HIV Prevalence, Risk Behaviors, Health Care Use, and Mental Health Status of Transgender Persons: Implications For Public Health Intervention
HIV Prevalence, Risk Behaviors, Health Care Use, and Mental Health Status of Transgender Persons: Implications For Public Health Intervention
HIV Prevalence, Risk Behaviors, Health Care Use, and Mental Health Status of Transgender Persons: Implications For Public Health Intervention
A B S T R A C T
Objectives. This study described
HIV prevalence, risk behaviors, health
care use, and mental health status of
male-to-female and female-to-male transgender persons and determined factors
associated with HIV.
Methods. We recruited transgender
persons through targeted sampling,
respondent-driven sampling, and agency
referrals; 392 male-to-female and 123
female-to-male transgender persons were
interviewed and tested for HIV.
Results. HIV prevalence among
male-to-female transgender persons was
35%. African American race (adjusted
odds ratio [OR] = 5.81; 95% confidence
interval [CI] = 2.82, 11.96), a history of
injection drug use (OR =2.69; 95% CI=
1.56, 4.62), multiple sex partners (adjusted OR = 2.64; 95% CI = 1.50, 4.62),
and low education (adjusted OR = 2.08;
95% CI=1.17, 3.68) were independently
associated with HIV. Among female-tomale transgender persons, HIV prevalence (2%) and risk behaviors were much
lower. Most male-to-female (78%) and
female-to-male (83%) transgender persons had seen a medical provider in the
past 6 months. Sixty-two percent of the
male-to-female and 55% of the femaleto-male transgender persons were depressed; 32% of each population had attempted suicide.
Conclusions. High HIV prevalence
suggests an urgent need for risk reduction
interventions for male-to-female transgender persons. Recent contact with
medical providers was observed, suggesting that medical providers could provide an important link to needed prevention, health, and social services. (Am
J Public Health. 2001;91:915921)
Kristen Clements-Nolle, MPH, Rani Marx, PhD, MPH, Robert Guzman, BA,
and Mitchell Katz, MD
Transgender is a term used to describe
individuals who have a persistent and distressing discomfort with their assigned gender.1 Such individuals were born anatomically
as one biological sex but live their lives to
varying degrees as the opposite sex. Qualitative research suggests that male-to-female and
female-to-male transgender individuals experience severe employment, housing, and health
care discrimination, and many engage in behaviors that put them at risk for HIV.26
High HIV prevalence has been found in
small published studies of male-to-female
sex workers recruited from street locations
in Atlanta, Ga7 (68%), and Tel Aviv, Israel8
(11%); a drug treatment center in Rome,
Italy9 (46%); and a clinic in Italy10 (57%). In
addition, research comparing male-to-female
sex workers with male and female sex workers in the same neighborhoods has consistently found higher HIV prevalence in transgender samples.7,8,1114
Male-to-female transgender persons not
currently involved in sex work were included
in 4 published studies that sampled in clinics,
communities, and prison systems.1519 HIV
prevalence was lower than in studies of maleto-female sex workers, but sexual and injection drug risk behaviors were common.1517
Two of these studies compared male-to-female
transgender persons with males, females recruited in the same settings, or both, and found
a higher prevalence of risk behaviors among
male-to-female transgender persons.18,19
Studies of male-to-female transgender
individuals have relied on small convenience
samples and predominantly have reported
data on individuals seeking HIV testing or
self-reporting HIV seropositivity.9,10,1519
Three studies included street-based sampling
methods but lacked power to determine independent predictors of HIV infection.7,8,17
No studies have quantitatively assessed HIV
risk among female-to-male transgender individuals. To address these limitations, we
conducted interviews and HIV testing with
Methods
Subjects and Recruitment
Our targeted sampling plan20 was based
on information gathered from focus groups
with 100 transgender persons,6 30 key informant interviews, and social mapping in
neighborhoods thought to have high concentrations of transgender persons. We used participatory research strategies whereby members of the target population were involved in
development, implementation, and interpretation of the study.21,22
Three male-to-female transgender individuals (African American, Filipina, and Latina) and 3 female-to-male transgender persons (1 Vietnamese and 2 White) were hired
as interviewers. Staff received 35 training sessions, including HIV counseling, interviewThe authors are with the San Francisco Department
of Public Health, San Francisco, Calif. Mitchell
Katz is also with the University of California, San
Francisco.
