Association for Public Opinion Research (AAPOR)13 guidelines.
Adults 18 years of age and older were included, with SGM respondents self-identified through a demographic item and a random sample of non-SGM respondents. The survey was designed to include 200 SGM and 400 non-SGM participants to provide sufficient power to compare responses across groups. Furthermore, we estimated that a sample of 175 to 200 SGM participants would yield enough people who would respond yes to a follow-up interview (assuming 66 affirmative responses out of a cohort of 200, with 95% confidence and 10% margin of error). Race and ethnicity data were collected to characterize the sample and assess for differences in ACP. Self-reported categories included Asian; Black Hispanic; Black non-Hispanic; Native American, American Indian, or Alaska Native; White Hispanic; White non-Hispanic; multiracial and other race.
The ACP survey items were adapted from the validated survey questions for ACP engagement in the study by Sudore et al,14 including an option to capture the role of discrimination in hesitancy to speak about wishes for EOL. We also included an item inviting SGM respondents to provide contact information for follow-up interviews.
In the second phase of the study, we conducted interviews to help explain the survey data and to learn more about the experiences of discrimination among SGM participants. In line with strong mixed-methods techniques, joint displays were used.
The qualitative interview guide was developed after the survey data collection period. Draft domains were based on review of the literature and were informed by the survey results. Domains included ACP discussions with clinicians, expectations, perception of quality and effects of ACP, and satisfaction with discussions; selecting MDMs, sharing preferences, and understanding of their role. The draft guide was revised based on feedback from the study expert advisory panel and members of the SGM community to further explore experiences of discrimination, which was reported at higher rates among SGM survey participants, and whether participants felt that their identified MDMs would be supported. Items were also included to ask to what extent participants felt their SGM identities influenced ACP discussions and experiences and whether they would like sexual orientation and/or gender identity (SOGI) information to be collected by clinicians (eAppendix 3 in the Supplement). The guide was pilot tested for length and clarity.15 Survey respondents who consented to a follow-up interview were scheduled within 3 to 4 months following the survey (January to March 2021). Interview participants were provided a $50 honorarium for their time.
Differences in simple proportions between SGM and non-SGM participants were evaluated using χ2 tests in SAS version 9.4 (SAS Institute). P < .05 was considered significant, and all tests were 2-tailed.
The research team met to discuss emerging concepts and update the interview guide. A preliminary codebook was developed using the interview guide, and two authors (A.J.R. and S.P.) independently tested it on 2 transcripts. These coded transcripts were reviewed line by line to discuss discrepancies and revised based on discussion with the research team and applied to all other transcripts. Twenty percent of transcripts were double-coded by A.J.R. and S.P. and compared to ensure fidelity to the codebook. To avoid bias, multiple members of the research team reviewed the data separately to gather alternative explanations from each researcher’s positionality: physician, social scientist, and research professionals. The team engaged in a process of thematic analysis16 of all coded transcripts and interview memos, guided by interpretive description.17
The survey included 201 respondents who self-identified as SGM (101 [50.2%] female; 22 [10.9%] Black, 37 [18.4%] Hispanic, and 140 [69.7%] White individuals), and 402 who did not self-identify as SGM (hereafter referred to as non-SGM; 199 [49.5%] female; 35 [8.7%] Black, 41 [10.2%] Hispanic, and 324 [80.6%] White individuals). Compared with non-SGM participants, SGM were slightly younger, at a mean (SD) of 45.7 (18.6) years, than non-SGM, at a mean (SD) of 53.7 (19.2) years and more likely to be single, never married (55 [27.4%] vs 58 [14.4%]). SGM participants were also slightly more likely to be uninsured (32 [19.0%] vs 37 [13.2%]) and to live in urban areas (116 [57.7%] vs 203 [50.5%]) (Table 1).
