If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address correspondence to: Janelle M. Downing, PhD, 915 Greene Street #349, Arnold School of Public Health, University of South Carolina, Columbia SC 29208
Division of Health Policy and Management, School of Public Health, University of Minnesota, Minneapolis, MinnesotaSchool of Nursing, Oregon Health & Science University, Portland, Oregon
Transgender people experience significant interpersonal and structural discrimination and stigma. However, little is known about the health of transgender people, and even less about the health of specific groups—including male-to-female, female-to-male, and gender-nonconforming transgender populations—despite the variation in social and biological characteristics across groups.
Methods
Data are from the 2014–2016 Behavioral Risk Factor Surveillance System, analyzed in 2017. The study population included 2,221 transgender and 523,080 cisgender respondents from 31 states and one territory. The authors estimated the prevalence and adjusted odds of chronic health conditions, health-related quality of life, disabilities, health behaviors, and health utilization among three transgender groups, when compared separately with cisgender males and cisgender females.
Results
An estimated 0.24% (95% CI=0.21, 0.27) identified as male-to-female; 0.14% (95% CI=0.12, 0.17) identified as female-to-male; and 0.10% (95% CI=0.08, 0.12) identified as gender-nonconforming. All transgender groups experience worse mental health and disabilities; few differences in healthcare access and utilization were observed. Gender-nonconforming people had higher odds of multiple chronic conditions, poor quality of life, and disabilities than both cisgender males and females. Female-to-male people had a higher odds of no exercise and cardiovascular disease compared with cisgender females.
Conclusions
Given the high burden of disabilities; poor mental health; and multiple chronic conditions among transgender (particularly gender-nonconforming) populations, supportive services and care coordination may be consequential levers for improving transgender health.
Introduction
Transgender and gender-nonconforming people (TGNC; people who do not identify with their assigned sex at birth, i.e., gender-minority people) experience interpersonal, structural, and cultural discrimination that has significant negative effects on their physical and mental health. This study contributes to the emerging evidence base on gender-minority health disparities, and is responsive to the National Academy of Medicine’s (formerly known as the Institute of Medicine) 2011 call for an expansion of all aspects of the evidence base for gender-minority health, given the scarcity of relevant research.
This study seeks to address three limitations of the existing literature. First, it uses nationally representative, population-level data that are generalizable beyond the sampling frame. Most prior research used clinical samples, which exclude transgender people who do not seek gender-confirming health care, or community samples that do not generalize across place. One population-level study using the Massachusetts Behavioral Risk Factor Surveillance (BRFSS) found few differences in health by gender identity.
A larger, 2014 BRFSS study of 19 states found that compared with cisgender individuals (those who identify with their assigned sex at birth), transgender individuals had worse general health and more days of poor physical and mental health, yet no differences were observed in the prevalence of chronic illness and health behaviors.
These population-level findings, although limited by small sample sizes, found fewer health disparities than studies using community and clinical samples, which documented high rates of psychological distress, HIV, sexually transmitted infections, disabilities, and risky health behaviors among transgender populations.
Second, this study evaluates transgender health in domains that have previously been unexamined in the literature. These gaps are due, in part, to the absence of gender identity measures in national health surveillance,
as well as inadequate funding for transgender-specific health research: between 1989 and 2011, NIH funded only 43 transgender-specific studies, most of which focused on HIV-related health issues.
Nearly half of studies on transgender health from 2008 to 2014 included measures of mental health and substance abuse, whereas less than 7% of studies included general health conditions, such as diabetes and cancer.
Third, most extant population-level studies did not have adequate samples to examine the health of transgender subpopulations: transgender women (male-to-female, MTF); transgender men (female-to-male, FTM); and gender-nonconforming (GNC) people.
This gap in the literature is significant, as the lived experiences of these populations can be markedly different, both in exposure to interpersonal and structural cissexism (discrimination and stigma against TGNC people), and the amount and quality of engagement with the healthcare system.
For example, the National Transgender Discrimination Survey found that GNC people faced more discrimination in medical care and worse health than other transgender groups.
Furthermore, few studies have compared transgender people with both cisgender males (CMs) and cisgender females (CFs).