Requests for reprints should be sent to Kristen Clements-Nolle, MPH, Epidemiology and Evaluation Section, San Francisco Department of Public Health, 25 Van Ness, #500, San Francisco, CA
94102 (e-mail: [email protected]).
This article was accepted July 29, 2000.
Note. All study protocols and materials received
approval from the Committee on Human Research
at the University of California, San Francisco, and
all participants provided informed consent. The views
expressed are those of the authors and do not necessarily reflect those of the funding agencies.
915
ing, confidentiality, harm reduction, street outreach, suicide prevention, and referrals. From
July through December 1997, staff recruited
individuals from street settings, bars, and social
gatherings catering to transgender persons, and
agencies referred clients to the study. We also
initiated respondent-driven sampling23; each
study participant received an incentive for eligible persons recruited.
Recruited individuals called a toll-free
number and were screened for eligibility. Individuals were eligible if they (1) were 18
years or older; (2) lived, worked, or socialized in San Francisco; (3) spoke English,
Spanish, Vietnamese, or Tagalog; and (4)
stated that their primary gender was transgender, transsexual, bigender, transvestite,
cross-dresser, intersexed, or the opposite sex
of that at birth.
Eligible subjects were scheduled for an
interview at 1 of 8 community-based organizations and could request the gender or
race/ethnicity of their interviewer. Interviewers identified and prevented 3 participants from reenrolling. A composite variable of unique personal identifiers also was
created; no duplicates were identified with
this variable.
Written informed consent was obtained
before interviewing and HIV testing. Two
weeks later, participants received HIV test
results and further counseling and referrals
from the same interviewer. Missed follow-up
appointments could be rescheduled, but staff
did not try to locate participants who missed
appointments. Participants received $40 for
the interview, $10 for the follow-up appointment, and $5 for up to 5 eligible persons they
referred.
Statistical Methods
Results
Participation
We screened 645 individuals, of whom
586 (91%) were eligible for inclusion in the
study. Of the eligible participants, 523 (89%)
completed the interview and HIV test. No demographic differences were found between eligible subjects who did and did not complete
the interview. Eight intersexed individuals
(born with ambiguous or both male and female
genitalia) were excluded from this analysis,
because they could not be classified as maleto-female or female-to-male persons.The final
sample comprised 392 male-to-female and 123
female-to-male transgender individuals.
The most common (nonmutually exclusive) ways that participants were recruited
for the study were respondent-driven sampling (39%), outreach by study staff (38%),
flyer recruitment (14%), and referrals from
agencies (10%). Male-to-female transgender
persons were more likely to report respondentdriven sampling recruitment (42% vs 32%,
P=.04), and female-to-male transgender persons were more likely to be recruited by interviewers (58% vs 32%, P<.001). There were
no other differences in demographics or HIV
prevalence by recruitment type. Seventy percent of the participants returned for their HIV
test results.
Sociodemographics
Most male-to-female and female-to-male
participants self-identified as transgender, the
June 2001, Vol. 91, No. 6
Female-to-Male
(n = 123), No. (%)
Pa
145 (37)
135 (34)
...
100 (26)
12 (3)
46 (37)
...
