Abstract
Objective
To examine the changes in health insurance coverage, access to care, and health services utilization among nonelderly sexual minority and heterosexual adults between pooled years 2013‐2014 and 2017‐2018.
Data Sources
Data on 3223 sexual minorities (lesbians, gay men, bisexual individuals, and other nonheterosexual populations) and 86 181 heterosexuals aged 18‐64 years were obtained from the 2013, 2014, 2017, and 2018 National Health Interview Surveys.
Study Design
Unadjusted and regression‐adjusted estimates compared changes in health insurance status, access to care, and health services utilization for nonelderly adults by sexual minority status. Regression‐adjusted changes were obtained from logistic regression models controlling for demographic and socioeconomic characteristics.
Principal Findings
Uninsurance declined for both sexual minority adults (5 percentage points, P < .05) and heterosexual adults (2.5 percentage points, P < .001) between 2013‐2014 and 2017‐2018. Reductions in uninsurance for sexual minority and heterosexual adults were associated with increases in Medicaid coverage. Sexual minority and heterosexual adults were also less likely to report unmet medical care in 2017‐2018 compared with 2013‐2014. Low‐income adults (regardless of sexual minority status) experienced relatively large increases in Medicaid coverage and substantial improvements in access to care over the study period. The gains in coverage and access to care across the study period were generally similar for heterosexual and sexual minority adults.
Conclusions
Sexual minority and heterosexual adults have experienced improvements in health insurance coverage and access to care in recent years. Ongoing health equity research and public health initiatives should continue to monitor health care access and the potential benefits of recent health insurance expansions by sexual orientation and sexual minority status when possible.
Keywords: health care access, LGBT health, sexual minorities
What this study adds.
Few studies have examined changes in health insurance coverage and access to care for sexual minority adults following health reform and the national legalization of same‐sex marriage.
Uninsurance declined by 5 percentage points for nonelderly sexual minority adults and 2.5 percentage points for nonelderly heterosexual adults between 2013‐2014 and 2017‐2018.
Gains in health insurance coverage and access to care between 2013‐2014 and 2017‐2018 were pronounced among low‐income sexual minority and heterosexual adults.
1. INTRODUCTION
Previous research has documented substantial disparities in health insurance coverage and access to care for sexual minorities (ie, lesbians, gay men, bisexual individuals, and other nonheterosexual populations). For example, numerous studies using US Census data on adults in same‐sex couples 1 , 2 , 3 , 4 and population‐based data on people who self‐identify as lesbian, gay, and bisexual (LGB) have found that sexual minorities were less likely to have private health insurance and more likely to report financial‐related barriers to medical care compared with their heterosexual peers. 5 , 6 , 7 These disparities in access to care may have existed for a variety of reasons. Prior to federal legalization of same‐sex marriage, many employers did not make available employer‐sponsored health insurance (ESI) to same‐sex spouses. 8 For those same‐sex spouses who did have access to ESI, this coverage was treated as taxable or imputed income at the federal level until 2014. Meanwhile, when compared to heterosexual individuals, sexual minority adults have been shown to be in worse overall health, have a higher prevalence of diagnoses for chronic conditions, and were more likely to participate in adverse health behaviors (ie, heavy smoking and heavy drinking) due to high levels of discrimination and stigma. 9 This combination of poor baseline health and elevated risk for disease may have made some sexual minority individuals more likely to have been denied coverage or charged higher premiums for nongroup coverage because of preexisting health conditions prior to health insurance reforms in the United States.
Beginning in 2014, the Affordable Care Act (ACA) expanded health insurance to over 20 million Americans through Medicaid expansions for low‐income families and through premium assistance to purchase private health insurance in the health insurance marketplaces. 10 , 11 Notably, however, states were given the option to expand Medicaid, and previous research has shown that low‐income LGB adults in states that expanded Medicaid have lower uninsurance rates. 12 The ACA also prohibited health insurers from denying coverage or charging higher premiums to people with preexisting health conditions. Meanwhile, around the same time, two US Supreme Court decisions removed financial and legal barriers to employer‐sponsored insurance (ESI) for same‐sex couples. First, in 2013, the Supreme Court ruled the federal definition of marriage between one man and one woman unconstitutional in the United States v. Windsor. 13 As a result, same‐sex couples could access the federal benefits of marriage, including tax exemptions for covering a dependent spouse on ESI. Although same‐sex marriages were recognized under federal law, not all states were required to legalize same‐sex marriages. In 2015, the US Supreme Court ruled same‐sex marriage a constitutional right across the nation in Obergefell v. Hodges, which required all states to recognize same‐sex unions and required more employers to cover same‐sex spouses as dependents. 14 , 15
More research is still needed on whether and how health insurance and access to care for sexual minority adults has changed over time. Previous studies have documented increased health insurance coverage for sexual minorities since these policy changes occurred, 6 , 16 , 17 , 18 but fewer studies have examined changes in access to care and health services utilization by sexual minority status in recent years. 18 It remains unclear whether access to care and health services utilization has improved for sexual minorities. This study builds on previous research using more recent nationally representative data to examine changes in health insurance status, access to care, and health services utilization among nonelderly sexual minorities and their heterosexual peers. We also add to the literature a better understanding of how health care access has changed in recent years among low‐income Americans by sexual minority status.
2. METHODS
This study analyzed changes in health insurance status, access to care, and health services utilization by sexual minority status between pooled years from the 2013‐2014 and 2017‐2018 National Health Interview Survey (NHIS). Conducted annually by the National Center for Health Statistics at the Centers for Disease Control and Prevention, the NHIS is a nationally representative health survey of the civilian, noninstitutionalized population that provides timely and comprehensive data on the nation's health. 19 The family core questionnaire records basic demographic, health, and disability information for each household member, while a single random adult in each household is selected for a detailed interview on more specific health information that includes health insurance coverage, access to health care, and health services utilization. Our study sample was drawn from the sample adult component of the NHIS, which was accessed through the University of Minnesota's IPUMS Health Series (a harmonized and publicly available version of the NHIS). 20
To identify sexual minority adults, we relied on a sexual orientation question that was added to the annual sample adult component of the NHIS in 2013. Adult respondents were asked which of the following categories best represents how they identify: lesbian or gay; straight, that is, not gay; bisexual; something else; I don't know the answer; or refuse. We classified sexual minority individuals as respondents who indicated their sexual orientation as lesbian, gay, bisexual, or something else. Our findings were similar with and without the sample of adults responding “something else” to the sexual orientation question. The comparison group was comprised of respondents indicating their sexual orientation as heterosexual. We excluded respondents that did not know the answer or refused to answer the sexual orientation question. We also restricted the analysis to nonelderly adults aged 18‐64 years as our primary interest does not involve coverage through the Medicare program (a group largely unaffected by the ACA). Our study was approved and exempted from review by the Vanderbilt University Institutional Review Board as all analyses were limited to publicly available, deidentified data.
