Barriers to family history knowledge and family communication among LGBTQ+ individuals in the context of hereditary cancer risk assessment

Abstract

Openness about identity as lesbian, gay, bisexual, transgender, queer, and other sexual orientations and gender identities (LGBTQ+) may cause strain on relationships between family members, which could lead to limited knowledge of cancer family history and reduced communication with family members. As a result, members of the LGBTQ+ community may have more difficulty accessing genetic counseling services for inherited cancer risk. We applied a mixed-methods approach to explore potential barriers to knowledge of cancer family history and family communication among participants of the Cancer Health Assessments Reaching Many (CHARM) study who self-identified as LGBTQ+. We assessed perceptions of family functioning and communication of genetic test results to family members using survey tools and supplemented these data with 20 in-depth interviews to further assess participant perspectives and experiences. LGBTQ+ participants were more likely to report unhealthy family functioning on the survey tool, and some interviewees endorsed that openness about their LGBTQ+ identity led to strained family relationships and reduced communication about their family history of cancer. Overall, this study identified barriers that may be faced by members of the LGBTQ+ community which could limit their ability to access genetic counseling services for inherited cancer risk.

INTRODUCTION

The two most prevalent hereditary cancer syndromes, hereditary breast and ovarian cancer syndrome (HBOC) and Lynch syndrome, affect an estimated 1 in 130 individuals in the US. Individuals with these syndromes face significantly elevated lifetime risks for cancer and are at risk to develop cancer at earlier ages compared to the general population (Chen et al., 2006; Manickam et al., 2018; Nelson et al., 2013). The National Cancer Institute recommends that individuals at increased risk for cancer be referred for genetic counseling for further risk assessment and consideration for genetic testing (P. D. Q. Cancer Genetics Editorial Board, 2002). Genetic counselors educate patients about their cancer risks and facilitate decision-making about clinical interventions to reduce cancer-related morbidity and mortality, such as prophylactic surgery and earlier and more frequent cancer surveillance (Daly, Pilarski, et al., 2016; Moyer, 2014; National Collaborating Centre for Cancer, 2013; Nelson et al., 2013). Oftentimes, risk assessment tools are used to guide referral decisions for genetic counseling services (Bellcross, Hermstad, Tallo, & Stanislaw, 2019; Kastrinos et al., 2017). These tools largely rely on family history information (primarily cancer type and age of onset) from first- and second-degree relatives. Once a pathogenic variant has been identified, individuals are encouraged to share this information with their biological relatives so that they can pursue genetic counseling for individualized risk assessment and management recommendations (Roberts et al., 2018). The realization of these goals is dependent on open family communication about cancer history and pathogenic variant status.

Members of the lesbian, gay, bisexual, transgender, queer, and other marginalized sexual orientations and gender identities (LGBTQ+) community represent a medically-underserved population (Hafeez, Zeshan, Tahir, Jahan, & Naveed, 2017). Individuals who identify as queer, questioning, and transgender are especially likely to report negative healthcare experiences (Macapagal, Bhatia, & Greene, 2016). Members of this community also report culturally insensitive communication from providers, insufficient social support, and increased social isolation (Boehmer, 2018; Bradford, Reisner, Honnold, & Xavier, 2013; Hill & Holborn, 2015; Hulbert-Williams et al., 2017; Kano et al., 2020). Despite the well-recognized disparities faced by this community in other healthcare settings there is currently a lack of research exploring disparities that affect individuals who identify as LGBTQ+ in hereditary cancer risk assessment and genetic testing.

Disclosure of LGBTQ+ identities within the family may significantly alter family dynamics, relationships, and communication (Heatherington & Lavner, 2008; Loftus, 2001). Negative reactions to this disclosure may result in strained relationships or estrangement from biological family members (Bradford et al., 2013; Heatherington & Lavner, 2008) which could impact health history communication and knowledge. As a consequence, the effectiveness of cancer risk assessment tools (many of which rely heavily on family history information) could be reduced. Not fully capturing cancer risk could limit referrals or access to genetic counseling and genetic testing for inherited cancer risk. To date, only one study has assessed family history knowledge in individuals that identify as LGBTQ+. This study assessed survey data from The Health of Women (HOW) Study and found that bisexual and lesbian women had greater odds of not knowing their family history of breast cancer [OR=2.59 (95% CI: 1.79–3.75) and OR=1.56 (95% CI: 1.07–2.29), respectively], ovarian cancer [OR=1.62 (95% CI: 1.23–2.14) and OR=1.27 (95% CI: 0.99–1.62), respectively], and colon cancer [OR=1.62 (95% CI: 1.23–2.13) and OR=1.46 (95% CI: 1.15–1.86), respectively] compared to heterosexual women (Roberts, Krakow, Wheldon, & Silver, 2019). Participants in this study were not asked directly about the cause of the lack of family history knowledge, or about how their sexual orientation may have impacted family relationships. As a result, the cause of this disparity is unknown.

