Abstract
This article introduces the multidimensional properties of social connectedness among sexual and gender minority (SGM) midlife and older adults and examines the relationship between these properties and general health. Data were analyzed from Aging With Pride: National, Aging, and Sexuality/Gender Study, including 2,450 SGM adults aged 50 and older in the United States. The structure, function, and quality of interpersonal relations as well as community-level activities and engagement were measured through a self-administered survey and an in-person interview. Findings indicated that SGM midlife and older adults, on average, had a large social network with high bridging potential and low density, consisting of more nonrelative family members than immediate family members. They also showed frequent availability of social support, moderate or higher satisfaction with interpersonal relations, and moderate SGM community engagement. Properties of social connectedness differed by gender, sexual identity, and gender identity, with SGM men, sexually diverse women, and transgender people showing distinct challenges in interpersonal relations. All aspects of social connectedness were positively associated with good general health, particularly network diversity, outdoor leisure activity engagement, and access to health-related decision support, controlling for age and chronic conditions. Care receiving and loneliness were negatively associated with good general health. Intervention development can target these factors to promote social and community connectivity and reduce the negative health effects of persistent social stressors. This study underscores the necessity of addressing all facets (i.e., structure, function, and quality) of interpersonal relations encompassing both immediate and chosen family as well as community-level social connectedness.
Keywords: social network; social support; lesbian, gay, bisexual, transgender, and questioning; social connectivity; social relations
Social connectedness has been identified as a vital resource that has far-reaching effects on physical (Holt-Lunstad, 2018) and mental health (Webster et al., 2022) and well-being (Pinquart & Sörensen, 2000) throughout the life course, leading to optimal aging (Bosnes et al., 2019). Older adults who foster and maintain social support and resources are better equipped to handle stressful life events such as bereavement and retirement, disabling chronic conditions, limited functionality, and decline in economic resources (Aldwin & Yancura, 2010; Lubben & Gironda, 2003; Thoma & Gee, 2019). Over the past few decades, considerable progress has been made in comprehending the multidimensional aspects of social connectedness, including interpersonal relations and community engagement (Antonucci et al., 2009; B. Cornwell et al., 2008) and its health consequences among older adults. Limited research suggests that the nature of social connectedness for sexual and gender minority (SGM) older adults may be unique from older adults in the general population (Kim et al., 2017). Yet, present SGM aging and health studies have not comprehensively explored all dimensions of social connectedness and their roles in health mechanisms.
This article describes how multidimensional social connectedness is conceptualized and assessed in the National Health, Aging, and Sexuality/Gender Study (NHAS), the first longitudinal study that aims to understand mechanisms of the health and well-being of SGM adults aged 50 and older (Fredriksen-Goldsen & Kim, 2017). Using baseline data, the article also provides a foundational description of SGM older adults’ social connectedness by gender, sexual identity, and gender identity; examines the relationship between social connectedness properties and general health; and explicates how these findings can be used to design interventions to promote social and community connectivity.
Conceptualization of Social Connectedness in the NHAS
The NHAS is guided by the health equity promotion model (HEPM), which elucidates mechanisms of health disparities and equity among SGM older adults from a life course developmental perspective that highlights both shared SGM life course experiences and individual differences in their health trajectories (Fredriksen-Goldsen & Kim, 2017; Fredriksen-Goldsen, Simoni, et al., 2014). The conceptual model asserts that distinct SGM experiences of social connectedness are considered social processes that operate as key modifiable pathways to health outcomes, interplaying with the psychological, behavioral, and biological processes.
SGM older adults have encountered challenges in developing and maintaining social connectedness. These challenges include historical contexts such as the AIDS pandemic (Rosenfeld et al., 2012) and SGM identity-related life experiences such as social exclusion, family rejection, identity stigma, and identity concealment (Brennan-Ing et al., 2014; Klein & Golub, 2016). Such obstacles can lead to limited social networks (Fredriksen-Goldsen, Kim, Bryan, et al., 2017). Yet, SGM individuals have managed to build support networks interpersonally and unite as a community. Underground communities started to expand for sexual and gender minorities during and after World War II, defying societal stigma (Canaday, 2011). The gay rights movement experienced a significant turning point with the Stonewall Riots in 1969. SGM activists and grassroots political organizations have transformed resistance into a pressing call for “emancipation” (Weststrate & McLean, 2010). The resistance gradually shifted toward an increasingly urgent demand for community integration. The HEPM posits social connectedness as a potential health-promoting pathway through which SGM older adults can navigate social resources and secure social support and a sense of community belonging, which in turn influence psychological health, health-promoting behaviors, and good physical health (Fredriksen-Goldsen, Kim, Bryan, et al., 2017).
With its framework, the NHAS acknowledges social connectedness as a combination of interpersonal relations and community engagement, drawing upon the convoy model of social relations (hereinafter referred to as the convoy model; Antonucci et al., 2014) as well as B. Cornwell et al.’s (2008) suggestion to expand the scope of social connectedness to encompass community-level connections. The convoy model suggests that individuals’ life experiences are intertwined with social networks of close and supportive others, and these interpersonal relations can be understood as multidimensional in terms of structure, function, and quality. It is important to distinguish community engagement from interpersonal relations in the context of the types and scope of resources obtained, while both play integral roles in social connectedness. Community engagement also requires a distinct set of skills and a higher level of commitment to social connectedness. The NHAS has incorporated the following properties of interpersonal relations and community engagement to gain a comprehensive understanding of how SGM older adults are socially connected or disconnected. The rationale behind the selection of these properties is briefly explained below.
Social Connectedness Property: Interpersonal Relations.
Dimension of Interpersonal Relations: Structure.
The structural aspect of interpersonal relations can be conceptualized as objective configurations of individuals who are closely connected in terms of composition. Social network size is one of the most commonly used indicators of the structure of interpersonal relations, and a large network is known to benefit the health and quality of life of older adults in general (Rafnsson et al., 2015) as well as SGM older adults (Fredriksen-Goldsen et al., 2015). SGM older adults with diverse types of relationships within their network (e.g., partner or spouse, close friends, immediate family) report better mental health (Kim et al., 2017). Recent studies have demonstrated that network diversity is independently linked to better cognitive and physical functioning as well as lower mortality rates among older adults, regardless of social network size (Ali et al., 2018).
Identity homophily can be defined as the tendency for network members to hold similar characteristics (McPherson et al., 2001). SGM older adults have often experienced social ostracism and exclusion due to their sexual and gender identity (Fredriksen-Goldsen et al., 2019). Their peer networks often provide a supportive environment for sharing similar experiences and emotional support (Hudson & Romanelli, 2020). Peer networks are essential in health promotion as research has shown these networks can supplement absent familial network support (Fiori et al., 2006). Yet, little empirical research has examined the effect of identity homophily on health and quality of life among SGM older adults. Among older adults, generally, peer homophily tends to be related to similarities in not only health-promoting behaviors but also health risk behaviors and depressive symptoms (Flatt et al., 2012).
Network density refers to the extent to which people in social networks are tied to each other. A greater level of network density is built upon close interpersonal relations where network members share information and resources and potentially can play a caregiving role when one of them needs assistance with daily activities (Hurlbert et al., 2000). Network bridging, on the other hand, refers to the potential of being able to connect people in a social network who do not have strong connections with each other or do not have any connections at all (B. Cornwell, 2009). A person who is in this role might have a greater chance to have connections across multiple communities and a capacity to access diversified information and resources and transfer them to their social ties (Burt, 1995) as well as have more positive health outcomes (B. Cornwell, 2009; Perry et al., 2022). Currently, there is a research gap in understanding the extent to which social ties in the network of SGM older adults are connected or disconnected and how being a bridging potential is related to health outcomes.
The volume of contact with network members objectively demonstrates the extent of connectivity via mutual interaction with social ties. Among SGM older adults, a greater volume of interaction with network members has been linked to an increased likelihood of interacting with diverse social ties (Kim et al., 2017), which in turn can lead to accessing a wider range of resources, including psychological, informational, and material resources, and better quality of life (Berkman et al., 2000). Nurturing a close-knit social bond on a consistent basis, regardless of the size and diversity of social network, can be beneficial. This is particularly applicable to SGM older adults who may encounter limited connections.
Dimension of Interpersonal Relations: Function.
Social ties play an important role in the provision of support to SGM older adults. Social support is a multidimensional construct, with each underlying dimension of support (e.g., informational, emotional, tangible, affectionate, companionship support; Sherbourne & Stewart, 1991) functioning distinctly to enhance the health and mental health of its recipient. For example, among SGM older adults, informational support (e.g., advice, suggestions, and direction; Cohen & Wills, 1985) has been identified as an important facilitator of help-seeking. Studies have found that health provider recommendations from trusted network members may increase access to and use of health services (Hudson & Romanelli, 2020; Matsuzaka et al., 2021). The provision of tangible support through the exchange of material resources is common among social networks of SGM communities and within chosen families—nonbiological kinship networks created through close relationships with friends and peers (Jackson Levin et al., 2020). Research has identified shared cohabitation, automobiles, costs, and skills exchanges (e.g., childcare) as forms of tangible support provided through mutual aid networks among disabled and chronically ill SGM adults (Jackson Levin et al., 2020). Emotional support is one of the most salient forms of support impacting the health of SGM adults. Not only does emotional support mitigate the deleterious effects of discrimination on help-seeking behaviors (Romanelli et al., 2018) but it is directly associated with decreased psychological distress among sexual minority older adults (Lyons, 2016). Finally, companionship support is given and provided through participation in leisure/recreational activities with others (Cohen & Wills, 1985). Generally, older adults active in social activities (e.g., volunteering) report improved health (Tang, 2009), while SGM adults specifically report improved mental health with participation in leisure, wellness, and physical activities (Fredriksen-Goldsen, Kim, Bryan, et al., 2017).
Another important functional property of social networks among SGM older adults is informal caregiving. As daily living activities are often more limited with aging, the need for informal caregiving likely increases over time among SGM older adults. Furthermore, while the role of close friends in terms of caregiving among SGM older adults is crucial (Shiu et al., 2016), care from families of choice might be limited given lack of legal standing, especially when health care decisions are required (Muraco & Fredriksen-Goldsen, 2011). It will be essential to understand whether existing social network ties can function within a caregiver role and whether health-related decision support is available to fully understand their informal care needs.
Dimension of Interpersonal Relations: Quality.
The qualitative aspect of interpersonal relations refers to individuals’ subjective evaluation of their interpersonal relations and the support they receive. The quality of relationships may vary significantly among social network members, and social networks can hold both negative and positive attributes (Antonucci et al., 2014). Whereas evidence supports the overall positive association between presence of support network and quality of life among older adults, additional evidence suggests that network ties can also be a source of discord and conflict (Connidis & Barnett, 2018).
Thus, satisfaction is a key property representative of network quality. Generally, this property represents positive feelings and attitudes toward the dyad partner (Fincham & Rogge, 2010) and plays a significant role in health promotion (McLaughlin et al., 2010; Whitton & Kuryluk, 2014). Research examining SGM individuals’ level of satisfaction with their network often focuses on relationships with immediate family. Satisfaction with partners remains an important area of research within SGM populations as evidence indicates that some SGM-specific factors, such as internalized stigma and outness as well as couple-level minority stigma, may directly impact this outcome (Ballester et al., 2021; LeBlanc & Frost, 2020). In addition, some SGM individuals can experience familial strain and rejection; therefore, extending research beyond significant others and biological family to assess satisfaction with other critical network members, including chosen family, is necessary (Fredriksen-Goldsen et al., 2019). This fills an important gap in the current literature, as little is known about the role of the quality of other network types, such as close friends and ex-partners, in health promotion and under certain circumstances where long-term care is needed.
Similar to satisfaction, emotional closeness with network members among SGM individuals is often considered in the context of familial/parental and partner relationships. It is posited that the level of emotional closeness is linked to well-being through increased belonging and esteem (B. Cornwell et al., 2009). Among familial relationships, disruption in emotional closeness among SGM individuals can result from identity disclosure avoidance (Horn & Wong, 2014) and anticipated or actual rejection of the family member’s sexual or gender identity (Katz-Wise et al., 2016). Meanwhile, internalized hetero/cissexism leads to emotional distancing in romantic relationships (Ballester et al., 2021). B. Cornwell et al. (2009) noted that, despite a focus on kin and romantic relationships, emotional closeness must be considered among other close contacts of older adults. Besides the direct association with health, close relationships are important in health promotion as contacts can help recognize health concerns and influence one’s help-seeking behaviors (Pescosolido et al., 1998).
