AIDS Care

Psychological and Socio-medical Aspects of AIDS/HIV

ISSN: 0954-0121 (Print) 1360-0451 (Online) Journal homepage: http://www.tandfonline.com/loi/caic20

HIV risks among gay- and non-gay-identified

migrant money boys in Shanghai, China
F. Y. Wong Ph.D. , Z. J. Huang , N. He , B. D. Smith , Y. Ding , C. Fu & D. Young
To cite this article: F. Y. Wong Ph.D. , Z. J. Huang , N. He , B. D. Smith , Y. Ding , C. Fu & D. Young
(2008) HIV risks among gay- and non-gay-identified migrant money boys in Shanghai, China,
AIDS Care, 20:2, 170-180, DOI: 10.1080/09540120701534707
To link to this article: http://dx.doi.org/10.1080/09540120701534707

Published online: 25 Apr 2008.

Submit your article to this journal

Article views: 234

View related articles

Citing articles: 32 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=caic20
Download by: [The University of British Columbia]

Date: 28 February 2016, At: 14:09

AIDS Care, February 2008; 20(2): 170!180

HIV risks among gay- and non-gay-identified migrant money boys in

Shanghai, China
F. Y. WONG1, Z. J. HUANG1, N. HE2, B. D. SMITH1, Y. DING2, C. FU2, & D. YOUNG1
Department of International Health, Georgetown University Nursing and Health Studies, Washington, US and 2Department
of Epidemiology, Fudan University School of Public Health, Shanghai and the Key Labouratory for Public Health safety,
Ministry of Education, China

Downloaded by [The University of British Columbia] at 14:09 28 February 2016

Abstract
Men having sex with men (MSM) now account for 7% of all HIV/AIDS cases in China and there is growing awareness that
internal rural-to-urban migration might shift the HIV epidemic within China by broadening social and sexual mixing. About
70% of HIV/AIDS infections are among rural residents, of whom 80% are males and 60% aged 16!29. This young, male,
rural-to-urban migrant population has been identified as the tipping point for the AIDS epidemic in China. A subgroup of
these migrants is the money boy population, i.e. those who engage in same-sex transactional sex for economic survival.
However, the literature addressing money boys is very limited. The present study aims to elucidate factors for preventing
substance abuse and HIV among two types of money boys gay-identified and non-gay-identified living in the Shanghai
metropolitan area. This work is conceptually underpinned by Chng et al.s (2003) tripartite model, which postulates that
risk behaviors (e.g. substance abuse) engaged in by transient or non-native individuals are often shaped and regulated by
factors in the home environment, migration experience and current environment. Results reveal gay and non-gay money
boys were not significantly different in age, income, marriage status and education. Both groups shared similar patterns of
substance use. Both groups had high self-reported depressive symptoms and low HIV knowledge. However, sexual
orientation differentially predicted HIV testing, with gay money boys more likely to be tested for HIV. Non-gay money boys
showed fewer sexual risks. HIV prevention targeting MSM (including money boys) within rapidly changing China is
discussed, as are methodologies and outreach strategies most effective for particular subgroups of MSM.

Introduction
More than two decades after the first reported case
of HIV/AIDS in mainland China, the epidemic has
finally exploded. The Joint United Nations Program
on AIDS (UNAIDS, 2004) gloomily predicts that by
2010 an estimated 8.5 million people in China will
be infected with HIV/AIDS. There is a growing
awareness that internal rural-to-urban migration
might also shift the HIV epidemic by broadening
the social and sexual mixing of the migrant population (He et al., 2006; Hong et al., 2006; Hu et al.,
2006; Li et al., 2004; Liu et al., 1998; Qiao et al.,
2000; Zheng et al., 2000).
In the years following the Chinese Communist
Partys rise to power, the government created a
nationwide household system (known as hukou)
that assigned every individual to a particular place
of residence and officially classified people as either
urban or rural residents. Those wishing to relocate
to another part of the country required official
approval, which allowed the central government to

effectively control the social, political, and economic lives of its people (Zhang, 2001). During the
heyday of the hukou system (between 1953 and the
late 1970s), the government was able to control
illicit drug use and trade as well as sexually
transmitted diseases (STDs) (Cohen et al., 1996).
In the early 1980s, however, the Chinese government undertook a series of structural economic
reforms that improved the efficiency of farming and
created a labour surplus of nearly 200 million rural
labourers. The reforms simultaneously spurred
rapid urban economic growth and a soaring demand for cheap urban labour. In response, millions
of peasants began migrating illegally to cities
throughout China. The city of Shanghai, Chinas
largest center of commerce and finance, became
a popular migrant destination. As of 2000, there
were 121 million migrants throughout China and
Shanghais metropolitan population included some
4.4 million migrants (China National Bureau of
Statistics, 2002).