Despite these results, there remain significant gaps in our understanding of the barriers to family history knowledge in the LGBTQ+ population and the extent to which openness about LGBTQ+ identity impacts communication about genetic risk in families. Further still, there are no studies assessing barriers in men who identify as LGBTQ+, an important population given men are generally less likely to discuss their health with others or share genetic risk information (Dean & Rauscher, 2018).

To fill these gaps in the literature, we performed a mixed-methods analysis (Creswell & Clark, 2017) to characterize health disparities and assess barriers and facilitators to the communication of hereditary cancer risk among individuals who identify as LGBTQ+. Through this study, we sought to determine if familial relationships strained by the disclosure of an LGBTQ+ identity may lead to (1) lower family health history knowledge and (2) reduced communication of genetic testing results back to at-risk family members. Understanding these dynamics is essential to providing the best possible care to this patient population.

METHODS

Study Overview

Our analysis was conducted using the Cancer Health Assessments Reaching Many (CHARM) study cohort. The CHARM study is one of the Clinical Sequencing Evidence-generating Research consortium (CSER) projects and focused on providing hereditary cancer risk assessment, genetic testing, and genetic counseling to participants from two large healthcare systems. Specifically, the CHARM study investigated methods to provide access to inherited cancer risk information to healthy 18–49-year-olds who speak either English or Spanish. Recruitment efforts actively engaged individuals belonging to populations historically underserved in healthcare, such as populations with low socioeconomic status, low educational attainment, racial and ethnic minority populations, and people who identify as LGBTQ+ (Kraft et al., 2020)

Study Population

Study participants were recruited from two large healthcare systems: Kaiser Permanente Northwest (KPNW), an integrated healthcare system in Portland, Oregon, and Denver Health, a federal qualified health center based in Denver, Colorado. Potential study participants were eligible for study enrollment if they screened positive on an electronic patient-facing family history risk assessment tool for HBOC and LS. This tool incorporated two clinically validated risk assessment tools: Breast Cancer Genetics Referral Screening Tool (B-RST™) 3.0 for HBOC syndrome (Bellcross et al., 2019) and PREMM₅™ for Lynch syndrome (Kastrinos et al., 2017). Participants with limited family history knowledge were also eligible for the study if they were adopted, had a small family size, or did not know their family health history. Following consent, participants received results from a targeted clinical exome which included genes associated with increased cancer risk. Participants could opt in for additional findings. Additional findings fell into two categories: medically actionable additional findings and carrier status for 14 recessive conditions.

Quantitative Data

CHARM participants completed an electronic, patient-facing risk assessment tool that was developed specifically for the study. This tool captured data necessary for inherited cancer risk assessment, including personal and family cancer history and sex assigned at birth [male/female]. In addition, CHARM participants were asked to complete three surveys during the course of the study: 1) a baseline survey administered following enrollment and consent, but before genetic testing results were disclosed; 2) a follow-up survey administered within one month of result disclosure; and 3) a second follow-up survey administered 6 months following result disclosure. The baseline survey included the General Functioning Subscale of the McMaster Family Assessment Device (GF6+ FAD), a validated tool that assesses an individual’s perception of their biological family in terms of interpersonal relationships and social environment (Boterhoven de Haan, Hafekost, Lawrence, Sawyer, & Zubrick, 2015). On the first follow-up survey, participants were asked to recall whether their genetic test results were positive or negative for cancer risk, carrier, or other medically actionable genes. On the second follow-up survey, participants were asked which of their family members and healthcare providers, if any, they had shared their results with.

Definition of LGBTQ+ Identity

On the baseline survey, participants selected the option that best described their sexual orientation [selection options: heterosexual, lesbian, gay, bisexual, not sure/questioning, another sexual orientation (with an open text field), and prefer not to answer] and their gender identity [selection options: female, male, transgender female, transgender male, non-binary/genderqueer, not sure/questioning, another gender identity (with an open text field), and prefer not to answer]. For the purpose of analysis, the study team used these selections to categorize participants as either LGBTQ+ or non-LGBTQ+. For the purpose of readability and clarity the authors will use these terms as defined here throughout the manuscript when referring to study participants. Participants were categorized as LGBTQ+ if their gender identity did not correspond to their reported sex assigned at birth (even if they did not self-identify as transgender) (National LGBT Health Education Center, 2016) or if they selected one of the following identities: lesbian, gay, bisexual, not sure/questioning, another sexual orientation, transgender female, transgender male, non-binary/genderqueer, not sure/questioning, or another gender identity. Participants were categorized as non-LGBTQ+ if their gender identity corresponded to their reported sex assigned at birth and they selected heterosexual for their sexual orientation. Survey data were collected and managed using REDCap electronic data capture tools hosted at KPNW (Harris et al., 2019; Harris et al., 2009).

