Differences in Cardiovascular Health at the Intersection of Race, Ethnicity, and Sexual Identity

This cross-sectional study uses National Health and Nutrition Examination Survey data to examine differences in cardiovascular health metrics at the intersection of race, ethnicity, and sexual identity.


Introduction
Disparities in cardiovascular health (CVH) exist across race and ethnicity categories and in sexual minority (SM; ie, lesbian, gay, and bisexual) adults, but the intersectional effect of multiple marginalized experiences is poorly understood. 1 Intersectionality is a framework describing how social categories (ie, race, sexual identity) interconnect to create different experiences of discrimination and privilege 2 and can affect CVH.For example, a 2017 study found that Black SM women had 3 times the prevalence of self-reported stroke compared with Black heterosexual women and 4.5 times the prevalence of self-reported stroke compared with White heterosexual women. 3Understanding the effect of multiple marginalized experiences on CVH enables development of tailored interventions to better identify and treat those most at risk.The American Heart Association's (AHA's) measure of ideal CVH, Life's Essential 8, is composed of 4 health behaviors (diet, physical activity, nicotine exposure, and sleep) and 4 health factors (body weight, blood lipid levels, blood glucose level, and blood pressure) 4 that have been associated with cardiovascular disease. 5A 2023 study using 2007-2016 National Health and Nutrition Examination Survey (NHANES) data found that bisexual female adults had lower CVH scores compared with heterosexual female adults. 6The objective of the current study was to investigate the association of sexual identity, race, and ethnicity with CVH.Of note, the term sexual identity is used in this article to mirror the language used for this categorization in NHANES; however, we acknowledge that it is not how one identifies that drives inequity but rather how one is categorized in the context of societal power structures.

Methods
This cross-sectional study, conducted from July 27 to September 6, 2023, was deemed exempt and informed consent was waived by the University of California San Francisco institutional review board because we used publicly available data with no attempt at contacting or identifying deidentified participants.The study was reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
We conducted a cross-sectional analysis of publicly available NHANES data from 2007 to 2016. 7xual identity data are not publicly available after 2016, limiting our ability to use more contemporaneous data in our analysis.Full details of the methods were previously published. 6We used self-reported questionnaire, physical examination, and dietary recall data to investigate sexual identity differences in CVH scores across racial and ethnic identities among noninstitutionalized, nonpregnant adults (aged 18-59 years) with no history of cardiovascular disease (heart attack, stroke, coronary heart disease, or heart failure) who had complete data for sexual identity, race, ethnicity, and CVH metrics.Participants self-identifying as lesbian, gay, bisexual, or "something else" were categorized as SM due to limited sample size.Self-reported race and ethnicity were categorized as non-Hispanic Black (hereafter, Black), Hispanic, non-Hispanic White (hereafter, White), and other (Asian, multiracial, or any other race and ethnicity).Following AHA recommendations, respondents received a score of 0 to 100 for each CVH metric, and the overall CVH score was calculated as the unweighted mean of these 8 components, with higher scores corresponding to better CVH. 4

Statistical Analysis
NHANES complex survey weights were applied in all analyses.We used an intercategorical approach to the analysis. 8This approach was chosen given the hypothesis that those who are marginalized based on race and ethnicity, sexual identity, or both experience barriers to CVH equity compared with White heterosexual individuals.Multiple imputation with chained equations was used for the approximately 5% of respondents with missing demographic data.For univariate analyses, t tests and χ 2 tests were used to compare SM participants across racial and ethnic identities with heterosexual adults.Sex-and race and ethnicity-stratified linear regression models were used to examine differences in CVH metrics and overall CVH score for SM adults across racial and ethnic identities, adjusting for age, survey year, and socioeconomic status (SES) factors (poverty-to-income ratio, educational level, health insurance coverage, and routine place for health care).We also performed sex-stratified linear regression models to examine differences in overall CVH, with White heterosexual adults as the reference category.We used nonstratified linear regression models with an interaction term between race and ethnicity and sexual identity to compare differences in CVH for SM adults of Black, Hispanic, or other race and ethnicity with their heterosexual counterparts.Given that sexual identity data may be missing not at random and therefore multiple imputation of these data may introduce further bias, 9 we also performed a sensitivity analysis comparing the mean overall CVH score between those with complete sexual identity data and those without sexual identity data across race and ethnicity categories.
Statistical significance was set at 2-sided P < .05for both univariate and multivariable analyses.

Discussion
In a nationally representative sample, we found that Black, Hispanic, and White SM female adults had lower overall CVH scores compared with their heterosexual counterparts.Black SM female individuals also had lower overall CVH scores compared with White heterosexual female individuals; there was no association with overall CVH for Hispanic SM female individuals, although this may have been due to limited power from the small sample size.We found no difference in overall CVH in SM male adults across racial and ethnic identities.In the fully adjusted models, overall CVH scores for Black SM female adults were 3.2 points lower compared with Black heterosexual female adults and 5.7 points lower compared with White heterosexual female adults.While the absolute difference between these results is small, the significant disparity when looking across both sexual identity and race and ethnicity categories is consistent with our hypothesis and with the tenets of intersectionality.
Additionally, there were differences in individual CVH metrics; for example, Black and White SM female individuals had less favorable nicotine exposure scores, while Hispanic SM female individuals had less favorable BMI and blood pressure scores compared with heterosexual female individuals of the same race and ethnicity.These differences in CVH metrics suggest the importance of withingroup investigations to assess the association of multiple marginalized experiences with health outcomes and is consistent with trends described in the current literature.Sexual minority individuals in marginalized racial and ethnic groups have higher odds of substance use disorder, 12,13 obesity, 14 and hypertension compared with White SM people. 15The etiology of these disparities is likely multifactorial, including disparate experiences of health care discrimination. 16

Limitations
This study has several limitations.The cross-sectional design of NHANES limits causal inference.
NHANES does not collect gender identity or expression, limiting our ability to identify transgender, Other race and ethnicity includes those who identified as Asian, multiracial, or any race and ethnicity other than Black, Hispanic, or White.The sexual minority category includes those who identified as lesbian, gay, bisexual, or "something else."The overall cardiovascular health score is the unweighted mean of the 8 cardiovascular health metrics.

Table 1 .
Sexual Identity Differences in CVH Across Strata of Race and Ethnicity Among 6033 Female Individuals Includes those who identified as Asian, multiracial, or any race and ethnicity other than Black, Hispanic, or White. b