Utilization of Recommended Preventive Health Screenings Between Transgender and Cisgender Older Adults in Sexual and Gender Minority Communities

Study Design and Sample

This study is a secondary analysis of data collected in 2010 as part of the Caring and Aging with Pride Study, funded by the National Institutes of Health and the National Institute on Aging. The cross-sectional survey of 2560 LGBT midlife and older adults 50 years of age and older was conducted through collaborations with 11 community-based organizations across the United States. Because the survey aimed to understand the health, aging and well-being of LGBT midlife and older adults, numerous types of questions were included regarding health and health behaviors, including utilization of screening services and preventive care. Many of these health related items were adapted from standard items in nationally representative surveys, such as the Behavioral Risk Factor Surveillance System (BRFSS), culturally tailored to capture the unique healthcare needs in the lesbian, gay, bisexual, and transgender community. For more in-depth information regarding the study design and methodology, see Fredriksen-Goldsen, et al. (2011).

We identified transgender older adults using two items with a skip pattern: “Are you transgender? [yes/no].” ”Yes” responses to this question were followed with an additional question asking whether they were “Female to Male (FTM),” “Male to Female (MTF),” or “Other.” Because the sample size for the “other” category was small (n = 9), they were not included in the analyses. The total sample size used for the analysis reported here is 2514 including 165 transgender (n for transgender women = 104; no for transgender men = 61) and 2349 cisgender older adults. All study procedures were reviewed and approved by the University of Washington Institutional Review Board.

Measurement

Analytic Strategy

We computed all statistics with Stata/IC for Windows (Version 14). We ran Little’s Missing Completely at Random (MCAR) test to determine if there was an underlying pattern of systemically missing data, which would introduce a “lot of noise” in the data and would require complex statistical adjustments to minimize the potential bias in the data. As the MCAR rate was only 1.8%, we applied complete-case analyses, as opposed to multiple imputation or deletion. We initially ran descriptive statistics to understand characteristics of the sample. We then applied bivariate statistics to compare transgender and cisgender LGB subsamples on the key background demographic characteristics. We employed bivariate statistics (t-tests or χ² tests) to compare screening service utilizations between transgender and cisgender LGB adults aged 50 and older. We used logistic multivariable regressions to conduct significance tests of the differences in prevalence of screening service utilizations between two groups after controlling for background characteristics (age, race/ethnicity, income, education) with cisgender LGB participants as the reference group. We further applied the same statistical tests to compare screening service utilizations between transgender women and transgender men with the former as the reference group.

Disparities in Utilization of Preventive Screenings

Overall, we found that compared to their cisgender LGB peers, transgender participants were significantly less likely to have accessed four of the eight preventive health screenings within the recommended period: mammogram, Pap smear, osteoporosis test, and PSA test. This is not surprising as research comparing transgender older adults to their cisgender LGB counterparts consistently finds that transgender older adults fare worse across key outcomes (Fredriksen-Goldsen, Cook-Daniels, et al., 2013; Hoy-Ellis & Fredriksen-Goldsen, 2017). In turn, population-based research suggests that compared to heterosexuals aged 50 and older, LGB adults have worse outcomes in multiple areas, including lower rates of mammography for lesbian and bisexual women, and lower rates of HIV testing for bisexual men (Fredriksen-Goldsen et al., 2012). Compared to heterosexual women aged 50 and older, lesbian and bisexual women are significantly more likely to delay needed care (Wallace et al., 2011).

Mammography

Breast cancer is the most common cancer among people assigned female at birth worldwide, and is a leading cause of premature death in women. Although there has been current interest in estimating harms associated with early screening, there is consensus among the USPSTF and professional organizations in the United States that biennial screening should start at age 55 (U.S. Preventative Services Task Force, 2016). Mammography screening is associated with a reduction in breast cancer mortality across a range of study designs. While we need more research in this area, there are potentially special considerations for older transgender women and transgender men in regards to mammography screening. Specifically, for transgender men who have breast tissue—no surgery or only chest reconstruction—mammography recommendations are as for cisgender females. For transgender women, mammography is recommended in patients over age 50 with additional risk factors, particular with use of estrogen and progestin over 5 years (Sonnenblick et al., 2018). Why transgender adults age 50 and older get mammograms at less than half the rate of their cisgender counterparts is likely due to a number of reasons. One reason may be that more than one in four transgender adults have experienced discrimination in healthcare settings or been denied health insurance (Bradford et al., 2012). Experiencing such discrimination may decrease the likelihood of ongoing attempts to access healthcare, or withdrawal from the healthcare system entirely (Ceres et al., 2018). Furthermore, insurance may not cover some transgender-related medical concerns (American Medical Association, 2008). Transgender men may not be aware that they remain at risk for breast cancer, even if they have had chest reconstruction surgery.

