Abstract
Objective
Sexual minority women (lesbian and bisexual) experience significant stigma, which may increase their cardiovascular disease (CVD) risk. The purpose of this study was to examine the prevalence of modifiable risk factors for CVD (including mental distress, health behaviors, blood pressure, glycosylated hemoglobin, and total cholesterol) and CVD in sexual minority women compared with their heterosexual peers.
Materials and Methods
A secondary analysis of the National Health and Nutrition Examination Survey (2001-2012) was conducted. Multiple imputation with chained equations was performed. Logistic regression models adjusted for relevant covariates were run. Self-report (medical history and medication use) and biomarkers for hypertension, diabetes, and high total cholesterol were examined.
Results
The final analytic sample consisted of 7,503 that included 346 sexual minority women (4.6%). Sexual minority women were more likely to be younger, single, have a lower income, and lack health insurance. After covariate adjustment, sexual minority women exhibited excess CVD risk related to higher rates of frequent mental distress (adjusted odds ratio [AOR], 2.05; 95% confidence interval [CI], 1.45–2.88), current tobacco use (AOR, 2.11; 95% CI, 1.53–2.91), and binge drinking (AOR, 1.66; 95% CI, 1.17–2.34). Sexual minority women were more likely to be obese (AOR, 1.61; 95% CI, 1.23–2.33) and have glycosylated hemoglobin consistent with prediabetes (AOR, 1.56; 95% CI, 1.04–2.34). No differences were observed for other outcomes.
Conclusions
Sexual minority women demonstrated increased modifiable risk factors for CVD, but no difference in CVD diagnoses. Several emerging areas of research are highlighted, in particular, the need for CVD prevention efforts that target modifiable CVD risk in sexual minority women.
Sexual minorities (lesbian, gay, and bisexual individuals) experience significant health disparities related to stigma (
Institute of Medicine, 2011
). A growing body of research indicates that sexual minorities are exposed to interpersonal and structural stigma that is associated with negative health outcomes (Meyer, 2003
), poor mental health (Collier, van Beusekom, Bos, & Sandfort, 2013
, Cramer, McNiel, Holley, Shumway, & Boccellari, 2011
), decreased life expectancy, and increased mortality (Hatzenbuehler, Bellatorre et al.,2014
). Approximately 80% of sexual minorities report experiencing some form of harassment (Katz-Wise and Hyde, 2012
) and 20% to 33% have experienced a hate crime in their lifetime (Burks et al., 2015
, Herek, 2009
). In 2016, sexual orientation motivated violence accounted for 17.7% of all hate crimes reported in the United States, representing a 2% increase compared with the previous year (U.S. Department of Justice, 2017
). Social policies represent forms of structural stigma that can also negatively impact the health of sexual minorities. Currently only 31 states have hate crime laws prohibiting bias-motivated violence against sexual minorities and there is no federal law that prohibits discrimination based on sexual orientation (U.S. Department of Justice. (2017). Hate crime statistics. Available: https://ucr.fbi.gov/hate-crime/2016/topic-pages/incidentsandoffenses.pdf. Accessed: November 26, 2017.
Human Rights Campaign, 2015
). Additional factors associated with health disparities in this population include inadequate training of health care providers, poverty, and lower rates of health insurance coverage (Human Rights Campaign. (2015). Marriage center. Available: http://www.hrc.org/campaigns/marriage-center. Accessed: January 25, 2016.
Institute of Medicine, 2011
).It is well-documented that, compared with their heterosexual counterparts, sexual minority women experience significant health disparities, such as higher rates of poor mental health (
King et al., 2008
, Pakula and Shoveller, 2013
, Plöderl and Tremblay, 2015
), obesity, and gynecological cancers (Institute of Medicine, 1999
); however, little is known about disparities in other chronic conditions, including cardiovascular disease (CVD; Lick et al., 2013
). CVD remains the leading cause of death worldwide (World Health Organization, 2014
) and approximately 90% of CVD risk is attributed to modifiable risk factors, including psychosocial factors, tobacco use, alcohol consumption, physical inactivity, diet, obesity, hypertension, diabetes, and lipids (World Health Organization. (2014). Global status report on noncommunicable diseases 2014. Available: http://apps.who.int/iris/bitstream/10665/148114/1/9789241564854_eng.pdf. Accessed: July 25, 2017.
Yusuf et al., 2004
). Stress is a recognized CVD risk factor (Jood et al., 2009
, Rosengren et al., 2004
, Steptoe and Kivimäki, 2013
, Teo et al., 2009
) that has a deleterious effect on the health of sexual minorities (Meyer, 2003
). Stress contributes to inflammation and endothelial dysfunction that increase CVD risk through mediated pathways (Cohen et al., 2012
, Xue et al., 2015
). Stress is also associated with negative health behaviors that increase CVD risk, including tobacco use, binge drinking, physical inactivity, and unhealthy dietary patterns (American Psychological Association, 2017
). Maladaptive coping strategies associated with stress, such as tobacco and alcohol use, predispose sexual minority women to increased risk for CVD compared with their heterosexual peers (Bloomfield et al., 2011
, Blosnich et al., 2013
). In addition, two recent systematic reviews concluded that sexual minority women demonstrate higher rates of obesity than heterosexual women (Caceres et al., 2017
, Eliason et al., 2015
).The National Academy of Medicine underscored the need for research on CVD in sexual minorities (
Institute of Medicine, 2011
). A recent systematic review examined CVD risk and CVD diagnoses in sexual minority and heterosexual adults (Caceres et al., 2017
). Overall, sexual minority women exhibited greater CVD risk compared with heterosexual women. Despite the strength of these findings the authors identified several limitations of the empirical literature. Although stress is posited as a main contributor to health disparities in sexual minorities (Lick et al., 2013
), only five studies included any measures of stress. Also, to date, few researchers have used biomarkers to examine CVD in sexual minorities. Most data are based on participant self-report, with only seven studies including biomarkers. The present study sought to address these research gaps.We hypothesized that sexual minority women would exhibit greater modifiable risk factors for CVD compared with heterosexual women. Thus, the purpose of this study, using data from the National Health and Nutrition Examination Survey (NHANES; 2001-2012), was to examine the prevalence of modifiable risk factors for CVD (including mental distress, tobacco use, alcohol consumption, physical inactivity, and dietary fat intake, hypertension, diabetes, and high total cholesterol) and CVD outcomes in sexual minority and heterosexual women.