2.1. Study outcomes
We compared several dimensions of health insurance status, access to care, and health services utilization by sexual minority status. First, we compared sources of health insurance, if any, by sexual minority status. Because respondents could report multiple sources of health insurance, respondents were assigned to the following hierarchy of coverage based on their responses to a series of health insurance questions: Medicare, employer‐sponsored health insurance, Medicaid, insurance from the individual market (ie, nongroup insurance purchased directly from an insurance company either on or off the marketplaces), and uninsured. We used this hierarchy to minimize overreporting of individual coverage, 21 , 22 to maximize the likelihood of actual coverage for nonelderly adults, and to define mutually exclusive categories. 23
We also compared a variety of measures related to access to care and health services utilization. We examined whether respondents indicated they had a usual source of medical care that was not a hospital. Five additional measures compared financial‐related barriers to care: delayed medical care due to cost, unmet medical care due to cost, unmet dental care due to cost, unmet prescription medications due to cost, and unmet mental health care due to cost. All five financial‐related barriers to care used the prior 12 months as a reference time frame. Finally, we compared two measures of health services utilization: reporting no office visit in the prior year and having an emergency room visit in the prior year. Of note, measures analyzed in this study were self‐reported but are routinely used to monitor access to health care in the United States. 19
2.2. Statistical analysis
We used descriptive statistics, two‐tailed t tests, and survey weights to characterize the study sample by sexual minority status for pooled years 2013‐2014 and 2017‐2018. We pooled 2013‐2014 as the baseline period (despite the fact that many ACA coverage provisions started in 2014) for two reasons: We wanted to measure changes in the study outcomes using robust sample sizes, and 2013 was the first year to ascertain sexual orientation in the NHIS. Next, we estimated survey‐weighted changes (not adjusted for covariates) in health insurance status, access to care, and health services utilization between 2013‐2014 and 2017‐2018. Then, we used multivariable logistic regression models to estimate adjusted changes in each outcome while controlling for demographic and socioeconomic characteristics, including age, race/ethnicity, sex, relationship status, the presence of a child in the household, educational attainment, employment status, family income relative to the federal poverty guidelines, self‐rated health status, and US Census region. The primary variable of interest was an indicator for pooled years 2017‐2018 (equal to one) vs 2013‐2014 (equal to zero). All independent variables are defined in Table 1. Results from the adjusted logistic regression models are presented as percentage point changes (ie, marginal effects) between 2013‐2014 and 2017‐2018, which were calculated from differencing the predicted probabilities of each outcome in 2017‐2018 compared with 2013‐2014. All logistic regression models included heterosexual and sexual minority adults, and interactions between pooled years and sexual minority status were included to test whether the changes for sexual minority adults were statistically different than the changes for heterosexual adults. Finally, we estimated unadjusted and adjusted changes for each health care outcome by sexual minority status—but we restricted the analysis to the subsample directly affected by Medicaid expansions under the ACA: individuals in households with family incomes beneath 138 percent of the federal poverty guidelines (FPG). We conducted all analyses in Stata version 14 (StataCorp. 2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP) using survey weights and the svy command to adjust standard errors for the complex survey design of the NHIS and to generate nationally representative estimates. 24 As recommended by the National Center for Health Statistics, 25 we used multiple imputations in Stata (via the mi suite of commands) to adjust for missing responses to family income. The margins and mimrgns commands were also used to generate adjusted estimates following multivariable logistic regression models. 26 , 27
TABLE 1.
Characteristics of nonelderly adults in the United States by sexual minority status
| Heterosexual | Sexual minority | |||
|---|---|---|---|---|
| 2013‐14 | 2017‐18 | 2013‐14 | 2017‐18 | |
| n = 51 499 | n = 34 682 | n = 1637 | n = 1586 | |
| Sexual orientation | ||||
| Heterosexual | 100.0 | 100.0 | 0.0** | 0.0** |
| Gay or lesbian | 0.0 | 0.0 | 64.4** | 49.5** |
| Bisexual | 0.0 | 0.0 | 28.5** | 38.4** |
| Something else | 0.0 | 0.0 | 7.1** | 12.2** |
| Sex | ||||
| Male | 49.2 | 49.5 | 45.8 | 43.2* |
| Female | 50.8 | 50.5 | 54.2 | 56.8* |
| Age, y | ||||