Correspondence: Frank Y. Wong, Ph.D., Department of International Health, Georgetown University Nursing and Health Studies, 3700
Reservoir Road, Box 571107, Washington, D.C. 20057, US. Tel: "1 (202) 687 0333. Fax: "1 (202) 784 3128. E-mail:
fyw@georgetown.edu
ISSN 0954-0121 print/ISSN 1360-0451 online # 2008 Taylor & Francis
DOI: 10.1080/09540120701534707

Downloaded by [The University of British Columbia] at 14:09 28 February 2016

Chinese money boys and HIV risk

In the field of HIV research, a growing body of
studies is reaching consensus regarding the effects of
migration on health (Anderson et al., 2003; He
et al., 2005; Hong et al., 2006; Li et al., 2004).
Researchers and policymakers alike have repeatedly
characterized Chinas rural-to-urban migrant population as the tipping point for the HIV epidemic in
China (Anderson et al., 2003; Ministry of Health,
UNAIDS, & WHO, 2006). The difficulty of conducting HIV surveillance within this transient
and marginalized population has been previously
documented (Lin et al., 2005). Among the 121
million migrants, 60% are males and 40% are aged
20!24 (Huang & Yang, 2000; Thomas, 1998). Most
migrants are either unmarried, or married but living
apart from their spouses, and have a low awareness
of HIV/AIDS (He et al., 2007). Without permanent
urban household registration status, these migrants
cannot access the social welfare benefits available to
urban residents. They are also difficult to reach via
traditional preventive health education, as they are
often deprived of access to healthcare (McCoy & Yu,
1999). In the meantime, studies have shown a high
prevalence of HIV among rural migrants. Multi-city
HIV surveillance data between 1995 and 2000 reveal
that over two-thirds of the HIV cases were found
among rural-to-urban migrants. In 2000, 85.4% of
Beijings and 74.4% of Shanghais new HIV infections occurred among migrants (Lin et al., 2005). It
has been reported that HIV prevalence was 1.8 times
higher among rural-to-urban migrants than among
non-migrant rural residents (Xu et al., 1998).
Men having sex with men in China: HIV and STIs
Studies on HIV risks among Chinese migrants tend
to focus on heterosexual populations. Worldwide,
roughly 5!10% of all HIV infections are spread
through sexual contact between men, though this
percentage varies considerably by country and region. In Asia, men who have sex with men (MSM)
are disproportionately affected by HIV (amFAR,
2006). Choi et al. (2003) found an HIV infection
rate of 3.1% among MSM in Beijing. The latest
statistics show that MSM account for 7% of all HIV
AIDS cases in China (Merli, 2006; Xia, 2006).
According to the UNAIDS report (2006), there is
increasing evidence of HIV outbreaks among Chinese MSM. Chinas chief administrative authority,
the Chinese State Council (1998), now considers
MSM to be a high-risk group. In response, the
National Center for AIDS/STD Control and Prevention of the Chinese Center for Disease Control
and Prevention (Chinese CDC) is conducting 17
ongoing projects targeting MSM (Wu, 2006). Still,
most national AIDS programmes fail to address the
specific needs of Chinese MSM, and the empirical or