Materials and Methods
A secondary analysis of NHANES (2001-2012) data was conducted. NHANES is a national cross-sectional survey used to monitor the health of the nation by estimating the prevalence of major diseases and risk factors (
Johnson et al., 2014
). NHANES is the largest national survey in the United States that collects information about sexual identity and the biomarkers of interest. NHANES uses a complex multistage probability sampling design to achieve a representative sample of individuals from across the United States (Johnson et al., 2014
). Data from the 2013-2014 NHANES release were not included because the measure of mental distress used from 2001-2012 was removed starting this cycle. This study was exempt by the Institutional Review Board of New York University.Sample Size
Inclusion criteria
Only participants between the ages of 18 and 59 were asked about sexual identity as part of the sexual behavior module in NHANES. All female adult participants who identified as sexual minority women (lesbian and bisexual), regardless of sexual behavior, were included in this study. Because we were primarily concerned with the impact of sexual identity on CVD risk, we only included participants who identified as heterosexual and reported no sexual behavior with women.
Exclusion criteria
The following participants were excluded: those who 1) responded “don't know,” “refused,” “something else,” or “not sure” to the sexual identity item, 2) had missing data for sexual identity, blood pressure, glycosylated hemoglobin (HbA1C), total cholesterol, or CVD outcomes, and 3) were heterosexual women who reported any history of sexual behavior with women.
Measures
Sexual identity
Sexual identity was measured with the item: “Do you think of yourself as heterosexual or straight, homosexual or lesbian, bisexual, something else, or not sure?” Women who identified as lesbian or bisexual were then categorized as sexual minority women.
Demographic and clinical characteristics
Demographic characteristics were selected based on social determinants of health that are significantly associated with increased CVD risk (
Havranek et al., 2015
). Age was a continuous variable ranging from 18 to 59 years. Race/ethnicity was coded as non-Hispanic white, non-Hispanic Black, Hispanic, and other race. The measure of income was the family income to poverty ratio provided by NHANES. The income to poverty ratio (range, 0-5) was calculated by dividing the total household income by the poverty threshold as published by the Federal Register for that specific survey year. Participants with an income to poverty ratio of less than 1 met the definition of poverty, whereas higher ratios indicated higher levels of income. Education was categorized as less than high school, high school, some college, or college graduate or greater. Additionally, relationship status (never married, married/partnered, widowed, divorced, separated) was examined. Family history of CVD and health insurance coverage were assessed as clinical characteristics. Family history of CVD was a dichotomous variable based on self-report of having a blood relative with a history of angina, heart attack, or stroke before the age of 50. Current health insurance coverage was also assessed.Modifiable risk factors for CVD
Mental distress was measured using the following item: “Now thinking about your mental health, which includes stress, depression, and problems with emotions, for how many days during the past 30 days was your mental health not good?” from the CDC Healthy Days/HRQOL-4 (
Centers for Disease Control and Prevention, 2011a
). This measure has been identified as a clinical marker of depression and anxiety disorders (Centers for Disease Control and Prevention. (2011a). Health-related quality of life. Available: http://www.cdc.gov/hrqol/faqs.htm#3. Accessed: July 7, 2017.
Strine et al., 2005
). Participants who reported 14 or more days of mental distress were considered to have frequent mental distress (Centers for Disease Control and Prevention, 2011a
). Several health behaviors were examined. Participants who reported tobacco use (on some days or every day) were considered current smokers. Binge drinking in women is commonly defined as consumption of four or more alcoholic drinks within a 2-hour period (Centers for Disease Control and Prevention. (2011a). Health-related quality of life. Available: http://www.cdc.gov/hrqol/faqs.htm#3. Accessed: July 7, 2017.
National Institute on Alcohol Abuse and Alcoholism, 2017
). However, binge drinking was not assessed in this manner in NHANES. Instead, binge drinking was defined as consumption five or more alcohol drinks within 1 day in the previous year. For physical activity, we calculated the average number of minutes of moderate- and vigorous-intensity aerobic activity in the past week from participants’ self-report. A binary measure of physical activity was created by determining if participants met physical activity recommendations for adults (either ≥150 minutes of moderate-intensity aerobic activity per week, ≥75 minutes of vigorous-intensity aerobic activity per week, or an equivalent combination of moderate- and vigorous-intensity aerobic activity per week; National Institute on Alcohol Abuse and Alcoholism. (2017). Drinking levels defined. Available: https://www.niaaa.nih.gov/alcohol-health/overview-alcohol-consumption/moderate-binge-drinking. Accessed: December 11, 2017.
Centers for Disease Control and Prevention, 2015
). Fat intake data were obtained from the dietary interview of NHANES estimated based on dietary recall of foods and beverages consumed during the previous 24-hour period. The diet measure used in this study was the unsaturated fat (sum of polyunsaturated fatty acid and monounsaturated fatty acid) to saturated fatty acid ratio. This ratio was dichotomized based on the Healthy Eating Index 2010 with an unsaturated to saturated fat ratio of 2.5 or greater considered adequate (Centers for Disease Control and Prevention. (2015). How much physical activity do adults need? Available: https://www.cdc.gov/physicalactivity/basics/adults/index.htm. Accessed: July 25, 2017.
Guenther et al., 2013
).Trained health technicians collected biomarker data (including body mass index, blood pressure, HbA1c, and total cholesterol) as part of the physical examination in NHANES. These procedures have been previously described (
Centers for Disease Control and Prevention, 2011b
). Overweight (body mass index ≥ 25.0 kg/m2) and obesity (body mass index ≥ 30.0 kg/m2) were defined based on established criteria (Centers for Disease Control and Prevention. (2011b). National Health and Examination Survey (NHANES): Anthropometry procedures manual. Available: https://www.cdc.gov/nchs/data/nhanes/2011-2012/manuals/Anthropometry_Procedures_Manual.pdf. Accessed: July 28, 2017.
