Neurocognitive health in LGBTQIA+ older adults: Current state of research and recommendations Provisionally Accepted

Riccardo Manca^{1, 2*} Alexander Moreno^{3, 4, 5} Alessandra Nicoletti⁶ Neil J. Henderson⁷ Jason D. Flatt⁸

• ¹Department of Life Sciences, Brunel University London, United Kingdom
• ²Department of Medicine and Surgery, University of Parma, Italy
• ³Department of Psychology, Université of Montréal, Canada
• ⁴Centre de recherche de l’institut universitaire de gériatrie de Montréal, CIUSSS du Centre-Sud-de-l’Île-de-Montréal, Canada
• ⁵Notre-Dame Hospital, Centre intégré universitaire de santé et de services sociaux du Centre-Sud-de-l’Île-de-Montréal (CCSMTL), Canada
• ⁶Department of Medical, Surgical Sciences and Advanced Technologies G.F. Ingrassia, University of Catania, Italy
• ⁷Department of Social Work, University of Western Cape, South Africa
• ⁸Department of Social and Behavioral Health, School of Public Health, University of Nevada, Las Vegas, United States

Over the last decade, research on the health of sexual and gender minorities (SGM), i.e., lesbian, gay, bisexual, transgender, queer, and people with other sexual orientations and forms of gender expression, has been expanding to explore cognitive decline and dementia risk. This field still represents a small niche, especially when compared to research on sexual health and HIV. Indeed, data on neurological disorders in SGM populations are very scarce. For instance, a recent review of 348 neurological studies found 60 focused on cognition of SGM people, only 6 of which (10%) investigated cognitive health unrelated to HIV (1). A possible reason is due to dementia being an issue concerning primarily older people (2). The SGM older adult population is extremely small, although somehow difficult to quantify since most aging cohort studies collect no information about sexual orientation and gender identity (SOGI). The 2021 population census in England and Wales (3) has shown that people aged 65+ are far less likely to self-identify as SGM than people aged 16-24 (<0.6% vs 1-4%). Similar estimations were reported in other Western and primarily English-speaking countries (4)(5)(6). Although difficult to assess, recent public debates, especially regarding gender minorities, might have also limited research on some SGM groups (7).Compared with heterosexual and cisgender groups, SGM people show higher rates of subjective cognitive decline (SCD), i.e., self-reported perception of cognitive deterioration (8). Although SCD is considered to be a risk factor for dementia, only a proportion of people with such diagnosis may be experiencing preclinical symptoms of a neurodegenerative disease (9). Several factors unrelated to neurodegeneration may affect the subjective perception of cognitive decline (10). This may explain inconsistencies regarding differences in objective measures of cognitive health (e.g., performance on cognitive tests and dementia risk) between SGM and non-SGM groups (11)(12)(13)(14)(15)(16)(17)(18)(19). Nonetheless, genderdiverse older adults appear to be an SGM subgroup with worse dementia risk profiles than heterosexual cisgender people (20)(21)(22)(23)(24).As a result, it is currently not possible to draw definite conclusions on whether SGM people may have an increased risk of cognitive decline (Table 1). This knowledge gap persists even though several other health disparities (e.g., higher risk of stress, anxiety, depression, suicide, functional limitations, hypercholesterolemia, heart attack) have been extensively documented in SGM groups.These disparities have been linked to the impact of adverse social environments (25). Older SGM individuals are likely to have lived through negative social conditions in their youth, since A first area of weakness is the striking lack of neuroscientific research addressing cognitive health of SGM older adults (1). To date, evidence on increased dementia risk is scarce and etiology is usually undetermined (17,20,21,23), thus leaving unanswered questions on whether non-neurodegenerative conditions may contribute to cognitive decline. Investigations specific to Alzheimer's disease found 9 times higher risk (adjusted odds ratio = 8.95, 95% CI = 4.25-18.83) in transgender and gender diverse people (22), but not in older adults in same-sex relationships (11).SCD is the outcome measure most commonly studied for aging SGM groups, often operationalized as a binary variable based on self-rated worsening in memory and/or thinking abilities (27)(28)(29)(30).Although this approach may provide useful exploratory findings to guide future investigations, the As a result, vulnerability to decline in specific cognitive domains (e.g., semantic memory, visuospatial attention, and social