Another qualitative property of interpersonal relations is social isolation or subjective feeling of social isolation, which has deleterious consequences on physical, psychological, and cognitive health (Cacioppo & Hawkley, 2009; E. Y. Cornwell & Waite, 2009). Structural social isolation typifies many older SGM adults’ social network experiences with their having a higher likelihood of living alone than their non-SGM counterparts (Fredriksen-Goldsen, Kim, Shui, et al., 2017), which is found to be significantly associated with loneliness via a restricted social network and lack of social support (Kim & Fredriksen-Goldsen, 2016).
Social Connectedness Property: Community Engagement.
SGM individuals are often embedded in a larger SGM community context, enabling access to a supportive, accepting network that ultimately promotes health through the provision of belonging support (Ceatha et al., 2019; Holt-Lunstad & Unchino, 2015). While SGM-specific community engagement might include this sense of belonging, participation in social activities, and contributing to the larger community through community activism (Hudson & Romanelli, 2020), activities with other network members through volunteer work, spiritual/religious pursuits, and outdoor leisure activities encompass positive community engagement opportunities. Though network size and closeness may decrease with age, for some, these types of community-level social activities may increase with age (B. Cornwell et al., 2008). Problematically, in many SGM communities, factors like ageism (Wight et al., 2015) may exclude SGM older adults from realizing the benefits of community engagement (e.g., belonging, resilience, mastery; Emlet et al., 2017). For example, research has found that older gay men cite facing social challenges, such as feeling invisible within the larger community. Still, little is known about to what extent SGM older adults are involved in community engagement activities. It is critical to fill this gap as community participation and engagement activities have been connected to better general health (Choi et al., 2016).
Social Connectedness by Sexual and Gender Identities
As highlighted by the HEPM (Fredriksen-Goldsen, Simoni, et al., 2014), in order to fully comprehend social connectedness among SGM older adults, it is imperative to acknowledge the intricate nature of intersecting social identities encompassing diverse gender and sexual and gender identities as well as age cohort. Despite this, most studies on social connectedness among SGM individuals have overlooked the diversity inherent within these communities (Institute of Medicine, 2011). Limited literature suggests that older lesbians and bisexual women and men have the highest likelihood of having children, while older gay men report the lowest likelihood (Brennan-Ing et al., 2014). Gay and bisexual older men are more likely to live alone and experience limited social networks and lack of social support (Fredriksen-Goldsen et al., 2013). Whereas transgender older adults are more likely to have children, are less likely to live alone, and have a larger social network than their cisgender counterparts, they show lower community belonging and availability of social support (Fredriksen-Goldsen, Cook-Daniels, et al., 2014).
To gain a comprehensive understanding of social connectedness in this context, it is crucial to conduct further research using demographically diverse data with reduced sampling bias. Prior studies relying on nonprobability sampling may not adequately depict the overall and subgroup distributions of social connectedness among SGM older adults. By addressing these limitations, we will be able to enrich our understanding of the distinct needs and strengths of SGM individuals with differing sexual and gender identities.
The Present Study
The present study introduces each measure of multidimensional social connectedness assessed in the NHAS; presents the distributions of the social connectedness properties by gender, sexual identity, and gender identity; and examines the associations between these properties and general health among SGM older adults.
Method
Data
In 2014, the NHAS sampled adults aged 50 and older who identified as lesbian, gay, bisexual, or transgender or were engaged in same-sex sexual behavior or a romantic relationship with someone of the same-sex or gender. They were recruited via aging agency contact lists and chain-referral sampling across all U.S. census divisions, and they were born between 1916 and 1964, with 1,092 participants aged 50–64 (born 1950–1964) and 1,358 participants aged 65 and older (born 1949 or earlier). A self-administered survey, available in English and Spanish, was employed in the first wave of 2014 and completed by 2,450 participants. Descriptive statistics for the sample are presented in Table 1. A subsample of 318 participants from Atlanta, Los Angeles, New York City, and Seattle metropolitan areas was randomly selected for an in-person interview, and they completed a social network battery of measures as well as life event inventories and biological, physical, functional, and cognitive assessments. This research was approved by the University of Washington’s Institutional Review Board. A more detailed description of the methods, including sampling strategies, response rates, and participant background characteristics, was published elsewhere (see Fredriksen-Goldsen & Kim, 2017).
Table 1.
Baseline Characteristics of the Sample: National Health, Aging, and Sexuality/Gender Study, 2014 Survey (N = 2,450)
| Variable | Unweighted n | M or % [95% CI] |
|---|---|---|
| Age | 2,450 (range: 50–98) | 61.4 [60.9, 61.9] |
| Gender | ||
| Women | 995 | 43.4 [40.2, 46.6] |
| Men | 1,389 | 50.7 [47.4, 53.9] |
| Gender diverse | 64 | 5.9 [4.4, 8.0] |
| Sexual identity | ||
| Gay/lesbian | 2,102 | 72.3 [69.0, 75.3] |
| Bisexual | 216 | 17.2 [14.6, 20.1] |
| Sexually diverse | 127 | 10.6 [8.5, 13.1] |
| Gender identity, transgender | 207 | 16.8 [14.3, 19.8] |
| Race/ethnicity | ||
| Non-Hispanic White | 1,888 | 76.9 [73.9, 79.7] |
| Hispanic | 188 | 9.8 [7.9, 12.1] |
| African American and Black | 216 | 9.1 [7.3, 11.3] |
| Other | 151 | 4.2 [3.0, 5.8] |
| Education | ||
| High school or less | 222 | 25.8 [22.6, 29.2] |
| Some college | 1,116 | 37.8 [34.8, 40.9] |
| College graduate or more | 1,106 | 36.4 [33.5, 39.5] |
| Household income, 200% FPL or below | 882 | 36.5 [33.4, 39.8] |
Note. Survey weights were applied to data analyses. FPL = federal poverty level; CI = confidence interval.
The NHAS developed study instruments to assess the key components of the HEPM. We employed standardized measures whenever possible, which were taken from existing aging and health studies so that the results could be compared to those for the general population. For measures specific to the SGM older adults, we either developed new measures or modified existing ones. The self-administered survey instrument was to be brief enough for older adult participants to complete in a reasonable amount of time, psychometrically robust, and culturally responsive to SGM older adult populations. For the measures of social connectedness, its multidimensional aspects and its likelihood of being associated with health outcomes were taken into account. In-person interviews were able to add to the interview guide additional measures of social connectedness that would require follow-up probing from the interviewer. The goal of developing a set of measures assessing social connectedness was to provide researchers with the potential to link multiple dimensions of social connectedness to health outcomes and empirically identify modifiable factors that would effectively reduce health disparities.
Measures of Social Connectedness in NHAS Self-Administered Survey
Interpersonal Relations.
Structure.
The NHAS self-administered survey measured social network composition and volume of contact. To measure social network composition, separate questions were asked about six different types of social ties. Survey participants were asked whether they were partnered/married (=1) or single (=0). They were also asked whether they were in contact with any living children including adopted or stepchildren and how many of them had a close relationship with, ranging from 0 to 10 (=10 or more). The same set of questions was applied to four other types of social ties: other immediate family members (e.g., brothers or sisters, parents, cousins, or grandchildren), ex-partners/spouses, friends, and neighbors. The network diversity index was calculated by counting how many different types of social ties participants reported they had, ranging from 0 to 6. Total social network size was the summed number of close social ties across the six types, ranging from 0 to 51. The ratio of nonrelative family members to immediate family members was computed by dividing the total network size of close ex-partners/spouses, friends, and neighbors by the total network size of current partner/spouse, children, and other immediate family members. NHAS survey measured the volume of contact with each type of social ties except for current partner/spouse by asking how often participants talk or communicate with them, and the response options included never to almost every day. Following the suggestions by B. Cornwell et al. (2008), we estimated the number of contacts per year in the social network by transforming the survey responses (i.e., 0 = never; 1 = less than once a year; 3 = a few times a year; 36 = a few times a month; 52 = once a week; 365 = almost every day) and then aggregated these estimates across the five network types, which ranged from 0 to 1,825.
Function.
To measure types of social support, the NHAS adopted the four-item abbreviated version (Gjesfjeld et al., 2008) of Medical Outcomes Study Social Support Scale (Sherbourne & Stewart, 1991). Participants were asked how often each type of support was available to them if they needed it on a 5-point Likert scale (0 = never; 4 = very often). The types of social support included in the scale were tangible support, emotional-informational support, positive social interaction support, and affectionate support. The Cronbach’s α was .86. To assess informal caregiving and care receiving, participants were asked whether they currently assist a partner, spouse, friend, or family member who has a health issue or other needs (1 = yes; 0 = no) and whether they receive any assistance from them because of a health issue or other needs (1 = yes; 0 = no).
Quality.
In the NHAS self-administered survey, the quality of social connectedness was measured by assessing the satisfaction with two main sources of interpersonal relations (i.e., personal relationships and friends) as well as feelings of social isolation. Utilizing two items from the social dimension of the brief version of the World Health Organization Quality of Life (Bonomi et al., 2000), participants were asked how satisfied they were with their personal relationships and how satisfied they were with the support they got from their friends on a 5-point Likert scale (1 = not at all; 5 = extremely). Loneliness was measured with the three-item University of California, Los Angeles loneliness scale (Hughes et al., 2004) assessing frequencies of feeling isolated from others, feeling lack of companionship, and feeling left out. A 5-point Likert scale was used (0 = never; 4 = very often). A summary score was created by averaging across the three items, with higher scores representing higher levels of perceived loneliness. The Cronbach’s α was .89.
Community Engagement.
Both community engagement activities in the larger society as well as SGM-specific community engagement were assessed in the NHAS self-administered survey. The dimensions of community engagement activities included attending spiritual or religious activities, attending club meetings or group activities, going out for enjoyment, and volunteering, and they have been widely used to examine how community activities outside home relate to health outcomes and the social and environmental dimensions of quality of life (Douglas et al., 2017; Freedman et al., 2011; Latham & Clarke, 2018). Participants were asked how often they did each of the community engagement activities in the past month. Possible responses included 0 = never, 1 = rarely, 2 = some days, 3 = most days, and 4 = every day. SGM community engagement was measured with a newly developed scale. The constructs assessed by this scale included a sense of belonging, social activity in the SGM community, a sense of trust and helpfulness, and making contributions to the SGM community, which were derived from multiple studies (Chavis & Pretty, 1999; Frost & Meyer, 2012; Lin & Israel, 2012). The scale has four items that cover these constructs. Participants were asked, “Think about the LGBT community. Indicate the extent to which you agree” to the four statements: “I feel part of the community,” “I am active or socialize in the community,” “I get help from the community,” and “I help other people in the community.” A 6-point Likert scale was used, which ranged from 0 = strongly disagree to 5 = strongly agree. All four items were satisfactorily loaded on one latent factor with a good model fit (comparative fit index = 1.00; standardized root-mean-square residual = .008; root-mean-square error of approximation = 0.00), and the Cronbach’s α was .86. A summary score was computed by averaging the items, with higher scores indicating greater levels of engagement in the SGM community.
Measures of Social Connectedness in NHAS In-Person Interview
The NHAS in-person interview asked a series of questions to further understand the structural, functional, and qualitative properties of core confidant network adapting the confidante network roster (B. Cornwell et al., 2009) from the National Social Life, Health, and Aging Project. The confidante network roster asked participants to name up to five people with whom they had discussed important matters (i.e., core confidante), and for each confidante they identified, follow-up questions were subsequently asked to examine their relationship to the participant, gender including gender nonbinary, sexual orientation, gender identity, cohabitation status, age, frequency of contact, and closeness. The participants were also asked if they would discuss health-related concerns with the identified confidantes.
Structure.
Core confidante network size was computed by counting the number of people whom participants identified as discussion networks in the confidante network roster. SGM identity homophily rate was assessed by computing the proportions of confidantes with SGM identity in the respondent’s network, and this approach is similar to that used in the previous studies (Ibarra, 1992). For instance, if the relative rate measure of sexual and gender identity shows that a participant has higher than 50% of core confidantes with an SGM identity, this would be interpreted as manifesting stronger preferences for SGM than cisgender heterosexuals as core confidantes. To measure network density and network bridging potential, additional questions assessed how frequently their identified confidantes talked to each other with eight possible response options from “every day” to “have never spoken to each other.” As suggested by B. Cornwell et al. (2009), the density (the extent to which confidantes know each other) was calculated as the proportion of pairs in which two confidantes are connected to each other compared to all existing pairs, assuming no connection if a participant reported “have never spoken to each other.” Being a network bridging potential was determined by assessing whether there were any two confidantes who lacked direct and indirect connections to each other in the absence of the respondent (B. Cornwell, 2009), meaning the respondent was the only person who could create a link between the two confidantes who did not know each other.