41 (33)
26 (21)
10 (8)
104 (27)
106 (27)
106 (27)
49 (13)
24 (6)
34 (1867)
12 (10)
14 (11)
82 (67)
9 (7)
5 (4)
36 (1961)
< .001
271 (69)
82 (21)
21 (5)
15 (4)
43 (35)
40 (33)
15 (12)
22 (18)
< .001
113 (29)
278 (71)
744 (557346)
207 (53)
185 (47)
255 (65)
119 (30)
137 (35)
5 (4)
118 (96)
1100 (1006000)
96 (78)
26 (21)
36 (29)
6 (5)
2 (2)
.09
.25
< .001
<.001
< .001
< .001
< .001
< .001
opposite gender of that at birth, or transsexual (Table 1). The median age of male-tofemale and female-to-male participants was
similar, but the 2 populations differed significantly on all other sociodemographic characteristics. Seventy-three percent of the maleto-female persons identified as non-White,
whereas two thirds of the female-to-male persons identified as White. Male-to-female transgender individuals were more likely than
female-to-male transgender individuals to
identify as heterosexual, to report prior incarceration, and to have unstable housing, low
education, and low monthly income. The most
common ways (nonmutually exclusive) that
male-to-female transgender individuals obtained money in the past 6 months included
part- or full-time employment (40%), sex work
(32%), Supplemental Security Income and Social Security Disability Insurance (29%), and
General Assistance (23%), whereas 81% of
the female-to-male were employed (not shown
in table).
June 2001, Vol. 91, No. 6
917
Adjusted OR
(95% CI)a
(n = 381)
106 (27)
104 (27)
106 (27)
73 (19)
23 (17)
65 (47)
31 (23)
18 (13)
Reference
5.81 (2.82, 11.96)
0.80 (0.37, 1.72)
1.10 (0.49, 2.48)
<.001
.56
.81
133 (34)
149 (38)
110 (28)
35 (26)
65 (47)
37 (27)
Reference
1.71 (0.95, 3.07)
0.93 (0.47, 1.84)
.07
.83
278 (71)
113 (29)
255 (65)
86 (63)
51 (37)
108 (79)
Reference
2.08 (1.17, 3.68)
1.12 (0.62, 2.03)
.01
.71
212 (54)
180 (46)
134 (34)
256 (65)
330 (84)
213 (54)
312 (80)
231 (59)
51 (37)
86 (63)
69 (50)
108 (79)
127 (93)
94 (69)
124 (91)
92 (67)
Reference
2.64 (1.50, 4.62)
2.69 (1.56, 4.62)
1.67 (0.94, 2.97)
2.34 (0.95, 5.78)
1.38 (0.82, 2.35)
0.82 (0.37, 1.82)
1.02 (0.59, 1.75)
<.001
<.001
.08
.06
.23
.62
.95
HIV+
(n = 137),
No. (%)
80 (20)
69 (18)
99 (25)
144 (37)
146 (37)
90 (62)
144 (37)
63 (44)
128 (33)
36 (28)
27 (20)
24 (18)
41 (30)
45 (33)
66 (48)
36 (55)
55 (40)
25 (45)
46 (34)
11 (24)
HIV
(n = 255),
No. (%)
Pa
32 (13)
17 (53)
17 (53)
10 (31)
<.001
.37
.58
.95
53 (21)
45 (18)
58 (23)
99 (39)
80 (31)
54 (68)
89 (35)
38 (43)
82 (32)
25 (30)
.10
<.001
.11
.30
.75
.78
.43
918
TABLE 4Health Care Use and Mental Health Status of Male-to-Female and
Female-to-Male Transgender Persons: San Francisco, Calif, 1997
Male-to-Female
Female-to-Male
(n = 392), No. (%) (n = 123), No. (%)
Health care use
Health insurance, current
None
202 (52)
Public
132 (34)
Private
56 (14)
Clinic or doctor visit, past 6 mo
306 (78)
Emergency department visit, past 6 mo
98 (25)
Hormone use, past 6 mo
288 (73)
b
Injected hormones, past 6 mo
121 (42)
Source of hormones, past 6 mob
Medical provider
204 (71)
Streets, black market, friends
84 (29)
Mental health status
Mental health hospitalization, ever
87 (22)
Suicide attempt, ever
127 (32)
Depression: CES-D score 16c
242 (62)
Pa
51 (41)
14 (11)
58 (47)
102 (83)
22 (18)
65 (53)
63 (97)
<.001
63 (97)
2 (3)
<.001
24 (20)
39 (32)
68 (55)
.50
.89
.17
.25
.10
<.001
<.001
Discussion
To our knowledge, this is the largest quantitative study to describe HIV risk, health care
use, and mental health status of male-to-female
and female-to-male transgender individuals.