cognition) has never been tested in SGM older adults.SCDOnly two studies compared grey matter alterations in older adults with dementia in same-sex and different-sex relationships (15,16). This could be due primarily to the lack of SOGI data on older adult samples. have been repeatedly exploited (13,14,(27)(28)(29)(30)(31)(34)(35)(36)(37)(38). Possible overlaps between samples across these studies with consequences on biased estimations of SOGI-related effects on cognitive outcome measures cannot be ruled out.Moreover, these databases might be affected by different types of bias, such as SOGI-related nonresponse (39), healthy volunteer (40), and survival biases (41), that may affect findings and their interpretations. These may have skewed sample compositions (e.g., mostly White, highly educated, healthier participants with greater social resources) explaining both inconsistent findings on dementia risk (i.e., either higher than or equivalent to non-SGM people) and better episodic memory performance of SGM people (15,36,42).
A high degree of methodological heterogeneity characterizes the definition of SGM groups. While most studies used self-identified sexual orientation (13,17,19,(27)(28)(29)(30)(31)(35)(36)(37)(38)43), some focused on either sexual attraction (42) or same-sex relationships (11,12,(14)(15)(16)18) due to lack of detailed SOGI data. Gender minorities have been identified by using either self-reported identity (21,24,44) or by selecting people with gender identity disorders who accessed gender-affirming care (20,22,23). Such differences limit the generalizability and comparability of findings. Since SOGI characteristics are multidimensional, it is possible that different measurement approaches may not equally detect health disparities in SGM populations. Indeed, some accounts are available on distinct associations between sexual orientation dimensions and both mental (45) and physical health (46), but not with cognitive outcome measures.
While most studies have attempted to quantify cognitive health differences between SGM and non-SGM people, little research has focused on hypothesis-driven factors driving potential disparities.Correro & Nelson (47) suggested that minority stress (48) could be a crucial risk factor for cognitivedecline, yet only one recent study investigated and found a negative impact of minority stress on fluid intelligence (but not on episodic memory and temporal orientation abilities) of non-heterosexual older adults (42). This is primarily due to the lack of data on risk/protective factors relevant to SGM populations in cohort studies and databases not designed for such purpose. Mental health conditions (e.g. anxiety and depression) are more common among the SGM individuals probably due to a greater burden of stigmatisation, victimisation, discrimination and barriers to accessing healthcare (49). Furthermore, more severe mental health issues in SGM people are consistently associated with higher risk of SCD (29,34,43), cognitive impairment (13), and grey matter loss (16). These findings suggest that some psycho-social risk factors are associated with worse cognitive health among SGM older adults. It is still unclear, instead, whether poorer mental health could have a differential impact on cognition of SGM and non-SGM older adults and, as a consequence, could lead to increased rates of dementia. By contrast, being married ( 14) and greater social support in lesbian women (19) seem to protect against cognitive decline, but replication of these findings is needed to draw conclusions on the extent of such effects.
Another issue is represented by the lack of diversity in quantitative investigations of SGM cognitive health, since all current studies have been carried out in Western countries (i.e., USA, Canada, and the UK) on samples comprising mostly White English-speaking people. Although these conditions favor comparability between studies, it also raises concerns regarding generalizability of findings across different socio-cultural and ethno-racial contexts. Moreover, it could lead to underestimating cognitive decline risk profiles specific to SGM subgroups, e.g., people of color and other ethno-racial minorities (28,34) for whom childhood sexual trauma seems to be a stronger predictor of SCD decline than for White SGM older adults (29). When ethnicity data are collected, people of color are systematically underrepresented in most of the studies published so far and often make up less than 15% of the sample. Such small sample sizes prevented clinically meaningful analyses (11,42) and led researchers to choose race/ethnicity as a covariate to control for, rather than investigating interaction/mediation effects with SOGI characteristics (12)(13)(14).
