"Decisions, Decisions": End-of-Life Planning for Sexual Minorities in the United States

Abstract

“Death may be inevitable, but a bad death is not” (Carr, 2019, p. 200). Dying patients and their caregivers frequently name end-of-life (EOL) decision-making as a core component of a “good death” or “dying well” (Byock, 1998; Steinhauser et al., 2000). And EOL planning is one of the few ways that people can ensure and retain their autonomy at the end-of-life. Given the growing yet under-researched LGBQ+ population in the United States, there is a dearth of research on sexual identity as an axis of stratification in relation to EOL planning behaviors and attitudes. Most research about EOL planning has emphasized racial/ethnic, (binary) gender, and (to a lesser extent) SES differences. For example, we know that Black and Latine respondents are significantly less likely to have a living will than White respondents, and we know that women have earlier and greater participation in EOL planning than men (Carr, 2019). However, this research focuses exclusively on different-sex couples or heterosexuals (or does not account for sexual identity or attempt to identify if respondents are sexual minorities). Consequently, very little is known about EOL planning for sexual minorities (SMs). Studies that consider SMs have been restricted to qualitative examinations of predominately White, affluent same-sex couples. Yet, SMs (i.e. people who do not identify as heterosexual or “straight”) are a diverse and growing population marked by stark health disparities and unique historical traumas, including the AIDS Epidemic of the 1980s. And this population faces unique minority stressors that include precarious legal environments in the EOL context, the consequences of which are not well understood. Minority stress theory (MST) posits that SMs should have lower rates of EOL planning as a consequence of these unique minority stressors. However, MST also suggests that structural discrimination in health care facing SMs (rooted in the AIDS Epidemic of the 1980s and the medicalization of sexuality and pathology ascribed to homosexuality) contributes to health disparities between SMs and non-SMs which may incentivize EOL planning among SMs. To test these competing hypotheses, I apply MST and SM structural discrimination perspectives in this novel investigation of variation between SMs and non-SMs, and variation within SMs, in EOL planning. Because SMs are a diverse population, this dissertation adopts an intersectional perspective, positing that examining SM status in isolation from other social identities may obscure important differences in how EOL planning is experienced. Further, this dissertation investigates differences in EOL planning among partnered SMs by adopting a gender-as-relational perspective, which suggests that EOL planning behaviors should vary based on the gender composition of couples (i.e. the combination of own gender and spouse or partner gender). Bridging theories of minority stress, structural discrimination, intersectionality, a gender-as-relational perspective and a “doing gender” perspective, I examine four broad research questions: (1) How does EOL planning vary by sexual identity?; (2) How does EOL planning differ vary by couple type?; (3) How are gender, SES, and race/ethnicity (i.e. established determinants of EOL planning among non-SMs) associated with EOL planning among SMs?; and (4) What are other sources of heterogeneity in EOL planning behaviors among SM Americans? I adopt a three-paper model for my dissertation to address these research questions by using: (1) cross-sectional and longitudinal data from the Health and Retirement Study (Chapters 3-4); and (2) cross-sectional data from a survey of approximately 1,200 SM adults that I designed and fielded via the Prolific platform (Chapter 5). Findings from Chapter 3 seek to answer research questions 1 and 3 by using self-reported sexual identity information to compare EOL planning behaviors between SMs and non-SMs; Findings from Chapter 4 seek to answer research questions 2-3 by using household data to identify and compare EOL planning behaviors between those in same-sex and different-sex couples; and findings from Chapter 5 seek to answer research questions 1, 3, and 4 by using data from an original survey of SM adults to explore variation among diverse SMs and to identify novel sources of heterogeneity in SMs’ EOL planning behaviors. Using cross-sectional data from the 2022 wave of the Health and Retirement Study (HRS), Chapter 3 finds that gay/lesbian Americans have significantly greater odds of engaging in formal EOL planning compared to heterosexual Americans, and gay/lesbian Americans also have significantly greater odds of engaging in informal EOL planning. Chapter 3 also finds that SMs identifying with “some other sexual identity” have significantly lower odds of engaging in formal and informal EOL planning compared to non-SMs. Using longitudinal data from the 2012-2018 waves of the HRS, Chapter 4 finds further evidence that EOL planning is unequally distributed among the U.S. population by SM status (as proxied by couple type). More specifically, findings from Chapter 4 show that those in same-sex couples devote considerable attention to multiple measures of formal EOL plans compared to those in different-sex couples. Using data from an original survey of SM adults, Chapter 5 finds very little evidence of differences by sexual identity in EOL planning behaviors when comparing across SM identities. However, findings from Chapter 5 identify several sources of heterogeneity in the EOL planning behaviors of SM Americans, including marital status, age group, minority stress, and social support. Synthesizing findings from Chapters 3-5 provides insights into how key sociodemographic variables (i.e. gender, SES, and race/ethnicity) are associated with EOL planning among SMs as well as between SMs and non-SMs in ways both consistent and divergent from the broader EOL planning literature. The results emphasize the importance of considering sexual identity and SM status in empirical investigations concerning the EOL context. Beyond extending our understanding of disparities in EOL planning, this dissertation contributes to the literature in several ways. Specifically, my dissertation: (1) provides new quantitative evidence on EOL planning among underexplored population groups by testing applications of multiple theoretical frameworks; (2) refines our understanding of social stratification by demonstrating the importance of examining sexual identity and gender as core axes of stratification rather than additive or secondary characteristics; (3) identifies how cultural norms surrounding gender and communication about death and dying apply to SMs; (4) identifies novel sources of heterogeneity in EOL planning among SMs; (5) provides an empirical foundation for future research on EOL planning behaviors and attitudes; and (6) advances knowledge about the increased diversity of families and growth of minority populations in a time of aging populations – illuminating potential vulnerabilities at the EOL among SMs and underscoring the importance of SM status and sexual identity to understanding how EOL planning is stratified. Lastly, this dissertation has broader impacts, such as the creation of a novel data source and fostering conversations among individuals, families, and communities regarding death and dying.