Our data show the diversity that exists within
the transgender community. For example, gender identification does not determine sexual
orientation. Male-to-female study participants
primarily had sex with the opposite gender (males) and self-identified as heterosexual, whereas female-to-male participants were
more likely to have sex with other males and
transgender individuals and self-identified as
gay or bisexual. We also found major differences between male-to-female and female-tomale participants with respect to demographic
and HIV risk characteristics. In particular,
male-to-female transgender persons were more
socioeconomically disadvantaged and had
higher HIV prevalence.
Our estimate of HIV prevalence among
male-to-female transgender persons is higher
than estimates from studies with gay men and
injection drug users of the same age in San Francisco.3134 Half of the male-to-female transgenJune 2001, Vol. 91, No. 6
cult to maintain an erection,16 and playing the receptive role is viewed as more feminine.3
Inconsistent condom use during receptive
anal sex was commonly reported, particularly
with main partners. This finding is consistent
with past research on male-to-female transgender persons7,38 and calls for interventions
that decrease barriers to condom use with main
partners. Risk reduction interventions also are
needed for female-to-male transgender persons who have sex with men, because condom
use was infrequent.
According to qualitative research, many
male-to-female transgender persons turn to sex
work because they face severe employment
discrimination24,6; this may account for the
high number of sexual partners and prevalence
of sex work among male-to-female transgender persons in our study. Street outreach and
more intensive HIV prevention interventions,
such as prevention case management, are urgently needed for sex workers and should include education, job training, and job placement. Given the high rate of incarceration
among our participants, jails and prisons may
be good settings for HIV prevention interventions for sex workers.
We found very low levels of HIV risk associated with hormone injection, probably because several clinics in San Francisco offer hormone therapy. However, nonhormonal injection
drug use was associated with HIV, and the prevalence of sharing syringes among injectors was
higher than that in studies with out-of-treatment
injection drug users in San Francisco.34,39,40 Public health providers need to ask transgender
clients about hormonal and nonhormonal injection drug practices and refer them to harm
reduction and treatment interventions. In addition, needle exchange programs should reach
out to transgender populations and provide both
hormone needles (22 g, 11/2 in) and nonhormone needles (27 g, 5/8 in).
Our study confirms that many transgender persons enter the medical system in pursuit
of hormones.1,15,16 In addition to providing hormone therapy, health care providers should
counsel and appropriately refer transgender
clients who are in need of HIV, substance abuse,
and mental health services. There is a particular need to assess the potential for depression
and suicide and intervene appropriately.
The prevalence of suicide attempts among
male-to-female and female-to-male transgender
persons in our study was much higher than that
found in US household probability samples
and a population-based sample of adult men
reporting same-sex partners.41,42 This finding
supports a previous study of male-to-female
transgender persons in the Netherlands, which
found that the number of suicides among maleto-female transgender persons was much higher
than the expected mortality for the Dutch male
American Journal of Public Health
919
population.43 Although we did not assess factors associated with suicide attempts, researchers have hypothesized that higher rates
of attempted suicide among gay youths compared with heterosexual youths may partially
be the result of increased discrimination and
victimization4446tragically, common experiences for transgender individuals.24,6
The primary limitation of our research was
the use of nonprobability sampling. Our findings may not generalize to other urban areas,
and there may be threats to internal validity if
certain sampling methods were more likely to
recruit high-risk individuals. Most traditional
random sampling approaches would not produce reliable samples, however, because the
transgender population has strong privacy concerns and has never been counted, and because
many transgenderpersons are marginally housed
or homeless.