171

scientific bases of these projects may be less than

optimal due to the limited existence of literature.
Compared to western countries, HIV-related prevention and intervention programmes targeting
MSM (non-commercial sex workers) is lacking in
China, mainly due to political, cultural and customary restrictions (Zhang & Chu, 2005). To our
knowledge, only six English-language empirical papers on HIV among Chinese non-money boy MSM
have been published (Choi et al., 2002, 2003, 2004;
He et al., 2006; Wang & Ross, 2002; Zhang et al.,
2000).1 Five of these papers used quantitative
methodologies and one used ethnographic and
qualitative approaches. One paper provided HIVinfection rates based on testing (Choi et al., 2003);
one paper provided self-reported HIV rates (Zhang
et al., 2000); all studies reported HIV-related knowledge, attitudes, beliefs and behaviours (KABB). For
example, in Beijing, Choi et al. (2002) found that
the major barriers to utilizing existing STD/HIV
services were: (1) lack of perceived risk, (2) lack of
privacy and anonymity, (3) uninteresting programme content and (4) distrust and questionable
credibility of providers.
A review of Chinese-language articles on the
MSM population published between 2002 and
2006 found 39 empirical studies (of varying scientific quality), all of which documented patterns of
findings similar to the English-language studies. A
sizeable proportion of these MSM had unprotected
anal intercourse with men and unprotected anal or
vaginal intercourse with women during the six
months prior to the study.
To the best of our knowledge, however, there are
few academic publications on Chinese money boys:
one published English-language empirical article
and four published Chinese-language empirical papers. Using an ethnographic approach, He et al.
(2007) found that, compared to general male
migrants, money boys were slightly better-off economically and more knowledgeable about HIV and
STIs; used alcohol less but drugs more; had higher
HIV/STI testing rates; and exhibited fewer HIV risk
behaviours. However, it was noted that money boys
who were new to the enterprise often lacked HIV
knowledge and protective skills. Moreover, many of
the money boys wanted to receive STI testing since
they perceived having such an infection as an
immediate threat to their vocation. About one-fifth
of money boys self-identified as heterosexuals and
claimed their commercial sexual activities as purely
business transactions. This sub-group (i.e. heterosexual-identified money boys) continues to engage in
heterosexual sexual activities in their private lives.
Among the four studies published in Chinese,
two employed venue-based approaches, one used
a combined CPOL (community popular opinion

Downloaded by [The University of British Columbia] at 14:09 28 February 2016

172

F. Y. Wong et al.

leader) and venue-based approach and one used

target sampling. Collectively, the studies documented prevalence ranges of 11.6!30.5% for any selfreported STI, 5.8!19.5% for syphilis and 0!2.1% for
HIV. Prevalence was 12.6% for chlamydia, 8.4% for
gonorrhea and 6.1% for herpes. All four studies
documented high-risk sexual behaviours (Ma et al.,
2004; Qi et al., 2006; Qian et al., 2005; Zhuang
et al., 2005).
Collectively, these studies suggest that MSM in
China are at risk for HIV. Given that many MSM
also have sexual relationships with women, these
men may be acting as a bridge for heterosexual
transmission. Meanwhile, the identification of two
distinct types of money boys*(1) gay-identified and
(2) non-gay-identified (or heterosexual-identified)*
by He et al. (2007) demands an in-depth examination of the possible differentiation effects of sexual
identification on HIV and sexual risks.
Yet, understanding of the growing, high-risk
money boy population is hindered by the limited
nature of the aforementioned studies. The present
study hopes to elucidate high-risk behaviours for
HIV transmission among Chinese money boys.
Specifically, it aims to describe and examine factors
for preventing substance abuse and HIV among two
types of money boys*gay-identified and non-gayidentified*living in the Shanghai metropolitan area.
Given the paucity of research on HIV risks among
migrant money boys in China, we have borrowed a
conceptual model on migration developed by Chng
et al. (2003) to guide this research. Although Chngs
model was developed to examine the experiences of
Asian immigrants and refugees and Asian immigrant
MSM in the US, we believe that it is reasonable to
assume that many of these experiences and processes
are common among most (if not all) immigrant,
refugees, non-native or transient populations
throughout the world. For example, migrant money
boys in Shanghai probably experience cultural isolation and/or discrimination (e.g. for being gay or not
speaking the local Shanghai dialect), just as many
Asian immigrants and refugees might experience in
dealing with cultural, racial and language barriers in
the US In turn, these processes and experiences may
pose similar stress on individual well-being, which
may result in similar negative coping mechanisms
and health outcomes.
In brief, Chng et al.s (2003) tripartite model
postulates that risk behaviours (e.g. substance abuse)
engaged in by transient or non-native individuals in a
society are often shaped and regulated by three sets of
factors: the home environment, migration experience
and current environment. Factors in the original
home environment (Impact Domain 1) include
cultural norms and understandings regarding sexual
mores, shame or stigma, sexual practices and sub-