Centers for Disease Control and Prevention, 2016
). Sexual minorities are less likely to access preventive care (Centers for Disease Control and Prevention. (2016). Defining adult overweight and obesity. Available: https://www.cdc.gov/obesity/adult/defining.html. Accessed: December 28, 2017.
Institute of Medicine, 2011
); therefore, we chose to examine hypertension, diabetes, and high total cholesterol using two methods: self-report and biomarkers. Participants completed a questionnaire reporting their medical history and current medication use. We classified participants as having hypertension, diabetes, or high total cholesterol if they reported 1) a medical history of hypertension, diabetes, or high total cholesterol, and/or 2) current use of medications to treat these conditions. We also examined biomarkers to determine the presence of hypertension, diabetes, high total cholesterol, and preclinical disease states based on established guidelines. A systolic blood pressure of 140 mm Hg or higher and/or a diastolic blood pressure of 90 mm Hg or higher were considered evidence of hypertension. Prehypertension was defined as a systolic blood pressure between 120 and 129 mm Hg and/or a diastolic blood pressure of greater than 80 to 89 mm Hg (James et al., 2014
). Diabetes was based on a HbA1c of 6.5% or greater and prediabetes was defined as an HbA1c between 5.7% and 6.4% (American Diabetes Association, 2017
). HbA1c is an indicator of glycemic status over the preceding 2 to 3 months (American Diabetes Association, 2017
). A total cholesterol of 240 mg/dL or higher was considered high and total cholesterol between 200 and 239 mg/dL was defined as borderline high (National Heart, 2016
).National Heart, Lung, and Blood Institute. (2016). How is high blood cholesterol diagnosed? Available: https://www.nhlbi.nih.gov/health/health-topics/topics/hbc/diagnosis. Accessed: December 12, 2017.
CVD
The presence of CVD diagnoses including angina, coronary heart disease, heart failure, myocardial infarction, and stroke was based on self-report. Participants who reported at least one of these conditions were considered to have CVD.
Statistical Analysis
Descriptive statistics
All statistical analyses were conducted in Stata version 15.1 (StataCorp, College Station, TX). Two-year sample weights for NHANES (2001-2012) were averaged and combined before conducting analyses.
Missing data
The three variables with the most missing data were physical activity (13.7%), binge drinking (11.9%), and income (4.3%). Sexual minority women were significantly less likely to have missing data for binge drinking (p < .001). Investigating missing data mechanisms is important to determine the statistical method for handling missing data. After conducting Little's test (
Little, 1988
), it was evident that data were not missing completely at random. Therefore, listwise deletion may lead to biased findings and multiple imputation is recommended. Missing at random occurs if the probability that data are missing does not depend on unobserved data, but may be explained by observed data. Analyses indicated that missing values were significantly associated with age and race, thus, the assumption that data were missing at random was deemed plausible.Multiple imputation
Multiple imputation was conducted in Stata version 15.1 (StataCorp). Multiple imputation is a simulation-based technique for handling missing data that is preferred over single imputation methods because these tend to overestimate variance. We used multiple imputation with chained equations because it does not assume data are normally distributed, which permits flexibility for imputation of non-normal and categorical data. Multiple imputation with chained equations imputes multiple variables iteratively through a sequence of univariate imputations with all variables, except the one being imputed, used to predict missing values (
StataCorp, 2013
).Multiple imputation consists of three steps: imputation, estimation, and pooling (
Kenward and Carpenter, 2007
). The imputation step consisted of generating an imputation model that incorporated outcome and structural variables in NHANES (sampling weights, strata, and cluster) to reduce bias (Johnson and Young, 2011
). Based on previous recommendations a total of 20 imputations were used (Dong and Peng, 2013
). The estimation and pooling steps can be thought of as the analysis stage. Imputation diagnostics were performed and results from imputed datasets were combined into a single imputed dataset for statistical analyses.Statistical Analyses
To assess differences in demographic and clinical characteristics, modifiable risk factors for CVD, and CVD between sexual minority and heterosexual participants bivariate analyses were conducted using the Student t test and design-adjusted Rao-Scott χ2 test, for continuous and categorical variables, respectively. A significance level of p < .05 was predetermined. We ran multiple logistic regression models for all outcomes with demographic characteristics included as covariates in all models. Additional covariate adjustment was decided a priori based on prior evidence. Mental distress was included as a covariate in logistic regression models that examined health behaviors because it is associated with each health behavior examined (tobacco use, alcohol consumption, physical activity, and dietary fat intake;
American Psychological Association, 2017
). Similarly, informed by established evidence, we added mental distress and health behaviors as covariates in logistic regression models for obesity, hypertension, diabetes, total cholesterol, their pre-clinical disease states, and CVD (Benjamin et al., 2017
, Havranek et al., 2015
).Results
The total number of potential participants after aggregating data across six NHANES cycles (2001-2012) was 31,393. After applying the inclusion and exclusion criteria described, the final analytic sample consisted of 7,503 women, of which 346 (4.6%) identified as sexual minority women. Descriptive statistics are shown in Table 1. Several sexual orientation differences in demographic and clinical characteristics were observed. Sexual minority women were significantly younger (p < .001) and reported a lower family income to poverty ratio (p < .001) than heterosexual women. Sexual minority women were also less likely to be currently married or living with a partner (p < .001) or to have health insurance coverage (p < .001). In addition, sexual minority women had significantly higher rates of frequent mental distress (p < .001), current tobacco use (p < .001), binge drinking (p < .001), and obesity (p < .03). No other significant differences were observed.