Function.
Health-related decision support was measured in the NHAS confidante network roster. Participants were asked how likely they would talk with each person listed as a core confidante in the roster about a health problem supposing they were concerned about or needed to make an important decision about their own medical treatment. The response options were very likely, somewhat likely, and not likely. To address the presence of health decision support, the proportion of confidantes whom the respondent would very likely talk to about important health-related decisions was computed.
Quality.
Emotional closeness with their confidantes was assessed by asking participants how close their relationship was with each of their confidantes who were identified in the roster, on a 4-point Likert scale (1 = not very close; 4 = extremely close). Emotional closeness was summarized by computing the average closeness across confidantes.
Health Outcome
General health was measured with a single-item question from the Medical Outcomes Health Survey (Ware et al., 1994) assessing overall self-reported health, and the responses were dichotomized into good (=good, very good, or excellent) versus poor (=fair or poor). The single-item measure of general self-reported health has been widely used in population-based health research, such as Behavioral Risk Factor Surveillance System (Pierannunzi et al., 2013), and has been found to have good reliability and robust correlations with multi-item measures of self-reported health (DeSalvo et al., 2006; Thombs et al., 2008).
Background Characteristics
Age was calculated in years as the difference between birth years and 2014. Current gender was coded as women (=0), men (=1), and gender diverse (=2) who selected “not listed above.” Sexual identity was coded as gay or lesbian (=0), bisexual (=1), and sexually diverse (=2) who selected “not listed above.” Gender identity was coded as transgender (=1) and cisgender (=0). Socioeconomic status was assessed using education (high school or less vs. some college vs. college graduate or more) and household income (200% of the federal poverty level or below vs. higher than 200% federal poverty level; U.S. Department of Health and Human Services, 2021). Race/ethnicity was coded as non-Hispanic White, Hispanic, African American or Black, and other racial and ethnic minorities. The number of disabling chronic conditions (Brault, 2010; Freedman et al., 2007) was computed based on 12 conditions. Participants were asked whether they had ever been told by a doctor that they had the following conditions: heart attack, heart disease, high blood pressure, arthritis, osteoporosis, diabetes, lung condition (including asthma), stroke, back pain, dementia or Alzheimer’s, cancer, and HIV/AIDS.
Statistical Analysis
Analyses were conducted using STATA/MP (Version 16.1; Stata Corp, College Station, Texas). Survey weights were computed for the NHAS survey sample to reduce sampling bias from the nonprobability sample by applying a two-step postsurvey adjustment (Lee & Valliant, 2009) utilizing external probability sample data as benchmarks (see Fredriksen-Goldsen & Kim, 2017, for detailed information regarding study methods.). The survey weights were applied to the analyses of the measures included in the NHAS survey, and the in-person interview data were not weighted. First, we estimated descriptive statistics of the social connectedness measures for the SGM participants by gender (woman, man, and gender diverse), sexual identity among women (lesbian, bisexual, and sexually diverse), sexual identity among men (gay, bisexual, and sexually diverse), and gender identity (transgender and cisgender). To investigate statistical differences in social connectedness by gender, we estimated the margins for each group based on fitted ordinary least squares or logistic regression models of gender effect on the properties of social connectedness, as appropriate, controlling for age, which is known to be correlated with social connectedness among older adults (B. Cornwell et al., 2008), and then performed pairwise comparisons estimated by Bonferroni-adjusted Wald tests. The same approaches were applied to examine differences in sexual identity and gender identity. To examine the contributions of the properties of social connectedness to good general health, first, we examined the relationship between each social connectedness variable and general health by applying multiple variable logistic regressions controlling for age and the number of disabling chronic conditions. Second, backward stepwise selection procedures were used to examine the most important properties of social connectedness in predicting good general health, reducing collinearity issues, and increasing statistical power. The selection of the predictors for the final logistic model was implemented by removing the predictors whose significance was less than .2 in the full model, including all predictors, age, and the number of disabling chronic conditions.
Results
Properties of Social Connectedness by Gender, Sexual Identity, and Gender Identity in NHAS Survey
Table 2 displays the distributions of properties of interpersonal relations and the findings of age-adjusted comparisons by gender, sexual identity, and gender identity. The summary of the findings is presented below.
Table 2.
Distributions of Properties of Interpersonal Relations by Gender, Sexual Identity, and Gender Identity and the Results of Age-Adjusted Pairwise Comparisons: National Health, Aging, and Sexuality/Gender Study, 2014 Survey (N = 2,450)
| Sexual identity | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Gender | Women | Men | Gender identity | |||||||||||||
| Total | Women | Men | Gender diverse | Joint testa p | Lesbian | Bisexual | Sexually diverse | Joint testa p | Gay | Bisexual | Sexually diverse | Joint testa p | Transgender | Cisgender | Wald test p | |
| Variable | M (SE) | M (SE) | M (SE) | M (SE) | M (SE) | |||||||||||
| Structure | ||||||||||||||||
| Diversity | 3.7 (0.0) | 3.9 (0.1) | 3.5 (0.1)b | 3.8 (0.2) | <.001 | 4.1 (0.1) | 3.8 (0.2) | 3.3 (0.3)c | .025 | 3.4 (0.1) | 3.8 (0.2)d | 3.6 (0.4) | .053 | 3.7 (0.1) | 3.7 (0.0) | .746 |
| Network size | 8.2 (0.2) | 8.7 (0.3) | 7.9 (0.3) | 7.5 (0.8) | .107 | 9.0 (0.3) | 9.1 (0.9) | 6.3 (0.9)c,e | .018 | 7.8 (0.3) | 8.5 (0.9) | 6.6 (0.9) | .335 | 6.8 (0.5) | 8.5 (0.2) | .004 |
| Volume of contact | 349.8 (10.5) | 370.7 (15.7) | 335.0 (14.8) | 335.9 (45.6) | .255 | 377.8 (19.2) | 366.0 (32.8) | 329.4 (44.4) | .613 | 316.2 (13.4) | 380.7 (48.2) | 348.2 (83.2) | .457 | 330.8 (27.5) | 353.6 (11.3) | .378 |
| Partner/spouse, % (SE) | 51.0 (1.7) | 55.4 (2.5) | 46.8 (2.3)b | 56.8 (7.9) | .036 | 60.3 (2.8) | 50.2 (6.6) | 36.0 (8.5)c | .018 | 46.3 (2.4) | 48.4 (6.8) | 49.9 (10.5) | .916 | 48.1 (4.7) | 51.6 (1.8) | .399 |
| Ratio of nonrelative family to immediate family | 2.3 (0.1) | 2.3 (0.1) | 2.4 (0.1) | 1.7 (0.4) | .305 | 2.4 (0.2) | 2.5 (0.3) | 1.2 (0.2)c,e | <.001 | 2.5 (0.1) | 2.4 (0.4) | 1.5 (0.3)d,f | .001 | 1.9 (0.1) | 2.4 (0.2) | .048 |
| Function | ||||||||||||||||
| Instrumental support | 2.4 (0.0) | 2.5 (0.1) | 2.4 (0.1) | 2.4 (0.2) | .262 | 2.7 (0.1) | 2.3 (0.2) | 2.0 (0.3)c | .009 | 2.4 (0.1) | 2.3 (0.2) | 2.5 (0.3) | .910 | 2.3 (0.1) | 2.5 (0.1) | .302 |
| Emotional-informational support | 2.8 (0.0) | 3.0 (0.1) | 2.7 (0.1)b | 2.9 (0.2) | .011 | 3.1 (0.1) | 3.1 (0.1) | 2.2 (0.2)c,e | <.001 | 2.7 (0.1) | 2.7 (0.2) | 2.7 (0.3) | .986 | 2.6 (0.1) | 2.9 (0.0) | .029 |
| Positive social interaction support | 2.9 (0.0) | 3.0 (0.1) | 2.9 (0.1) | 2.7 (0.2) | .074 | 3.1 (0.1) | 3.0 (0.1) | 2.5 (0.2)c | .037 | 2.9 (0.1) | 2.9 (0.1) | 2.7 (0.3) | .733 | 2.7 (0.1) | 3.0 (0.0) | .036 |
| Affectionate support | 2.6 (0.0) | 2.7 (0.1) | 2.6 (0.1) | 2.5 (0.3) | .238 | 2.9 (0.1) | 2.7 (0.2) | 2.1 (0.3)c | .025 | 2.5 (0.1) | 2.9 (0.2) | 2.5 (0.3) | .136 | 2.3 (0.2) | 2.7 (0.1) | .022 |
| Caregiving, % (SE) | 32.0 (1.6) | 34.2 (2.5) | 28.3 (2.0) | 49.1 (8.1)g | .010 | 34.2 (2.8) | 34.0 (6.4) | 33.0 (7.9) | .969 | 27.7 (2.1) | 33.7 (6.4) | 23.0 (7.9) | .468 | 34.0 (4.5) | 31.6 (1.6) | .506 |
| Care receiving, % (SE) | 24.6 (1.4) | 25.2 (2.2) | 23.4 (1.9) | 27.7 (7.0) | .706 | 25.3 (2.5) | 25.1 (5.8) | 25.7 (7.3) | .998 | 21.5 (1.9) | 32.1 (6.2) | 30.0 (8.5) | .167 | 26.7 (4.0) | 24.2 (1.5) | .532 |
| Quality | ||||||||||||||||
| Satisfaction with personal relationships | 3.5 (0.0) | 3.6 (0.1) | 3.5 (0.1) | 3.3 (0.2) | .243 | 3.7 (0.1) | 3.6 (0.2) | 3.0 (0.2)c,e | .002 | 3.5 (0.1) | 3.4 (0.1) | 3.6 (0.3) | .932 | 3.3 (0.1) | 3.6 (0.0) | .009 |
| Satisfaction with support from friends | 3.6 (0.0) | 3.8 (0.1) | 3.5 (0.0)b | 3.5 (0.2) | .001 | 3.8 (0.1) | 3.8 (0.1) | 3.5 (0.2) | .268 | 3.6 (0.1) | 3.4 (0.1) | 3.5 (0.2) | .686 | 3.5 (0.1) | 3.7 (0.0) | .208 |
| Loneliness | 1.7 (0.0) | 1.6 (0.1) | 1.7 (0.1) | 1.9 (0.2) | .336 | 1.5 (0.1) | 1.8 (0.1) | 2.1 (0.2)c | .016 | 1.7 (0.1) | 2.0 (0.2) | 1.6 (0.2) | .445 | 1.9 (0.1) | 1.7 (0.0) | .050 |
Note. Survey weights were applied to data analyses. To examine differences in properties of interpersonal relations by gender, sexual identity among women, sexual identity among men, and gender identity, we implemented pairwise Wald tests with Bonferroni adjustments. These tests were performed by comparing the margins estimated from age-adjusted ordinary least squares or logistic regression models, as appropriate. SE = standard error.
Joint test denotes a Wald test for joint hypotheses for comparisons of more than two groups (i.e., comparisons by gender and sexual identity).
Results of pairwise Wald test: The difference from women was significant (p < .05).
Results of pairwise Wald test: The difference from lesbians was significant (p < .05).
Results of pairwise Wald test: The difference from gay men was significant (p < .05).
Results of pairwise Wald test: The difference from bisexual women was significant (p < .05).
Results of pairwise Wald test: The difference from bisexual men was significant (p < .05).
Results of pairwise Wald test: The difference from men was significant (p < .05).
Interpersonal Relations: Structure.
SGM older adults, on average, had nearly four different types of social ties, and their average network size was 8.2. About 10.7% have only one or zero people in their network. The total volume of contact across immediate family members, friends, ex-partners/spouses, and neighbors was 349.8 on average, indicating that SGM older adults communicated with their social ties almost every day. Nearly half were partnered or married. The average ratio of nonrelative family members to immediate family members in their close social ties was 2.3 meaning that, for every immediate family member that an SGM older adult had in their social network, they had over two other nonrelative family members.