| 18‐25 | 17.5 | 16.5 | 25.2** | 27.6** |
| 26‐34 | 19.4 | 19.9 | 21.8 | 27.4** |
| 35‐49 | 31.3 | 31.3 | 28.2* | 23.4** |
| 50‐64 | 31.8 | 32.3 | 24.8** | 21.6** |
| Race/ethnicity | ||||
| Non‐Hispanic white | 63.0 | 60.4 | 66.2 | 63.7 |
| Non‐Hispanic black | 12.2 | 12.3 | 13.5 | 11.4 |
| Hispanic | 17.0 | 18.1 | 13.6* | 16.5 |
| Non‐Hispanic other | 7.7 | 9.0 | 6.6 | 8.2 |
| Missing data | 0.2 | 0.2 | 0.1 | 0.2 |
| Citizenship | ||||
| US citizen | 90.0 | 90.5 | 95.1** | 95.2** |
| Noncitizen | 9.9 | 9.2 | 4.9** | 4.6** |
| Missing data | 0.1 | 0.3 | 0.0** | 0.2 |
| Relationship status | ||||
| Married or living with a partner | 61.4 | 61.5 | 40.0** | 41.5** |
| Separated/divorced/widowed | 12.8 | 12.0 | 8.4** | 8.5** |
| Never married | 25.6 | 26.3 | 51.3** | 49.8** |
| Missing data | 0.2 | 0.1 | 0.4 | 0.2 |
| Child < 18 y present in household | ||||
| No | 57.0 | 58.0 | 76.5** | 77.0** |
| Yes | 43.0 | 42.0 | 23.5** | 23.0** |
| Educational attainment | ||||
| Less than high school | 12.4 | 10.7 | 9.8* | 8.3* |
| High school graduate | 25.0 | 23.6 | 20.2* | 21.4 |
| Some college | 20.6 | 19.1 | 23.1 | 24.4* |
| ≥Bachelor's degree | 30.1 | 34.4 | 34.8* | 36.0 |
| Missing data | 11.8 | 12.2 | 12.1 | 9.9* |
| Employment status | ||||
| Full‐time | 53.0 | 56.7 | 48.8* | 50.6* |
| Part‐time | 17.0 | 16.7 | 18.5 | 21.1* |
| Unemployed | 5.9 | 3.8 | 10.0* | 5.5* |
| Not in labor force | 22.9 | 21.6 | 22.1 | 21.4 |
| Missing data | 1.2 | 1.2 | 0.6* | 1.4 |
| Family income relative to federal poverty guidelines | ||||
| 0%‐138% | 21.4 | 17.8 | 24.4* | 25.5** |
| 139%‐399% | 40.4 | 37.7 | 37.8 | 35.2 |
| ≥401% | 38.2 | 44.5 | 37.8 | 39.3* |
| Health status | ||||
| Excellent | 31.8 | 31.2 | 28.8 | 26.5* |
| Very good | 32.6 | 33.6 | 31.4 | 30.8 |
| Good | 24.7 | 25.0 | 26.7 | 28.8* |
| Poor, fair | 10.9 | 10.1 | 13.2* | 13.8* |
| Missing data | 0.1 | 0.0 | 0.0* | 0.2 |
Data are from the 2013‐2014 and 2017‐2018 National Health Interview Survey, adults aged 18‐64 years. *P < .05; **P < .001 indicate statistically significant differences in means between sexual minorities and heterosexuals of the same pooled years based on two‐tailed t tests.
3. RESULTS
3.1. Descriptive statistics
Table 1 presents descriptive characteristics stratified by sexual minority status. Our final analytic sample size included 89 404 adults (53 136 in 2013‐2014 and 36 268 in 2017‐2018) [larger sample sizes in 2013‐2014 reflect past efforts to increase sample sizes in most states for state‐level analyses conducted at the NCHS]. 28 We obtained our sample after excluding nonelderly respondents that either did not know the answer (n = 539) or refused to answer (n = 549) the sexual orientation question. Our final sample included 3223 nonelderly sexual minorities and 86 181 nonelderly heterosexuals. Compared with heterosexual adults, sexual minorities were more likely to be younger, US citizens, and never married. Approximately 42 percent of heterosexual adults had a child in the household, and about 23 percent of sexual minorities had a child in the household. Educational attainment was relatively similar between sexual minorities and heterosexuals, but sexual minority adults were slightly more likely to report lower levels of family income and poor/fair health. Interestingly, over the study period, sexual minorities were less likely to describe their sexual orientation as gay or lesbian and more likely to self‐identify as bisexual or something else.
3.2. Results for all nonelderly adults by sexual minority status
Figure 1 presents the prevalence of uninsurance and select health care access measures for nonelderly sexual minority and heterosexual adults in 2013‐2014 and in 2017‐2018. Approximately 14 percent of nonelderly sexual minority and heterosexual adults remained uninsured in pooled years 2017‐2018. Seven percent of heterosexual adults reported unmet medical care due to cost in 2017‐2018, while 11 percent of sexual minority adults reported unmet medical care due to cost in the same period. Table 2 reports the unadjusted and adjusted changes in health insurance coverage, access to care, and health services utilization for all nonelderly adults by sexual minority status. After controlling for sociodemographic factors, uninsurance declined approximately 2.5 percentage points (P < .05) between 2013‐2014 and 2017‐2018 for heterosexual adults, and this decline was related to a 2.9 percentage point increase in Medicaid coverage. Heterosexual adults also experienced significant (P < .05) declines in reporting unmet medical care due to cost (−1.0 percentage points), unmet prescription medications due to cost (−1.0 percentage points), no office visits in the prior year (−1.4 percentage points), and an increase in having at least one emergency room visit in the prior year (1.3 percentage points) after adjusting for sociodemographic characteristics.
FIGURE 1.
Prevalence of uninsurance and access to care by sexual minority status, nonelderly adults
TABLE 2.