stance use/abuse. These norms, understandings and

practices are modified by the nature of migration
experience (Impact Domain 2), which for some
migrants may include trauma and the endurance of
prolonged socioeconomic and psychological hardship. These reconfigured norms, understandings,
and practices are further reshaped by the situation
that these populations face as they recreate their lives
in the new environment (Impact Domain 3).
Methods
Study population and sampling methods
This cross-sectional survey study examined the
prevalence and characteristics of alcohol and drug
use/abuse and HIV/AIDS KABB, as well as migration experience and sexual self-identity (e.g. gay
versus non-gay-identified) among a cohort of money
boys in Shanghai, China. To be eligible for the study,
participants must have: (1) been aged 18 and above;
(2) been not native to Shanghai; (3) been able to give
verbal and written (in Mandarin) consent; (4) ever
had sex with men; and (5) self-identified as a money
boy at the time of participation. A total of 239
money boys were recruited between April 1, 2006
and June 30, 2006 using two methods: (1) a type of
snowball sampling methodology modeled after Respondent-Driven Sampling (RDS) and (2) a community leader from a local non-governmental
organization (NGO) that serves MSM in Shanghai.
Specifically, three seeds (two gay-identified migrant money boys and one non-gay-identified money
boy) were selected by the research team. Each seed
was then asked to recruit up to three of his peers, and
each eligible participant (excluding the three seeds)
was asked two questions: (1) to identify the nature of
his relationship to the seed; and (2) to identify how
long he had known the seed. While the number of
seeds and waves of recruitment were limited due to
resource constraints, one gay-identified seed resulted
in 130 participants (in seven waves); the other gayidentified seed yielded 41 participants (in three
waves). Selection of the non-gay-identified seed
yielded 12 participants (in two waves). The community leader from the NGO was a highly visible
community leader among gays and MSM in Shanghai and highly regarded by members of the money
boy community.
Recruited participants were verbally informed of
the nature and purpose of the study, survey procedures, the sensitive nature of the questions, confidentiality parameters, payment for participation
(US$40 for the quantitative survey), risks and
benefits (including referrals for other services
needed), and the freedom to cease participation
at any time without penalty. When respondents

Chinese money boys and HIV risk

Downloaded by [The University of British Columbia] at 14:09 28 February 2016

verbally indicated an understanding of these issues,

they signed a consent form. They were also given a
copy of the Research Subjects Bill of Rights. All
consent and human subject forms and procedures
had been reviewed by both of the appropriate
academic institutional review boards in the US and
China.
All eligible participants agreed to take part in the
study without any expressed reservations. In one
instance, an individual presented a fake referral
coupon. He was subsequently disqualified from
participating.
Survey techniques
The self-administered questionnaire included seven
sections: (1) demographic information, (2) migration experiences, (3) HIV/AIDS KABB, (4) substance use/abuse, (5) sexual behaviour risk, (6) HIV
and STD testing and other healthcare services
access, and (7) depression.2 The survey took approximately 30 to 75 minutes (and an average of
45 minutes) for participants to complete. A research
assistant was onsite to answer any questions related
to the survey.

Variables
Outcome variables. Major outcomes included: (1)
HIV/AIDS knowledge, (2) problem drinking, (3)
30-day any drug use, (4) an HIV risk behaviour
index, (5) HIV/STD testing and access to healthcare
and (6) depression as measured by the Center of
Epidemiology Study-Depression (CES-D) (Radloff,
1977). Problem drinking was defined by scoring two
or more on the CAGE Assessment for alcohol abuse
(Mayfield et al., 1974). In separate questions,
respondents were asked about their use of marijuana, heroin, opium, cocaine, ecstasy, methamphetamine, stimulants and tranquilizers in the past
30 days. Hard drug use was classified as having
used cocaine, heroin or opium at least once in the
past 30 days. Soft drug use was classified as having
used marijuana, ecstasy, methamphetamine, stimulants or tranquilizers in the past 30 days. The HIV
risk behaviour index was created using self-reported
sexual behaviours: (1) ever had anal sex with a
customer without a condom; (2) had anal sex with a
main partner/casual partner without a condom in the
past 30 days; (3) not always used a condom when
having oral sex in the past 30 days; (4) ever had sex
without a condom because you were under the
influence of alcohol/drugs; or (5) drank alcohol or
used drugs immediately before sex. If a participant
had engaged any of these risk behaviours, he was
coded as exhibiting risky sexual behaviour.