We found that network diversity and relationship status were significantly associated with gender among SGM older adults, whereas network size, volume of contact, and the ratio of nonrelative family members to immediate family members were not. The pairwise tests with Bonferroni adjustment revealed that men reported a significantly lower level of network diversity (contrast = −0.4, p < .001) and a lower likelihood of having a partner or spouse (contrast = −0.1, p = .025) compared to women. The age-adjusted estimates of the structural properties of interpersonal relations for gender-diverse individuals did not significantly differ from those for women and men.
Among SGM women, sexual identity was significantly associated with all of the structural properties of interpersonal relations except for the volume of contact. Sexually diverse women had a lower network size (contrast = −2.7, p = .011; contrast = −2.8, p = .045, respectively) and a lower ratio of nonrelative family members to immediate family members (contrast = −1.3, p < .001; contrast = −1.3, p = .002, respectively) than lesbians and bisexual women and a lower level of network diversity (contrast = −0.7; p = .022) and a lower likelihood of having a partner or spouse (contrast = −0.2, p = .016) than lesbians. We did not observe any significant differences between lesbians and bisexual women.
Among SGM men, sexual identity was associated with network diversity at p = .053 and the ratio of nonrelative family members to immediate family members. Pairwise tests indicated that bisexual men had a higher network diversity (contrast = 0.4, p = .036) compared to gay men, but network diversity for sexually diverse men did not differ from those for gay and bisexual men. The ratio of nonrelative family members to immediate family members for sexually diverse men was smaller than those for gay (contrast = −1.1, p < .001) and bisexual men (contrast = −1.1, p = .039). The other structural properties of interpersonal relations were not associated with sexual identity among older SGM men.
Transgender older adults showed a smaller network size and a smaller ratio of nonrelative family members to immediate family members than cisgender older adults while the other structural properties did not differ by gender identity.
Interpersonal Relations: Function.
SGM older adults reported the availability of instrumental and affectionate support as sometimes to fairly often and emotional-informational and positive social interaction support as fairly often. A third of the participants were providing informal caregiving support, and a quarter were receiving informal care from their social network.
Significant gender differences in the availability of emotional-informational support and the provision of informal care were observed. Specifically, men reported less frequent availability of emotional-informational support (contrast = −0.3, p = .006) than women while gender-diverse persons did not differ significantly from either women or men. The likelihood of providing informal care for gender-diverse persons was significantly higher (contrast = 0.2, p = .017) than that for men, while there was no significant difference between women and men or between women and gender-diverse persons. No other functional properties of interpersonal relations were significantly associated with gender.
Among SGM women, the availabilities of the four types of social support were significantly associated with sexual identity, whereas caregiving and care receiving were not. According to pairwise tests, sexually diverse persons showed less frequent availabilities in instrumental (contrast = −0.7, p = .018), positive social interaction (contrast = −0.6, p = .024), and affectionate support (contrast = −0.8, p = .013) compared to lesbians, and less frequent emotional-informational support compared to both lesbians (contrast = −0.8, p < .001) and bisexual women (contrast = −0.8, p = .002). No significant differences between lesbians and bisexual women were observed. Among SGM men, sexual identity was not significantly associated with any functional properties of interpersonal relations.
In terms of gender identity, transgender older adults had significantly lower levels of emotional-informational (contrast = −0.3, p = .029), positive social interaction (contrast = −0.2, p = .036), and affectionate support (contrast = −0.4, p = .022). Gender identity was not significantly associated with instrumental support, caregiving, and care receiving.
Interpersonal Relations: Quality.
Participants in the survey indicated moderate to high levels of satisfaction with their personal relationships and support from their friends at a level ranging from “moderate” and “very much.” They also reported feeling lonely sometimes, on average. See Table 2.
Satisfaction with support from friends was significantly associated with gender, but satisfaction with personal relationships and loneliness were not. According to pairwise tests, men had a significantly lower level of satisfaction with support from their friends (contrast = −0.3, p = .001) than women. Gender-diverse individuals did not show any significant difference in satisfaction levels when compared to either men or women.
Sexual identity among SGM women was significantly associated with satisfaction with personal relationships and loneliness. Sexually diverse women reported a significantly lower level of satisfaction with personal relationships than both lesbians (contrast = −0.8, p = .001) and bisexual women (contrast = −0.6, p = .038) and a significantly higher level of loneliness (contrast = 0.6, p = .018) than lesbians. Sexual identity was not associated with satisfaction with support from friends among SGM women. Among SGM men, sexual identity was not significantly associated with any qualitative properties of interpersonal relations.
Transgender older adults showed a lower level of satisfaction with personal relationships (contrast = −0.3, p = .009) and a higher level of perceived loneliness (contrast = 0.2, p = .050) than cisgender older adults. Satisfaction with support from friends was not associated with gender identity.
Community Engagement.
As delineated in Table 3, SGM older adults reported participating in outdoor leisure activities on some days, on average, during the past month, while they reported low levels of participation in club or group activities, volunteer activities, and spiritual and religious activities. The mean score of SGM community engagement was 3.9, which is above a moderate level of engagement.
Table 3.
Distributions of Community Engagement by Gender, Sexual Identity, and Gender Identity and the Results of Age-Adjusted Pairwise Comparisons: National Health, Aging, and Sexuality/Gender Study, 2014 Survey (N = 2,450)
| Sexual identity | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Gender | Women | Men | Gender identity | |||||||||||||
| Total | Women | Men | Gender diverse | Joint test p | Lesbian | Bisexual | Sexually diverse | Joint test p | Gay | Bisexual | Sexually diverse | Joint test p | Transgender | Cisgender | Wald test p | |
| Variable | M (SE) | M (SE) | M (SE) | M (SE) | M (SE) | |||||||||||
| Club meetings or group activities | 1.4 (0.0) | 1.5 (0.0) | 1.4 (0.0) | 1.4 (0.2) | .705 | 1.4 (0.1) | 1.4 (0.1) | 1.7 (0.2) | .275 | 1.4 (0.0) | 1.5 (0.1) | 1.2 (0.2) | .482 | 1.5 (0.1) | 1.4 (0.0) | .713 |
| Outdoor leisure activities | 2.0 (0.0) | 2.1 (0.0) | 2.0 (0.0) | 2.0 (0.1) | .681 | 2.1 (0.0) | 2.0 (0.1) | 2.2 (0.2) | .820 | 2.1 (0.0) | 2.0 (0.1) | 1.7 (0.2) | .240 | 2.0 (0.0) | 2.1 (0.1) | .977 |
| Volunteer | 1.3 (0.0) | 1.3 (0.1) | 1.2 (0.0) | 1.5 (0.2) | .074 | 1.3 (0.1) | 1.3 (0.1) | 1.6 (0.2) | .286 | 1.2 (0.1) | 1.2 (0.2) | 1.3 (0.3) | .904 | 1.5 (0.1) | 1.2 (0.0) | .016 |
| Spiritual/religious activities | 1.0 (0.0) | 1.1 (0.0) | 1.0 (0.0) | 1.1 (0.2) | .242 | 1.1 (0.1) | 1.1 (0.1) | 1.2 (0.1) | .598 | 0.9 (0.0) | 1.2 (0.1) | 1.1 (0.2) | .072 | 1.1 (0.1) | 1.0 (0.0) | .303 |
| SGM community engagement | 3.9 (0.0) | 3.9 (0.1) | 3.8 (0.1) | 4.1 (0.2) | .426 | 3.9 (0.1) | 3.8 (0.2) | 4.4 (0.2)a | .047 | 3.9 (0.1) | 3.9 (0.1) | 3.2 (0.2)b,c | .009 | 4.1 (0.1) | 3.8 (0.0) | .036 |
Note. Survey weights were applied to data analyses. To examine differences in properties of community engagement by gender, sexual identity among women, sexual identity among men, and gender identity, we implemented pairwise Wald tests with Bonferroni adjustments. These tests were performed by comparing the margins estimated from age-adjusted ordinary least squares or logistic regression models, as appropriate. Joint test denotes a Wald test for joint hypotheses for comparisons of more than two groups (i.e., comparisons by gender and sexual identity). SE = standard error.
Results of pairwise Wald test: The difference from lesbians was significant (p < .05).
Results of pairwise Wald test: The difference from gay men was significant (p < .05).
Results of pairwise Wald test: The difference from bisexual men was significant (p < .05).
No significant gender differences were observed in the four types of community engagement activities (i.e., club or group activities, outdoor leisure activities, volunteer activities, and spiritual and religious activities) and SGM community engagement. SGM community engagement was significantly associated with sexual identity for both women and men, whereas the community engagement activities were not. Pairwise tests revealed that sexually diverse women had a higher level of SGM community engagement (contrast = 0.6, p = .030) than lesbians while bisexual women did not differ from either lesbians or sexually diverse women in that regard. Sexually diverse men showed a lower level of SGM community engagement than gay (contrast = −0.7, p = .005) and bisexual men (contrast = −0.7, p = .019) with no significant difference between gay and bisexual men.
As for gender identity, transgender older adults showed higher levels of volunteer activities (contrast = 0.3, p = .016) and SGM community engagement (contrast = 0.3, p = .036) compared to cisgender older adults, and no significant differences by gender identity were observed in the other community engagement activities.
Association Between Social Connectedness Properties and General Health in NHAS Survey
Table 4 demonstrates the findings from separate logistic regression models where each property of social connectedness was individually entered as a predictor of general health, controlling for age and the number of disabling chronic conditions as control variables. Most predictors, with a few exceptions, were positively and significantly associated with good general health. Care receiving and loneliness were negatively associated with good general health. The volume of contact, the ratio of nonrelative family members to immediate family members, and caregiving were not associated with general health.
Table 4.
Multiple Variable Logistic Regression of Good General Health on Properties of Social Connectedness: National Health, Aging, and Sexuality/Gender Study, 2014 Survey (N = 2,450)
| Individual logistic regressiona | Backward logistic regressionb | |||
|---|---|---|---|---|
| Variable | AOR [95% CI] | p | AOR [95% CI] | p |
| Interpersonal relations: structure | ||||
| Diversity | 1.44 [1.23, 1.68] | <.001 | 1.26 [1.03, 1.55] | .026 |
| Network size | 1.06 [1.02, 1.10] | .007 | 0.96 [0.92, 1.01] | .084 |
| Volume of contact | 1.00 [1.00, 1.00] | .220 | Removed | |
| Partner/spouse, % | 1.96 [1.39, 2.75] | <.001 | Removed | |
| Ratio: nonrelative family/immediate family | 1.04 [0.97, 1.12] | .303 | Removed | |
| Interpersonal relations: function | ||||
| Instrumental support | 1.17 [1.05, 1.30] | .006 | Removed | |
| Emotional-informational support | 1.31 [1.14, 1.51] | <.001 | 1.15 [0.93, 1.41] | .192 |
| Positive social interaction support | 1.57 [1.33, 1.86] | <.001 | Removed | |
| Affectionate support | 1.37 [1.21, 1.55] | <.001 | Removed | |
| Caregiving, % | 0.93 [0.66, 1.31] | .673 | Removed | |
| Care receiving, % | 0.39 [0.27, 0.55] | <.001 | 0.27 [0.18, 0.41] | <.001 |
| Interpersonal relations: quality | ||||
| Satisfaction with personal relationship | 1.56 [1.32, 1.84] | <.001 | Removed | |
| Satisfaction with support from friend | 1.64 [1.36, 1.98] | <.001 | Removed | |
| Loneliness | 0.61 [0.51, 0.72] | <.001 | 0.76 [0.61, 0.94] | .013 |
| Community engagement | ||||
| Club meetings or group activities | 1.55 [1.29, 1.87] | <.001 | 1.24 [0.98, 1.57] | .068 |
| Outdoor leisure activities | 2.06 [1.60, 2.65] | <.001 | 1.66 [1.25, 2.21] | .001 |
| Volunteer | 1.45 [1.22, 1.72] | <.001 | Removed | |
| Spiritual/religious activities | 1.21 [1.03, 1.43] | .020 | Removed | |
| SGM community engagement | 1.35 [1.18, 1.54] | <.001 | Removed | |
Note. Survey weights were applied to data analyses. SGM = sexual and gender minority; AOR = adjusted odds ratio; CI = confidence interval.
Separate logistic regression models for each property of social connectedness were tested after controlling for age and the number of disabling chronic conditions.