Pap Smear

Cervical cancer deaths in the United States have decreased since the implementation of widespread cervical cancer screening in the form of Pap smears. Most cases of cervical cancer occur in women who have not been appropriately screened and there is convincing evidence that appropriate screening can reduce cervical cancer mortality (McGraw & Ferrante, 2014). There are potential special considerations for Pap Smears in transgender men based on gender-related surgical history. Specifically, following total hysterectomy - if prior history of high-grade cervical dysplasia and/or cervical cancer, annual Pap smears of the vaginal cuff (constructed by suturing together the top part of the vagina, close to where the cervix was previously located, usually following a complete hysterectomy) (Thompson et al., 2020). Pap smears are recommended until three normal tests are documented and then every two to three years thereafter. If ovaries were removed but uterus/cervix remains intact, it is recommended that Pap smear guidelines for natal females be followed (Weyers et al., 2021). As these guidelines demonstrate, clinical guidelines for transgender men getting Pap smears are complex. It is understandable that transgender older men might be less likely to access this important preventive screening. Beyond the already highlighted disparities in income and health insurance, lack of culturally sensitive care likely plays a role. In a large national study comparing transgender adults aged 50 and older to their cisgender LGB peers, transgender participants were significantly more likely to fear accessing healthcare services and to conceal their identity from providers (Fredriksen-Goldsen, Cook-Daniels, et al., 2013). The American Medical Association (2019) has made clear that healthcare recipients’ failure to disclose—and providers’ failure to ask—about sexual and gender minority identities increases the risk for substandard or even harmful healthcare services (2009). From a health equity perspective, lack of affirming healthcare and healthcare spaces (e.g., clinic waiting rooms) specific to breast and Pap smear screening can be a significant barrier to preventive screenings (National Academies of Sciences, Engineering and Medicine, 2020).

Osteoporosis Test

Osteoporosis in older adults is a major health concern and osteoporotic fractures, particularly hip fractures, are associated with chronic pain and disability, loss of independence, decreased quality of life, and increased mortality. There is evidence that bone measurement tests predict short-term risk for osteoporotic fractures in women and men (Mikolajewicz et al., 2020). While more research in the area is needed, older transgender women and transgender men have potentially special considerations obtaining a dual-energy X-ray absorptiometry (DEXA) scan due to the long-term effects of hormone use and gonadectomy (Dobrolińska et al., 2019). Particularly, for transgender women, screening can be considered for agonadal patients who have been off estrogen for over 5 years and for transgender men who have taken testosterone for over 5 years and who have not had oophorectomy (Dobrolińska et al., 2019). DEXA scans are recommended for individuals over age 50, or even earlier if additional risk factors are present. Unfortunately, the medical model rarely actually takes into account social factors in the etiology of disease, and consequently preventive screening recommendations. The Health Equity Promotion Model (HEPM) highlights how allostatic load—the net effect of “wear and tear” on the body—is an adverse health pathway that increases the risk for the development of chronic health conditions (Fredriksen-Goldsen, Simoni, et al., 2014). Allostatic load increases as a function of exposure to structural, institutional, interpersonal, and intrapersonal stigma. Transgender older adults are among the most marginalized social groups in the nation, and consequently are exposed to more types and degrees of discrimination, victimization, and exclusion. Thus, the HEPM would suggest that transgender older adults would have significantly increased allostatic load due to lifelong discrimination and marginalization and greatly benefit from indicated intervention measures to increase preventive screening utilization (Almeida et al., 2009).

A matched-case study of Black and White women aged 45 to 54-years old found that after controlling for background characteristics, Black women were on average seven years older biologically than White women, which the researchers attributed to the chronic stressors of racism in Black women’s lives (Geronimus et al., 2010). Similar evidence for accelerated aging associated with chronic depression has also been established (Wolkowitz et al., 2011), as has evidence that transgender adults may have among the highest rates of depression in the country (Fredriksen-Goldsen, Cook-Daniels, et al., 2013). Thus, not only are transgender older adults lagging woefully behind their cisgender peers in getting DEXA scans, evidence supports that they should be screened for osteoporosis years earlier than standard guidelines recommend.