Table 1Descriptive Statistics for Demographic/Clinical Characteristics, Modifiable Risk Factors, and Cardiovascular Disease in Women (N = 7,503)
| Demographic and Clinical Characteristics | Heterosexual Women (n = 7,157) | Sexual Minority Women (n = 346) | p Value |
|---|---|---|---|
| Age (mean) | 40.1 | 34.7 | <.001 |
| Family income to poverty ratio (mean) | 2.6 | 2.0 | <.001 |
| Race/ethnicity | .26 | ||
| Non-Hispanic White | 69.7 | 72.9 | |
| Non-Hispanic Black | 11.8 | 13.0 | |
| Hispanic | 13.0 | 9.2 | |
| Other | 5.5 | 4.9 | |
| Education | .39 | ||
| Less than high school | 13.7 | 16.7 | |
| High school | 21.0 | 19.2 | |
| Some college | 34.3 | 37.7 | |
| College graduate or greater | 31.0 | 26.4 | |
| Relationship status | <.001 | ||
| Married/partnered | 66.3 | 40.9 | |
| Widowed | 1.8 | 0.1 | |
| Divorced | 11.1 | 13.8 | |
| Separated | 2.9 | 2.3 | |
| Never married | 17.9 | 42.9 | |
| CVD family history | 23.9 | 27.2 | .36 |
| Health insurance coverage | 81.9 | 72.5 | <.001 |
| Modifiable risk factors | |||
| Frequent mental distress | 14.6 | 28.3 | <.001 |
| Current tobacco use | 21.6 | 44.5 | <.001 |
| Binge drinking | 26.9 | 46.2 | <.001 |
| Meets physical activity recommendations | 30.5 | 34.6 | .69 |
| Adequate fat intake | 19.1 | 18.1 | .72 |
| Overweight (BMI ≥ 25 kg/m2) | 61.7 | 63.6 | .69 |
| Obesity (BMI ≥ 30 kg/m2) | 34.1 | 43.0 | .03 |
| Hypertension self-report | 23.8 | 18.6 | .11 |
| Hypertension (SBP ≥ 140 and/or DBP ≥ 90) | 8.4 | 6.8 | .51 |
| Pre-hypertension (SBP 120-129 and/or DBP 80-89) | 30.4 | 27.6 | .23 |
| Diabetes self-report | 5.3 | 6.5 | .46 |
| Diabetes (HbA1c ≥ 6.5%) | 4.1 | 2.7 | .29 |
| Prediabetes (HbA1c 5.7%-6.4%) | 14.1 | 15.7 | .56 |
| High total cholesterol self-report | 24.1 | 17.6 | .06 |
| High total cholesterol (≥240 mg/dL) | 15.3 | 12.5 | .36 |
| Borderline high total cholesterol (200-239 mg/dL) | 30.1 | 24.9 | .10 |
| Cardiovascular disease | |||
| Cardiovascular disease | 3.3 | 2.8 | .67 |
Abbreviations: BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; HbA1c, glycosylated hemoglobin.
∗ p < .05.
The significant differences between sexual minority and heterosexual women that were noted in bivariate analyses for frequent mental distress, current tobacco use, binge drinking, and obesity remained significant after covariate adjustment. These results are presented in Table 2. Sexual minority women reported significantly higher rates of frequent mental distress (AOR, 2.05; 95% CI, 1.45–2.88), current tobacco use (AOR; 95% CI, 2.11, 1.53–2.91), and binge drinking (AOR, 1.66; 95% CI, 1.17–2.34) compared with heterosexual women. No differences in physical activity or dietary fat intake were observed. Sexual minority women did not display a significant difference in being overweight, but were significantly more likely to be obese (AOR, 1.61; 95% CI, 1.23–2.33). In terms of diabetes, sexual minority women had higher but not statistically significant self-report of diabetes (AOR, 1.82; 95% CI, 0.89–3.72). Although they did not demonstrate higher HbA1C values consistent with diabetes (AOR, 0.90; 95% CI, 0.35–2.31), sexual minority women had higher rates of prediabetes even after adjusting for body mass index (AOR, 1.56; 95% CI, 1.04–2.34). Despite higher CVD risk related to several risk factors, sexual minority women did not exhibit significant differences in subjective or objective measures of hypertension, total cholesterol, or CVD.
Table 2Sexual Orientation Differences in Modifiable Risk Factors and Cardiovascualr Disease in Women (N = 7,503)
| OR (95% CI) | AOR (95% CI) | |
|---|---|---|
| Frequent mental distress | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 2.30 (1.67–3.17) | 2.05 (1.45–2.88), |
| Current tobacco use | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 2.91 (2.20–3.86) | 2.11 (1.53–2.91), |
| Binge drinking | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 2.33 (1.67–3.25) | 1.66 (1.17–2.34), |
| Meets physical activity recommendations | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 1.10 (0.76–1.60) | 1.04 (0.69–1.56) |
| Adequate fat intake | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 0.94 (0.64–1.36) | 0.92 (0.62–1.34) |
| Overweight (BMI ≥ 25 kg/m2) | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 1.08 (0.74–1.57) | 1.24 (0.84–1.85) |
| Obesity (BMI ≥ 30 kg/m2) | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 1.46 (1.04–2.05) | 1.61 (1.12–2.33), |
| Hypertension self-report | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 0.73 (0.49–1.08) | 0.76 (0.49–1.18) |
| Hypertension (SBP >140 and/or DBP >90) | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 0.79 (0.38–1.62) | 1.12 (0.52–2.41) |
| Pre-hypertension (SBP 120-129 and/or DBP 80-89) | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 0.83 (0.62–1.13) | 1.02 (0.75–1.39) |
| Diabetes self-report | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 1.25 (0.68–2.28) | 1.82 (0.89–3.72) |
| Diabetes (HbA1c ≥ 6.5%) | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 0.64 (0.28–1.50) | 0.90 (0.35–2.31) |
| Prediabetes (HbA1c 5.7–6.4%) | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 1.13 (0.74–1.74) | 1.56 (1.04–2.34), |
| High total cholesterol self-report | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 0.67 (0.45–1.01) | 0.86 (0.47–1.28) |
| High total cholesterol (≥240 mg/dL) | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 0.79 (0.48–1.32) | 1.04 (0.61–1.78) |
| Borderline high total cholesterol (200–239 mg/dL) | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 0.77 (0.56–1.05) | 0.88 (0.62–1.25) |
| Cardiovascular disease | ||
| Heterosexual | Reference | Reference |
| Sexual minority | 0.84 (0.38–1.85) | 0.69 (0.29–1.66) |
Abbreviations: AOR, adjusted odds ratio; BMI, body mass index; CI, confidence interval; DBP, diastolic blood pressure; HbA1c, glycosylated hemoglobin; OR, odds ratio; SBP, systolic blood pressure.
The reference group is heterosexual women.
∗ p < .05.
† Adjusted for age, race, education, relationship status, and income.