Changes in health insurance coverage, access to care, and health services utilization for nonelderly heterosexual and sexual minority adults between pooled years 2013/2014 and 2017/2018
| Unadjusted percentage point change (95% CI) | Adjusted percentage point change (95% CI) | |||
|---|---|---|---|---|
| Heterosexual | Sexual minority | Heterosexual | Sexual minority | |
| Health insurance coverage | ||||
| Employer | 3.1 (1.8 to 4.3)** | 2.8 (−1.8 to 7.3) | −0.2 (−1.0 to 0.6) | 1.7 (−2.0 to 5.3) |
| Individual | −0.5 (−1.0 to −0.04)* | −2.6 (−4.8 to −0.4)* | −0.5 (−0.9 to −0.03)* | −2.3 (−4.5 to −0.2)* |
| Medicaid | 1.3 (0.5 to 2.1)* | 4.5 (1.2 to 7.8)* | 2.9 (2.3 to 3.5)** | 4.2 (1.7 to 6.7)* |
| Medicare | 0.2 (−0.1 to 0.6) | 1.0 (−1.0 to 2.9) | 0.5 (0.2 to 0.8)** | 1.2 (−0.3 to 2.7) |
| Uninsured | −4.1 (−5.0 to −3.2)** | −5.6 (−8.9 to −2.3)* | −2.5 (−3.1 to −1.8)** | −5.0 (−7.9 to −2.0)** |
| Access to care | ||||
| No usual source of medical care | −1.3 (−2.3 to −0.4)* | −0.4 (−4.4 to 3.6) | −0.3 (−1.1 to 0.6) | −0.6 (−4.2 to 3.0) |
| Delayed care due to cost | −0.9 (−1.5 to −0.3)* | −1.0 (−4.2 to 2.1) | −0.2 (−0.7 to 0.3) | −1.0 (−3.6 to 1.6) |
| Unmet medical care due to cost | −1.6 (−2.1 to −1.1)** | −2.5 (−5.2 to 0.3) | −1.0 (−1.4 to −0.5)** | −2.1 (−4.4 to −0.1)* |
| Unmet dental care due to cost | −1.7 (−2.4 to −0.9)** | −3.7 (−7.0 to −0.3)* | −0.6 (−1.2 to 0.04) | −3.4 (−6.2 to −0.5)* |
| Unmet prescription medications due to cost | −1.7 (−2.2 to −1.1)** | −2.5 (−5.4 to 0.3) | −1.0 (−1.5 to −0.5)** | −2.4 (−4.8 to 0.02) |
| Unmet mental health care due to cost | 0.3 (0.05 to 0.6)* | 2.1 (−0.3 to 4.4) | 0.5 (0.2 to 0.8)** | 1.3 (−0.4 to 3.0) |
| Health services utilization | ||||
| No office visit in the prior year | −2.2 (−3.1 to −1.3)** | −2.2 (−6.0 to 1.5) | −1.4 (−2.1 to −0.6)** | −1.7 (−5.3 to 1.9) |
| Emergency room visit in the prior year | 0.4 (−0.4 to 1.2) | 7.0 (3.1 to 10.9)**† | 1.3 (0.6 to 2.0)* | 6.2 (2.6 to 9.8)*† |
Adjusted changes are estimated from logistic regression models controlling for age, sex, race/ethnicity, citizenship, relationship status, the presence of children in the household, educational attainment, employment status, family income relative to federal poverty guidelines, health status, and US Census region. *P < .05; **P < .001. † P < .05 compared with changes for heterosexual adults.
Source: Nonelderly adults (n = 86 181 heterosexual; 3223 sexual minority) in the 2013, 2014, 2017, and 2018 National Health Interview Survey (NHIS).
Table 2 also reports changes in health insurance, access to care, and health services utilization for all nonelderly sexual minority adults. After controlling for sociodemographic characteristics, uninsurance significantly (P < .05) declined 5.0 percentage points for sexual minorities between pooled years 2013‐2014 and 2017‐2018—which corresponded to a significant 4.2 percentage point increase in Medicaid coverage and a nonsignificant 1.7 percentage point increase in employer‐sponsored insurance. Meanwhile, sexual minority adults experienced significant (P < .05) declines in unmet medical care due to cost (−2.1 percentage points) and unmet dental care due to cost (−3.4 percentage points) after adjusting for sociodemographic characteristics. There were no statistically significant changes in having a usual source of care, delayed medical care due to cost, unmet prescription medications due to cost, unmet mental health care due to cost, and having an office visit in the prior year for sexual minority adults. Between pooled years 2013‐2014 and 2017‐2018, sexual minority adults were significantly (P < .05) more likely to have at least one emergency room visit in the prior year (6.2 percentage points). Changes in emergency room visits across the study period were significantly (P < .05) larger for sexual minority adults compared with the corresponding changes for heterosexual adults.
3.3. Results for low‐income adults by sexual minority status
The next set of analyses were restricted to low‐income nonelderly adults in families with incomes less than 138 percent of the federal poverty guidelines by sexual minority status. Figure 2 presents the prevalence of uninsurance and health care access for low‐income sexual minority and heterosexual adults in 2013‐2014 and in 2017‐2018. Approximately 26 percent of low‐income heterosexual adults and 20 percent of low‐income sexual minorities remained uninsured in pooled years 2017‐2018. During the same period, the prevalence of unmet medical care, dental care, and prescription medications due to cost remained above 10 percent for heterosexual and sexual minority adults alike. Table 3 reports the unadjusted and adjusted changes in health insurance coverage, access to care, and health services utilization for low‐income heterosexual adults. Low‐income heterosexual adults experienced a statistically significant 9.3 percentage point reduction in uninsurance, a significant 7.5 percentage point increase in Medicaid coverage, and a nonsignificant 1.3 percentage point increase in employer‐sponsored insurance. After controlling for sociodemographic characteristics, low‐income heterosexual adults experienced statistically significant reductions in having no usual source of care (−4.6 percentage points), delayed medical care due to cost (−3.6 percentage points), unmet medical care due to cost (−4.1 percentage points), unmet dental care due to cost (−4.2 percentage points), unmet prescription medications due to cost (−4.1 percentage points), and having no office visits in the prior year (−2.6 percentage points).
FIGURE 2.
Prevalence of uninsurance and access to care by sexual minority status, low‐income nonelderly adults < 138% FPG
TABLE 3.