Prostate-Specific Antigen Test

Older transgender individuals are also significantly less likely (39% vs. 74%) to have gotten a PSA test for prostate cancer than their cisgender peers. PSA tests may be even more problematic due to lack of knowledge about transgender healthcare issues among healthcare providers, as well as insurance coverage and other issues previously discussed. We would expect to see rates of PSA testing among transgender women as recommendations are universal in that anyone with a prostate gland (which is not removed in male-to-female gender-affirming surgery) be tested after age 50.

Early detection of prostate cancer is easier to treat and reduces the likelihood of potential metastization to other parts of the body. The risk of developing prostate cancer increases as a function of age, particularly after 50, as well as being African American (Mayo Foundation for Medical Education and Research [MFMER], 2021). Transgender older adults are among the most marginalized of populations and consequently, due to a multiplicity of intersectional social identities and positionalities experience “stress proliferation” (LeBlanc et al., 2015). Essentially, transgender older adults are at greater risk for exposure to more types and greater degrees of chronic stress, the consequences of which also accumulates across the life course (Fredriksen-Goldsen, Simoni, et al., 2014). Multiple studies indicate that long-term chronic stress can play a significant role in the development of cancerous tumors and promote cancer development (Dai et al., 2020).

While we would hope from a health equity perspective that rates of preventive screening utilization among transgender older adults would be equivalent to older adults in the general population, the reality is far different. As previously discussed, health disparities among transgender older adults have been consistently documented. Large community-based samples consistently find that transgender older adults experience significant social and economic deprivation and exposure to significant social determinants of health disparities (Fredriksen-Goldsen, Cook-Daniels, et al., 2013; James et al., 2016). Limited population-based samples mirror these findings. Analyses of Behavioral Risk Factor Surveillance System (BRFSS) data found that compared to the general cisgender adult population, gay and bisexual cisgender men were significantly more likely to meet PSA-screening guidelines, transgender women were significantly less likely to meet those guidelines (Ma et al., 2021). The authors note that findings such as this highlight the need for more research on the social determinants of transgender health disparities.

Interestingly, transgender and cisgender older adults did not differ significantly on utilization of annual physical exams, stool tests, HIV tests, or colonoscopies every 10 years. The lack of significant differences in utilization of these screenings may be easier to understand. Annual physical exams are typically part of the core of healthcare services across age groups, race/ethnicity, and sex/gender. As such, their “routineness” is built into healthcare systems and typically do not incur extra costs. Annual stool tests are not invasive, are relatively inexpensive, and can be done in the privacy and security of one’s own home. Similarities in rates of colonoscopy between transgender and cisgender participants may be a function of time and age. The USPSTF recommendations for colonoscopy are every 10 years for adults aged 50 to 75-years old, so there is much more “temporal room” for transgender and cisgender older adults to have received a recommended screening. Although we could not locate any peer-reviewed studies that specifically examined life expectancy by gender identity/expression, studies that have made comparisons on other variables—such as key health outcomes—overwhelmingly report transgender older adults as being significantly younger than their cisgender counterparts (see, for example, Downing & Przedworski, 2018; Fredriksen-Goldsen, Cook-Daniels, et al., 2013).

The confluence of cultural and historical factors that may also be intersecting with structural and institutional heterosexism may explain similar rates of HIV testing. Since the 1980s when the healthcare institution first began to recognize and understand it, HIV/AIDS in the U.S. continues to be characterized as a “gay disease.” Generally, members of LGBT communities, regardless of sexual orientation or gender identity/expression have long recognized their increased risk and acted accordingly, for example, widespread, routine HIV testing. At the same time, recognizing the potential for pandemic resurgence, HIV testing and care is widely available and typically free or low-cost, an important factor for those experiencing economic marginalization. Many HIV service providers are also located in community settings, and are typically more affirming compared to traditional medical settings. Concurrently, gay and transgender identities in medical settings are synonymous with HIV, with HIV screening questions routinely embedded in history taking when a non-heterosexual and or transgender identity are known.

Implications for Practice

Our findings point to important practice implications. One challenge for practitioners is a consequence of the near hyper-specialization of healthcare provision in the United States today. While it is laudable that there are increasing numbers of healthcare providers that offer affirming care to transgender individuals (National Academies of Sciences, Engineering and Medicine, 2020), such care typically focuses on medical interventions intended to make bodies more congruent with identities (e.g., hormone replacement therapy, mastectomy, and genital reconstruction). Medical risks related to aging and knowledge of indicated preventive screenings among affirming care providers may be limited or even absent. Conversely, providers who specialize, for example, in geriatrics are likely to have limited or even a complete lack of knowledge of transgender bodies. Understanding why transgender and cisgender older adults do not differ significantly on utilization of annual physical exams, stool tests, HIV tests, or colonoscopies could provide insights that healthcare providers and stakeholders, regardless of specialization or setting, might use to increase utilization of other preventive screenings.