A backward stepwise logistic regression was performed after controlling for age and the number of disabling chronic conditions, and the predictors with a significance level less than .2 were removed from the model.
Table 4 also delineates the results of stepwise backward logistic regressions evaluating the most important predictors of general health after controlling for age and the number of disabling chronic conditions. In the final model, network diversity and outdoor leisure activities were significantly associated with a higher likelihood of good general health, and care receiving and loneliness were associated with a lower likelihood of good general health. Although social network size, emotional-informational support, and club meetings or group activities were selected for the final model, they did not remain significant.
Properties of Social Connectedness by Gender, Sexual Identity, and Gender Identity and Their Association With General Health in NHAS In-Person Interview
Table 5 presents the distributions of properties of confidant networks among NHAS in-person interview respondents and the findings of age-adjusted comparisons by gender, sexual identity, and gender identity. The average size of core confidantes was 4.0. The network density, on average, was .61, meaning that approximately 60% of all pairs of respondents’ confidantes knew each other. About 39% had network bridging potential that could connect core confidantes who might not be connected without the respondent. In terms of SGM identity homophily, the average proportion of confidantes with an SGM identity in their core confidante network was 50%. The average number of core confidantes who could provide health-related decision support was 2.5. The average emotional closeness to confidantes was 2.9, which can be interpreted as “very close.”
Table 5.
Distributions of Properties of Core Confidant Network by Gender, Sexual Identity, and Gender Identity and the Results of Age-Adjusted Pairwise Comparisons: National Health, Aging, and Sexuality/Gender Study, 2014 In-Person Interview (N = 318)
| Sexual identity | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Gender | Women | Men | Gender identity | |||||||||||||
| Total | Women | Men | Gender diverse | Joint test p | Lesbian | Bisexual | Sexually diverse | Joint test p | Gay | Bisexual | Sexually diverse | Joint test p | Transgender | Cisgender | Wald test p | |
| Variable | M (SE) | M (SE) | M (SE) | M (SE) | M (SE) | |||||||||||
| Network size | 4.0 (0.1) | 4.0 (0.1) | 4.1 (0.1) | 3.8 (0.7) | .915 | 4.0 (0.1) | 4.3 (0.4) | 3.3 (0.8) | .329 | 4.0 (0.1) | 4.4 (0.2) | 4.0 (0.6) | .348 | 3.7 (0.4) | 4.1 (0.1) | .306 |
| Density | 0.6 (0.0) | 0.6 (0.0) | 0.6 (0.0) | 0.5 (0.2) | .720 | 0.6 (0.0) | 0.5 (0.1) | 0.6 (0.2) | .218 | 0.6 (0.0) | 0.6 (0.1) | 0.5 (0.1) | .723 | 0.5 (0.1) | 0.6 (0.0) | .131 |
| Identity homophily | 0.5 (0.0) | 0.5 (0.0) | 0.6 (0.0) | 0.4 (0.1) | .299 | 0.5 (0.0) | 0.4 (0.1) | 0.4 (0.1) | .325 | 0.6 (0.0) | 0.4 (0.1)a | 0.7 (0.1) | .016 | 0.5 (0.1) | 0.5 (0.0) | .567 |
| Bridging potential, % (SE) | 39.1 (2.8) | 38.3 (4.5) | 38.6 (3.7) | 60.0 (21.9) | .584 | 37.0 (4.8) | 45.5 (15.0) | 50.0 (25.0) | .731 | 38.1 (4.0) | 40.9 (10.5) | 42.9 (18.7) | .939 | 58.8 (11.9) | 37.9 (2.90) | .078 |
| Health-related decision support, number | 2.5 (0.1) | 2.7 (0.1) | 2.4 (0.1) | 2.2 (0.7) | .184 | 2.8 (0.1) | 2.3 (0.4) | 1.8 (0.3) | .205 | 2.4 (0.1) | 2.5 (0.3) | 1.9 (0.5) | .657 | 2.4 (0.4) | 2.5 (0.1) | .582 |
| Emotional closeness | 2.9 (0.0) | 2.9 (0.1) | 2.8 (0.0) | 2.9 (0.2) | .221 | 3.0 (0.1) | 2.8 (0.2) | 2.8 (0.5) | .386 | 2.9 (0.0) | 2.8 (0.1) | 2.3 (0.3)a | .057 | 2.8 (0.2) | 2.9 (0.0) | .253 |
Note. To examine differences in properties of core confidant network by gender, sexual identity among women, sexual identity among men, and gender identity, we implemented pairwise Wald tests with Bonferroni adjustments. These tests were performed by comparing the margins estimated from age-adjusted ordinary least squares or logistic regression models, as appropriate. Joint test denotes a Wald test for joint hypotheses for comparisons of more than two groups (i.e., comparisons by gender and sexual identity). SE = standard error.
Results of pairwise Wald test: The difference from gay men was significant (p < .05).
In the analysis of the in-person interview data, none of the properties of the confidant network were associated with gender, sexual identity, and gender identity with two exceptions. SGM identity homophily for bisexual men was significantly lower than for gay men. The level of emotional closeness for sexually diverse men was significantly lower than gay men.
Good general health was associated with higher numbers of core confidantes providing health-related decision support (AOR = 1.29; 95% CI [1.05, 1.60]; p = .015), and the other properties were not significantly associated with general health.
Discussion
Despite the increased attention to the positive impact of social connectedness in later life in enhancing health and well-being, one of the foremost barriers to SGM aging and health research has been the dearth of in-depth research using tailored social connectedness measures. Incorporating the multidimensional aspects of social connectedness of SGM older adults in the NHAS self-administered survey and in-person interview provides the opportunity to fill this critical research gap among SGM adults aged 50 and older. To the best of our knowledge, this is one of the few studies to comprehensively examine interpersonal relations as well as community-level engagement using a large sample of demographically diverse SGM older adults across the United States, with the application of survey weights to reduce sampling bias.
Our findings suggest that, on average, SGM older adults have cultivated resourceful and positive social connectedness over their lifetime. However, within-group heterogeneity of social connectedness properties was observed by gender, sexual identity, and gender identity among SGM older adults, yielding the necessity of addressing the distinct needs of subgroups in SGM research and intervention efforts. Study findings on the relationship between social connectedness properties and general health also suggest that all facets (i.e., structure, function, and quality) of interpersonal relations as well as community-level social connectedness should be addressed in SGM health research. As posited by the HEPM, SGM older adults in this study have demonstrated remarkable resilience in fostering health-promoting social connectedness, despite experiences of marginalization such as rejection, discrimination, and victimization (Brennan-Ing et al., 2014; Rosenfeld et al., 2012). Social connectedness may be especially beneficial for SGM older adults in later life as escalating risks of multimorbidity and disability may increase the need for informal caregiving and social support (Fredriksen-Goldsen, Kim, Shui, et al., 2017; Metlife Mature Market Institute, 2010). These findings will serve as a foundation for further in-depth understanding of the distinct nature of social connectedness and its impact on health equity, particularly in SGM older adult populations.
Consistent with the previous studies (Orel, 2017), we found that the networks of SGM older adults in the NHAS, on average, are extensive and diverse, more connected to nonrelative family members than immediate family. Knauer (2016) contended that SGM older adults have built chosen families amid the challenging context of social exclusion. Chosen families primarily consist of social ties formed through friendship, providing various forms of support and informal care not available from their immediate family members. Indeed, informal caregiving and care receiving emerged as important functions of participants’ social connectedness in older age. Our findings suggest that positive social interaction, emotional-informational, affectionate, and instrumental support are often available within SGM older adults’ diverse social networks, as are opportunities for a larger community engagement. The prevalent practice of chosen families and SGM community engagement may be linked to shared historical experiences, such as the AIDS pandemic during the 1980s and early 1990s. In response to the escalating health care crisis and lack of social resources, SGM communities throughout the country mobilized to provide support to those who lacked adequate support from their biological families. As a result, reciprocal care within SGM communities has been formed as the cultural norm (Aronson, 1998). Such collective efforts toward seeking solidarity and social justice may have promoted personal growth, sense of belonging, and value sharing (Roberts & Christens, 2021; Szymanski et al., 2023). Still, it is crucial to explicate the distinct roles that social network members play in providing social support and to identify limitations and barriers to securing social support in later life. Chosen family members may face impediments when dealing with medical systems as services are typically designed for heterosexual partners or other biological family members (Muraco & Fredriksen-Goldsen, 2011). Also, parallel aging trajectories of chosen family members may limit the ability to attend to rising needs for social support and informal care (Knauer, 2016).
Whereas the NHAS participants expressed satisfaction with their personal relationships and support from their friends at a slightly better than moderate level, it is important to note many still reported feelings of loneliness. In addition, participants who reported a severely limited social network (e.g., over 10% of the NHAS participants having only one or zero people in their network) may be at considerable risk of loneliness. Loneliness is documented to be associated with the onset of chronic diseases including cardiovascular disease and depression (Courtin & Knapp, 2017) as well as mortality (Wang et al., 2023). However, the previous studies suggest that improved function of support networks and the quality of social support can alleviate loneliness, even in cases of small social networks (Kim & Fredriksen-Goldsen, 2016). Targeting improved function and quality, especially among SGM older adults with restricted networks, provides an encouraging opportunity for interventions that close the gap between desired and experienced social connection. Importantly, it remains critical to identify those who have no support network to link them with community services and support.
This study reinforces the HEPM’s emphasis on accounting for heterogeneity within groups when analyzing social connectedness among SGM older adults. Whereas SGM older adults have historically been portrayed as a homogeneous group, the NHAS illustrates the need to examine the multiple aspects of social connectedness by gender, sexual identity, and gender identity, among demographically diverse SGM older adults. The findings showed that among SGM older adults, men experience relatively weaker social connectedness in the structure, function, and quality of their interpersonal relations than women. Findings related to structural properties are consistent with the previous research; SGM men’s networks are documented to be smaller and less diverse than SGM women displaying lower likelihoods of having a partner or spouse, having children, and disclosing their sexual or gender identity to a neighbor (Erosheva et al., 2016). Results also indicate that SGM men have weaker social connectedness in the function and quality, in particular, availability of emotional support and satisfaction with support from friends. These factors are known to buffer the deleterious effect of stressors on health-promoting behaviors and health outcomes (Bryan et al., 2017; Lyons, 2016).
A dearth of research details the differences in social connectedness by sexual identity among SGM older adults, as the majority of extant research on SGM aging and health excludes those who are sexually diverse or do not identify as lesbian, gay, or bisexual. The present study begins to address this gap, finding that sexually diverse women reported weaker social connectedness across the dimensions of interpersonal relations than lesbians and bisexual women, and sexually diverse men were less engaged in SGM communities than gay and bisexual men. These findings highlight the importance of understanding sexually diverse women and men’s experience of and access to communities and resources, especially as the derivation of many community resources centered on the needs of lesbians and gay men and may still be primarily oriented toward these subgroups. A recent population-based study suggests that sexually diverse adults experience the greatest disparities in health care access (Fredriksen-Goldsen et al., 2022), which likely leads to poor health outcomes in later life. As social support and resources play a critical role in enhancing help-seeking behaviors and facilitating entry into formal health care systems (Hudson & Romanelli, 2020), it is crucial in the services and interventions to address the distinct challenges that sexually diverse older adults experience in acquiring support and building social connectedness.
Consistent with the prior research (Fredriksen-Goldsen, Cook-Daniels, et al., 2014), this study found that, among SGM older adults, transgender older adults are more likely to experience limited social support when compared to their cisgender counterparts. However, unlike earlier studies that observed a larger social network size among transgender older adults (Fredriksen-Goldsen, Cook-Daniels, et al., 2014), we observed that the network size of transgender older adults is smaller than that of cisgender older adults when only considering close ties as the measure of social network. Furthermore, the NHAS data revealed that satisfaction with personal relationships and the ratio of nonrelative family members to immediate family members are lower among transgender older adults, suggesting that they may lack intimate partners and close friends to share their experiences with and alleviate their feelings of loneliness. Despite disadvantages in interpersonal and close connectedness, transgender older adults demonstrate resilience in terms of engaging in more volunteer activities and other SGM community opportunities. For transgender communities, connection with others through shared historical contexts and experiences of identity development and actualization may engender a collective identity that contributes to a sense of belongingness (Barr et al., 2016). Further research is warranted to better understand why the potentially protective community-level connectedness does not translate to close interpersonal connectedness for this population.