This study uses the most comprehensive sample of transgender adults in the U.S. to date, pooling data from 2014 to 2016 probability samples of 31 states and one territory. Because of the structural and interpersonal discrimination faced by transgender people,
the study authors hypothesized that each transgender group would have worse outcomes, greater health risks, and lower healthcare access and utilization than cisgender people. As a result of variation in how gender identity was measured in prior studies, no ex-ante hypothesis exists for how and if the magnitude of differences vary across transgender groups.
Methods
This study used data from 2014, 2015, and 2016 BRFSS, an initiative by the Centers for Disease Control to collect state-specific data on behavioral and health factors among non-institutionalized adults using telephone (landline and cellular) survey methods. More than 400,000 interviews are completed each year. The median response rate was between 47.7% and 48.7% for landlines and 40.5% and 47.2% for cellular in included study years.
The sample included data from residents of 31 states and one U.S. territory (Guam) that completed the sexual orientation and gender identity (SOGI) module during 2014–2016 (Figure 1). Ten states administered the SOGI module in all 3 years, 13 states and Guam in 2 years, and eight states in 1 year.
Figure 1Percent of state population identified as transgender using BRFSS 2014–2016.
Compared with states that did not administer the SOGI, respondents in states with the SOGI module were more likely to be female, older, less non-Hispanic white, wealthier, have lower educational attainment, and be living with a partner and have at least one child in the household (Appendix Table 1, available online).
Measures
Respondents were asked: Do you consider yourself to be transgender? Those who responded yes were prompted to indicate male-to-female, female-to-male, gender-nonconforming, do not know, I’m not sure, or refused to answer (Appendix Figure 1, available online). Five measures of gender identity were created: transgender MTF, transgender FTM, and transgender GNC. Respondents who did not identify as transgender were considered CF if they were female sex, and CM if they were male sex. In 2014 and 2015, the sex of respondents was assigned by the interviewer based on the sound of participant voices. In 2016, sex was assessed by asking respondents, Are you male or female?
Demographics included age; race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, other/multiple); educational attainment (did not complete high school, high school graduate, some college/technical school, college/technical school graduate); income (<$15,000, $15,000–$24,999, $25,000–$34,999, $35,000–$49,999, ≥$50,000); employment status (employed, unemployed, not looking for work, retired, unable to work); health insurance; housing tenure (owner, renter, other arrangement); marital status (married/partnered, divorced/separated/widowed, single); presence of children in the household; and sexual orientation (heterosexual, lesbian/gay, bisexual).
Five most prevalent chronic conditions (i.e., >10% of the sample) were included. Respondents were asked, Has a doctor, nurse, or other professional ever told you that you had any of the following: (1) coronary heart disease or myocardial infarction (CHD/MI), (2) asthma, (3) arthritis (some form of arthritis, rheumatoid arthritis, gout, lupus, or fibromyalgia), (4) diabetes (excluding pregnancy-related), or (5) depressive disorder (depression, major depression, dysthymia, or minor depression)? Respondents were classified as having multiple chronic conditions if they reported two or more of the prior five conditions; cancer (excluding skin cancer); stroke; chronic obstructive pulmonary disease; emphysema; chronic bronchitis; or kidney disease.
Four binary measures of Health Related Quality of Life Measures were included: (1) poor/fair general health (reference: good or better health); (2) frequent poor physical health (≥14 days of poor physical health, including physical illness and injury in the prior 30 days); (3) frequent mental distress (≥14 days of poor mental health including stress, depression, and problems with emotions) in the past 30 days; and (4) frequent limitations (≥14 days during which poor physical or mental health led to activity restrictions [such as self-care, work, recreation] in the past 30 days).
Three of the most prevalent (i.e., >10%) functional types of disabilities were included: disability in mobility (having serious difficulty walking or climbing stairs); disability in cognition (serious difficulty concentrating, remembering, and making decisions because of physical, mental, or emotional conditions); and disability in independent living (having difficulty doing errands alone, such as visiting a doctor’s office or shopping, because of physical, mental, or emotional conditions). The authors defined multiple disabilities as having at least one of the prior three disabilities; disability in vision (being blind or having serious difficulty seeing); or disability in self-care (difficulty dressing or bathing).