‡ Adjusted for age, race, education, relationship status, income, and frequent mental distress.
§ Adjusted for age, race, education, relationship status, income, insurance, cardiovascular disease family history, frequent mental distress, current tobacco use, binge drinking, physical activity, and fat intake.
‖ Adjusted for age, race, education, relationship status, income, insurance, frequent mental distress, insurance, cardiovascular disease family history, current tobacco use, binge drinking, physical activity, fat intake, and body mass index.
¶ Adjusted for age, race, education, relationship status, income, insurance, frequent mental distress, cardiovascular family history, current tobacco use, binge drinking, physical activity, fat intake, body mass index, hypertension, diabetes, and high cholesterol.
Discussion
This study contributes to the growing body of research on CVD in sexual minority women. Our findings are consistent with previous studies that demonstrated sexual minority women had higher rates of frequent mental distress (
Farmer et al., 2016
, Fredriksen-Goldsen et al., 2012
, Shilo and Mor, 2014
), tobacco use (Blosnich et al., 2013
, Emory et al., 2015
, Fredriksen-Goldsen et al., 2011
), and binge drinking (Fredriksen-Goldsen, K., Kim, H., & Emlet, C. (2011). The aging and health report. Available: http://www.diverseelders.org/wp-content/uploads/2012/07/aging_and_healthreport_disparities_LGBT1.pdf. Accessed: July 27, 2017.
Boehmer et al., 2012
, Coulter et al., 2015
, Gonzales and Henning-Smith, 2017
, Hughes et al., 2010
) than heterosexual women. We did not observe differences in physical activity and diet, which is consistent with the empirical literature. Although most studies indicate there are no physical activity differences (Blosnich et al., 2014
, Case et al., 2004
, Fredriksen-Goldsen et al., 2013
, Garland-Forshee et al., 2014
, Hatzenbuehler et al., 2013
, Hatzenbuehler and Slopen et al., 2014
, Matthews and Lee, 2014
), a small number of studies suggest sexual minority women have lower rates of physical activity compared with heterosexual women (Everett and Mollborn, 2013
, Herrick and Duncan, 2018
). However, a recent study found that sexual minority women had higher rates of aerobic activity but also reported more sedentary behaviors than heterosexual women (VanKim et al., 2017
). Moreover, several studies assert sexual minority women demonstrate similar (Dilley et al., 2010
, Matthews and Lee, 2014
) or worse diet quality (Minnis et al., 2015
, Roberts et al., 2003
, Valanis et al., 2000
) compared with heterosexual women. A notable exception is a recent analysis of the Nurses’ Health Study, which revealed that sexual minority women had better diet quality than heterosexual women (VanKim et al., 2016
).In the present study, sexual minority women also demonstrated higher rates of objectively measured obesity compared with heterosexual women. Two recent systematic reviews identified that most studies rely on self-reported height and weight to determine presence of obesity in this population (
Caceres et al., 2017
, Eliason et al., 2015
). Studies that have objectively measured obesity in sexual minority women report conflicting findings. Although some studies report no difference in obesity between sexual minority and heterosexual women (Clark et al., 2015
, Farmer et al., 2013
, Strutz et al., 2014
), others have identified significantly higher rates of obesity in sexual minority women (Everett and Mollborn, 2013
, Kinsky et al., 2016
). This area is important for further research, because both sexual minority and heterosexual women have been shown to underreport their body mass index (Richmond et al., 2012
).Sexual minority women in the present study did not have a higher prevalence of hypertension, high total cholesterol, or CVD, which supports findings from two recent systematic reviews (
Caceres et al., 2017
, Simoni et al., 2017
). The systematic review conducted by Caceres et al., 2017
identified that, of the 17 studies that examined diabetes, only 2 identified higher rates of self-reported diabetes in sexual minority women (Diamant et al., 2000
, Dilley et al., 2010
). Although we did not observe a significant difference for diabetes, sexual minority women had significantly higher rates of prediabetes than heterosexual women. This finding is significant, because approximately 18% of individuals with prediabetes develop diabetes within 2 years (Glauber et al., 2018
). Overall, few studies have used biomarkers (HbA1c and/or fasting glucose) to assess glycemic status in sexual minority women. Most researchers have found no significant difference in glycemic status between sexual minority and heterosexual women (Clark et al., 2015
, Hatzenbuehler et al., 2013
, Hatzenbuehler and Slopen et al., 2014
). However, these studies were all analyses of data from the National Longitudinal Study of Adolescent to Adult Health, which included young adults (mean age of <30 years old). Although Kinsky et al., 2016
found significantly higher mean fasting glucose in sexual minority women (35-65 years old), it is important to note they did not examine whether participants met criteria for diabetes or prediabetes.Study Limitations
This study has several limitations that must be considered. Although NHANES is a nationally representative dataset, its cross-sectional design limits causal inference, which is a noted weakness of cardiovascular research with sexual minorities (
Caceres et al., 2017
). Another limitation was that this analysis was limited to young and middle-aged adults, because sexual identity was not assessed in participants over the age of 59 in NHANES. In addition, mental distress was assessed with a nonspecific measure of acute stress because NHANES does not include measures of chronic stress. It is possible that the random sampling methods used by population-based surveys such as NHANES do not produce representative samples of sexual minorities. Sexual minorities who participate in population-based surveys may be inherently different from those who do not. Another limitation is that we examined individual risk factors rather than creating a composite score of CVD risk, such as the Framingham or American College of Cardiology/American Heart Association risk scores. It is possible that sexual minority women exhibit elevations in several CVD risk factors that may not result in CVD morbidity or mortality. Also, to achieve sufficient statistical power we combined lesbian and bisexual women for all analyses. This weakness is significant and should be addressed in future studies.Implications for Practice and/or Policy
Research
Our findings provide future directions for research on CVD in sexual minority women and add to the nascent body of research highlighting hyperglycemia as a CVD risk factor in this population. Prospective studies are needed to establish the temporality of modifiable risk factors and CVD diagnosis. Although few researchers have used objective measures to investigate CVD in sexual minority women, our findings indicate there is a need to use a combination of subjective and objective measures. In particular, we recommend more research using objective measures of obesity and diabetes. In addition, neuroendocrine, coagulation, and inflammatory biomarkers related to CVD should be included in future research to understand if stress in sexual minorities differentially affects these pathways. Despite growing evidence of increased CVD risk, there is a dearth of research on subclinical markers of CVD in sexual minority women. Recent studies indicate subclinical measures, including carotid artery intima thickness and coronary artery calcification, may reveal CVD in asymptomatic adults that is undetected by traditional biomarkers (
Budoff et al., 2013
, Rampersaud et al., 2008
, - Rampersaud E.