Consistent with the convoy model, our findings showed that all aspects of the social connectedness properties were positively associated with general health. Social network size and availability of social support have been empirically proven to be predictors of health outcomes among SGM older adults (Fredriksen-Goldsen, Cook-Daniels, et al., 2014; Fredriksen-Goldsen et al., 2013). The findings also demonstrate the importance of other properties of interpersonal relations, including structure, function, and quality as well as the contribution of community engagement to fully understand the health-promoting role of social factors in the HEPM. The backward stepwise regression model of this study suggests that diversity (i.e., connections to various types of social ties) is likely more important than network size in promoting health. Diverse networks may permit access to expanded information and social resources. Results also suggest that interventions to reduce loneliness and promote outdoor leisure activities could be effective in improving the general health of SGM older adults. Care receiving was negatively associated with general health even after controlling for disabling chronic conditions. However, longitudinal research may clarify the temporal relationship between care receiving, health determinants, and general health.
The NHAS in-person network results from participants’ confidante network roster provide valuable insights into the relationships between social ties among SGM older adults. Interestingly, the network density seems to be relatively small, yet the likelihood of bridging potential for the NHAS sample is much larger when compared to a previous study of older adults in general (B. Cornwell et al., 2009), suggesting that many SGM older adults belong to multiple clusters that may not share common linkages. Some possible benefits of bridging include opportunities for accessing various types of resources and control over their utilization (Burt, 1995).
Another notable finding from the in-person interview data is that health-related decision support was significantly associated with general health. When having someone with whom they can openly share their health concerns, SGM older adults may have a greater chance to receive informal care support and access additional resources needed for medical decisions. While chosen family, primarily comprised of close friends, is a central segment of SGM older adults’ social support networks (Wardecker & Matsick, 2020), they may not have the legal authority to help support medical decision making for SGM older adults. To be sure, more than 30% of SGM older adults do not have a durable power of attorney for health care (Fredriksen-Goldsen et al., 2011). It is imperative to ensure SGM older adults establish advance directives and secure support for medical decision making.
Limitations
This foundational research has several limitations. First, the measures were based on self-report. There could have been recall error when respondents were asked to count or list their network ties (Brewer & Webster, 1999). Second, we did not measure negative aspects of social connectedness, such as relational strain. It would be beneficial in future research to examine the health consequences of negative aspects of social connectedness as studies show inconclusive results (Antonucci et al., 2014). Third, the cross-sectional nature of the current data limits the ability to examine temporal relationships between social connectedness and health outcomes. It also hinders conducting time-series analysis on policy changes that affect social connectedness of the contemporary cohort of SGM populations, such as state- and federal-level policies on the legalization of civil marriage among same-sex couples (Freedom to Marry, 2016). Fourth, although we applied survey weights to mitigate sampling bias, there remain certain hard-to-reach segments within the older SGM population that may not be adequately represented, particularly individuals who have experienced significant disconnection from their SGM peers and aging services. Thus, it is important to acknowledge that the findings of this study may not be entirely applicable to the broader and hard-to-reach diverse older SGM population. Fifth, it is crucial to acknowledge the limitations of self-administered surveys in comprehensively assessing diverse forms of intimate relationships and family structures among SGM older adults. As addressed by Reczek (2020), to gain a deeper understanding of the complexities of SGM family dynamics, further research is needed, delving more deeply into how individuals with various and intersecting SGM identities perceive and define their family connections. Additional research is also needed to better understand the characteristics of core confidantes in each of the disconnected clusters and what types of support and resources SGM older adults can access from specific social clusters. Sixth, the intersectionality of gender, sexual identity, and gender identity was not fully examined due to small cell sizes of some subgroups (e.g., comparisons by sexual identity among gender-diverse participants; comparisons by gender and sexual identity among transgender participants). Also, the study sample was predominantly non-Hispanic White, and the findings may not necessarily be extrapolated to other racial and ethnic SGM populations as studies suggest that some racial and ethnic minority individuals often access sources of support from their biological family or own racial and ethnic communities without disclosing their SGM identity (Gómez et al., 2005; Moradi et al., 2010). Such experiences may give rise to distinct patterns in social connectedness among different populations. Future studies oversampling hard-to-reach subgroups and applying innovative analytical approaches, such as excess intersectional disparity (Jackson et al., 2016), would expand our understanding of more diversities and their relationship to SGM social connectedness. Last, the mode of communication with social ties was also not addressed in this study. As the use of online communication is dramatically increasing among older adults (Zickuhr & Madden, 2012), we need more research to understand how it supports social connections among SGM aging communities.
Implications
Despite the limitations, the NHAS data and findings can provide valuable guidance for researchers seeking to delve into distinct features of social connectedness among SGM older adults and the intricate interplay of social connectedness properties in health promotion. These factors should be integral considerations in future studies investigating the influence of social connectedness on other health outcomes including cognitive impairment, multimorbidity, and disability, which have been found to be more prevalent among SGM older adults(Fredriksen-Goldsen et al., 2022; Fredriksen-Goldsen, Kim, Shui, et al., 2017).
Furthermore, the translation of this research to enhance the social connectedness of SGM older adults creates important opportunities in both practice and policy development and implementation (Fredriksen-Goldsen, Kim, McKenzie, et al., 2017). This study provides timely insights into much-needed support to enhance social connection in these communities that can be implemented moving forward. Several modifiable factors were identified in this study that are directly linked to the study’s findings regarding the structure, function, and quality of SGM interpersonal relations and the need for greater community engagement to improve the health and well-being of SGM older adults. Given the diversity in social connectedness among differing gender, sexual identity, and gender identity groups, tailored interventions addressing the elevated needs of these subgroups, particularly SGM men, sexually diverse women, and transgender people, are suggested. For instance, implementing strategies to expand and diversify support networks and cultivate emotional support among these subgroups can overcome the identified network limitations that potentially impact information and resource access, satisfaction, and ultimately health.
The importance of diversity within the structure of interpersonal relations, including both immediate family and other nonrelative family members has many programmatic and policy-related implications. To date, most programs serving older adults in need are based on an outdated assumption that those available to support older adults are primarily a spouse and/or adult children (Fredriksen-Goldsen, Shuman, et al., 2023). Yet, a more inclusive definition of family is needed. In our investigation of the function of interpersonal relations in the provision of informal care, it was evident that relying on traditional conceptions of family caregiving support (the care provided by a spouse and/or an adult child) did not reflect the experience of SGM older adults. These findings support the development of the first randomized controlled trial for SGM older adults living with dementia and a care partner incorporating the important role of other nonrelative chosen family members identified as a care support person in their lives (Fredriksen-Goldsen, Teri, et al., 2023). As we move forward, it is critical for programs and policies to allow and support SGM older adults’ definitions of their families and care networks.
Despite having built interpersonal networks over their lifetime, our findings indicate that some SGM older adults are at risk with low-quality social connectedness, social isolation, and diminished satisfaction with their social relationships. For example, geographic dispersion characterizes the social networks of many SGM older adults (Yarker & Buffel, 2022), threatening social connectedness through isolation and loneliness (Hoy-Ellis et al., 2016). To respond, multilevel policies, multimodal programs, and targeted outreach are needed. Expanded development of tailored social and community opportunities that leverage resources, ranging from online to senior housing communities, may attend to the unique social connectivity needs of SGM older adults. Another potential obstacle to enhancing community engagement among SGM older adults is ageism, which may be internalized or experienced at the community level. Wight et al. (2015), for example, found relatively high levels of internalized ageism among older gay men, which can accompany depressive symptoms. In addition, many SGM communities themselves offer few programs designed specifically for SGM older adults, further restricting their opportunities for social connection. Educational and human service programs within these communities need to provide specialized programs and services to create visibility for SGM older adults and illustrate their strengths and resilience. Health care access programs also need to both educate providers and support the diverse social connections in these communities.
Extant research also recommends the development of programs that promote engagement between SGM older adults and younger SGM generations. Accordingly, Aging With Pride translational programs create cross-generational, evidence-based solutions to building stronger communities (Fredriksen-Goldsen, 2016). The NHAS was designed with a high level of engagement with community partners (Fredriksen-Goldsen & Kim, 2017). This approach enhances the ability to apply such findings for community-based infrastructure development and policy development and implementation.
Conclusion
This research advances the knowledge of social connectedness of SGM older adults by assessing multidimensional aspects of interpersonal relations and community engagement guided by the HEPM (Fredriksen-Goldsen, Simoni, et al., 2014) and social connectedness models (Antonucci et al., 2014; B. Cornwell et al., 2009). Although SGM older adults experience marginalization, they also navigate and access resources. With the NHAS social connectedness data, researchers can examine the distinct nature of SGM older adults’ social relationships and community connectivity and identify modifiable factors that promote or impede health equity. Such factors can be addressed through program and policy development that alleviates the negative health consequences of ongoing social stressors.
Public Policy Relevance Statement.
Limited knowledge exists regarding the multidimensional aspects of social connectedness of sexual and gender minority older adults. This study delineates the distinct nature of their interpersonal relationships and community engagement and identifies social connectedness factors that can contribute to health equity. By addressing these factors, culturally tailored programs and policies could foster resilience and mitigate the negative health consequences of ongoing social stressors.
Acknowledgments
This work was supported by the National Institute on Aging of the National Institutes of Health (Grant R01 AG026526; Karen Fredriksen-Goldsen: principal investigator). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors have no conflicts of interest to disclose.