Four health behaviors were examined: no exercise (no physical activity/exercise during the past 30 days other than regular job); current smoking (smoking cigarettes every day or some days); heavy episodic drinking (four [for female sex identified] or five [for male sex identified] or more drinks on at least one occasion in the past 30 days; this variable was assigned based on interviewer or self-reported sex, irrespective of gender identity); and obesity (BMI ≥30).
Six binary measures of healthcare access and utilization were included: (1) had a routine primary care visit in the past year; (2) had a routine dental visit in the past year; (3) delayed accessing medical care because of cost; (4) has a primary care provider; (5) ever tested for HIV; and (6) influenza vaccine in the past year.
Statistical Analysis
Data were pooled and reweighted to be representative of each state in the sample. Survey weights in each year were multiplied by the number of all respondents surveyed in that year divided by the number of respondents surveyed during the period.
These recalibrated weights were used to calculate the pooled (period) prevalence of the transgender population.
The weighted distribution of demographic and health variables was estimated for each gender identity group (additional demographic statistics presented in Appendix Table 3 [available online] for individuals who reported don’t know/not sure, refused, or missing); Pearson chi-square tests were used to compare groups.
Logistic regressions were used to estimate adjusted odds of each health measure for the three TGNC groups, first using CMs as the reference, and then using CFs as the reference. Adjusted models controlled for age, race/ethnicity, relationship status, educational attainment, health insurance coverage, and state of residence. An additional analysis was conducted to estimate the ORs for each TGNC group controlling for age only (Appendix Table 4, available online). All analyses were conducted separately for each outcome using Stata, version 13, with pooled survey weights in February 2018.
Results
A total of 525,301 respondents were asked the transgender identity question between 2014 and 2016 in BRFSS, of which 0.48% (95% CI=0.44, 0.53) of respondents identified as transgender. By gender identity, 0.24% (95% CI=0.21, 0.27) identified as MTF, 0.14% (95% CI=0.12, 0.17) identified as FTM, 0.10% (95% CI=0.08, 0.12) identified as GNC; whereas 50.9% (95% CI=50.6, 51.2] were CF, and 47.1% (95% CI=46.8, 47.4) were CM; and 0.60% (n=2,964) of respondents answered they did not know or were not sure, and 0.80% (n=4,281) refused. A higher proportion of young, Latinx, and divorced/separated people reported they did not know or were not sure (Appendix Table 3, available online).
State-specific proportions of transgender respondents ranged from 0.82% (95% CI=0.58, 1.17) in West Virginia to 0.26% (95% CI=0.13, 0.49) in Wyoming (Figure 1, Appendix Table 2 [available online] provides proportions in remaining sample states).
Compared with cisgender respondents, transgender respondents were younger, more racially/ethnically diverse, and more likely to identify as lesbian, gay, or bisexual (Table 1). Transgender people had lower educational attainment and income and were more likely to be unemployed, never married, and uninsured. Despite these commonalities, differences between transgender groups existed. Notably, FTM transgender people were the most socioeconomically disadvantaged: This group was the least likely to have a college degree, be homeowners, make >$50,000, or to have health insurance. FTM transgender people were also the most likely to have children living in their household.