Of the 201 SGM survey participants, 75 agreed to enroll in the qualitative portion of the study, and 25 interviews were completed. Differences between those who did and did not participate in interviews appear in the eTable in the Supplement. Participants self-identified as gay (8 [32.0%]), lesbian (5 [20.0%]), bisexual (7 [28.0%]), pansexual (3 [12.0%]), queer (3 [12.0%]), asexual (1 [4.0%]), and polyamorous (1 [4.0%]). Regarding gender identity, the sample included 2 transgender participants (8.0%), 1 nonbinary participant (4.0%), 20 nontransgender participants (80.0%), and 2 who reported unsure (8.0%). Most reported White race (22 [88.0]%) and had high levels of education (23 [92.0%] had at least some college). Overall, 9 (360%) described themselves as single, 4 (16.0%) single living with a partner, 10 (40.0%) married, and 2 (8.0%) divorced (Table 2). Most described their health as at least fair (84%) (eAppendix 1 in the Supplement). Of the 75 who initially agreed, the study team started with the earliest respondents until saturation was reached. Our goal was to have a diverse sample without focusing exclusively on one sexual or gender minority group; experiences were common across the diverse groups interviewed (eAppendix 4 in the Supplement). With the rigorous coding processes the study team had in place, there was a very strong sense of saturation, meaning no new themes emerged, by the time we reached 25 interviews.
SGM and non-SGM participants did not differ significantly in response to survey questions related to readiness to sign official papers naming an MDM or readiness to talk to an MDM about the kind of care they would want at the end of life (Table 3). In response to the question, “If you haven’t spoken about your wishes, why haven’t you?” SGM participants were significantly more likely to select “I don’t see the need” than non-SGM participants (72 [73.5%] vs 131 [57.2%]; P = .005) and “I feel discriminated against” (12 [12.2%] vs 6 [2.6%]; P < .001). The response to the discrimination item was explored further in interviews.
Three main themes and 2 subthemes were identified (Table 4). These themes captured the impact of discrimination on ACP encounters and beliefs about the usefulness of ACP documents for future care.
Participants reported some situations in which SGM status is very relevant to care, and relevant to ACP, but wanted to first be assured that the information will be managed in a safe and respectful way. While some described a desire to normalize the process of asking about, for example, partners and preferred pronouns, others worried that they may be discriminated against based on that information or did not want to be defined by it.
Participant 20 described hesitation to disclose SGM status to clinicians. “If the reason they were asking [about SOGI] is their absolute written ethical policy was we ask because we want to accommodate the LGBTQ [lesbian, gay, bisexual, transgender, and queer] community, then the answer is yes. If it were for any other reason, the answer is hell no.”
Some proactively address this by noting SGM status early on, while others opt to avoid disclosing. Participant 25 said, “I guess it’s always on the back of my mind that it could be a problem for someone, and that’s why I tend to proactively say before I see someone new, ‘I'm gay,’ in case it’s a problem.”
In general, SGM people described facing an additional burden in establishing a good clinical relationship that would facilitate quality ACP discussions. As participant 17 explained, “It’s influenced the difficulty I have had in finding a primary care physician that I feel comfortable talking with about certain health concerns.” One participant expressed fear over addressing ACP discussions in the context of their identity as an SGM person, “I am really, really scared. I never thought of those two things [referring to SGM and ACP discussions] at the same time” (participant 23).
Participants described multiple identities as important to their experiences of health care and ACP, including race, socioeconomic status, and disability, as well as the regional or cultural context in which their identities are understood. Within the United States and across the health care system, some communities may feel more accepting and safe than others. Discrimination, or fear of it, may occur within multiple layers. Others described how different parts of their personal identities intersect with SGM identities and how that impacted ACP.
Many participants pointed to regional or urban/rural differences. For example, participant 3 stated that “not normative in Oklahoma is interesting because you either learn to be very good at hiding it, or you learn to really, really just not care. That’s not the right word. You definitely care. But acknowledge that other people don’t like you.”
Some participants described how different parts of their personal identities intersect with SGM identities. For example, participant 9 said, “I’m autistic. So I am actually far more likely to get issues with advocacy over being autistic than I am over being miscellaneous flavors of queer. But the way those things intersect is very often relevant.”