There are clear implications for transgender older adults who postpone or avoid treatment: delayed care can mean that preventable and treatable illnesses are not appropriately diagnosed in a timely manner, and chronic health conditions worsen. This last point speaks to perhaps the most striking implication of our analyses; readily available protocols detailing the specific concerns of a transgender older adult population in relation to preventative health screening recommendations do not exist, or provide conflicting information. Difficulty in being able to easily locate reliable and understandable health literature can be a major barrier to accessing preventative health screening recommendations. Because transgender older adults potentially have a unique medical history of long-term hormone use and may have a unique surgical history including, but not limited to, chest reconstruction and gonadectomy, it is important that care providers understand the specific health risks accounted for in this population. Without understanding how to accurately assess risk, providers will not be able to interpret existing national guidelines for preventative health screening in the general older adult population. In addition to obtaining education specific to transgender health and well-being, healthcare providers and professional bodies should collaborate with other knowledgeable providers, for example, UCSF’s Center of Excellence for Transgender Health (CETH), as well as older members of the transgender community to develop and validate a set of readily available preventative health screening recommendations specific to transgender midlife and older adults. The CETH mission is to advance health equity for transgender and gender non-binary communities via education and training, capacity building, and ongoing research that is community driven. Medical care settings and programs must integrate transgender sensitive care into their professional standards and develop trainings to ensure competencies in these areas. These are some important ways that healthcare environments can be modified to encourage and support culturally relevant preventative screenings in these communities.

Comparing Transgender Women and Transgender Men

Among transgender participants, after controlling for background characteristics, transgender men were significantly less likely than transgender women to have had a stool test, or a colonoscopy. On one hand, we might not expect differences in utilization of these preventive screenings as both seek to detect colorectal cancers, which are not sex-specific. On the other hand, there may be a sociocultural reason for the difference. Research on gender differences in perceptions of stool tests’ efficacy and ease of use suggest that women tend to perceive colorectal cancers as a “men’s disease” (Friedemann-Sánchez et al., 2007). Transgender women have most often been socialized as boys and/or men for some portion of their lives; for those aged 50 and older this may have been a significant portion of their lives. The opposite is true for transgender men. Transgender women and men’s cultural perceptions of relative risk of colorectal cancers may be more reflective of heterosexist socialization, irrespective of gender identity. Another possibility may be differential self-stigma related to transgender bodies, Transgender men in particular may be more likely to engage in unhealthy behaviors in attempts to camouflage their bodies from too close of scrutiny (Fredriksen-Goldsen et al., 2019)—for example, intentionally carrying extra weight to lessen visibility of breasts or pelvic configuration (Center of Excellence for Transgender Health, 2014). Considering actual and feared discrimination in healthcare settings in the context of colonoscopy procedures (e.g., knowing that one’s ano-genital area will be fully exposed while one is sedated) may arouse intense anxiety such that transgender men may perceive foregoing preventive screening procedures as the preferable alternative.

Implications for Practice

Beyond knowledge of sex-linked differences among the transgender population in general, healthcare and other care providers should have education in psychosocial differences between transgender women and men. At the individual level, assessment should also include an understanding of the timing and trajectory of gender transitions across the life course. Such information may provide insights into degrees and duration of earlier life socialization around gender, and how that may drive perceived risks in the present.

Overall, we found that older transgender women and men were significantly less likely to receive breast, cervical, and prostate cancer screenings and osteoporosis screenings compared to cisgender LGB older adults. Our results also suggest that compared to older transgender men, transgender women are significantly more likely to utilize colorectal cancer screenings. As this is an exploratory study, and there is so little other extant research in this area, we have more questions than answers. It may be that many of the answers to our questions regarding differential preventive screening utilization may be more psychosocial and sociocultural in nature. Indeed, the National Academies of Sciences, Engineering, and Medicine (2020) have recently noted a dearth of research on social determinants of transgender health beyond already identified factors (e.g., discrimination, rejection, economic marginalization, lack of access to culturally responsive providers). These drive ancillary

… Social needs such as trauma, housing insecurity, financial strain, and social isolation, particularly among groups such as … older adults. It is important for researchers, care providers, and policy makers to develop and evaluate targeted efforts to address social determinants of health and meet social needs for SGD people (National Academies of Sciences, Engineering & Medicine, 2020, p. 357).