Table 1Demographic Characteristics of Transgender and Cisgender Adults, 2014–2016 BRFSS
Characteristics
Male-to-female, %
Female-to-male, %
Gender-nonconforming, %
Cisgender female, %
Cisgender male, %
Total, n (%)
1,073 (0.24)
699 (0.14)
449 (0.10)
297,810 (50.91)
218,021 (47.11)
Age, years (p<0.001)
18–24
18.2
21.4
26.4
11.3
12.9
25–34
13.7
17.8
18.5
15.4
16.7
35–44
15.6
22.6
14.9
16.0
16.7
45–54
17.1
11.2
8.9
17.7
18.1
55–64
20.8
12.1
14.7
17.4
17.3
≥65
14.8
14.8
14.7
22.1
18.3
Race/ethnicity (p<0.001)
White
56.0
48.9
52.0
62.8
62.7
Black
13.2
15.2
14.6
11.2
10.4
Asian
7.6
3.9
4.4
5.8
4.9
Other/multiracial
4.0
3.6
7.1
2.4
2.7
Hispanic/Latinx
17.4
26.0
20.8
16.3
14.0
Education (p<0.001)
Less than high school
26.9
27.4
16.5
13.1
14.0
High school graduate
34.2
39.4
30.0
27.7
30.4
Some college
25.4
22.6
35.1
32.3
29.0
College graduate
13.0
9.8
18.5
26.8
26.3
Income (p<0.001)
<$15,000
17.0
16.9
15.8
10.3
7.5
$15,000–$24,999
19.2
20.2
19.0
15.1
12.9
$25,000–$34,999
13.6
13.9
9.4
9.0
8.7
$35,000–$49,999
9.6
9.8
11.7
11.4
12.6
$≥50,000
27.5
22.5
33.2
38.4
46.3
Employment status (p<0.001)
Employed
56.3
50.4
45.1
50.2
64.4
Unemployed
8.3
8.1
8.0
5.1
5.8
Not looking for work
8.7
15.5
16.3
17.8
5.4
Retired
15.3
12.4
18.3
19.1
17.6
Unable to work
10.3
9.7
11.7
7.2
6.0
Health insurance (p<0.001)
Insured
81.4
76.0
84.1
90.1
87.0
Uninsured
17.3
23.1
13.2
9.6
12.4
Marital status (p<0.001)
Married/partnered
52.5
43.0
44.0
54.4
57.9
Divorced/separated/widowed
18.0
22.0
17.0
24.4
15.4
Never married
29.2
33.9
37.9
20.7
26.2
Sexual orientation (p<0.001)
Straight/heterosexual
74.6
69.8
52.3
93.1
94.1
Lesbian or gay
4.0
11.4
11.1
1.2
2.0
Bisexual
11.5
10.2
22.9
2.4
1.4
Homeowner (p<0.001)
56.1
50.2
57.6
69.6
68.4
At least one child in the household (p<0.001)
30.0
46.4
26.9
38.6
34.4
Note: 2014–2016 pooled data are weighted; does not sum to 100% because “missing” for each variable was not reported here.
Table 2 shows the prevalence of each health outcome across groups. Table 3 shows the adjusted odds of transgender groups, separately compared with CMs and CFs. Compared with CMs, MTF transgender people had higher odds of diagnosed depression (AOR=2.02, 95% CI=1.52, 2.69); compared with CFs, this group had higher odds of CHD/MI (AOR=2.07, 95% CI=1.37, 3.13) and lower odds of asthma (AOR=0.55, 95% CI=0.39, 0.76). FTM transgender people had higher odds of diagnosed depression than both CMs (AOR=3.14, 95% CI=2.07, 4.77) and CFs (AOR=1.58, 95% CI=1.03, 2.42). They also had higher odds of arthritis (AOR=1.57, 95% CI=1.15, 2.13) and multiple chronic conditions (AOR=1.88, 95% CI=1.32, 2.67) than CMs, and higher odds of CHD/MI (AOR=1.90, 95% CI=1.24, 2.9) than CFs. Compared with both CMs and CFs, GNC transgender people had higher odds of CHD/MI (AOR=2.31, 95% CI=1.10, 4.84, and AOR=6.42, 95% CI=2.33, 17.70, respectively); diagnosed depression (AOR=4.31, 95% CI=2.69, 6.89, and AOR=2.29, 95% CI=1.43, 3.67, respectively); and multiple chronic conditions (AOR=2.90, 95% CI=1.71, 4.93, and AOR=2.16, 95% CI=1.27, 3.68, respectively).