173

Predictor: Sexual identification. Self-reported sexual

identification was defined based on the question:
Do you think you are a true homosexual (Tong Zhi)?
Yes or No. If the participant answered No, the selfreported sexual identification was listed as non-gayidentified. Those who answered Yes were listed as
gay-identified.
Socio-demographic variables in migration tripartite
model. Measures on the tripartite model in migration
experience were grouped in three domains: (1) status
in hometown (occupation at home and reasons for
leaving hometown); (2) migration experiences (cities
traveled to before Shanghai; years away from hometown; and if participant had received a job offer in
Shanghai before coming); and (3) current situation
in Shanghai (living conditions; experience with
discrimination; amount of stress felt; and satisfaction
with life). Discrimination, life stress and satisfaction
in Shanghai, respectively, was measured by a single
item.
Statistical analysis
Chi-square analyses were used to test for associations between gay identity, migration experiences
and four domains of health risks: (1) problem
drinking, (2) drug use, (3) HIV sexual-risk behaviours and (4) depression. Logistic regression analyses were used to provide crude odds ratios ORs)
and examine the independent effects of gay identity
on different health risk factors and migration characteristics. Adjusted OR and 95% confidence intervals (CI) were computed by using the beta
coefficients and standard errors obtained from the
multi-variable logistic analyses.
Results
Demographic characteristics
Table I shows the demographic, socioeconomic
status and the migration characteristics of participants in this survey study. A total of 239 money boys
from Shanghai, China participated. About one-fifth
(19.7%; n #47) of the money boys self-identified as
non-gay in the survey. The average participant age
was 25.2 years old. More than 10% of the money
boys were married or divorced. A narrow majority of
participants (50.6%) ended their formal education at
the high school level and 7.1% had a college
education, while 41% had less than a high school
education. More than one-fifth (21.3%) of participants earned less than 1,000 Yuan monthly and
more than one-third (34.7%) earned at least 3,000
Yuan monthly (US$1 #!8 Yuan).

174

F. Y. Wong et al.

Table I. Demographic characteristics of gay vs. non-gay-identified money boys, Shanghai, China, 2006.

Downloaded by [The University of British Columbia] at 14:09 28 February 2016

All money boys

%

Gay-identified money boys

%

Non-gay-identified money boys

%

P-value

N
Age (mean "/$ SD)

239
25.2 "/ $ 4.7

192
25.7 "/$ 4.9

Marriage
Never married
Married or divorced

89.5
10.5

88.0
12.0

95.7
4.3

0.120

Education
Cannot read
Elementary school
Middle school
High school
College (or more)

0.8
4.6
36.4
50.6
7.1

1.0
4.7
36.5
51.0
6.8

0.0
4.4
37.0
50.0
8.7

0.950

Monthly income
B1000 Yuan
1000!2999.99 Yuan
3000!4999.99 Yuan

21.3
43.5
34.7

21.9
43.2
34.9

19.6
45.7
34.8

0.931

34.3
39.8
10.5
15.5

33.9
39.6
10.9
15.6

36.2
40.4
8.5
14.9

0.962

56.8
16.7
2.6
34.9

66.0
8.5
10.6
31.9

0.252
0.16
0.014
0.699

Number of cities migrated to before Shanghai

0
25.5
1!2
44.7
3!5
25.5
6"
4.3

28.2
45.2
21.8
4.8

14.9
42.6
40.4
2.1

0.038

Years away from home

0!1
2!5
6!9
10"

4.4
30.6
48.9
16.1

4.6
52.3
40.9
2.3

0.016

22.9
77.1

19.2
80.8

0.577

47
23.1 "/$ 3.1

B0.0001

Status in hometown
Occupation in hometown
Student
Small business
Farmer
Looking for work*