Changes in health insurance coverage, access to care, and health services utilization for nonelderly low‐income heterosexual and sexual minority adults <138% FPG between pooled years 2013/2014 and 2017/2018
| Unadjusted percentage point change (95% CI) | Adjusted percentage point change (95% CI) | |||
|---|---|---|---|---|
| Heterosexual | Sexual minority | Heterosexual | Sexual minority | |
| Health insurance coverage | ||||
| Employer | 2.0 (−0.2 to 4.2) | 3.1 (−4.4 to 10.7) | 1.3 (−0.4 to 3.0) | 0.4 (−5.2 to 6.1) |
| Individual | 0.6 (−0.2 to 1.4) | −3.3 (−7.0 to 0.3)† | 0.5 (−0.3 to 1.3) | −3.2 (−6.3 to −0.1)*† |
| Medicaid | 6.1 (3.6 to 8.6)** | 10.8 (2.7 to 19.0)* | 7.5 (5.4 to 9.5)** | 12.9 (5.9 to 19.9)** |
| Medicare | 1.2 (0.1 to 2.3)* | 1.2 (−3.3 to 5.7) | 0.8 (−0.1 to 1.7) | 1.5 (−2.3 to 5.3) |
| Uninsured | −9.9 (−12.3 to −7.5)** | −11.9 (−19.4 to −4.3)* | −9.3 (−11.3 to −7.4)** | −12.5 (−20.0 to −5.1)* |
| Access to care | ||||
| No usual source of medical care | −5.1 (−7.2 to −3.1)** | −3.6 (−11.7 to 4.6) | −4.6 (−6.5 to −2.7)** | −4.3 (−12.3 to 3.6) |
| Delayed care due to cost | −3.4 (−4.9 to −1.9)** | −9.1 (−16.5 to −1.8)* | −3.6 (−5.0 to −2.2)** | −9.5 (−16.2 to −2.8)* |
| Unmet medical care due to cost | −3.8 (−5.3 to −2.4)** | −9.4 (−16.0 to −2.9)* | −4.1 (−5.4 to −2.7)** | −9.3 (−15.5 to −3.1)* |
| Unmet dental care due to cost | −4.5 (−6.3 to −2.7)** | −16.1 (−23.7 to −8.4)**† | −4.2 (−5.9 to −2.6)** | −16.6 (−24.2 to −8.9)**† |
| Unmet prescription medications due to cost | −4.2 (−5.6 to −2.8)** | −10.3 (−17.0 to −3.6)* | −4.1 (−5.3 to −2.8)** | −10.2 (−16.5 to −3.8)* |
| Unmet mental health care due to cost | −0.5 (−1.3 to 0.3) | −0.1 (−5.5 to 5.3) | −0.4 (−1.2 to 0.3) | −0.8 (−5.0 to 3.4) |
| Health services utilization | ||||
| No office visit in the prior year | −3.2 (−5.3 to 1.1)* | −1.1 (−8.7 to 6.5) | −2.6 (−4.5 to −0.8)* | −1.4 (−9.3 to 6.5) |
| Emergency room visit in the prior year | 1.5 (−0.3 to 3.4) | −2.8 (−11.2 to 5.7) | 2.2 (0.5 to 3.9)* | −2.2 (−10.1 to 5.7) |
Adjusted changes are estimated from logistic regression models controlling for age, sex, race/ethnicity, citizenship, relationship status, the presence of children in the household, educational attainment, employment status, family income relative to federal poverty guidelines, health status, and US Census region.
Abbreviation: FPG, federal poverty guidelines.
P < .05; **P < .001. † P < .05 compared with changes for heterosexual adults.
Source: Nonelderly adults (n = 19 671 heterosexual; 911 sexual minority) in the 2013, 2014, 2017, and 2018 National Health Interview Survey (NHIS).
Table 3 also reports the changes in health insurance coverage, access to care, and health services utilization for low‐income nonelderly sexual minority adults. After controlling for sociodemographic characteristics, low‐income sexual minority adults experienced a 12.5 percentage point decrease in uninsurance (P < .05) alongside a 12.9 percentage point increase in Medicaid coverage, a 3.2 percentage point reduction in individual (nongroup) coverage, and a nonsignificant 0.4 percentage point increase in employer‐sponsored insurance. Changes in individual coverage were statistically different for low‐income sexual minorities than the changes in individual coverage for low‐income heterosexual adults. Low‐income sexual minorities also experienced reductions in delayed medical care due to cost (−9.5 percentage points), unmet medical care due to cost (−9.3 percentage points), unmet dental care due to cost (−16.6 percentage points), and unmet prescription medications due to cost (−10.2 percentage points). Changes in unmet dental care due to cost were significantly (P < .05) larger for low‐income sexual minority adults relative to the changes for low‐income heterosexual adults over the study period. Changes in health insurance coverage, access to care, and health services utilization (presented in Tables 2 and 3) were generally similar in direction and magnitude but with less precision after excluding the sample of adults responding “something else” to the sexual orientation question (data are available in Appendix Tables A1 and A2).
4. DISCUSSION
In this study, we used nationally representative data from the National Health Interview Survey to document recent changes in health insurance coverage, access to care, and health services utilization for sexual minorities and heterosexuals in the United States. We found that uninsurance declined for nonelderly sexual minority and heterosexual adults between pooled years 2013‐2014 and 2017‐2018—which coincided with gains in Medicaid coverage. Our analysis is also one of the first to document changes in access to care for low‐income adults by sexual minority status and across a period of major health insurance reforms: Low‐income sexual minority and heterosexual adults were less likely to delay or forgo medical care, dental care, and prescription medications due to cost after the implementation of the Affordable Care Act in pooled years 2017‐2018, compared with 2013‐2014.
Overall, these results show that there have been significant improvements in Americans’ ability to access health insurance coverage and health care in the years following the implementation of the ACA. The fact that the gains we observed in health insurance coverage through Medicaid suggests that Medicaid expansion was successful in broadening health care access regardless of sexual minority status. We are not able to examine variation by state and Medicaid expansion status in our analysis, but future research should explore whether gains in health insurance coverage differed by sexual minority status between Medicaid expansion and nonexpansion states. Further, while these results document improvements in health care access, it is notable that 26 percent of low‐income heterosexual adults and 20 percent of low‐income sexual minority adults were still uninsured in 2017‐2018 (presented in Figure 2). More efforts are needed to ensure adequate health insurance coverage and access to care for those individuals living below 138 percent of the federal poverty guidelines.
We also found that the changes in having an emergency room visit between 2013‐2014 and 2017‐2018 were larger for sexual minority adults compared with the changes among heterosexual adults. This suggests that there may be gaps in having a primary care provider or medical home among sexual minorities. Indeed, having a usual source of care did not change much between 2013‐2014 and 2017‐2018 for sexual minority adults (Table 2). If sexual minority adults are increasingly likely to receive medical care in emergency rooms, then physicians and trainees in emergency medicine should receive training on the unique health care needs of LGBT populations. Unfortunately, most emergency medicine residency programs do not incorporate curricula on LGBT health 29 despite the fact that some emergency medicine physicians find LGBT patients challenging in emergency room settings. 30 More research should evaluate and identify best practices for delivering LGBT health training for health care specialists, including in emergency settings.