- Bielak L.F.
- Parsa A.
- Shen H.
- Post W.
- Ryan K.A.
- Mitchell B.D.
- et al.
The association of coronary artery calcification and carotid artery intima-media thickness with distinct, traditional coronary artery disease risk factors in asymptomatic adults.
American Journal of Epidemiology. 2008; 168: 1016-1023
Zaid et al., 2017
). Also, researchers should include measures of chronic and minority stressors (including stigma, expectations of rejection, internalized homophobia, etc.) to assess their association with CVD in sexual minority women. Because participants in the present study were young (mean age of 38.4 years), it is necessary to examine whether the higher rates of mental distress, current tobacco use, binge drinking, obesity, and prediabetes we observed persist in older sexual minority women.Practice
Because inadequate training of health care providers has been identified to contribute to health disparities in this population, sexual minority health should be incorporated into health professions curricula, because few programs currently include this content (
Carabez et al., 2015
, Obedin-Maliver et al., 2011
). As electronic health records increasingly include sexual orientation items, it is imperative for clinicians to appreciate the importance of collecting these data. Furthermore, there is a need to ensure clinicians assess sexual orientation in a manner that promotes patient trust. These findings indicate that clinicians should screen sexual minority women for modifiable risk factors for CVD, including stress, tobacco use, binge drinking, obesity, and hyperglycemia. Thus, there is a need to focus on lifestyle modification to reduce CVD risk in this population. Clinicians and public health practitioners should develop initiatives to reduce modifiable risk factors for CVD in sexual minority women.Policy
Government agencies concerned with health disparities in this population should advocate for the inclusion of sexual orientation items in national surveys. Although several national surveys have procedures for oversampling underserved populations, such as racial/ethnic minorities and older adults, presently none oversample sexual minorities. This action would be important to providing larger sample sizes to study health disparities in this population.
We were unable to examine the impact of state-level policies of CVD risk in sexual minority women because NHANES does not release state-level data. However, the impact of structural stigma on the health of sexual minority women cannot be underestimated. There is a lack of measures to assess structural stigma in sexual minorities (
Hatzenbuehler, Bellatorre et al.,2014
). Most research in this area has focused on interpersonal stigma, which does not adequately capture the full extent of stigma that sexual minorities are exposed to (Fredriksen-Goldsen et al., 2014
). Future studies should examine the impact of interpersonal and structural stigma on CVD risk in sexual minority women. In addition, policies should be enacted to ensure that health care professionals are aware of the health issues that impact sexual minorities. One example of such a policy is the mandatory lesbian, gay, bisexual, and transgender cultural competency training for licensed health care professionals enacted by the District of Columbia City Council in 2016 (Chibbaro, 2016
).- Chibbaro L.
D.C. Council passes LGBT “cultural competency” bill.
2016
http://www.washingtonblade.com/2016/02/03/d-c-council-passes-lgbt-cultural-competency-bill/
Date accessed: January 20, 2017
Conclusions
This study contributes to the nascent body of research examining CVD risk in sexual minorities. Overall, this sample of sexual minority women exhibited excess risk for CVD related to higher rates of frequent mental distress, current tobacco use, binge drinking, obesity, and prediabetes. Even though few differences in other CVD outcomes were observed in sexual minority women, there is evidence of an increased risk for CVD in this population. Several emerging areas of research are highlighted, including the need to further explore hyperglycemia in this population, use of measures of chronic and minority stress, oversampling of sexual minorities in population-based surveys, and the incorporation of novel biomarkers to examine CVD disparities in this population. These findings can help to inform the development of primary and secondary CVD prevention efforts.
Supplementary Data
- Data Profile
References
- Standards of medical care in diabetes - 2017.Diabetes Care. 2017; 40: s4-s128
- Stress and health disparities: Contexts, mechanisms, and interventions among racial/ethnic minority and low socioeconomic status populations. Washington, DC.Available, 2017 (Accessed: October 1, 2017)
- Heart disease and stroke statistics-2017 update: A report from the American Heart Association.Circulation. 2017; 135: e146-e603
- International differences in alcohol use according to sexual orientation.Substance Abuse. 2011; 32: 210-219
- A systematic review of the aetiology of tobacco disparities for sexual minorities.Tobacco Control. 2013; 22: 66-73
- Health inequalities among sexual minority adults: Evidence from ten U.S. states, 2010.American Journal of Preventive Medicine. 2014; 46: 337-349
- Adult health behaviors over the life course by sexual orientation.American Journal of Public Health. 2012; 102: 292-300
- Progression of coronary calcium and incident coronary heart disease events: MESA (Multi-Ethnic Study of Atherosclerosis).Journal of the American College of Cardiology. 2013; 61: 1231-1239
- Frequency nature and correlates of hate crime victimization experiences in an urban sample of lesbian, gay, and bisexual community members.Journal of Interpersonal Violence. 2015; : 1-19
- A systematic review of cardiovascular disease in sexual minorities.American Journal of Public Health. 2017; 107: e13-e21
- “Never in all my years…”: Nurses’ education about LGBT health.Journal of Professional Nursing. 2015; 31: 323-329
- Sexual orientation, health risk factors, and physical functioning in the Nurses’ Health Study II.Journal of Women’s Health (2002). 2004; 13: 1033-1047
Centers for Disease Control and Prevention. (2011a). Health-related quality of life. Available: http://www.cdc.gov/hrqol/faqs.htm#3. Accessed: July 7, 2017.
Centers for Disease Control and Prevention. (2011b). National Health and Examination Survey (NHANES): Anthropometry procedures manual. Available: https://www.cdc.gov/nchs/data/nhanes/2011-2012/manuals/Anthropometry_Procedures_Manual.pdf. Accessed: July 28, 2017.