Reasons for leaving hometown (multiple responses allowed)

Earn more income
58.6
No freedom at home
15.1
Nothing to do at home
4.2
Want to see the world
34.3
Migration experience

4.5
34.8
47.3
13.4

Received a job offer before arriving in Shanghai

Yes
22.2
No
77.8
Current situation in Shanghai
Living arrangements
Alone
Co-worker(s) or friend(s)
Partner or spouse
Other

46.4
40.2
8.4
5.0

47.4
38.5
8.3
5.7

42.6
46.8
8.5
2.1

0.613

Experienced discrimination?
Yes
No/dont know

53.1
46.9

55.2
44.8

44.7
55.3

0.195

Is life stressful?
Yes
No/dont know

39.3
61.7

38.0
62.0

44.7
55.3

0.402

Satisfied with life?

No
Yes/doesnt matter

22.6
77.4

21.4
78.6

27.7
72.3

0.354

*Within the Chinese context, looking for work is considered to be an activity related to employment. In other words, it is not synonymous
with the notion of being unemployed or having no job as typically understood in western cultures.

Downloaded by [The University of British Columbia] at 14:09 28 February 2016

Chinese money boys and HIV risk

Table I also shows characteristics of participants
migration experiences related to participants status
in their hometowns, migration experience and current situation in Shanghai per the tripartite model.
All of the participants had migrated from rural areas
and settled in Shanghai. The two subgroups worked
similar occupations in their hometowns: one-third
(34.3%) were students; nearly 40% worked in small
business and roughly 10% in farming; about 15%
were unemployed and looking for work.
In identifying a reason for migration, a majority
(58.6%) of participants named the possibility of
earning more income, one-third (34.3%) indicated a
desire to see the world and 15.1% cited a lack of
freedom in their hometowns. However, the data also
revealed some statistically significant differences
between the two subgroups. Non-gay-identified
participants were four times more likely (10.6%
versus 2.6%; p #0.014) to cite nothing to do at
home as the motivation for migrating.
Of all participants, only 22.2% had received a job
offer in Shanghai before migrating there. About onequarter (25.5%) of all participants had moved to
Shanghai directly from their hometowns, while
44.7% had lived in one or two other cities, 25.5%
had lived in three to five cities and 4.3% had lived in
six or more cities before arriving in Shanghai. The
data revealed some statistically significant differences
between the two subgroups (p #0.038), as non-gayidentified participants were more likely to have
migrated to three or more cities (42.5% versus
26.6%) and gay-identified money boys were nearly
twice as likely to have migrated to Shanghai directly
from their hometowns (28.2% versus 14.9%).
Of all participants, 4.5% had been living outside
their hometowns for one year or less, 34.8% for two
to five years, nearly half (47.3%) for six to nine years
and 13.4% for ten or more years. Here too, however,
we found some statistically significant differences
(p #0.016). Gay-identified money boys were also
seven times more likely to have lived away from
home for ten or more years (16.1% versus 2.3%).
Questions concerning participants current situation in Shanghai did not reveal significant differences
between the two subgroups. Nearly half (46.4%) of
all participants lived alone and 40.2% lived with at
least one co-worker or friend. Most participants
(53.1%) believed that they personally had experienced some discrimination, while 39.3% found life
in Shanghai to be stressful and more than threequarters (77.4%) were satisfied with life in Shanghai.
HIV/AIDS KABB
Overall, there was no significant difference between
gay-identified and non-gay-identified money boys in
their HIV/AIDS KABB. Despite differences in