4.1. Limitations
There were several limitations to using the NHIS for this study. All responses to the NHIS were self‐reported, which can lead to response and recall bias when describing health insurance status and access to care. Ascertaining sexual minority status during in‐person interviews with NHIS surveyors may discourage some respondents from reporting accurate sexual orientation information. Preliminary research analyzing the quality of sexual orientation data in the NHIS found item nonresponse (ie, “refused to answer”) higher among respondents with lower levels of education, respondents residing in rural areas, and respondents not completing the survey in English. 31 Together, our results may be biased to the extent that sexual minority individuals from marginalized backgrounds are missing in this analysis. Relatedly, we also included respondents indicating their sexual orientation as “something else” despite some debate in the literature on how to treat this sample. We followed the suggestions of Eliason et al 2016, 32 and we classified this category within our sample of sexual minorities (rather than discard them from the sample). The inclusion of adults describing their sexual orientation as “something else” did not change the main findings of our study, but they notably represent a growing percentage of the sexual minority population alongside bisexual adults. Ongoing research should examine how sociodemographic factors and policy contexts influence whether and how adults describe their sexual orientation in national surveys.
While the NHIS is a well‐validated, nationally representative survey, the final sample consisted of relatively small numbers of nonelderly sexual minorities compared with heterosexuals. Because of small sample sizes, we were cautious about exploring subgroup or intersectional analyses, including comparisons by specific sexual orientations (eg, lesbian, gay, and bisexual), race/ethnicity, or socioeconomic status. In order to examine the experiences of subgroups in future analyses, the NCHS should consider oversampling sexual minorities in the NHIS. Further, the relatively small sample size of sexual minority adults in our analysis (compared with heterosexual adults) likely impacted the frequency in which we detected statistically significant differences over the study period; thus, absolute changes, confidence intervals, and statistical significance should be considered when interpreting our results. Relatedly, we combined data from 2013‐2014 for the baseline period in order to have larger samples of sexual minority adults. However, approximately half the states expanded Medicaid in 2014, which may underestimate the changes across the study period that are reported here.
Finally, this descriptive analysis documents the cross‐sectional changes in health insurance status and health care access by sexual minority status and does not investigate the underlying causal mechanisms. Having detailed longitudinal data on the same individuals over time or qualitative data that describe sexual minority experiences in seeking and obtaining health care would be useful to tease out how health insurance coverage and access to care has changed for sexual minorities. Meanwhile, other confounding variables may be missing from our cross‐sectional analysis. For example, our study does not include missing factors that may explain the changes and differences reported here, such as incidents of discrimination and state‐level policy environments such as Medicaid expansion. Unfortunately, the NHIS public use files do not provide information on state of residence, but future research should examine how state‐level Medicaid expansions have impacted sexual minority health, access to care, and health disparities.
5. CONCLUSION
Health insurance coverage and access to care improved between pooled years 2013‐2014 and 2017‐2018 for nonelderly sexual minority and heterosexual adults alike. These improvements appear to have been driven primarily by increases in Medicaid enrollment, especially for low‐income adults. We also found that increases in emergency room utilization were larger for sexual minority adults than the corresponding increases among heterosexual adults. Thus, emergency rooms and their providers should receive adequate training to ensure that LGBT patients are welcomed and treated with respect. Additionally, future studies should continue to explore how access to care evolves under varying policy contexts at the intersections of sexual minority status, gender, race/ethnicity, and socioeconomic status. Ultimately, access to timely, affordable, and high‐quality care is one potential mechanism for ensuring health equity for sexual minorities. This study uncovers important gains in coverage and access to care for all nonelderly adults, but more research is needed to fully identify and eliminate barriers to care for sexual minority adults.
Supporting information
Author Matrix
ACKNOWLEDGMENTS
Joint Acknowledgment/Disclosure Statement: Gilbert Gonzales received support from the Robert Wood Johnson Foundation during this project. All authors (Gilbert Gonzales, Carrie Henning‐Smith, and Jesse Ehrenfeld) have no financial conflicts of interest to report.
1.
TABLE A1.
Changes in health insurance coverage, access to care, and health services utilization for nonelderly heterosexual and sexual minority adults between pooled years 2013/2014 and 2017/2018
| Unadjusted percentage point change (95% CI) | Adjusted percentage point change (95% CI) | |||
|---|---|---|---|---|
| Heterosexual | Sexual minority | Heterosexual | Sexual minority | |
| Health insurance coverage | ||||
| Employer | 3.1 (1.8 to 4.3)** | 3.2 (−1.6 to 8.0) | −0.2 (−1.0 to 0.6) | 2.0 (−1.8 to 5.7) |
| Individual | −0.5 (−1.0 to −0.04)* | −2.7 (−5.0 to −0.4)* | −0.5 (−0.9 to −0.03)* | −2.4 (−4.7 to −0.2)* |
| Medicaid | 1.3 (0.5 to 2.1)* | 4.3 (0.9 to 7.7)* | 2.9 (2.3 to 3.5)** | 4.6 (1.9 to 7.3)* |
| Medicare | 0.2 (−0.1 to 0.6) | 0.7 (−1.3 to 2.7) | 0.5 (0.2 to 0.8)** | 1.1 (−0.6 to 2.7) |
| Uninsured | −4.1 (−5.0 to −3.2)** | −5.5 (−8.8 to −2.2)* | −2.5 (−3.1 to −1.8)** | −5.1 (−8.1 to −2.1)** |
| Access to care | ||||
| No usual source of medical care | −1.3 (−2.3 to −0.4)* | −1.2 (−5.3 to 2.9) | −0.3 (−1.1 to 0.6) | −1.4 (−5.1 to 2.3) |
| Delayed care due to cost | −0.9 (−1.5 to −0.3)* | −0.9 (−4.2 to 2.5) | −0.2 (−0.7 to 0.3) | −1.0 (−3.8 to 1.8) |
| Unmet medical care due to cost | −1.6 (−2.1 to −1.1)** | −2.6 (−5.5 to 0.3) | −1.0 (−1.4 to −0.5)** | −2.5 (−4.9 to 0.02) |
| Unmet dental care due to cost | −1.7 (−2.4 to −0.9)** | −3.8 (−7.3 to −0.2)* | −0.6 (−1.2 to 0.04) | −3.6 (−6.6 to −0.6)* |
| Unmet prescription medications due to cost | −1.7 (−2.2 to −1.1)** | −2.5 (−5.3 to 0.3) | −1.0 (−1.4 to −0.5)** | −2.5 (−4.8 to −0.2)* |
| Unmet mental health care due to cost | 0.3 (0.05 to 0.6)* | 1.6 (−0.7 to 3.8) | 0.5 (0.2 to 0.8)** | 0.9 (−0.7 to 2.5) |
| Health services utilization | ||||
| No office visit in the prior year | −2.2 (−3.1 to −1.3)** | −2.8 (−6.6 to 1.0) | −1.4 (−2.1 to −0.6)** | −2.3 (−6.1 to 1.4) |
| Emergency room visit in the prior year | 0.4 (−0.4 to 1.2) | 7.0 (2.9 to 11.1)*† | 1.3 (0.6 to 2.0)* | 6.3 (2.5 to 10.1)*† |
Adjusted changes are estimated from logistic regression models controlling for age, sex, race/ethnicity, citizenship, relationship status, the presence of children in the household, educational attainment, employment status, family income relative to federal poverty guidelines, health status, and US Census region. *P < .05; **P < .001. † P < .05 compared with changes for heterosexual adults.