Centers for Disease Control and Prevention. (2015). How much physical activity do adults need? Available: https://www.cdc.gov/physicalactivity/basics/adults/index.htm. Accessed: July 25, 2017.
Centers for Disease Control and Prevention. (2016). Defining adult overweight and obesity. Available: https://www.cdc.gov/obesity/adult/defining.html. Accessed: December 28, 2017.
- D.C. Council passes LGBT “cultural competency” bill.2016 (Available:)http://www.washingtonblade.com/2016/02/03/d-c-council-passes-lgbt-cultural-competency-bill/Date accessed: January 20, 2017
- Disparities in long-term cardiovascular disease risk by sexual identity: The National Longitudinal Study of Adolescent to Adult Health.Preventive Medicine. 2015; 76: 26-30
- Chronic stress, glucocorticoid receptor resistance, inflammation, and disease risk.Proceedings of the National Academy of Sciences of the United States of America. 2012; 109: 5995-5999
- Sexual orientation and gender identity/expression related peer victimization in adolescence: A systematic review of associated psychosocial and health outcomes.Journal of Sex Research. 2013; 50: 299-317
- Evidence of syndemics and sexuality-related discrimination among young sexual-minority women.LGBT Health. 2015; 2: 1-10
- Mental health in violent crime victims: Does sexual orientation matter?.Law and Human Behavior. 2011; 36: 1-11
- Health behaviors, health status, and access to and use of health care: A population-based study of lesbian, bisexual, and heterosexual women.Archives of Family Medicine. 2000; 9: 1043-1051
- Demonstrating the importance and feasibility of including sexual orientation in public health surveys: Health disparities in the Pacific Northwest.American Journal of Public Health. 2010; 100: 460-467
- Principled missing data methods for researchers.SpringerPlus. 2013; 2: 1-17
- A systematic review of the literature on weight in sexual minority women.Women’s Health Issues. 2015; 25: 162-175
- Intragroup variance in lesbian, gay, and bisexual tobacco use behaviors: Evidence that subgroups matter, notably bisexual wom.Nicotine & Tobacco Research. 2015; : 1494-1501
- Differences in hypertension by sexual orientation among U.S. young adults.Journal of Community Health. 2013; 38: 588-596
- Gay acres: Sexual orientation differences in health indicators among rural and nonrural individuals.The Journal of Rural Health. 2016; 32: 321-331
- A population-based study of cardiovascular disease risk in sexual-minority women.American Journal of Public Health. 2013; 103: 1845-1850
- Disability among lesbian, gay, and bisexual adults: Disparities in prevalence and risk.American Journal of Public Health. 2012; 102: e16-21
- Health disparities among lesbian, gay, and bisexual older adults: Results from a population-based study.American Journal of Public Health. 2013; 103: 1802-1809
- The health equity promotion model: Reconceptualization of lesbian, gay, bisexual, and transgender (LGBT) health disparities.Am J Orthopsychiatry. 2014; 84: 653-663
Fredriksen-Goldsen, K., Kim, H., & Emlet, C. (2011). The aging and health report. Available: http://www.diverseelders.org/wp-content/uploads/2012/07/aging_and_healthreport_disparities_LGBT1.pdf. Accessed: July 27, 2017.
- Sexual orientation and sex differences in adult chronic conditions, health risk factors, and protective health practices, Oregon, 2005-2008.Preventing Chronic Disease. 2014; 11: E136
- A simple model for predicting two-year risk of diabetes development in individuals with prediabetes.Permanente Journal. 2018; 22: 1-6
- Health disparities by sexual orientation: Results and implications from the Behavioral Risk Factor Surveillance System.Journal of Community Health. 2017; 42: 1163-1172
- Update of the Healthy Eating Index: HEI-2010.Journal of the Academy of Nutrition and Dietetics. 2013; 113: 569-580
- Structural stigma and all-cause mortality in sexual minority populations.Social Science & Medicine. 2014; 103: 33-41
- Sexual orientation disparities in cardiovascular biomarkers among young adults.American Journal of Preventive Medicine. 2013; 44: 612-621
- Stressful life events, sexual orientation, and cardiometabolic risk among young adults in the United States.Health Psychology. 2014; 33: 1185-1194
- Social determinants of risk and outcomes for cardiovascular disease.Circulation. 2015; 132: 873-898
- Hate crimes and stigma-related experiences among sexual minority adults in the United States: Prevalence estimates from a national probability sample.Journal of Interpersonal Violence. 2009; 24: 54-74
- A systematic scoping review of engagement in physical activity among LGBTQ+ adults.Journal of Physical Activity & Health. 2018; 15: 226-232
- Sexual victimization and hazardous drinking among heterosexual and sexual minority women.Addictive Behaviors. 2010; 35: 1152-1156
Human Rights Campaign. (2015). Marriage center. Available: http://www.hrc.org/campaigns/marriage-center. Accessed: January 25, 2016.
- Lesbian health: Current assessment and directions for the future.Author, Washington, DC1999
- The health of lesbian, gay, bisexual, and transgender people: Building a foundation for better understanding.Author, Washington, DC2011
- 2014 Evidence-based guideline for the management of high blood pressure in adults.JAMA. 2014; 311: 507
- National Health and Nutrition Examination Survey: Sample design, 2011 – 2014.Vital Health Stat. 2014; 2: 1-33
- Toward best practices in analyzing datasets with missing data: Comparisons and recommendations.Journal of Marriage and Family. 2011; 73: 926-945
- Self-perceived psychological stress and ischemic stroke: A case-control study.BMC Medicine. 2009; 7: 53
- Victimization experiences of lesbian, gay, and bisexual individuals: A meta-analysis.Journal of Sex Research. 2012; 49: 142-167
- Multiple imputation: Current perspectives.Statistical Methods in Medical Research. 2007; 16: 199-218
- A systematic review of mental disorder, suicide, and deliberate self harm in lesbian, gay and bisexual people.BMC Psychiatry. 2008; 8: 70
- Risk of the metabolic syndrome in sexual minority women: Results from the ESTHER Study.Journal of Women’s Health. 2016; 25: 784-790
- Minority stress and physical health among sexual minorities.Perspectives on Psychological Science. 2013; 8: 521-548
- A test of missing completely at random for multivariate data with missing values.Journal of the American Statistical Association. 1988; 83: 1198-1202
- A profile of North Carolina lesbian, gay, and bisexual health disparities, 2011.American Journal of Public Health. 2014; 104: e98-e105
- Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence.Psychological Bulletin. 2003; 129: 674-697
- Differences in chronic disease behavioral indicators by sexual orientation and sex.Journal of Public Health Management Practice. 2015; 22: S25-S32
National Heart, Lung, and Blood Institute. (2016). How is high blood cholesterol diagnosed? Available: https://www.nhlbi.nih.gov/health/health-topics/topics/hbc/diagnosis. Accessed: December 12, 2017.