Considering the limitations highlighted above, we provide a few recommendations for researchers interested in investigating neurocognitive aging and decline in SGM groups.Boost neuroscientific investigations: more clinically informed studies favoring integrations of methods from clinical neuroscience and social sciences are encouraged to clarify in depth any neurocognitive health disparities between SGM and non-SGM older adults. Crucial aspects to address are the possible biological causes (i.e., dementia etiologies) driving such disparities and whether they may differ across SGM subgroups, such as transgender people who have a higher multiple sclerosis risk (50). Moreover, investigating diversified indices of self-reported and objective cognitive health and biomarkers (e.g., cerebrospinal fluid and blood) will provide useful insights to clinicians to tackle neurocognitive deterioration in SGM people.Improved and targeted studies: future studies should implement appropriate and inclusive practices for collection of SOGI data, e.g., self-identification, two-step questions for gender identity, culturally appropriate choices (26,51), that may boost opportunities to investigate health outcomes in SGM populations. Diversified study designs, such as case-control studies with smaller yet justified sample sizes could address specific questions around cognitive decline (e.g., comparing SCD vs cognitive deficits on neuropsychological tests) and brain health in SGM populations. More targeted hypothesisdriven investigations would also improve engagement and recruitment of SGM populations in clinical trials to address overlooked issues regarding treatment response in diverse samples (52).Impact of different SOGI dimensions: the operationalization of SOGI characteristics should be carefully considered, since comparing SGM subgroups identified using different strategies (e.g., based on self-identification, sexual behavior or medical diagnosis) may highlight unique risks for cognitive decline. Investigations with less represented SGM groups (e.g., asexual, bisexual, intersex, non-binary and ethno-racial minority people) should also be encouraged, since their health and experiences may be unique in terms of exposure to both biological and social factors. Eventually, this knowledge could provide useful evidence to train clinicians and researchers with the aim to improve research engagement of and provide much needed appropriate care to diverse SGM older adults (53,54).Testing hypothesis on risk/protective factors: although a few studies found that poor mental health may affect cognitive health in SGM groups, more investigations are needed to ascertain what factors (e.g., minority stressors) may explain heterogeneity in cognitive decline and dementia risk within SGM groups. Moreover, the impact of psycho-social factors (e.g., depression, socio-economic status, etc.) on both cognitive performance and brain metrics (e.g., volumes and functional connectivity) should be compared between SGM and non-SGM samples to clarify group-specific effects that could explain disparities. In fact, previous investigations found that depression was associated with worse cognitive performance equally in heterosexual and non-heterosexual older adults (42) and that psychological distress did not explain higher rates of SCD in SGM people (55). Future studies should also address the role of protective and resilience factors, either general (e.g., cognitive reserve (56))or specific for SGM older adults (e.g., social support (19) and identity affirmation (57)).Improve diversity of samples: addressing issues regarding intersectionality between SOGI and other socio-demographic determinants of health (58) will contribute to expanding this field, potentially in a global perspective. Cross-cultural investigations of SGM older adults living across diverse countries are needed to determine what factors may influence the risk of cognitive deficits and inform collaborative public health policies. Although this may not be feasible in nations where SGM people still face challenging social environments (e.g., incarceration, violence, and stigma), potential risks for cognitive health could be ascertained in SGM populations with a history of migration and former asylum seekers.
In this paper we have provided an overview of the current knowledge and controversies on neurocognitive health of older SGM people and some suggestions for addressing unanswered questions. Advancing our understanding of potential challenges faced by SGM older adults and the consequences of risk/protective factors on their cognition and brain health will prove a fundamental asset to remove established barriers between health-care users and clinicians (59) and to inform highquality standard care practices for diverse aging populations (60).