These results exclude sexual minorities describing their sexual orientation as “something else."
Source: Nonelderly adults (n = 86 181 heterosexual; 2906 sexual minority) in the 2013, 2014, 2017, and 2018 National Health Interview Survey (NHIS).
TABLE A2.
Changes in health insurance coverage, access to care, and health services utilization for nonelderly low‐income heterosexual and sexual minority adults < 138% FPG between pooled years 2013/2014 and 2017/2018
| Unadjusted percentage point change (95% CI) | Adjusted percentage point change (95% CI) | |||
|---|---|---|---|---|
| Heterosexual | Sexual minority | Heterosexual | Sexual minority | |
| Health insurance coverage | ||||
| Employer | 2.0 (−0.2 to 4.2) | 2.6 (−5.2 to 10.5) | 1.3 (−0.3 to 3.0) | 0.1 (−5.8 to 5.9) |
| Individual | 0.6 (−0.2 to 1.4) | −3.7 (−7.7 to 0.3)† | 0.5 (−0.3 to 1.3) | −3.5 (−6.9 to −0.03)*† |
| Medicaid | 6.1 (3.6 to 8.6)** | 11.4 (2.8 to 20.1)* | 7.5 (5.4 to 9.6)** | 13.9 (6.5 to 21.3)** |
| Medicare | 1.2 (0.1 to 2.3)* | 1.4 (−3.5 to 6.3) | 0.8 (−0.1 to 1.7) | 1.4 (−2.8 to 5.7) |
| Uninsured | −9.9 (−12.3 to −7.5)** | −11.8 (−19.9 to −3.6)* | −9.4 (−11.3 to −7.4)** | −12.4 (−20.4 to −4.4)* |
| Access to care | ||||
| No usual source of medical care | −5.1 (−7.2 to −3.1)** | −2.1 (−10.9 to 6.6) | −4.6 (−6.5 to −2.8)** | −2.8 (−11.3 to 5.7) |
| Delayed care due to cost | −3.4 (−4.9 to −1.9)** | −8.4 (−15.5 to 1.3) | −3.6 (−5.0 to −2.2)** | −8.7 (−16.0 to −1.4)* |
| Unmet medical care due to cost | −3.8 (−5.3 to −2.4)** | −9.4 (−16.0 to −2.9)* | −4.1 (−5.4 to −2.7)** | −9.0 (−15.6 to −2.4)* |
| Unmet dental care due to cost | −4.5 (−6.3 to −2.7)** | −14.7 (−23.1 to −6.3)*† | −4.2 (−5.9 to −2.6)** | −16.0 (−24.2 to −7.8)**† |
| Unmet prescription medications due to cost | −4.2 (−5.6 to −2.8)** | −9.4 (−16.7 to −2.1)* | −4.0 (−5.3 to −2.8)** | −9.8 (−16.5 to −3.1)* |
| Unmet mental health care due to cost | −0.5 (−1.3 to 0.3) | 0.3 (−5.4 to 6.0) | −0.4 (−1.2 to 0.3) | −0.7 (−5.1 to 3.7) |
| Health services utilization | ||||
| No office visit in the prior year | −3.2 (−5.3 to 1.1)* | −1.3 (−9.3 to 6.8) | −2.6 (−4.5 to −0.8)* | −1.1 (−9.8 to 7.5) |
| Emergency room visit in the prior year | 1.5 (−0.3 to 3.4) | −3.0 (−12.1 to 6.1) | 2.2 (0.5 to 3.9)* | −2.7 (−11.3 to 5.9) |
Adjusted changes are estimated from logistic regression models controlling for age, sex, race/ethnicity, citizenship, relationship status, the presence of children in the household, educational attainment, employment status, family income relative to federal poverty guidelines, health status, and US Census region. *P < .05; **P < .001. † P < .05 compared with changes for heterosexual adults.
Abbreviation: FPG, federal poverty guidelines.
These results exclude sexual minorities describing their sexual orientation as “something else.”
Source: Nonelderly adults (n = 19 671 heterosexual; 785 sexual minority) in the 2013, 2014, 2017, and 2018 National Health Interview Survey (NHIS).