Implications for Research

Moving forward, there are important areas of inquiry to be addressed. We know that transgender women and men share common experiences and perceptions of healthcare inequities, from discrimination (actual and feared), to lack of providers’ knowledge of trans-specific healthcare issues and needs. What we need to know is transgender individuals’ perceptions of a) healthcare literacy in general of the benefits and risks of preventive screenings, and b) perceived barriers to full utilization of such preventive screenings. While these may be similar to healthcare access in general, there may be unique differences. For example, it may be that transgender individual’s perception of immediate needs far outweigh the potential benefits of preventive screenings predicated on future health outcomes. Moving forward we must also address the need for research with transgender non-binary older adults. We do have some non-binary older adults in our sample, but too few for analyses. Just as sexual and gender minority (SGM) populations overall are expanding, so too are those who identify as non-binary, as evidenced by the National Institutes of Health recent expansion of the definition of SGM (National Institutes of Health, 2019). Similarly, we need to gather data from physicians and other care providers’ perceptions and knowledge of provision of preventive screenings for the transgender population. This will be further complicated by the current dominant discourse regarding age and aging.

America is ageist, both in the larger cultural context (Allen et al., 2021) and within transgender communities. Ageism may increase the risk for alienation—a deeply held sense of lack of belongingness and disconnection from community (Estes & DiCarlo, 2019). The metaphorical spotlight (and concurrent resources) overwhelmingly focuses on younger transgender populations, at the expense of those who are midlife and older. This is a disservice to both age groups, as the life course perspective demonstrates that early life experiences, whether for good or for ill accumulate over time to find amplified expression later in life. Equally important is for researchers to recognize and attend to the within-group heterogeneity of transgender midlife and older adults, which includes assessing similarities and differences between transgender women and men, as well as factors such as race/ethnicity, social positionalities and intersectionality, as well as (but not limited to) historic and ongoing social determinants of health and healthcare utilization. The Healthy People 2030 national initiative to reduce health disparities explicitly supports this recommendation in its call for all states and U.S. territories to incorporate sexual orientation and gender identity questions in their respective Behavioral Risk Factor Surveillance System (BRFSS) surveys (U.S. Department of Health & Human Services, 2020c). The Healthy People 2030 initiative also calls for an “increase [in] the number of nationally representative, population-based surveys that collect data on (or for) lesbian, gay and bisexual populations” (U.S. Department of Health & Human Services, 2020a), and “…on (or for) transgender populations” (U.S. Department of Health & Human Services, 2020b).

Similarly, from an equity perspective it will be extremely important to include transgender individuals, communities, agencies, and programs in the research endeavor. Several frameworks would foster and support such research. For example, community-based participatory research (CBPR) seeks to involve community stakeholders as equal partners in research endeavors through collective, reflective and systematic investigations, such as Manson and colleagues (2004) research with the Eastern Band Cherokee Indians. Participatory action research (PAR) is an approach that is typically single-issue oriented, yet iterative, wherein community stakeholders in partnership with researchers collaborate to address a specific issue, in context, and coupled with critical consciousness engage social change to take equitable action in response to that issue. PAR holds promise in tackling issues that many older adults face, yet is rarely employed with this population (Corrado et al., 2020). Another framework for critically informed, action-oriented research is the plan–do–study–act (PDSA) approach. PDSA attempts to assess the effectiveness of interventions in healthcare settings, and recognizes the centrality of local context in improving quality of constituent systems, such as preventive screenings. As such, PDSA is a cyclical process that builds upon each iteration to foster change in healthcare delivery embedded within the context of social systems (Taylor et al., 2014).

Limitations

Although we were unable to include geopolitical region in this study, the National Academies identifies this as a priority research area (Health and Medicine Division, 2011). In addition to the preliminary, exploratory nature of our study, there are other limitations. The data is cross-sectional and from a community-based survey, so findings cannot be generalized. Although there were fewer transgender men than transgender women in our study, this is likely reflective of the transgender population overall as prevalence data suggests there are fewer transgender individuals who identify as men versus those who identify as women (Zucker, 2017). Furthermore, we collected the data for this secondary analysis just over a decade ago; rates of change at multiple levels across multiple domains (e.g., age of first awareness of gender identity [and sexual orientation], age of initiation of affirming therapy) have been increasing rapidly over the last two decades.