Qualitative Interviews

CHARM participants who selected an LGBTQ+ identity on the baseline survey, had received their genetic testing results, and spoke English were eligible for a semi-structured, open-ended interview. Participants who indicated that they were not sure of or were questioning their sexual orientation or gender identity were ineligible since they may not have disclosed their identity to their family members. Potential participants were recruited by email with up to three follow-up phone calls by interview staff (JLS, JD). We developed an interview guide [See Supplemental Information] to elicit participant knowledge of family cancer history, experiences with family communication around family cancer history, barriers and facilitators that impact family communication, and sharing genetic test results with family members. In addition, we asked participants to describe their sexual orientations and gender identities and elaborate on how these identities influenced family communication about cancer history and genetic risk. We asked the CHARM study team, which included a patient advocate who identifies as LGBTQ+, and members of a KPNW employee resource group composed of individuals who identify as LGBTQ+ to review and provide feedback on interview questions for content and framing. This review informed modifications to the wording and tone of the recruitment materials and interview questions for cultural competence and sensitivity.

Interviews were conducted by two trained qualitative researchers (JLS and JD) via telephone. These interviews were, on average, 33 minutes in length [Range: 21–53 minutes] and were recorded and professionally transcribed verbatim. De-identified transcripts were uploaded into NVivo, a qualitative analysis software program (QSR International Pty Ltd, 2018), to facilitate coding and summarization. We developed our coding scheme by having each interviewer review a subset of transcripts. Codes denoted topics that arose from participants’ responses to the interview questions (e.g., plans to share CHARM genetic testing results with family members) and novel responses that emerged during the interview discussion (e.g., any future communication actions participants were considering). Coding commenced using an iterative process where two researchers (JLS and JD) independently coded 2–3 transcripts at a time, then discussed together to refine the codebook and application of codes to interviews. This process continued until the code list was static and all transcripts were coded. Then, using the query functions of NVivo, summary reports of coded text were produced and reviewed in partnership by JLS and JD to further group coded text into higher level categories or themes. Summary data from the interviews were compared and contrasted with survey data during meetings with the core study team (JEH, MS, SK, KM, LA, JLS, JD, BR) in an ongoing review process for discussion and refinement of interpretation, resulting in a final set of themes presented in this paper (Bernard & Ryan, 2010; Corbin & Strauss, 2008; Creswell & Clark, 2017; Patton, 2002).

Statistical Analyses

Descriptive statistics were used to summarize demographic characteristics of study participants and survey variables. Unadjusted models were used to compare continuous (t-tests) and categorical (chi square tests or Fisher’s exact tests, when appropriate) variables between LGBTQ+ and non-LGBTQ+ participants. An a priori threshold of p<0.05 was used to indicate statistical significance. All statistical analyses were performed in SAS software Version 9.4 (SAS Institute Inc., Cary, NC).

RESULTS

Participant Characteristics

Between August 2018 and April 2020, 832 CHARM participants completed the baseline survey. Of those, 31 had missing data for gender or sexual orientation. Demographics of LGBTQ+ and non-LGBTQ+ CHARM participants who responded to the survey are presented in Table 1. Of the 801 with gender and sexual orientation information, 133 (17%) self-identified as LGBTQ+, with a majority (79%) selecting lesbian, gay, or bisexual. Fourteen participants (11%) selected “another sexual orientation” and were given the option to self-describe. Of these, 9 participants self-described as “pansexual.” Twenty-three participants reported their gender identity as either transgender or non-binary/genderqueer. Of these, 3 participants identified as heterosexual, 2 as lesbian, 3 as gay, 7 as bisexual, and 8 as “another sexual orientation” (4 wrote in “pansexual”). LGBTQ+ participants were significantly younger (mean=32.5 years, SD=8.2 years vs. 36.9 years, SD=8.1 years, respectively; p<0.01), had a higher level of education (p<0.01), and were more likely to identify as white or European American (p<0.01) compared to non-LGBTQ+ participants. There was no significant difference for income (p=0.17) between these two groups (Table 1).

Table1.