Table 2Health Outcomes, Health Behaviors, and Healthcare Access and Services by Gender Identity
Characteristics
Male-to-female,n(%) (n=1,073)
Female-to-male,n(%) (n=699)
Gender-nonconforming,n(%) (n=449)
Cisgender female,n(%) (n=297,810)
Cisgender male,n(%) (n=218,021)
Chronic conditions
Multiple conditions
611 (48.1)
423 (46.7)
250 (50.7)
165,888 (45.8)
104,956 (38.3)
CHD or MI
131 (8.0)
72 (6.6)
57 (17.8)
18,862 (4.8)
22,818 (9.0)
Asthma
130 (9.9)
117 (13.6)
82 (19.1)
45,954 (15.8)
23,525 (11.4)
Arthritis
350 (24.0)
270 (24.3)
126 (21.7)
116,724 (29.6)
63,357 (21.3)
Diabetes
200 (14.5)
127 (11.1)
66 (13.0)
38,060 (10.5)
30,395 (11.4)
Depression
243 (24.2)
201 (31.1)
152 (38.2)
65,728 (21.1)
29,304 (12.5)
HRQOL-4 measures
Poor/fair health
260 (18.0)
192 (21.0)
132 (33.0)
55,360 (18.5)
37,786 (16.7)
Poor physical health
177 (15.3)
142 (13.7)
76 (17.1)
41,665 (13.0)
25,277 (10.7)
Mental distress
153 (15.1)
134 (25.2)
103 (28.1)
33,372 (12.7)
17,613 (9.3)
Limitations
131 (22.9)
96 (18.7)
70 (28.0)
27,046 (15.4)
16,381 (14.9)
Disabilities
Mobility
242 (17.6)
194 (17.7)
105 (19.9)
57,564 (15.5)
29,814 (11.4)
Cognition
174 (19.2)
147 (17.4)
110 (32.2)
31,029 (11.1)
18,387 (8.9)
Independent living
111 (10.2)
103 (11.0)
81 (20.8)
26,269 (7.9)
11,439 (5.0)
≥1 disability
373 (33.0)
280 (32.7)
193 (45.6)
81,435 (24.3)
46,186 (19.5)
Health behavior
No exercise
312 (30.8)
251 (38.7)
129 (34.1)
80,182 (26.2)
49,257 (22.4)
Current smoker
213 (45.7)
136 (62.3)
90 (46.4)
41,204 (39.8)
35,394 (39.3)
Heavy episodic drinker
160 (42.6)
73 (30.5)
73 (36.7)
25,680 (23.8)
39,113 (36.7)
Obese
334 (31.3)
232 (26.9)
143 (29.0)
81,464 (29.6)
66,155 (29.9)
Healthcare access and utilization
No primary care in prior year
254 (29.7)
156 (27.4)
119 (32.3)
60,034 (24.5)
59,355 (33.1)
No dental visit in prior year
280 (47.6)
165 (39.2)
105 (47.1)
56,333 (28.6)
47,218 (33.8)
No visit because of cost
149 (20.6)
94 (26.1)
70 (18.7)
29,916 (13.5)
18,276 (11.2)
No primary health provider
185 (25.3)
112 (34.2)
90 (23.3)
29,565 (15.4)
40,102 (25.7)
Never tested for HIV
697 (60.9)
448 (56.8)
271 (55.3)
204,928 (61.9)
147,991 (64.6)
No influenza vaccine in past year
612 (63.4)
376 (67.4)
264 (69.3)
148,664 (56.6)
122,644 (64.2)
Note: The proportions (%) are weighted using survey weights for 2014–2016 BRFSS.