Our use of social mapping and targeted
sampling helped ensure that important subpopulations were sampled. Respondent-driven
sampling capitalizes on the ability of members
of a hidden population to recruit their peers
and reduces many biases associated with traditional snowball sampling.23 Despite the use
of such methods, female-to-male participants
were difficult to recruit. Female-to-male transgender individuals were more likely to be recruited through personal contacts with study
staff than through respondent-driven sampling.
This may partially account for the demographic
differences we observed between male-tofemale and female-to-male participants but is
unlikely to account for the large difference in
HIV prevalence.
Despite these limitations, our data describe
an important population in San Francisco with
many needs. Similar studies in other US cities
are needed to assess the role of transgender individuals in local HIV epidemics. Our ability
to recruit this hard-to-reach population was
likely the result of engaging in participatory research and hiring a diverse group of male-tofemale and female-to-male transgender staff.
Similar peer-based approaches could be used to
provide culturally appropriate HIV, substance
use, mental health, education, employment, and
other social services for transgender persons. In
addition, because transgender individuals access the health care system for hormone therapy, medical providers can provide an important link to needed services.
Contributors
K. Clements-Nolle, R. Marx, and M. Katz conceived
and designed the study. K. Clements-Nolle and R.
Guzman implemented the study protocols and supervised data collection, coding, and entry. K.
Clements-Nolle analyzed the study data. R. Marx and
M. Katz assisted with data analysis and interpretation. K. Clements-Nolle conducted the literature re-
920
Acknowledgments
This study was supported by Centers for Disease
Control and Prevention Cooperative Agreement
U62CCU902017-12 and California Department of
Health Services, Office of AIDS Grant 97-10787.
We would like to thank Ari Bachrach, Nikki
Calma, Carla Clynes, Nashanta Stanley, Matt Rice,
and Doan Thai for conducting data collection, counseling, testing, and referrals; Vince Guilin and Scott
Ikeda for field coordination; the Transgender Community Advisory Board for assistance with protocol,
survey, and materials development; the San Francisco Department of Public Health Laboratory for
HIV test result processing; Art DeGuzman, Stella
Cheung, Irene Lee, and Belinda Van for data entry
and management; Aida Flandez for assistance with
manuscript preparation; and Nancy Padian, PhD,
Mike Pendo, MPH, and Doug Sebesta, PhD, for manuscript review. We would also like to express our appreciation to the following agencies who assisted
in recruitment and provided space to conduct interviews and HIV testing: Asian & Pacific Islander
Wellness Center, Center for Special Problems, Filipino Task Force on AIDS, FTM International, GlideGoodlett HIV/AIDS Program, Instituto Familiar de
la Raza, New Village, Proyecto ContraSIDA Por
Vida, Southeast Asian Community Center, Tenderloin AIDS Resource Center, and Tom Waddell Clinic.
References
1. White JC, Townsend MH. Transgender medicine: issues and definitions. J Gay Lesbian Med
Assoc. 1998;2:13.
2. Bockting WO, Robinson BE, Rosser BR. Transgender HIV prevention: a qualitative needs assessment. AIDS Care. 1998;10:505525.
3. Boles J, Elifson KW. The social organization of
transvestite prostitution and AIDS. Soc Sci Med.
1994;39:8593.
4. Pang H, Pugh K, Catalan J. Gender identity disorder and HIV disease. Int J STD AIDS. 1994;5:
130132.
5. Yep GA, Pietri M. In their own words: communication and the politics of HIV education for
transgenders and transsexuals in Los Angeles.
In: Elwood WN, ed. Power in the Blood: A
Handbook on AIDS, Politics, and Communication. Mahwah, NJ: Lawrence Erlbaum Associates Inc; 1998:199214.