National Institute on Alcohol Abuse and Alcoholism. (2017). Drinking levels defined. Available: https://www.niaaa.nih.gov/alcohol-health/overview-alcohol-consumption/moderate-binge-drinking. Accessed: December 11, 2017.
- Lesbian, gay, bisexual, and transgender–related content in undergraduate medical education.JAMA. 2011; 306: 971-977
- Sexual orientation and self-reported mood disorder diagnosis among Canadian adults.BMC Public Health. 2013; 13: 209
- Mental health of sexual minorities: A systematic review.International Review of Psychiatry. 2015; 27: 367-385
- The association of coronary artery calcification and carotid artery intima-media thickness with distinct, traditional coronary artery disease risk factors in asymptomatic adults.American Journal of Epidemiology. 2008; 168: 1016-1023
- Sexual orientation and bias in self-reported body mass index.Obesity. 2012; 20: 1-15
- Cardiovascular disease risk in lesbian women.Women’s Health Issues. 2003; 13: 167-174
- Association of psychosocial risk factors with risk of acute myocardial infarction in 11 119 cases and 13 648 controls from 52 countries (the INTERHEART study): Case-control.Lancet. 2004; 364: 953-962
- The impact of minority stressors on the mental and physical health of lesbian, gay, and bisexual youths and young adults.Health & Social Work. 2014; 39: 161-171
- Disparities in physical health conditions among lesbian and bisexual women: A systematic review of population-based studies.Journal of Homosexuality. 2017; 64: 32-44
- Stata multiple imputation reference manual: Release 13.Stata Press, College Station, TX2013
- Stress and cardiovascular disease: An update on current knowledge.Annual Review of Public Health. 2013; 34: 337-354
- Health-related quality of life, health risk behaviors, and disability among adults with pain-related activity difficulty.American Journal of Public Health. 2005; 95: 2042-2048
- Health disparities among young adult sexual minorities in the U.S.American Journal of Preventive Medicine. 2014; 48: 76-88
- Potentially modifiable risk factors associated with myocardial infarction in China : The INTERHEART China study.Heart. 2009; 95: 1857-1864
U.S. Department of Justice. (2017). Hate crime statistics. Available: https://ucr.fbi.gov/hate-crime/2016/topic-pages/incidentsandoffenses.pdf. Accessed: November 26, 2017.
- Sexual orientation and health: Comparisons in the Women’s Health Initiative sample.Archives of Family Medicine. 2000; 9: 843-853
- Dietary patterns during adulthood among lesbian, bisexual, and heterosexual women in the Nurses’ Health Study II.Journal of the Academy of Nutrition and Dietetics. 2016; S2212-2672: 31195-31199
- Physical activity and sedentary behaviors among lesbian, bisexual, and heterosexual women: Findings from the Nurses’ Health Study II.Journal of Women’s Health. 2017; 26: 1077-1085
World Health Organization. (2014). Global status report on noncommunicable diseases 2014. Available: http://apps.who.int/iris/bitstream/10665/148114/1/9789241564854_eng.pdf. Accessed: July 25, 2017.
- Investigating the role of acute mental stress on endothelial dysfunction: A systematic review and meta-analysis.Clinical Research in Cardiology. 2015; 104: 310-319
- Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): Case-control study.Lancet. 2004; 364: 937-952
- Coronary artery calcium and carotid artery intima media thickness and plaque: Clinical use in need of clarification.Journal of Atherosclerosis and Thrombosis. 2017; 24: 227-239
Biography
Billy A. Caceres, PhD, RN, AGPCNP-BC, is a Post-Doctoral Research Fellow in Comparative and Cost-Effectiveness at Columbia University School of Nursing. His research focuses on the impact of stress on cardiovascular disease risk and management in vulnerable populations across the lifespan.
Abraham A. Brody, PhD, GNP-BC, is Associate Professor and Associate Director of the Hartford Institute for Geriatric Nursing at NYU College of Nursing. He is an expert in the care of older adults.
Perry N. Halkitis, PhD, MS, MPH, is Dean and Professor at the Rutgers School of Public Health. He has focused a significant amount of his research on HIV/AIDS and drug abuse as well as the impact of psychiatric and psychosocial factors of mental health among men who have sex with men.
Caroline Dorsen, PhD, FNP-BC, is Assistant Professor NYU College of Nursing with a focus on health disparities in vulnerable populations including sexual and gender minorities.
Gary Yu, DrPH, MPH, is Associate Research Scientist at NYU College of Nursing. His research is on the health of men who have sex with men in the United States and male sex workers in Vietnam.
Deborah A. Chyun, PhD, RN, FAHA, FAAN, is Dean and Professor at the University of Connecticut School of Nursing with extensive research in cardiovascular disease prevention and management in adults with diabetes.
Article info
Publication history
Published online: April 13, 2018
Accepted:
March 13,
2018
Received in revised form:
March 6,
2018
Received:
October 14,
2017
Footnotes
Supported by a pre-doctoral institutional training grant (NCATS TL1TR001447) from the NYU-HHC Clinical and Translational Institute and a post-doctoral T32 fellowship from the National Institute of Nursing Research (NINR T32NR013454) in Comparative and Cost-Effectiveness Research for Nurse Scientists to BAC. The sponsor had no role in the study design, data analysis, interpretation of data, writing the report, or decision to submit the report for publication.
Identification
Copyright
© 2018 Jacobs Institute of Women's Health. Published by Elsevier Inc.