Table 3Adjusted Odds of Health Outcomes, Health Behavior, and Healthcare Access and Utilization by Gender Identity
Variable
Male-to-female (n=1,073)
Female-to-male (n=699)
Gender-nonconforming (n=449)
Ref: Cisgender male, AOR (95% CI)
Ref: Cisgender female, AOR (95% CI)
Ref: Cisgender male, AOR (95% CI)
Ref: Cisgender female, AOR (95% CI)
Ref: Cisgender male, AOR (95% CI)
Ref: Cisgender female, AOR (95% CI)
Chronic conditions
Multiple condition
1.085 (0.810, 1.453)
0.793 (0.592, 1.064)
1.876** (1.321, 2.666)
1.418 (0.996, 2.019)
2.899** (1.706, 4.926)
2.159** (1.269, 3.675)
CHD/MI
1.008 (0.666, 1.526)
2.068** (1.366, 3.133)
0.907 (0.595, 1.382)
1.895** (1.240, 2.894)
2.305* (1.098, 4.841)
6.415** (2.325, 17.702)
Asthma
0.814 (0.587, 1.127)
0.545** (0.391, 0.759)
1.092 (0.720, 1.657)
0.765 (0.502, 1.165)
1.549 (0.874, 2.745)
0.982 (0.548, 1.759)
Arthritis
1.215 (0.852, 1.734)
0.834 (0.584, 1.191)
1.566** (1.147, 2.138)
1.093 (0.800, 1.494)
1.340 (0.759, 2.367)
0.919 (0.520, 1.627)
Diabetes
1.129 (0.761, 1.675)
1.425 (0.960, 2.116)
1.037 (0.740, 1.452)
1.333 (0.951, 1.869)
1.367 (0.794, 2.353)
1.728* (1.003, 2.978)
Depression
2.024** (1.522, 2.693)
1.061 (0.796, 1.412)
3.141** (2.068, 4.770)
1.578* (1.031, 2.416)
4.306** (2.690, 6.893)
2.290** (1.429, 3.671)
HRQOL-4 measures
Poor/fair health
0.794 (0.558, 1.129)
0.728 (0.511, 1.038)
0.968 (0.671, 1.396)
0.844 (0.589, 1.210)
2.514** (1.432, 4.414)
2.189** (1.242, 3.859)
Poor physical health
1.213 (0.836, 1.760)
1.026 (0.706, 1.491)
1.134 (0.807, 1.595)
0.997 (0.709, 1.401)
1.831* (1.106, 3.032)
1.559 (0.938, 2.589)
Limitations
1.620* (1.063, 2.469)
1.624* (1.062, 2.484)
1.119 (0.635, 1.973)
1.206 (0.685, 2.124)
2.598** (1.331, 5.074)
2.665** (1.360, 5.222)
Mental distress
1.545* (1.087, 2.196)
1.026 (0.721, 1.461)
2.864** (1.753, 4.681)
1.808* (1.091, 2.996)
3.424** (2.053, 5.712)
2.139** (1.270, 3.603)
Disabilities
Mobility
1.454** (1.001, 2.110)
1.122 (0.772, 1.632)
1.604** (1.125, 2.287)
1.250 (0.876, 1.786)
2.166*** (1.293, 3.626)
1.686* (1.004, 2.833)
Cognition
1.898** (1.347, 2.674)
1.434* (1.014, 2.028)
1.609* (1.072, 2.415)
1.253 (0.833, 1.885)
4.689** (2.873, 7.654)
3.656** (2.237, 5.975)
Independent living
1.893*** (1.242, 2.889)
1.187 (0.773, 1.821)
1.943*** (1.323, 2.852)
1.262 (0.859, 1.853)
5.144*** (3.113, 8.497)
3.361** (2.034, 5.554)
≥1 disability
1.677*** (1.207, 2.330)
1.329 (0.955, 1.849)
1.700*** (1.173, 2.463)
1.403 (0.965, 2.038)
3.627*** (2.281, 5.768)
2.929** (1.836, 4.675)
Health behavior
No exercise
1.224 (0.902, 1.659)
1.006 (0.740, 1.368)
1.850** (1.305, 2.621)
1.510* (1.062, 2.147)
1.874* (1.090, 3.223)
1.583 (0.919, 2.727)
Current smoker
0.969 (0.650, 1.444)
1.124 (0.799, 1.582)
1.864* (1.083, 3.208)
1.616 (0.925, 2.825)
1.006 (0.559, 1.812)
1.003 (0.630, 1.598)
Heavy episodic drinker
1.148 (0.765, 1.723)
1.813** (1.263, 2.603)
0.548 (0.281, 1.072)
1.032 (0.631, 1.688)
0.696 (0.388, 1.248)
1.456 (0.886, 2.394)
Obese
0.981 (0.951, 1.011)
0.956 (0.686, 1.332)
0.941 (0.676, 1.309)
0.798 (0.560, 1.137)
0.781 (0.548, 1.111)
0.985 (0.620, 1.565)
Healthcare access and services
No primary care visit
0.780 (0.516, 1.180)
1.144 (0.755, 1.734)
0.625* (0.419, 0.933)
0.866 (0.581, 1.292)
0.919 (0.591, 1.430)
1.392 (0.897, 2.161)
No dental visit