References
- Aldwin C, & Yancura L (2010). Effects of stress on health and aging: Two paradoxes. California Agriculture, 64(4), 183–188. 10.3733/ca.v064n04p183 [DOI] [Google Scholar]
- Ali T, Nilsson CJ, Weuve J, Rajan KB, & Mendes de Leon CF (2018). Effects of social network diversity on mortality, cognition and physical function in the elderly: A longitudinal analysis of the Chicago Health and Aging Project (CHAP). Journal of Epidemiology and Community Health, 72(11), 990–996. 10.1136/jech-2017-210236 [DOI] [PubMed] [Google Scholar]
- Antonucci TC, Ajrouch KJ, & Birditt KS (2014). The convoy model: Explaining social relations from a multidisciplinary perspective. The Gerontologist, 54(1), 82–92. 10.1093/geront/gnt118 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Antonucci TC, Birditt KS, & Akiyama H (2009). Convoys of social relations: An interdisciplinary approach. In Bengston VL, Gans D, Pulney NM, & Silverstein M (Eds.), Handbook of theories of aging (2nd ed., pp. 247–2620). Springer. [Google Scholar]
- Aronson J (1998). Lesbians giving and receiving care: Stretching conceptualizations of caring and community. Women’s Studies International Forum, 21(5), 505–519. 10.1016/S0277-5395(98)00065-X [DOI] [Google Scholar]
- Ballester E, Cornish MA, & Hanks MA (2021). Predicting relationship satisfaction in LGBQ + people using internalized stigma, outness, and concealment. Journal of GLBT Family Studies, 17(4), 356–370. 10.1080/1550428X.2021.1923104 [DOI] [Google Scholar]
- Barr SM, Budge SL, & Adelson JL (2016). Transgender community belongingness as a mediator between strength of transgender identity and well-being. Journal of Counseling Psychology, 63(1), 87–97. 10.1037/cou0000127 [DOI] [PubMed] [Google Scholar]
- Berkman LF, Glass T, Brissette I, & Seeman TE (2000). From social integration to health: Durkheim in the new millennium. Social Science & Medicine, 51(6), 843–857. 10.1016/S0277-9536(00)00065-4 [DOI] [PubMed] [Google Scholar]
- Bonomi AE, Patrick DL, Bushnell DM, & Martin M (2000). Validation of the United States’ version of the World Health Organization Quality of Life (WHOQOL) instrument. Journal of Clinical Epidemiology, 53(1), 1–12. 10.1016/S0895-4356(99)00123-7 [DOI] [PubMed] [Google Scholar]
- Bosnes I, Nordahl HM, Stordal E, Bosnes O, Myklebust TÅ, & Almkvist O (2019). Lifestyle predictors of successful aging: A 20-year prospective HUNT study. PLOS ONE, 14(7), Article e0219200. 10.1371/journal.pone.0219200 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Brault M (2010). Americans with disabilities: 2005. Current Population Reports, 70–117. [Google Scholar]
- Brennan-Ing M, Seidel L, Larson B, & Karpiak SE (2014). Social care networks and older LGBT adults: Challenges for the future. Journal of Homosexuality, 61(1), 21–52. 10.1080/00918369.2013.835235 [DOI] [PubMed] [Google Scholar]
- Brewer DD, & Webster CM (1999). Forgetting of friends and its effects on measuring friendship networks. Social Networks, 21(4), 361–373. 10.1016/S0378-8733(99)00018-0 [DOI] [Google Scholar]
- Bryan AEB, Kim H-J, & Fredriksen-Goldsen KI (2017). Factors associated with high-risk alcohol consumption among LGB older adults: The roles of gender, social support, perceived stress, discrimination, and stigma. The Gerontologist, 57(Suppl. 1), S95–S104. 10.1093/geront/gnw100 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Burt RS (1995). Social capital, structural holes and the entrepreneur. Revue Francaise De Sociologie, 36(4), 599–628. 10.2307/3322448 [DOI] [Google Scholar]
- Cacioppo JT, & Hawkley LC (2009). Perceived social isolation and cognition. Trends in Cognitive Sciences, 13(10), 447–454. 10.1016/j.tics.2009.06.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Canaday M (2011). The straight state. https://press.princeton.edu/books/paperback/9780691149936/the-straight-state
- Ceatha N, Mayock P, Campbell J, Noone C, & Browne K (2019). The power of recognition: A qualitative study of social connectedness and wellbeing through LGBT sporting, creative and social groups in Ireland. International Journal of Environmental Research and Public Health, 16(19), Article 3636. 10.3390/ijerph16193636 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chavis DM, & Pretty GMH (1999). Sense of community: Advances in measurement and application. Journal of Community Psychology, 27(6), 635–642. 10.1002/(SICI)1520-6629(199911)27:6<635::AID-JCOP1>3.0.CO;2-F [DOI] [Google Scholar]
- Choi NG, DiNitto DM, & Marti CN (2016). Social participation and self-rated health among older male veterans and non-veterans. Geriatrics & Gerontology International, 16(8), 920–927. 10.1111/ggi.12577 [DOI] [PubMed] [Google Scholar]
- Cohen S, & Wills TA (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98(2), 310–357. 10.1037/0033-2909.98.2.310 [DOI] [PubMed] [Google Scholar]
- Connidis IA, & Barnett AE (2018). Family ties and aging (3rd ed.). Sage Publications. [Google Scholar]
- Cornwell B (2009). Network bridging potential in later life: Life-course experiences and social network position. Journal of Aging and Health, 21(1), 129–154. 10.1177/0898264308328649 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cornwell B, Laumann EO, & Schumm LP (2008). The social connectedness of older adults: A national profile. American Sociological Review, 73(2), 185–203. 10.1177/000312240807300201 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cornwell B, Schumm LP, Laumann EO, & Graber J (2009). Social Networks in the NSHAP Study: Rationale, measurement, and preliminary findings. The Journals of Gerontology: Series B, 64(Suppl. 1), i47–i55. 10.1093/geronb/gbp042 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cornwell EY, & Waite LJ (2009). Social disconnectedness, perceived isolation, and health among older adults. Journal of Health and Social Behavior, 50(1), 31–48. 10.1177/002214650905000103 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Courtin E, & Knapp M (2017). Social isolation, loneliness and health in old age: A scoping review. Health & Social Care in the Community, 25(3), 799–812. 10.1111/hsc.12311 [DOI] [PubMed] [Google Scholar]
- DeSalvo KB, Fisher WP, Tran K, Bloser N, Merrill W, & Peabody J (2006). Assessing measurement properties of two single-item general health measures. Quality of Life Research, 15(2), 191–201. 10.1007/s11136-005-0887-2 [DOI] [PubMed] [Google Scholar]
- Douglas H, Georgiou A, & Westbrook J (2017). Social participation as an indicator of successful aging: An overview of concepts and their associations with health. Australian Health Review, 41(4), 455–462. 10.1071/AH16038 [DOI] [PubMed] [Google Scholar]
- Emlet CA, Shiu C, Kim H-J, & Fredriksen-Goldsen K (2017). Bouncing back: Resilience and mastery among HIV-positive older gay and bisexual men. The Gerontologist, 57(Suppl. 1), S40–S49. 10.1093/geront/gnw171 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Erosheva EA, Kim HJ, Emlet C, & Fredriksen-Goldsen KI (2016). Social networks of lesbian, gay, bisexual, and transgender older adults. Research on Aging, 38(1), 98–123. 10.1177/0164027515581859 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fincham FD, & Rogge R (2010). Understanding relationship quality: Theoretical challenges and new tools for assessment. Journal of Family Theory & Review, 2(4), 227–242. 10.1111/j.1756-2589.2010.00059.x [DOI] [Google Scholar]
- Fiori KL, Antonucci TC, & Cortina KS (2006). Social network typologies and mental health among older adults. The Journals of Gerontology: Series B, 61(1), 25–32. 10.1093/geronb/61.1.P25 [DOI] [PubMed] [Google Scholar]
- Flatt JD, Agimi Y, & Albert SM (2012). Homophily and health behavior in social networks of older adults. Family & Community Health, 35(4), 312–321. 10.1097/FCH.0b013e3182666650 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen Goldsen KI, Jen S, & Muraco A (2019). Iridescent life course: LGBTQ aging research and blueprint for the future—A systematic review. Gerontology, 65(3), 253–274. 10.1159/000493559 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen Goldsen KI, Romanelli M, Hoy-Ellis CP, & Jung H (2022). Health, economic and social disparities among transgender women, transgender men and transgender nonbinary adults: Results from a population-based study. Preventive Medicine, 156, Article 106988. 10.1016/j.ypmed.2022.106988 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen-Goldsen KI (2016). The future of LGBT+ aging: A blueprint for action in services, policies, and research. Generations, 40(2), 6–15. [PMC free article] [PubMed] [Google Scholar]
- Fredriksen-Goldsen KI, Cook-Daniels L, Kim H-J, Erosheva EA, Emlet CA, Hoy-Ellis CP, Goldsen J, & Muraco A (2014). Physical and mental health of transgender older adults: An at-risk and underserved population. The Gerontologist, 54(3), 488–500. 10.1093/geront/gnt021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen-Goldsen KI, Emlet CA, Kim H-J, Muraco A, Erosheva EA, Goldsen J, & Hoy-Ellis CP (2013). The physical and mental health of lesbian, gay male, and bisexual (LGB) older adults: The role of key health indicators and risk and protective factors. The Gerontologist, 53(4), 664–675. 10.1093/geront/gns123 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen-Goldsen KI, Jung H, Kim H-J, Petros R, & Emlet C (2022). Disparities in subjective cognitive impairment by sexual orientation and gender in a national population based study of U.S. adults, 2013–2018. Journal of Aging and Health, 34(4–5), 519–528. 10.1177/08982643211046466 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen-Goldsen KI, & Kim H-J (2017). The science of conducting research with LGBT older adults—An introduction to Aging with Pride: National Health, Aging, and Sexuality/Gender Study (NHAS). The Gerontologist, 57(Suppl. 1), S1–S14. 10.1093/geront/gnw212 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen-Goldsen KI, Kim H-J, Bryan AEB, Shiu C, & Emlet CA (2017). The cascading effects of marginalization and pathways of resilience in attaining good health among LGBT older adults. The Gerontologist, 57(Suppl. 1), S72–S83. 10.1093/geront/gnw170 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen-Goldsen KI, Kim H-J, Emlet CA, Muraco A, Erosheva EA, Hoy-Ellis CP, Goldsen J, & Petry H (2011). The aging and health report: Disparities and resilience among lesbian, gay, bisexual, and transgender older adults. Institute for Multigenerational Health. [Google Scholar]
- Fredriksen-Goldsen KI, Kim H-J, McKenzie GL, Krinsky L, & Emlet CA (2017). Plan of action for real-world translation of LGBTQ health and aging research. LGBT Health, 4(6), 384–388. 10.1089/lgbt.2017.0185 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen-Goldsen KI, Kim H-J, Shiu C, Goldsen J, & Emlet CA (2015). Successful aging among LGBT older adults: Physical and mental health-related quality of life by age group. The Gerontologist, 55(1), 154–168. 10.1093/geront/gnu081 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen-Goldsen KI, Kim H-J, Shui C, & Bryan AEB (2017). Chronic health conditions and key health indicators among lesbian, gay, and bisexual older US adults, 2013–2014. American Journal of Public Health, 107(8), 1332–1338. 10.2105/AJPH.2017.303922 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen-Goldsen KI, Shuman A, Jen S, & Jung HH (2023). Stress and resilience among sexual and gender diverse caregivers. Research on Aging, 45(9–10), 654–665. 10.1177/01640275231156191 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen-Goldsen KI, Simoni JM, Kim H-J, Lehavot K, Walters KL, Yang J, Hoy-Ellis CP, & Muraco A (2014). The health equity promotion model: Reconceptualization of lesbian, gay, bisexual, and transgender (LGBT) health disparities. American Journal of Orthopsychiatry, 84(6), 653–663. 10.1037/ort0000030 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fredriksen-Goldsen KI, Teri L, Kim H-J, La Fazia D, McKenzie G, Petros R, Jung HH, Jones BR, Brown C, & Emlet CA (2023). Design and development of the first randomized controlled trial of an intervention (IDEA) for sexual and gender minority older adults living with dementia and care partners. Contemporary Clinical Trials, 128, Article 107143. 10.1016/j.cct.2023.107143 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Freedman VA, Kasper JD, Cornman JC, Agree EM, Bandeen-Roche K, Mor V, Spillman BC, Wallace R, & Wolf DA (2011). Validation of new measures of disability and functioning in the National Health and Aging Trends Study. The Journals of Gerontology: Series A, Biological Sciences and Medical Sciences, 66(9), 1013–1021. 10.1093/gerona/glr087 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Freedman VA, Schoeni RF, Martin LG, & Cornman JC (2007). Chronic conditions and the decline in late-life disability. Demography, 44(3), 459–477. 10.1353/dem.2007.0026 [DOI] [PubMed] [Google Scholar]
- Freedom to Marry. (2016). Winning the Freedom to Marry nationwide: The inside story of a transformative campaign. https://www.freedomtomarry.org/pages/how-it-happened
- Frost DM, & Meyer IH (2012). Measuring community connectedness among diverse sexual minority populations. Journal of Sex Research, 49(1), 36–49. 10.1080/00224499.2011.565427 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gjesfjeld CD, Greeno CG, & Kim KH (2008). A confirmatory factor analysis of an abbreviated social support instrument: The MOS-SSS. Research on Social Work Practice, 18(3), 231–237. 10.1177/1049731507309830 [DOI] [Google Scholar]