Gonzales G, Henning‐Smith C, Ehrenfeld JM. Changes in health insurance coverage, access to care, and health services utilization by sexual minority status in the United States, 2013‐2018. Health Serv. Res.2021;56:235–246. 10.1111/1475-6773.13567
REFERENCES
- 1. Gonzales G, Blewett LA. National and state‐specific health insurance disparities for adults in same‐sex relationships. Am J Public Health. 2014;104(2):e95‐e104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Gonzales G, Ortiz K. Health insurance disparities among racial/ethnic minorities in same‐sex relationships: an intersectional approach. Am J Public Health. 2015;105(6):1106‐1113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Buchmueller T, Carpenter CS. Disparities in health insurance coverage, access, and outcomes for individuals in same‐sex versus different‐sex relationships, 2000‐2007. Am J Public Health. 2010;100(3):489‐495. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Ash MA, Badgett MVL. Separate and unequal: the effect of unequal access to employment‐based health insurance on same‐sex and unmarried different‐sex couples. Contemp Econ Policy. 2006;24(4):582‐599. [Google Scholar]
- 5. Dahlhamer JM, Galinsky AM, Joestl SS, Ward BW. Barriers to health care among adults identifying as sexual minorities: a US national study. Am J Public Health. 2016;106(6):1116‐1122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Skopec L, Long SK. Lesbian, gay, and bisexual adults making gains in health insurance and access to care. Health Aff. 2015;34(10):1769‐1773. [DOI] [PubMed] [Google Scholar]
- 7. Hsieh N, Ruther M. Despite increased insurance coverage, nonwhite sexual minorities still experience disparities in access to care. Health Aff. 2017;36(10):1786‐1794. [DOI] [PubMed] [Google Scholar]
- 8. Claxton G, Rae M, Panchal N, et al. Employer Health Benefits: 2014 Annual Survey. Menlo Park, CA: Kaiser Family Foundation; 2014. [Google Scholar]
- 9. Gonzales G, Przedworski J, Henning‐Smith C. Comparison of health and health risk factors between lesbian, gay, and bisexual adults and heterosexual adults in the United States. JAMA Intern Med. 2016;176(9):1344‐1351. [DOI] [PubMed] [Google Scholar]
- 10. Frean M, Gruber J, Sommers BD. Premium subsidies, the mandate, and Medicaid expansion: coverage effects of the Affordable Care Act. J Health Econ. 2017;53:72‐86. [DOI] [PubMed] [Google Scholar]
- 11. Blumenthal D, Abrams M, Nuzum R. The Affordable Care Act at 5 years. N Engl J Med. 2015;372(25):2451‐2458. [DOI] [PubMed] [Google Scholar]
- 12. Gonzales G, Henning‐Smith C. The Affordable Care Act and health insurance coverage for lesbian, gay, and bisexual adults: analysis of the Behavioral Risk Factor Surveillance System. LGBT Health. 2017;4(1):62‐67. [DOI] [PubMed] [Google Scholar]
- 13. Supreme Court of the United States . United States v. Windsor. 77. 2013.
- 14. Supreme Court of the United States . Obergefell v. Hodges. 576 US. 2015.
- 15. Landers S. Civil rights and health‐beyond same‐sex marriage. N Engl J Med. 2015;373(12):1089‐1092. [DOI] [PubMed] [Google Scholar]
- 16. Tran LD. Moderate effects of same‐sex legislation on dependent employer‐based insurance coverage among sexual minorities. Med Care Res Rev. 2016;73(6):752‐768. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Gonzales G. Association of the New York state Marriage Equality Act with changes in health insurance coverage. JAMA. 2015;314(7):727. [DOI] [PubMed] [Google Scholar]
- 18. Dawson L, Kates J, Damico A. The Affordable Care Act and Insurance Coverage Changes by Sexual Orientation. Menlo Park, CA: Kaiser Family Foundation; 2018. [Google Scholar]
- 19. National Center for Health Statistics . Health, United States, 2016: With Chartbook on Long‐Term Trends in Health. Hyattsville, MD: National Center for Health Statistics; 2017. [PubMed] [Google Scholar]
- 20. Blewett LA, Drew JAR, Griffin R, King ML, Williams KCW. IPUMS Health Surveys: National Health Interview Survey, Version 6.2. Minneapolis: University of Minnesota; 2016. [Google Scholar]
- 21. Boudreaux M, Ziegenfuss JY, Graven P, Davern M, Blewett LA. Counting uninsurance and means‐tested coverage in the American Community Survey: a comparison to the current population survey. Health Serv Res. 2011;46(1 PART 1):210‐231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Office of the Assistant Secretary for Planning and Evaluation . Understanding Estimates of the Uninsured: Putting the Differences in Context. Washington, DC: Office of the Assistant Secretary for Planning and Evaluation; 2005. [Google Scholar]
- 23. Hest R. SHADAC’s Coverage Hierarchy for American Community Survey (ACS) Estimates on State Health Compare Author. Minneapolis, MN: State Health Access Data Assistance Center (SHADAC); 2020. [Google Scholar]
- 24. StataCorp . Stata Statistical Software: Release 14. College Station, TX: StataCorp LP; 2015. [Google Scholar]
- 25. National Center for Health Statistics . Multiple imputation of family income and personal earnings in the National Health Interview Survey: methods and examples. Hyattsville, MD: National Center for Health Statistics; 2018. [Google Scholar]
- 26. Williams R. Using Stata’s margins command to estimate and interpret adjusted predictions and marginal effects. Stata J. 2016;12(2):308‐331. [Google Scholar]
- 27. Klein D. MIMRGNS: Stata module to run margins after mi estimate. Stata Software Components. 2014.
- 28. National Center for Health Statistics . National Health Interview Survey: Survey Description, 2017. Hyattsville, MD: National Center for Health Statistics; 2018. [Google Scholar]
- 29. Moll J, Krieger P, Moreno‐Walton L, et al. The prevalence of lesbian, gay, bisexual, and transgender health education and training in emergency medicine residency programs: What do we know? Acad Emerg Med. 2014;21(5):608‐611. [DOI] [PubMed] [Google Scholar]
- 30. Moll J, Krieger P, Heron SL, Joyce C, Moreno‐Walton L. Attitudes, behavior, and comfort of emergency medicine residents in caring for LGBT patients: what do we know? AEM Educ Train. 2019;3(2):129‐135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. National Center for Health Statistics . A Brief Quality Assessment of the NHIS Sexual Orientation Data. Atlanta, GA: National Center for Health Statistics; 2014. [Google Scholar]
- 32. Eliason MJ, Radix A, McElroy JA, Garbers S, Haynes SG. The “something else” of sexual orientation: measuring sexual identities of older lesbian and bisexual women using National Health Interview Survey questions. Women’s Health Issues. 2016;26:S71‐S80. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Author Matrix