1.369 (0.917, 2.044)
1.754** (1.172, 2.626)
1.02 (0.567, 1.836)
1.328 (0.737, 2.394)
1.613 (0.855, 3.043)
2.176* (1.153, 4.105)
No visit because of cost
1.612*** (1.065, 2.395)
1.159 (0.772, 1.740)
2.147** (1.065, 2.395)
1.543 (0.880, 2.707)
2.147*** (1.051, 3.171)
1.368 (0.789, 2.371)
No primary health provider
0.821 (0.528, 1.277)
1.551 (0.993, 2.421)
1.095 (0.671, 1.786)
2.011** (1.226, 3.298)
0.774 (0.461, 1.301)
1.538 (0.919, 2.575)
Never tested for HIV
0.864 (0.640, 1.166)
0.985 (0.728, 1.331)
0.847 (0.538, 1.335)
0.989 (0.625, 1.564)
0.717 (0.466, 1.101)
0.842 (0.545, 1.301)
No flu shot in past year
0.872 (0.649, 1.170)
1.147 (0.853, 1.542)
0.921 (0.658, 1.290)
1.235 0.880,1.734
1.126 (0.716, 1.772)
1.465 (0.929, 2.311)
Note: Two models (one with cisgender male, one with cisgender female) were fit separately for each outcome were fit separately with survey weights, adjusted for age, race/ethnicity, relationship status, educational attainment, health insurance coverage, and state of residence. Boldface indicates statistical significance (*p<0.05; **p<0.01; ***p<0.001).
MTF transgender people had higher odds of frequent limitations compared with CMs (AOR=1.62, 95% CI=1.06, 2.47) and CFs (AOR=1.62, 95% CI=1.06, 2.48), and frequent mental distress (AOR=1.55, 95% CI=1.09, 2.20) compared with CMs. FTM transgender people had higher odds of mental distress than CMs (AOR=2.86, 95% CI=1.75, 4.68) and CFs (AOR=1.81, 95% CI=1.09, 3.00). GNC people had odds that were 1.83–3.42 higher than both CMs and CFs, except for poor physical health compared with CFs.
The three transgender groups had higher odds of the four measured disabilities than CMs. GNC transgender people additionally had higher odds of the four measures than CFs, whereas MTF people had higher odds of reporting disabilities in cognition.
MTF transgender people did not differ from cisgender people for any health behaviors, except for having higher odds of heavy episodic drinking when compared with CFs (AOR=1.81, 95% CI=1.26, 2.60). FTM transgender people had higher odds of no exercise than CMs (AOR=1.85, 95% CI=1.31, 2.62) and CFs (AOR=1.51, 95% CI=1.06, 2.15), as well as higher odds of currently smoking than CMs (AOR=1.86, 95% CI=1.08, 3.21). GNC people had higher odds of no exercise compared with CMs (AOR=1.87, 95% CI=1.09, 3.22).
All three transgender groups had higher odds of delaying accessing medical care because of cost than CMs. MTF and GNC transgender people had higher odds of no dental visit compared with CFs. FTM people had lower odds of no primary care visit in the last year than CMs, yet higher odds of not having a primary care provider compared with CFs.
Discussion
This is one of the first studies to use a national probability sample to compare a comprehensive panel of health outcomes between transgender and cisgender populations, and the first to examine MTF, FTM, and GNC transgender people to both CMs and CFs. Given that 0.48% of the study sample identified as transgender, this translates to an estimated 1.5 million transgender Americans, of whom 770,000 identify as MTF, 458,000 as FTM, and 332,000 as GNC. These results are similar to prior estimates of the transgender population, which was between 0.35% and 0.53% of the population.