6. Clements-Nolle K, Wilkinson W, Kitano K,
Marx R. HIV prevention and health service
needs of the transgender community in San
Francisco. In: Bockting W, Kirk S, eds. Transgender and HIV: Risks, Prevention, and Care.
Binghamton, NY: The Haworth Press Inc; 2001.
7. Elifson KW, Boles J, Posey E, Sweat M, Darrow W, Elsea W. Male transvestite prostitutes
and HIV risk. Am J Public Health. 1993;83:
260262.
8. Modan B, Goldschmidt R, Rubinstein E, et al.
Prevalence of HIV antibodies in transsexual and
female prostitutes. Am J Public Health. 1992;
82:590592.
9. Spizzichino L, Casella P, Zaccarelli M, Rezza
G, Venezia S, Gattari P. HIV infection among
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
foreign people involved in HIV-related risk activities and attending an HIV reference centre
in Rome: the possible role of counseling in reducing risk behavior. AIDS Care. 1998;10:
473480.
Galli M, Esposito R, Antinori S, et al. HIV-1 infection, tuberculosis, and syphilis in male transsexual prostitutes in Milan, Italy. J Acquir Immune Defic Syndr. 1991;4:10061007.
Tirelli U, Vaccher E, Bullian P, et al. HIV-1 seroprevalence in male prostitutes in northeast
Italy. J Acquir Immune Defic Syndr. 1988;1:
414415.
Tirelli U, Rezza G, Giuliani M, et al. HIV seroprevalence among 304 female prostitutes from
four Italian towns. AIDS. 1989;3:547548.
Elifson KW, Boles J, Sweat M. Risk factors associated with HIV infection among male prostitutes. Am J Public Health. 1993;83:7983.
Leonard T, Sacks J, Franks A, Sikes RK. The
prevalence of human immunodeficiency virus,
hepatitis B, and syphilis among female prostitutes in Atlanta. J Med Assoc Ga. 1988;77:
162164.
Moriarty HJ, Thiagalingam A, Hill PD. Audit of
service to a minority client group: male to female transsexuals. Int J STD AIDS. 1998;9:
238240.
Valenta LJ, Elias AN, Domurat ES. Hormone
pattern in pharmacologically feminized male
transsexuals in the California State prison system. J Natl Med Assoc. 1992;84:241250.
Sykes DL. Transgendered people: an invisible
population. Calif HIV/AIDS Update. 1999;12:
8085.
Stephens T, Scott C, Braithwaite RL. Transsexual orientation in HIV risk behaviours in an adult
male prison. Int J STD AIDS. 1999;10:2831.
Nemoto T, Luke D, Mamo L, Ching A, Patria J.
HIV risk behaviors among male-to-female transgenders in comparison with homosexual or bisexual males and heterosexual females. AIDS
Care. 1999;11:297312.
Watters JK, Biernacki P. Targeted sampling: options for the study of hidden populations. Soc
Probl. 1989;36:416430.
Schwab M, Syme SL. On paradigms, community participation, and the future of public health.
Am J Public Health. 1997;87:20492051.
Cornwall A, Jewkes R. What is participatory research? Soc Sci Med. 1995;41:16671676.
Heckathorn DD. Respondent-driven sampling: a
new approach to the study of hidden populations. Soc Probl. 1997;44:174199.
Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385401.
Mulrow CD, Williams JW, Geretty MB, Ramirez
G, Montiel OM, Kerber C. Case-finding instruments for depression in primary care settings. Ann Intern Med. 1995;122:913921.
Weissman MM, Sholomskas D, Pottenger M,
Prusoff BA, Locke BZ. Assessing depressive
symptoms in five psychiatric populations: a validation study. Am J Epidemiol. 1977;106:
203214.
Gallo D, George JR, Fitchen JH, Goldstein AS,
Hindahl MS. Evaluation of a system using oral
mucosal transudate for HIV-1 antibody screening and confirmatory testing. JAMA. 1997;277:
254258.
Mansergh G, Marks G. Age and risk of HIV in-
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
921