SGM participants described concerns over whether their MDMs or medical preferences for EOL care would be supported, by both their family and by the health care system. Participant 6 explained, “I’ve discussed it with my spouse, and she already said, ‘Yes. If you land up in the hospital with something that’s long-term or terminal or something, I’ll just bring your syringe every couple of weeks and give you your testosterone.’ Because the last thing I want to do is maybe get sick for six months and die not looking like the person I lived my life as. … I think it’s very, very important for clinicians to be able to say, ‘Look, if it could lengthen your life by eight months, would you want us to withhold your hormone therapy or vice versa?’ … I think that they need to ask really detailed questions like that.”
While many expressed that families were supportive of their relationships, not all did. For example, participant 10 stated that “even though my mother treated my daughter as one of our family, because she wasn't my birth child—she was my lesbian partner’s birth child—she was always treated just a little bit differently. And so there’s a rift in my family because of that.”
Furthermore, there was concern about how MDMs would be treated or whether they would be respected by the health care system generally. Participant 20 said, “Oh, yeah, if they tried to go over our wishes and try to go to my family, I would not be happy. … But I guess it is something that you have to have a concern about that they try to get your next of kin. I just hope that those forms I fill out are enough.”
Some who were married described the legal aspect of marriage being important because it meant that nobody could deny someone of the same gender the right to make decisions on their behalf. Others worried that the lack of a legal marriage may mean their partner may not be allowed to make decisions. Participant 7 said, “What I fear is my boyfriend and I aren’t married, so I don’t know if my sexual orientation, with us not being married—whether him being allowed to be a primary decision-maker or him having power of attorney would be more difficult than it would be in a heterosexual couple.”
Another participant described the sense of comfort derived from knowing their relationship had legal status. “Since M is my spouse now, I don’t worry about it as much as before she was my spouse.”
While relatively few participants described formal ACP in a clinical setting, many had discussed EOL wishes with important others and considered and identified (sometimes officially) MDMs. This finding, that most of these discussions occur outside of clinician settings, is consistent with patterns seen among non-SGM people as well.18
Many described detailed discussions with MDMs regarding preferences. For example, participant 5 described their discussion with their spouse: “I know that while quality of life is important to both of us and that is in our plan, if it comes to the point where you’re hooked up on life support and life support is the only thing that's keeping you going, unplug it. …That’s not something that either one of us want, to be kept here, our bodies just kept alive, waiting. So we’ve discussed that.”
Through the use of joint display analysis, we used side by side comparison of the survey and interview data19 to assess for confirmation, expansion, or discordance between the data sets (eFigure in the Supplement). Survey results indicate that SGM and non-SGM individuals did not differ significantly regarding naming and talking to MDMs. However, there were important differences in why they have not spoken to their clinician, and challenges with clinical encounters related to being SGM were described in the qualitative interviews (eFigure in the Supplement). Survey and interview data were concordant except for SGM participants citing they “don’t see the need” to speak about wishes if they had not. While 73% of survey respondents who had not spoken about wishes denied the need for these discussions, in interviews that asked why they did not feel the need to document their wishes—regardless of age—participants described concerns about whether preferences would be honored and MDMs would be supported. Younger participants who had spoken about wishes often described more complicated health statuses that motivated documentation. Participants offered recommendations on how to improve the ACP experience among SGM and clinicians. Recommendations focused on taking an honest and direct approach to discussions, promoting inclusion, and avoiding judgment (eAppendix 2 in the Supplement).
This national mixed-methods study found that discrimination and fear of discrimination in clinical settings may be an important factor influencing how SGM people engage in (or avoid) health care and ACP discussions. This extends to relationships with clinicians and comfort having ACP or confidence that personal wishes would be supported. In interviews, participants reported a spectrum of preferences regarding whether they wanted clinicians to ask for SOGI data. Many described clinical encounters in which acceptance, understanding, and support of SGM people was not clearly expressed by clinicians or health care organizations. The important connecting factor was a need to be assured that they would be treated safely and respectfully.