- Gómez CA, Mason B, & Alvarado NJ (2005). Culture matters: The role of race and ethnicity in the sexual lives of HIV-positive gay and bisexual men. In Halkitis PN, Gómez CA, & Wolitski RJ (Eds.), HIV+ sex: The psychological and interpersonal dynamics of HIV-seropositive gay and bisexual men’s relationships (pp. 87–100). American Psychological Association. 10.1037/11109-006 [DOI] [Google Scholar]
- Holt-Lunstad J (2018). Why social relationships are important for physical health: A systems approach to understanding and modifying risk and protection. Annual Review of Psychology, 69(1), 437–458. 10.1146/annurev-psych-122216-011902 [DOI] [PubMed] [Google Scholar]
- Holt-Lunstad J, & Unchino BN (2015). Social support and health. In Glanz K, Rimer BK, & Viswanath K (Eds.), Health behavior: Theory, research, & practice (5th ed., pp. 183–204). Jossey-Bass. [Google Scholar]
- Horn AJ, & Wong YJ (2014). Fathering gay sons: A typology of fathering concerns and clinical recommendations. Professional Psychology, Research and Practice, 45(4), 247–257. 10.1037/a0037632 [DOI] [Google Scholar]
- Hoy-Ellis CP, Ator M, Kerr C, & Milford J (2016). Innovative approaches address aging and mental health needs in LGBTQ communities. Generations, 40(2), 56–62. [PMC free article] [PubMed] [Google Scholar]
- Hudson KD, & Romanelli M (2020). “We are powerful people”: Health-promoting strengths of LGBTQ communities of color. Qualitative Health Research, 30(8), 1156–1170. 10.1177/1049732319837572 [DOI] [PubMed] [Google Scholar]
- Hughes ME, Waite LJ, Hawkley LC, & Cacioppo JT (2004). A short scale for measuring loneliness in large surveys—Results from two population-based studies. Research on Aging, 26(6), 655–672. 10.1177/0164027504268574 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hurlbert JS, Haines VA, & Beggs JJ (2000). Core networks and tie activation: What kinds of routine networks allocate resources in nonroutine situations? American Sociological Review, 65(4), 598–618. 10.1177/000312240006500407 [DOI] [Google Scholar]
- Ibarra H (1992). Homophily and differential returns: Sex-differences in network structure and access in an advertising firm. Administrative Science Quarterly, 37(3), 422–447. 10.2307/2393451 [DOI] [Google Scholar]
- Institute of Medicine. (2011). The health of lesbian, gay, bisexual, and transgender people: Building a foundation for better understanding. The National Academies Press. [PubMed] [Google Scholar]
- Jackson JW, Williams DR, & VanderWeele TJ (2016). Disparities at the intersection of marginalized groups. Social Psychiatry and Psychiatric Epidemiology, 51(10), 1349–1359. 10.1007/s00127-016-1276-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jackson Levin N, Kattari SK, Piellusch EK, & Watson E (2020). “We just take care of each other”: Navigating ‘chosen family’ in the context of health, illness, and the mutual provision of care amongst queer and transgender young adults. International Journal of Environmental Research and Public Health, 17(19), Article E7346. 10.3390/ijerph17197346 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Katz-Wise SL, Rosario M, & Tsappis M (2016). Lesbian, gay, bisexual, and transgender youth and family acceptance. Pediatric Clinics of North America, 63(6), 1011–1025. 10.1016/j.pcl.2016.07.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kim H-J, & Fredriksen-Goldsen KI (2016). Living arrangement and loneliness among lesbian, gay, and bisexual older adults. The Gerontologist, 56(3), 548–558. 10.1093/geront/gnu083 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kim H-J, Fredriksen-Goldsen KI, Bryan AEB, & Muraco A (2017). Social network types and mental health among LGBT older adults. The Gerontologist, 57(Suppl. 1), S84–S94. 10.1093/geront/gnw169 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Klein A, & Golub SA (2016). Family rejection as a predictor of suicide attempts and substance misuse among transgender and gender non-conforming adults. LGBT Health, 3(3), 193–199. 10.1089/lgbt.2015.0111 [DOI] [PubMed] [Google Scholar]
- Knauer N (2016). LGBT older adults, chosen family, and caregiving. The Journal of Law and Religion, 31(2), 150–168. 10.1017/jlr.2016.23 [DOI] [Google Scholar]
- Latham K, & Clarke PJ (2018). Neighborhood disorder, perceived social cohesion, and social participation among older Americans: Findings from the National Health & Aging Trends Study. Journal of Aging and Health, 30(1), 3–26. 10.1177/0898264316665933 [DOI] [PubMed] [Google Scholar]
- LeBlanc AJ, & Frost DM (2020). Couple-level minority stress and mental health among people in same-sex relationships: Extending minority stress theory. Society and Mental Health, 10(3), 276–290. 10.1177/2156869319884724 [DOI] [Google Scholar]
- Lee S, & Valliant R (2009). Estimation for volunteer panel web surveys using propensity score adjustment and calibration adjustment. Sociological Methods & Research, 37(3), 319–343. 10.1177/0049124108329643 [DOI] [Google Scholar]
- Lin YJ, & Israel T (2012). Development and validation of a psychological sense of LGBT Community Scale. Journal of Community Psychology, 40(5), 573–587. 10.1002/jcop.21483 [DOI] [Google Scholar]
- Lubben JE, & Gironda M (2003). Centrality of social ties to the health and well-being of older adults. In Berkman B & Harootyan LK (Eds.), Social work and health care in an aging society: Education, policy, practice, and research (pp. 319–350). Springer. [Google Scholar]
- Lyons A (2016). Social support and the mental health of older gay men: Findings from a national community-based survey. Research on Aging, 38(2), 234–253. 10.1177/0164027515588996 [DOI] [PubMed] [Google Scholar]
- Matsuzaka S, Romanelli M, & Hudson KD (2021). “Render a service worthy of me”: A qualitative study of factors influencing access to LGBTQ-specific health services. SSM—Qualitative Research in Health, 1, Article 100019. 10.1016/j.ssmqr.2021.100019 [DOI] [Google Scholar]
- McLaughlin D, Vagenas D, Pachana NA, Begum N, & Dobson A (2010). Gender differences in social network size and satisfaction in adults in their 70s. Journal of Health Psychology, 15(5), 671–679. 10.1177/1359105310368177 [DOI] [PubMed] [Google Scholar]
- McPherson M, Smith-Lovin L, & Cook JM (2001). Birds of a feather: Homophily in social networks. Annual Review of Sociology, 27(1), 415–444. 10.1146/annurev.soc.27.1.415 [DOI] [Google Scholar]
- Metlife Mature Market Institute. (2010). Still out, still aging: The MetLife study of lesbian, gay, bisexual, and transgender baby boomers.
- Moradi B, Wiseman MC, DeBlaere C, Goodman MB, Sarkees A, Brewster ME, & Huang YP (2010). LGB of color and white individuals’ perceptions of heterosexist stigma, internalized homophobia, and outness: Comparisons of levels and links. The Counseling Psychologist, 38(3), 397–424. 10.1177/0011000009335263 [DOI] [Google Scholar]
- Muraco A, & Fredriksen-Goldsen K (2011). “That’s what friends do”: Informal caregiving for chronically ill midlife and older lesbian, gay, and bisexual adults. Journal of Social and Personal Relationships, 28(8), 1073–1092. 10.1177/0265407511402419 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Orel NA (2017). Families and support systems of LGBT elders. Annual Review of Gerontology & Geriatrics, 37(1), 89–109. 10.1891/0198-8794.37.89 [DOI] [Google Scholar]
- Perry BL, McConnell WR, Peng S, Roth AR, Coleman M, Manchella M, Roessler M, Francis H, Sheean H, & Apostolova LA (2022). Social networks and cognitive function: An evaluation of social bridging and bonding mechanisms. The Gerontologist, 62(6), 865–875. 10.1093/geront/gnab112 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pescosolido BA, Gardner CB, & Lubell KM (1998). How people get into mental health services: Stories of choice, coercion and “muddling through” from “first-timers.” Social Science & Medicine, 46(2), 275–286. 10.1016/S0277-9536(97)00160-3 [DOI] [PubMed] [Google Scholar]
- Pierannunzi C, Hu SS, & Balluz L (2013). A systematic review of publications assessing reliability and validity of the Behavioral Risk Factor Surveillance System (BRFSS), 2004–2011. BMC Medical Research Methodology, 13(1), Article 49. 10.1186/1471-2288-13-49 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pinquart M, & Sörensen S (2000). Influences of socioeconomic status, social network, and competence on subjective well-being in later life: A meta-analysis. Psychology and Aging, 15(2), 187–224. 10.1037/0882-7974.15.2.187 [DOI] [PubMed] [Google Scholar]
- Rafnsson SB, Shankar A, & Steptoe A (2015). Longitudinal influences of social network characteristics on subjective well-being of older adults: Findings from the ELSA study. Journal of Aging and Health, 27(5), 919–934. 10.1177/0898264315572111 [DOI] [PubMed] [Google Scholar]
- Reczek C (2020). Sexual- and gender-minority families: A 2010 to 2020 decade in review. Journal of Marriage and Family, 82(1), 300–325. 10.1111/jomf.12607 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Roberts LM, & Christens BD (2021). Pathways to well-being among LGBT adults: Sociopolitical involvement, family support, outness, and community connectedness with race/ethnicity as a moderator. American Journal of Community Psychology, 67(3–4), 405–418. 10.1002/ajcp.12482 [DOI] [PubMed] [Google Scholar]
- Romanelli M, Lu W, & Lindsey MA (2018). Examining mechanisms and moderators of the relationship between discriminatory health care encounters and attempted suicide among U.S. transgender help-seekers. Administration and Policy in Mental Health, 45(6), 831–849. 10.1007/s10488-018-0868-8 [DOI] [PubMed] [Google Scholar]
- Rosenfeld D, Bartlam B, & Smith RD (2012). Out of the closet and into the trenches: Gay male Baby Boomers, aging, and HIV/AIDS. The Gerontologist, 52(2), 255–264. 10.1093/geront/gnr138 [DOI] [PubMed] [Google Scholar]
- Sherbourne CD, & Stewart AL (1991). The MOS social support survey. Social Science & Medicine, 32(6), 705–714. 10.1016/0277-9536(91)90150-B [DOI] [PubMed] [Google Scholar]
- Shiu C, Muraco A, & Fredriksen-Goldsen K (2016). Invisible care: Friend and partner care among older lesbian, gay, bisexual, and transgender (LGBT) adults. Journal of the Society for Social Work and Research, 7(3), 527–546. 10.1086/687325 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Szymanski DM, Goates JD, & Strauss Swanson C (2023). LGBQ activism and positive psychological functioning: The roles of meaning, community connection, and coping. Psychology of Sexual Orientation and Gender Diversity, 10(1), 70–79. 10.1037/sgd0000499 [DOI] [Google Scholar]
- Tang F (2009). Late-life volunteering and trajectories of physical health. Journal of Applied Gerontology, 28(4), 524–533. 10.1177/0733464808327454 [DOI] [Google Scholar]
- Thoma MV, & Gee SLM (2019). Successful aging in individuals from less advantaged, marginalized, and stigmatized backgrounds. Clinical Psychology in Europe, 1(3), 1–12. 10.32872/cpe.v1i3.32578 [DOI] [Google Scholar]
- Thombs BD, Ziegelstein RC, Stewart DE, Abbey SE, Parakh K, & Grace SL (2008). Usefulness of a single-item general self-rated health question to predict mortality 12 months after an acute coronary syndrome. European Journal of Cardiovascular Prevention and Rehabilitation, 15(4), 479–481. 10.1097/HJR.0b013e328300b717 [DOI] [PubMed] [Google Scholar]
- U.S. Department of Health and Human Services. (2021). Prior HHS poverty guidelines and federal register references. https://aspe.hhs.gov/topics/poverty-economic-mobility/poverty-guidelines/prior-hhs-poverty-guidelines-federal-register-references
- Wang F, Gao Y, Han Z, Yu Y, Long Z, Jiang X, Wu Y, Pei B, Cao Y, Ye J, Wang M, & Zhao Y (2023). A systematic review and meta-analysis of 90 cohort studies of social isolation, loneliness and mortality. Nature Human Behaviour, 7(8), 1307–1319. 10.1038/s41562-023-01617-6 [DOI] [PubMed] [Google Scholar]
- Wardecker BM, & Matsick JL (2020). Families of choice and community connectedness: A brief guide to the social strengths of LGBTQ older adults. Journal of Gerontological Nursing, 46(2), 5–8. 10.3928/00989134-20200113-01 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ware JE, Kosinski M, & Keller SD (1994). SF-36 physical and mental health summary scales: A user’s manual. The Health Institute, New England Medical Center. [Google Scholar]
- Webster NJ, Antonucci TC, & Ajrouch KJ (2022). Linked lives and convoys of social relations. Advances in Life Course Research, 54, Article 100502. 10.1016/j.alcr.2022.100502 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Weststrate NM, & McLean KC (2010). The rise and fall of gay: A cultural-historical approach to gay identity development. Memory, 18(2), 225–240. 10.1080/09658210903153923 [DOI] [PubMed] [Google Scholar]
- Whitton SW, & Kuryluk AD (2014). Associations between relationship quality and depressive symptoms in same-sex couples. Journal of Family Psychology, 28(4), 571–576. 10.1037/fam0000011 [DOI] [PubMed] [Google Scholar]
- Wight RG, LeBlanc AJ, Meyer IH, & Harig FA (2015). Internalized gay ageism, mattering, and depressive symptoms among midlife and older gay-identified men. Social Science & Medicine, 147, 200–208. 10.1016/j.socscimed.2015.10.066 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yarker S, & Buffel T (2022). Involving socially excluded groups in age-friendly programs: The role of a spatial lens and co-production approaches. Journal of Aging & Social Policy, 34(2), 254–274. 10.1080/08959420.2022.2049571 [DOI] [PubMed] [Google Scholar]
- Zickuhr K, & Madden M (2012, June 6). Older adults and internet use. Pew Research Center: Internet, Science & Tech. https://www.pewresearch.org/internet/2012/06/06/older-adults-and-internet-use/ [Google Scholar]