CHARM participant demographics

Domain	Values	non-LGBTQ+ CHARM Participants N=668	LGBTQ+ CHARM Participants N=133	Interviewed CHARM Participants N=20
Age (in years)	Mean (SD)	36.9 (8.1)	32.5 (8.2)	32.0 (7.8)
	Range	18–50	18–49	18–48
Sex assigned at birth	Male	137 (20.5%)	37 (27.8%)	8 (40.0%)
	Female	531 (79.5%)	96 (72.2%)	12 (60.0%)
Gender identity	Male	137 (20.5%)	29 (21.8%)	6 (30.0%)
	Female	531 (79.5%)	81 (60.9%)	11 (55.0%)
	Transgender female	0	2 (1.5%)	0
	Transgender male	0	6 (4.5%)	2 (10.0%)
	Non-binary/Genderqueer	0	15 (11.3%)	1 (5.0%)
	Not sure/questioning	0	0	0
	Another gender identity	0	0	0
Sexual orientation	Heterosexual	668 (100%)	4 (3.0%)	1 (5.0%)
	Lesbian	0	21 (15.8%)	3 (15.0%)
	Gay	0	28 (21.1%)	8 (40.0%)
	Bisexual	0	56 (42.1%)	5 (25.0%)
	Not sure/questioning	0	10 (7.5%)	0
	Another sexual orientation	0	14a (10.5%)	3b (15.0%)
Race/ethnicityc	Native American	12 (1.8%)	2 (1.5%)	1 (5%)\
	Asian	36 (5.4%)	3 (2.3%)	0
	Black	35 (5.2%)	6 (4.5%)	0
	Pacific Islander	3 (0.5%)	0	0
	Middle Eastern	4 (0.6%)	1 (0.8%)	0
	White	288 (43.1%)	81 (60.9%)	15 (75.0%)
	Hispanic	231 (34.6%)	26 (19.6%)	3 (15.0%)
	Selected > 1 category	58 (7.9%)	13 (9.8%)	1 (5.0%)
	Unknown	1 (0.2%)	1 (0.8%)	0
Incomed	<$39,999	247 (37.4%)	60 (45.5%)	10 (50.0%)
	$40–79,999	233 (35.3%)	44 (33.3%)	8 (40.0%)
	≥$80,000	181 (27.4%)	28 (21.2%)	2 (10.0%)
Education	Some high school or less	90 (13.5%)	5 (3.8%)	0
	High school or associate collegee degree	314 (47.0%)	68 (51.1%)	9 (45.0%)
	Bachelors or graduate degree	264 (39.5%)	60 (45.1%)	11 (55.0%)

^a14 write-in responses for “another sexual orientation”: 9 pansexual, 3 queer, 1 queer/pansexual, 1 bi/pan, polyamorous

^b3 write in responses for “another sexual orientation”: 1 pansexual, 1 queer, 1 bi/pan, polyamorous

^cRace/ethnicity data missing for 9 participants

^dIncome data missing for 8 participants

^eIncludes occupational, technical, or vocational programs

Survey Data

Eligibility for Study

Twenty-four percent of LGBTQ+ participants were eligible for participation based on lack of knowledge of family health history, compared to 20% of non-LGBTQ+ participants, which was not statistically significant (p=0.24) (Table 2).

Table 2.

CHARM quantitative data

Domain	Values	non-LGBTQ+ CHARM Participants N=668	LGBTQ+ CHARM Participants N=133	Interviewed CHARM Participants N=20
Eligibility for study	Positive screen on RAT	537 (80.4%)	101 (75.9%)	12 (60.0%)
	Limited family history knowledge	131 (19.6%)	32 (24.1%)	8 (40.0%)
Family relationships a	Healthy family relationships	521 (81.2%)	86 (63.8%)	9 (45.0%)
	Unhealthy family relationships	121 (18.9%)	47 (36.2%)	11 (55.0%)
Family communication	Communicated study result with:
	At least 1 family member	331/407 (81.3%)	68/80 (85.0%)	11/16 (68.7%)
	Parent	260/294 (88.4%)	56/63 (88.9%)	10/11 (90.9%)
	Sibling	233/298 (78.2%)	41/56 (73.2%)	7/9 (77.8%)
	Children	135/206 (65.5%)	10/15 (40.0%)	0/5 (0.0%)
Provider communication	Yes	94 (23.2%)	16 (19.8%)	3/17 (17.7%)
	Not yet, but I plan to	199 (49.0%)	40 (49.4%)	7/17 (41.2%)
	No, and I don’t plan to	113 (27.8%)	25 (30.9%)	1/17 (41.2%)

^aFamily relationship scores missing for 32 participants

RAT = Risk assessment tool

Family Functioning

LGBTQ+ participants were more likely than non-LGBTQ+ participants to have GF6+ FAD scores > 2.0, indicating unhealthy family functioning [36% vs. 19%, respectively (p<0.01)] (Table 2). The rate of GF6+ FAD scores > 2.0 did not vary significantly between various sexual orientations [gay (37%), lesbian (29%), and bisexual (32%)] but was increased in individuals who selected “another sexual orientation” (60%) and “not sure/questioning” (60%) for their sexual orientation. Among participants who selected a gender identity other than cisgender (N=23), 8 (40%) had GF6+ FAD scores > 2.0.