These findings are consistent with literature regarding fear and discrimination preventing SOGI disclosure and delaying entry to care20 and suggest that the health care system could make improvements to better support the ACP needs of SGM. Clinicians can take proactive measures to promote inclusivity and signal to patients that they will provide care in a nonjudgmental manner. A study of nurses found a lack of knowledge about advance directives for SGM, difficulties in having conversations, and a need for education and training in ACP for this population.21 Demeester et al22 offer a conceptual model to promote shared decision-making for SGM and historically marginalized populations, which includes 6 organization-level drivers (workflows, health information technology, organizational structure and culture, resources and clinic environment, education and training, and incentives and disincentives). Contextual changes are designed to establish a safe environment, build trust, and reduce stigma.
Participants in this study described approaches clinicians can implement to establish a supportive environment to engage in ACP, including an honest and direct approach, promoting inclusion, and avoiding judgement. The potential to mitigate anticipated bias can have important impacts. An analysis of transgender and gender nonconforming individuals23 found that those with higher levels of confidence that health care professionals will treat them with dignity and respect at EOL had increased odds of perceiving that they were aging successfully.
In addition to assuring the clinical environment has a zero-tolerance policy for discrimination, organizations should have current information and training available on state and federal policies relevant to EOL planning among SGM people.24 Findings from this study indicate that SGM individuals have documented MDMs at higher rates than non-SGM individuals, yet still worry about whether their preferences and MDMs will be supported. Health care organizations must go beyond documentation and ensure that patients can be confident that their information is recorded and interpreted appropriately.
This study has limitations. First, all SGM people were included in one group, but the diversity of identities within the SGM umbrella makes it difficult to capture all experiences in depth. The survey instrument also limited the gender item to a male and female binary. Second, among those survey participants who agreed to follow-up interviews, our subsample was predominantly White with high education levels. Those who participated in interviews may not reflect the experiences of other SGM people.
The results of this study indicate that most SGM participants have either already talked to someone about their EOL wishes or are planning to; in interviews, we found that many EOL discussions occurred outside the clinical setting. Clinicians should acknowledge that many SGM patients have already thought about preferences and determine how MDMs should be involved and included in planning processes using language that gives SGM patients and those who care for them space to answer in a way that fits them and their circumstances.
Accepted for Publication: June 5, 2022.
Published: July 20, 2022. doi:10.1001/jamanetworkopen.2022.22993
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Reich AJ et al. JAMA Network Open.
Corresponding Author: Amanda Jane Reich, PhD, MPH, Center for Surgery and Public Health, Brigham and Women’s Hospital, 1620 Tremont St, Boston, MA 02120 (email@example.com).
Author Contributions: Dr Reich had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Reich, Perez, Fleming, Ladin, Weissman, Candrian.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Reich, Perez, Candrian.
Critical revision of the manuscript for important intellectual content: Reich, Fleming, Gazarian, Manful, Ladin, Tjia, Semco, Prigerson, Weissman, Candrian.
Statistical analysis: Manful.
Obtained funding: Weissman.
Administrative, technical, or material support: Reich, Perez, Ladin, Semco, Prigerson, Weissman, Candrian.
Supervision: Prigerson, Weissman.
Conflict of Interest Disclosures: Dr Reich reported receiving grants from Brigham and Women’s Hospital (National Institutes of Health grant No. R017034) during the conduct of the study. Dr Fleming reported personal fees from Brigham and Women’s Hospital during the conduct of the study. Dr Gazarian reported receiving grants from National Institute of Nursing Research during the conduct of the study. Dr Manful reported receiving grants from the National Institutes of Health during the conduct of the study. Dr Weissman reported receiving grants from the National Institutes of Health during the conduct of the study. No other disclosures were reported.
Funding/Support: The work was supported by grant R017034 from the National Institutes of Health.
Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.