Health and Health Risks Among Sexual Minority Women: An Examination of 3 Subgroups

Abstract

We used 2001–2010 National Health and Nutrition Examination Survey data to examine insurance status, source of routine care, cigarette and alcohol use, and self-rated health among lesbian, bisexual, and heterosexual women who have sex with women, compared with heterosexual women who do not have sex with women. We found higher risks of being uninsured among lesbian and bisexual women, worse self-rated health among bisexual women, higher alcohol use among bisexual and heterosexual women who have sex with women, and higher smoking across all subgroups.

Sexual minority women (SMW), whether defined by sexual identity (e.g., lesbian or bisexual) or sexual behavior (i.e., same-sex sexual activity), face numerous health risks, including substance use,^1–3 mental health disorders,^4–6 and poorer physical health,^7–10 as well as barriers to quality health care,^11,12 compared with sexual nonminority women. Little research, however, has examined the health of different subpopulations of SMW.¹³ Studies often combine lesbian and bisexual women in analysis, obscuring meaningful differences.^14–17 Research also frequently overlooks heterosexual women who have sex with women (WSW), who may experience distinct health risks.^18–21

In this study, we used information about sexual behavior and sexual identity to further understand differences among SMW. Specifically, we examined health and health risks among 3 subgroups of women: lesbian, bisexual, and heterosexual WSW, compared with heterosexual women who do not have sex with women.

METHODS

We used 2001–2010 National Health and Nutrition Examination Survey (NHANES) data.²² We categorized SMW, aged 20 to 59 years, into 3 subgroups: (1) lesbian WSW, (2) bisexual WSW, and (3) heterosexual WSW. Heterosexual women who did not report sex with women (heterosexual non-WSW) were the reference group. Because of small sample sizes, we excluded women who indicated their sexual identity as “something else” or “not sure,” and lesbian or bisexual women who did not have sex with women. We combined years of data (2001–2010) to increase the precision of the estimates (n = 5868).

Outcome measures were self-rated health, insurance, having a source of routine care, heavy alcohol use (average ≥ 7 drinks per week in past year), binge drinking (≥ 5 drinks per day at least once in past year), and cigarette use (≥ 100 cigarettes over lifetime). We estimated relative risks by using generalized linear model regressions with Poisson distribution and robust variance estimators, an analytic approach demonstrated to reliably estimate relative risks for binary outcomes.^23,24 Regression models adjusted for age, race, ethnicity, education, poverty, relationship status, and survey year. We weighted analyses and adjusted standard errors to account for the complex sampling design.²⁵

RESULTS

In the weighted sample, 9.3% of women were a sexual minority (Table 1). Sexual minority women were more likely to be younger than heterosexual women. Bisexual women were more likely to be living in poverty than other groups. Heterosexual women were the most likely to be married.

TABLE 1—

Characteristics of Weighted Sample of Sexual Minority Women: National Health and Nutrition Examination Survey, United States, 2001–2010

Characteristic	Lesbian WSW (n = 84),a %	Bisexual WSW (n = 180),b %	Heterosexual WSW (n = 290),c %	Heterosexual Non-WSW (Ref; n = 5314),d %	Pe
Total	1.5	2.8	5.0	90.7
Age, y					< .001
20–29	28.3	43.5	29.3	22.4
30–39	28.7	28.6	25.3	25.2
40–49	27.6	18.1	27.0	28.9
50–59	15.4	9.8	18.4	23.5
Race/ethnicity					.237
Black, non-Hispanic	13.0	14.2	12.5	12.1
Hispanic or Latina	8.8	9.2	7.5	12.4
Other or multiracial	3.5	3.3	6.8	5.0
White, non-Hispanic	74.8	73.3	73.3	70.5
Education					.257
< high school	11.3	19.4	13.2	13.7
High-school graduate	18.3	22.5	16.4	22.8
Some college	42.0	36.9	42.7	34.9
College graduate	28.4	21.2	27.8	28.6
Poverty statusf					.011
Below FPL	13.6	23.1	14.8	13.0
At or above FPL	86.4	76.9	85.2	87.0
Relationship status					< .001
Married	0.0	26.6	46.3	60.8
Single	58.2	34.1	17.4	15.5
Living with partner	21.0	19.8	12.8	7.5
Widowed or divorced	20.8	19.6	23.5	16.2

Note. FPL = federal poverty line; WSW = women who have sex with women. Percentages are weighted to be representative of US civilian, noninstitutionalized population.

^aWomen who self-identified as lesbian and reported sex with women.

^bWomen who self-identified as bisexual and reported sex with women.

^cWomen who self-identified as heterosexual or straight and reported sex with women.

^dWomen who self-identified as heterosexual or straight and reported no sex with women.

^eBased on the Pearson χ² test.