Sharing Genetic Testing Results

At the time of this analysis, 535 participants had received their genetic test results and completed the 6-month follow-up survey. The proportions of participants who had either shared or planned to share their test results with at least one family member were similar between LGBTQ+ participants and non-LGBTQ+ participants (90% and 91%, respectively; p=0.77) (Table 2). All findings were similar when limited to participants who had a positive genetic testing result (data not shown).

Interview Data

We recruited 34 LGBTQ+ participants to participate in a qualitative interview, of whom 20 (59%) consented. The demographics of the interview cohort are presented in Table 1. While LGBTQ+ participants who were interviewed were not significantly different than LGBTQ+ participants who were not interviewed with respect to age (p=0.78), education (p=0.82), and income (p=0.49), , a higher proportion of the interview cohort had GF6+ FAD scores > 2.0 (p=0.06). There was no statistically significant difference between self-reported race/ethnicity between the two groups (p=0.30), 60% of LGBTQ+ participants who were not interviewed selected “White” for their race/ethnicity compared to 75% of LGBTQ+ participants who were interviewed. Of the 34 participants recruited for interviews, 25 (74%) had selected “White” for their race/ethnicity suggesting that the loss of diversity occurred due to our eligibility screening for interviews prior to recruitment. Though race/ethnicity was not part of our eligibility screening for interviews, the inclusion criteria of English as the preferred language may have limited the diversity of the eligible study population. Seven (35%) participants had positive test results (i.e., one or more pathogenic variants) returned to them. Of these, 3 participants were found to be carriers of a recessive condition, 4 were heterozygous for a pathogenic variant in MUTYH, and 2 were heterozygous for a pathogenic variant in a gene associated with increased cancer risk.

Among the 20 participants interviewed, 4 (25%) indicated a different sexual orientation and/or gender identity during the interview than they endorsed on the baseline survey: one participant selected lesbian on the baseline survey but identified as bisexual during the interview; one selected bisexual on the baseline survey but identified as pansexual during the interview; one selected gay on the baseline survey but identified as queer during the interview; and one selected female as their sex assigned at birth on the baseline survey but reported as transgender female during the interview.

During the interviews, the majority of participants (n=18) indicated that they had disclosed their sexual orientation and/or gender identity to their relatives. One participant had only disclosed their sexual orientation to their spouse, and another participant had not shared their sexual orientation and gender identity with any of their relatives. From the interview responses, four themes emerged relating to the negative impacts of openness about LGBTQ+ identity on family relationships and communication. Individually, these themes were not common across the interview cohort, but collectively they represent the experiences of most of the interviewees.

Theme 1: Family’s reaction to LGBTQ+ identity can strain relationship

Of the participants who had disclosed their LGBTQ+ identity to their relatives, 5 (28%) reported a negative reaction from family members that resulted in tension in their relationship. One interviewee described:

“I think my sister knew when I was gay when I was teenager. But it took me over ten years to tell people about it - it was not all at once. The last few people I told I was gay was my parents. And they didn’t take it well. It took twenty years for them to calm down about it. They still wanted grandkids from me. I mean they still want me to marry, I hate to say, a woman still. So I think they still think I could still be straight. It’s denial.” (Interview #14)

In another case, the interviewee shared:

“I disclosed to my brother in 1999 or 2000. [Brother’s reaction] is more like a mockery. It’s not sincere…as long as I’m not flamboyant or obviously gay, he [brother] is fine with it…I asked him recently, ‘what would you do if you saw me holding a guy’s hand…’ and he is like ‘I would cross the street.’ He is like, ‘I just don’t agree [with it] … So there’s a separation of trust and a separation of identity that I think we would have to get patched to have the trust to where if something was to happen regarding issues surrounding being gay, I could come to him and be like, look, this is what I have.” (Interview 17)

An additional four interviewees (22%) reported positive reactions from their parents and siblings, but negative ones from second-degree relatives. In many of these cases, negative reactions were reported only on one side of the family and were often attributed to ideological differences about sexual orientation.