Mental health and medical health disparities in 5135 transgender veterans receiving healthcare in the Veterans Health Administration: a case–control study.
transgender people experienced a higher burden of poor mental health than cisgender people. This is likely due to minority stress resulting from interpersonal and structural cissexist discrimination. A majority of Americans believe that a person’s gender identity cannot differ from the sex they were assigned at birth, effectively precluding the gender self-determination of transgender people.
Every year, numerous jurisdictions pass or consider bills that discriminate against transgender people. Consistent with previous research, there was variation in mental health outcomes across transgender groups
; more research is needed to understand the possibly differential impact of local, state, and national anti-discrimination and healthcare provision policies on the mental health of transgender people.
Transgender people also faced a higher burden of disability than cisgender people. This is consistent with recent findings that 71% of gender-minority Medicare beneficiaries were entitled to disability and experienced more disabilities compared with non–gender minority beneficiares.
These findings highlight the urgent need to understand both the causes of disability among TGNC people and the intersectional experiences of transgender people with disabilities in order to improve culturally competent care.
FTM and GNC transgender respondents had higher odds of multiple chronic conditions. These groups might face issues in care coordination across systems that are not organizationally linked,
Transgender people with multiple chronic conditions might be more likely to have an unmet need for caregiving; although 90% of the in-home long-term care are unpaid family or informal caregivers,
few differences in preventive care utilization were observed, except for delaying care because of cost compared with CMs. This is consistent with prior research on preventive care in Medicare data
and could be attributed to more healthcare use among this group compared with cisgender peers. These results differ from a prior estimate using the 2014–2015 BRFSS data that found GNC people reported delaying care because of cost and not having a primary care visit in the last year compared with CFs.
Research on how health access and insurance varies over time and place is critical.
Despite being more socioeconomically advantaged than both FTM and MTF transgender people, GNC transgender people experienced the most disparities in health outcomes. It is unclear why this population appears to be at greatest risk of poor health. One possibility is that GNC people are more likely to be perceived as transgender, putting them at greater risk of discrimination and stigma. However, the 2015 U.S. Transgender Survey found that GNC people were least likely to be perceived as transgender by others.
there were little differences in chronic health conditions or health behaviors of MTF respondents compared with CMs. This is likely to be an artifact of the survey design that was a telephone survey of non-institutionalized adults. Transgender women face deep poverty, homelessness,
Racial/ethnic disparities in history of incarceration, experiences of victimization, and associated health indicators among transgender women in the U.S.
Thus, most marginalized transgender women are likely to be “missing,” which biases the results toward the null. Alternatively, age of transition and chronic disease progression linked to biological sex could have confounded results.
Limitations
This study has several limitations. This sample includes a majority, but not all states. However, the population not sampled was demographically similar; therefore, these estimates are likely representative of the U.S. population. Further, until 2016, BRFSS respondents’ sex was attributed by the interviewer using vocal timbre,
which limits the ability to identify biological sex (only in 2016 were respondents asked to report their sex). This means that sex-specific variables, such as alcohol use, might be more inaccurate for transgender participants. Next, it would have been informative to compare respondents’ sex at birth, gender expression, and stigma over the life course.
Another limitation is that BRFSS did not include any questions about transgender-specific health care, which prohibited this study from describing differences in these healthcare experiences by gender identity. Finally, it is plausible that the manner in which gender identity was assessed created exposure misclassification bias, whereby some respondents with gender-minority identities other than those included in BRFSS might have been misclassified as cisgender, refused to answer, or did not know/were not sure. National surveys should consider following Gender Identity in U.S. Surveillance Group recommendations.
Despite these limitations, these findings are important for clinicians seeking to understand how to prevent and detect chronic diseases early in transgender patients, address issues of healthcare access and coordination of care, and improve cultural competency for transgender patients.
Acknowledgments
No financial disclosures were reported by the authors of this paper.
Mental health and medical health disparities in 5135 transgender veterans receiving healthcare in the Veterans Health Administration: a case–control study.
Racial/ethnic disparities in history of incarceration, experiences of victimization, and associated health indicators among transgender women in the U.S.
31870-1&id=gr1.jpg){kind=link}