^fBased on ratio of family income to poverty, as defined by the Department of Health and Human Services poverty guidelines, specific to family size, year, and state.²²

As shown in Table 2, compared with heterosexual non-WSW, bisexual women were more likely to report fair or poor self-rated health (adjusted relative risk [ARR] = 1.8), and lesbian women were less likely to report having a source of care (ARR = 1.7). The risk of being uninsured was greater for women with a sexual minority identity (lesbian, ARR = 1.7; bisexual, ARR = 1.6), whereas there was no difference in risk among heterosexual women, regardless of sexual behavior. Both bisexual and heterosexual WSW had higher risk of heavy alcohol use (ARR = 1.8 and 1.7, respectively) and binge drinking (ARR = 1.5 and 1.6, respectively). All subgroups had higher risk of cigarette use than heterosexual non-WSW.

TABLE 2—

Adjusted Relative Risks and Predicted Probabilities of Health Indicators Among Sexual Minority Women: National Health and Nutrition Examination Survey, United States, 2001–2010

	Lesbian WSWa		Bisexual WSWb		Heterosexual WSWc		Heterosexual Non-WSWd (Ref)
Health Indicator	ARR (95% CI)	Predicted %	ARR (95% CI)	Predicted %	ARR (95% CI)	Predicted %	ARR	Predicted %
Fair or poor self-rated health	1.2 (0.7, 1.8)	16.1	1.8* (1.2, 2.6)	24.9	1.2 (0.9, 1.7)	17.1	1.0	13.9
No insurance coverage	1.7* (1.2, 2.5)	29.5	1.6* (1.2, 2.0)	26.7	1.0 (0.7, 1.3)	16.8	1.0	17.1
No source of routine care	1.7* (1.1, 2.8)	17.2	1.0 (0.5, 1.7)	9.4	0.9 (0.6, 1.4)	9.0	1.0	9.9
Heavy alcohol usee	1.5 (0.9, 3.0)	11.0	1.8* (1.2, 2.9)	13.1	1.7* (1.2, 2.6)	12.6	1.0	7.1
Binge drinkingf	1.2 (0.8, 1.8)	34.1	1.5* (1.1, 1.9)	39.9	1.6* (1.3, 2.0)	44.0	1.0	28.7
Lifetime cigarette use	1.3* (1.1, 1.6)	56.1	1.6* (1.3, 1.9)	66.8	1.7* (1.6, 1.9)	72.6	1.0	42.2

Note. ARR = adjusted relative risk; CI = confidence interval; WSW = women who have sex with women. Relative risks were calculated by using a general linearized model regression with a modified Poisson with robust variance estimator.²³ Relative risks and predicted probabilities are adjusted for age, race/ethnicity, educational attainment, poverty status, relationship status, and survey year. Estimates and confidence intervals account for probability of selection, stratification, and clustering.

^aWomen who self-identified as lesbian and reported sex with women.

^bWomen who self-identified as bisexual and reported sex with women.

^cWomen who self-identified as heterosexual or straight and reported sex with women.

^dWomen who self-identified as heterosexual or straight and reported no sex with women.

^eMore than 7 alcoholic drinks per week, on average, in the past year.

^fDrinking 5 or more alcoholic drinks in a day at least once in the past year.

*P < .05.

DISCUSSION

Lesbian, bisexual, and heterosexual WSW experience differences in health and risk factors compared with sexual nonminority women. The task for future research is to untangle the mechanisms through which the intersection of sexual identity and sexual behavior lead to these disparities. We do not believe that sexual orientation per se explains these findings. Instead, differences in the social circumstances of these women’s lives are likely to be responsible.

Our results provide avenues for future exploration. We found that only women with a sexual minority identity (i.e., lesbian and bisexual women) were more likely to be uninsured. Many employers do not offer insurance benefits to same-sex partners of lesbian and bisexual employees.²⁶ Moreover, Medicaid eligibility based on family status may exclude many low-income SMW. However, only lesbian women were less likely to have a source of routine care, suggesting additional barriers to access for this group.

Past research also suggests that minority stress (i.e., the unique stressors experienced because of having a marginalized identity relative to society’s heterosexual cultural norm) may put sexual minority women at risk for negative health and health behaviors.²⁷ Our findings for lesbian and bisexual women are consistent with this research. However, minority stress theory has generally overlooked heterosexual WSW, a group we also found at high risk of negative health behaviors.

Our findings also suggest that, to be most effective, health research and practice may need to account for the unique experiences of various subgroups of SMW. Only bisexual women reported worse self-rated health, and only lesbian women were more likely to lack a source of care. This suggests that these groups should not be regarded as synonymous categories. Furthermore, heterosexual WSW, who we found to be at increased risk of cigarette and alcohol use, may not benefit from interventions that rely on a sexual minority identity framing. Instead, they may require separate consideration in policy and practice.

Finally, our results should be considered in light of limitations. The sample sizes were small, so we could not look at trends over time or outcomes in more recent years. Our sample excluded sexual minority women who were not sexually active, limiting the generalizability of results. Moreover, the NHANES sexual identity question may have conflated sexual identity with sexual attraction. It is unclear, however, whether this resulted in any significant misidentification of women’s sexual identity.^28,29