“… my mom accepted me for who I was…on my father’s side, my aunt who is still alive, I feel like it has impacted our communication. We don’t talk and we were very close at one time.” (Interview 8)

By contrast, one interviewee reported favorable reactions to their LGBTQ+ identity from family members:

“I disclosed voluntarily. My family is pretty comfortable with everything so I didn’t really hesitate to share that … I’m guessing my sisters probably were more aware of it before I ever said anything then … At the time I actually identified as lesbian. And I came out to my mom when she was sick. And she, actually, is the one who sort of indicated, ‘well, just because you like to sleep with girls doesn’t mean you’re necessarily gay.’ [Laughs] I think she was more aware that I was probably bisexual than even I was at the time.” (Interview 6)

Theme 2: Openness about LGBTQ+ identity can impact health-related communication within families

When asked if LGBTQ+ identity had impacted communication about health-related matters with their families 6 (30%) reported that it had. In one case, openness about LGBTQ+ identity improved communication within the family:

“I would say it probably broadened that kind of [health] communication – whole new level of trust. If they accept me for who I am then there’s probably not much else could say that would change that…it makes us very open.” (Interview 15)

Other interviewees reported that sharing their LGBTQ+ identity had created barriers to communication within their families:

“Mom was accepting … it didn’t bother them at all [mom and siblings] … My dad still absolutely hates the idea and won’t even talk about it - I don’t want to tell him anything about myself anymore … I won’t talk to him about anything that’s important at all … I don’t have that kind of trust to feel comfortable sharing … but for everybody else, I don’t really think that impacts anything at all.” (Interview 19)

“By the time I did come out a number of them [family members] had no interest in communicating with me anymore, or disowned me…My grandparents initially tried to kind of ‘save me’ before they chose to abandon me…I think an uncle told me that ‘I’m basically dead to them’…I chose not to communicate or try to discuss or locate any of that [health] information. Although it may have been interesting to find out, there was no point for me to officially try to communicate with them on it. Because, I mean, it would just be more disruptive for me in my personal life…emotional safety kind of stuff.” (Interview 20)

Two participants reported that they did not perceive their sexual orientation or gender identity to directly impact communication with their family about “universal” health matters (meaning conditions that are not linked to sexual activity or are commonly associated with the LGBTQ+ community). However, they did perceive that it would limit communication about health-related matters that could be linked to their sexual activity.

“If I was [HIV] positive he wouldn’t know that because of knowing how you get it… but with cancer I think he’d be more receptive because it’s a universal conversation. Whereas anything related to sexual orientation is not.” (Interview 17)

Theme 3: Openness about LGBTQ+ identity can limit knowledge of family history

Prior to joining the CHARM study, 6 participants (30%) stated that they knew their family history of cancer, 6 (30%) said they had some knowledge of their family history of cancer, and 8 (40%) reported limited or no knowledge of their family history. Eighteen participants (90%) did not cite openness about their LGBTQ+ identity as a reason for limited family history knowledge:

“I actually had asked my doctor if there was any type of genetic testing that I could possibly do just because I do have a background in my family of cancer…I knew it [family history] pretty well - like who has been diagnosed and who has passed away from cancer and things like that, so I would say I was pretty informed …I did ask my mom a little bit more [about family health history] to kind of to explain more, because I was a little bit younger on some of those instances, so I kind of just asked for more information.” (Interview 9)

However, two participants (10%) specifically cited their LGBTQ+ identity (and the resulting strain caused by being open with this identity) as a barrier to family history knowledge. One participant noted that their parents are deceased and has limited contact with their brother given the brother’s bias against their sexual orientation; thus, they did not reach out to their brother for any family history information. The second participant does not have a relationship with their father, which limited family history knowledge from their father’s family:

“It’s kind of limited [family health history]. I knew that my grandmother on my mom’s side and my grandmother on my dad’s side both died from cancer … so like getting into the study I’ve been more interested in finding out more about my father’s side of the family. But that’s been a little harder than I’ve imagined. I was trying to go through my mom because again I haven’t spoken to my dad in, well, I don’t even know how long it’s been. So I was trying to go through my mom, but she doesn’t really have information on his family.” (Interview 13)

Theme 4: Openness about LGBTQ+ identity can affect sharing of test results with family members

At the time of their interview, 15 participants (75%) had shared their results with at least some of their family members and 5 (25%) had not. One interviewee specifically cited their LGBTQ+ identity as the reason for not sharing their results:

“My family…I consider to be abusive in ways…so I didn’t communicate anything [participation in study or results] with them.” (Interview #20)

Another interviewee shared their results (heterozygous for a pathogenic variant in MUTYH) with their mother, but not their father, due to their LGBTQ+ identity. In this instance, they assumed that their mother would relay the information to their father, indicating that other family members may act as intermediaries who facilitate intrafamilial communication:

“My mom is really cool. I like her a lot. My dad and I kind of have a rocky relationship because he’s very, ‘you’re doing it my way or the highway type thing’, plus he didn’t really like me being gay… [The CHARM study] found a genetic mutation that I might have a better chance for colon cancer… [shared result] with my mom, grandmother and my aunt, my sister because I’m closer to them. I get along with all of them very well, so I told them… If they ask about it I’ll tell them. But I’m not going to run up to them and tell them [referring to some extended family members].” (Interview 12)

Another interviewee stated that they did not have an issue sharing their negative result, but that they would rely on a third party (such as a healthcare provider) to communicate more serious results with their family members:

“I think if something terrible happened, like I did get cancer, I wouldn’t be the one calling to tell him. I would have the doctor or a counselor do it because I wouldn’t think he’d be receptive to me to do it. It would be more blaming. Whereas if a professional told him, he might be more willing to listen.” (Interview 17)

Of the interviewees who did not cite their LGBTQ+ identity as a barrier to disclosing their test results, one shared:

“Maybe I told my Mom. I’m not sure. Definitely my husband …There’s really no one that I would specifically decide not to. It’s hard for me to imagine bringing up with my aunts or uncles or my more extended family. But anyone in my nuclear family, my parents or my brother, I would feel perfectly fine bringing it [results] up.” (Interview 1)

DISCUSSION

We used a mixed methods approach to explore the impacts of LGBTQ+ identity on family history knowledge and health-related communication in families. In our cohort, participants who identified as LGBTQ+ were more likely to have unhealthy family functioning than non-LGBTQ+ participants. From our thematic analysis of 20 interviews we found that openness about LGBTQ+ identity can strain relationships between family members, impact communication within families, limit family history knowledge, and limit how genetic test results are shared within some families. These findings illustrate several barriers that patients who identify as LGBTQ+ may face which could impact their ability to access or fully benefit from genetic counseling services in a hereditary cancer setting.

From the interviews we learned that many of our participants had one or more family members who had negative reactions to their LGBTQ+ identity. The strain resulting from these reactions was likely a factor contributing to the higher level of unhealthy family functioning we found in this population. Strained relationships with family members have been widely reported in the LGBTQ+ community, particularly in youth and adolescents, underscoring the importance of assessing family support when working with this population (Bregman, Malik, Page, Makynen, & Lindahl, 2013; Homma & Saewyc, 2007; Katz-Wise, Rosario, & Tsappis, 2016; Needham & Austin, 2010). Foreknowledge of these dynamics could help genetic counselors anticipate some of the barriers their LGBTQ+ patients may face within their families.

In many cases, participants reported that the intrafamilial tension resulting from their LGBTQ+ identity impacted communication within their families. There were multiple examples of individuals who endorsed reduced communication or no communication at all with family members due to their LGBTQ+ identity. One of the most poignant examples of this was seen in an interview where the participant stated that they would rely on a third party (such as a healthcare provider) to communicate serious health information to one or more family members due to the belief that these relatives would not take the information seriously if it came directly from the participant. This participant attributed the lack of credibility with their family members to their LGBTQ+ identity. While sometimes true, it is not universal that members of the LGBTQ+ community have decreased communication with their families. One participant in our study, for example, stated that disclosing their LGBTQ+ identity strengthened feelings of trust in their family and, thereby, improved communication. These findings underscore a core tenet of genetic counseling - to always ask without assumption.

Family history is critically important in hereditary cancer risk assessment; and, in many cases, the referral to a genetic counselor for this assessment is prompted by a characteristic pattern of cancer in a family. Further still, many healthcare and insurance providers strictly rely on family history criteria to determine eligibility for genetic testing. Therefore, inaccurate or incomplete family history information could limit an individual’s ability to access genetic counseling and/or genetic testing for inherited cancer risk. The proportions of LGBTQ+ participants and non-LGBTQ+ participants that were eligible for the CHARM study based on limited family history information were not significantly different; however, we learned through the interviews that, despite meeting inclusion criteria, some LGBTQ+ participants were unable to collect some or all of their family history information. This was especially apparent with one participant who stated that they relied on their mother to collect and relay the paternal family history of cancer. For this participant, there was a family history of cancer on both sides of the family (maternal and paternal). The maternal family history of cancer was sufficient for them to be eligible for the study, but the participant was equally concerned about the paternal family history of cancer, about which they had no concrete details. The participant endorsed that this family history knowledge gap was the result of the strained relationship with their father caused by their LGBTQ+ identity. This type of dynamic, if not identified by a genetic counselor during a patient visit, may leave the impression of a negative or less striking family history of cancer, which could impair risk assessment and lead to inappropriate testing recommendations. Therefore, it is important for genetic counselors to assess family dynamics which may limit family history knowledge, especially in the LGBTQ+ population where strained family relationships occur more commonly.

While genetic counselors typically support communication of a positive test result through means such as family letters, the ultimate responsibility of disseminating genetic test results within a family falls to the individual who was tested (American Society of Clinical Oncology, 2003). Moreover, studies show that results are most commonly shared with female, first-degree relatives (Conley et al., 2020; Healey et al., 2017; McGivern et al., 2004). Uptake of cascade testing for genetic variants identified in families is low, and previously identified barriers to sharing include poor communication, emotional factors, misunderstanding of results and geographically or emotionally distant family relationships (Daly, Montgomery, Bingler, & Ruth, 2016; Healey et al., 2017; Landsbergen, Verhaak, Kraaimaat, & Hoogerbrugge, 2005). In our survey data, we found no significant difference between the rates of sharing test results among LGBTQ+ participants compared to non-LGBTQ+ participants. Even LGBTQ+ participants in the interview cohort who had high GF6+FAD scores reported sharing their positive (and in many cases negative) test results with one or more family members. For example, one participant reported sharing their positive test result (in this case heterozygosity for a pathogenic variant in MUTYH) with their mother who, in turn, was to share that information with the father who had a history of colorectal cancer. This finding suggests that even in cases of strained familial relationships due to LGBTQ+ identity, it is still possible that genetic test results will be shared with other at-risk family members. As a result, genetic counselors could expect that their efforts to facilitate sharing of genetic test results within a family (such as writing family letters) may be just as effective for their LGBTQ+ patients as with their non-LGBTQ+ patients.

In addition to describing the impacts of LGBTQ+ identity, our study demonstrates the importance of using inclusive language in patient-facing materials. In developing our survey materials, we sought to be as inclusive in our terminology as possible with the goal of capturing all gender identities and sexual orientations in a sensitive and respectful manner. Despite these intentions, several participants selected “another sexual orientation” on the survey and wrote in terms not included as options (e.g., queer and pansexual). In addition, during the interviews, four (20%) participants described their LGBTQ+ identity differently than they had on the baseline survey. These findings are consistent with those from another study, which concluded that questions about sexual orientation and gender identity should allow for nuance and fluidity when describing these identities (Suen et al., 2020). In practice, this finding underscores the importance of using open-ended questions and mirroring patient language when discussing sexual orientation and gender identity during a genetic counseling visit.

Study Limitations and Future Directions

Even though we recruited a large proportion of LGBTQ+ participants in the CHARM Study, we only conducted in-depth interviews with 20 of these participants. The majority of these 20 interviews were with individuals who identified as white or European American, were from urban areas, and were relatively young and well educated. The themes that emerged from our interviews most likely do not reflect the diversity of experiences and opinions of the LGBTQ+ community and should be the basis for further investigation and research. Also, we did not ascertain the reason(s) for family dysfunction in participants with GF6+FAD scores >2.0. Following our structured interview guide, interviewers specifically asked about impacts on family relationships related to LGBTQ+ identity, but they did not investigate other possible contributors to these dynamics. The findings from our analysis should serve as preliminary evidence supporting the need to conduct a broader survey of the LGBTQ+ community with the goal of better understanding the unique issues faced by this population.

Conclusion

We found that CHARM study participants who identify as LGBTQ+ were more likely to have unhealthy family functioning than non-LGBTQ+ participants. We learned that some participants specifically cited their sexual orientations and/or gender identities as sources of strain on their relationships with their family members, which led to decreased knowledge about family history of cancer and reduced health-related communication. Without accurate and complete family history information it may be more difficult to access genetic counseling and testing services for hereditary cancer risk assessment. As a result, members of the LGBTQ+ community, who, at baseline, belong to a population that has been historically underserved in healthcare, may encounter more obstacles to getting the care they need. Inability to access genetic counseling services could result in the delay of life-saving cancer screening or risk-reduction strategies. Therefore, it is incumbent on genetic counselors to recognize the possible limitations of a negative family history during a hereditary cancer risk assessment in this population; and, when possible, reduce barriers to accessing genetic counseling and testing services for those with limited family history knowledge.

WHAT IS KNOWN ABOUT THIS TOPIC

Members of the LGBTQ+ community experience a number of healthcare disparities. While other studies have described issues relating to access to care, provider trust, and poor health outcomes, few have explored the specific issues faced by this patient population in the context of genetic counseling and genetic testing.

WHAT THIS PAPER ADDS TO THE TOPIC

This study describes the impact of LGBTQ+ identity on family history knowledge and health-related communication within families. This is the first study to describe how openness about LGBTQ+ identity may hinder access to genetic counseling services, and it explores possible downstream limitations on risk assessment and access to genetic testing in the context of evaluations for inherited cancer risk.

Recommendations.

1. Acknowledge that individuals who identify as LGBTQ+ may have strained relationships with their families which could impact family history knowledge and health-related communications.

2. Consider that incomplete family history knowledge for individuals who identify as LGBTQ+ may result in inaccurate assessment of cancer risk, which can impact access to genetic counseling services and cancer genetic testing.

3. Tailor patient resources and communication tools, such as family letters, to support LGBTQ+ patients in sharing test results within their families.

4. Facilitate open communication about LGBTQ+ identity by not making assumptions about sex assigned at birth, gender identity, and sexual orientation.

5. Use inclusive language on all patient-facing materials.