175

duration of stay in Shanghai, participants generally

had quite limited knowledge of HIV/AIDS. For
example, only 54.4% knew that there is no cure for
HIV. More than one-third (37.7%) did not know if
HIV/AIDS could be cured, while 8% believed that a
cure exists. More than one-third (38.5%) also
believed that mosquito bites could transmit HIV,
while 28% were unsure. More than one-fifth (21%)
of participants either believed that, or did not know
if, shaking hands with an HIV-infected person could
transmit the virus, while 23% either believed that, or
did not know if, eating with an HIV-infected person
could transmit the virus. Not quite half (45.6%) did
not know if antibiotics could prevent HIV and 16.7%
believed they could.
Sexual risk behaviours
Table II summarizes sexual risk behaviours. Not
surprisingly, we found a high prevalence of risky
sexual behaviours among participants. In the previous 30 days, nearly two-thirds (64.4%) of participants reported consistently having protected sex
with clients, while 21.8% had unprotected sex once
or twice. Slightly more than half (50.2%) indicated
that they had unprotected intercourse (without a
condom) at least once in the previous 30 days with
their main sexual partner (wife, lover or male
partner), while 84.5% had unprotected sex at least
once in their life.
Only three questions concerning sexual risk behaviours revealed statistically significant differences
between the two subgroups. In the previous 30
days, gay-identified participants were significantly
more likely to have taken part in oral sex (92.2%
versus 76.6%; p #0.009) and to have done so
without using a condom (66.2% versus 55.3%).
They were also more likely to have taken part in anal
sex (81.1% versus 74.5%; p #0.005) and to have
done so without using a condom during the same
period (28.1% versus 12.7%). However, non-gayidentified money boys were more than three times as
likely (10.6% versus 3.1%; p#0.03) to have had sex
without a condom due to the influence of drugs.
Substance use
There is no statistical difference between the two
subgroups in alcohol and other drug use (hence the
results are not shown in the Tables). A majority
(79.9%) of participants ever used cigarettes or
tobacco and 10% fitted the criteria for alcoholdependence (based on the CAGE assessment).
More than 12% reported ever having used any
drugs, 8% reported ever having used hard drugs
and 10% reported ever having used soft drugs.

176

F. Y. Wong et al.

Table II. Sexual risk behaviours by gay versus non-gay-identified money boys, Shanghai, China, 2006.
All money boys
%

Gay-identified money boys

%

Non-gay-identified money boys

%

1. Have you ever had unprotected sex (i.e. sex without use of a condom)?
Yes
84.52
84.90
No
15.48
15.10
Dont know
0.00
0.00

82.98
17.02
0.00

P-value

0.7447

Downloaded by [The University of British Columbia] at 14:09 28 February 2016

2. In the past 30 days, how many times would you say you had unprotected sex (e.g. without a condom) with a main sex partner (e.g. wife,
lover, male partner)?
None
49.79
47.92
57.45
0.2415
Yes ("1)
50.21
52.08
42.55
3. In the past 30 days, how many sexual partners did you have?
0
3.77
3.13
1!2
6.28
5.21
3!4
55.23
57.29
5"
34.73
34.38

6.38
10.64
46.81
36.17

4. In the past 30 days, how many times would you say you had unprotected sex (e.g. without a condom) with clients?
0
64.44
63.02
70.21
1!2
21.76
21.88
21.28
3!4
5.86
6.77
2.13
5"
4.18
5.21
0.00
Yes ("1)
3.77
3.13
6.38

0.3050

0.2663

5. In the past 30 days, how often would you say you had oral sex using a condom?
Always
25.10
26.04
Not always
64.02
66.15
No oral sex
10.88
7.81

21.28
55.32
23.40

0.0090

6. In the past 30 days, how often would you say you had anal sex using a condom?
Always
61.92
61.98
Not always
25.10
28.13
No anal sex
12.97
9.90

61.70
12.70
25.53

0.0050

7. Have you ever had sex without a condom because you were under the influence of alcohol?
Yes
34.31
32.81
No
65.69
67.19
Dont know
0.00
0.00

40.43
59.57
0.00

0.3244

8. Have you ever had sex without a condom because you were under the influence of drugs?
Yes
4.60
3.13
No
95.40
96.88
Dont know
0.00
0.00

10.64
89.36
0.00

0.0276

Depression and sexual violence

The findings in Table III reveal a high prevalence of
depression among all money boys. Based on the
CES-D scale, 60.7% showed a moderate or high
probability of depression (CES-D "# 16). The
percentages differed somewhat between the two
subgroups: nearly half (48.9%) of non-gay-identified
money boys showed a low probability of depression,
compared to only 37% of their gay-identified counterparts (though this difference was not statistically
significant). Gay-identified money boys reported a
higher prevalence of sexual violence compared to
non-gay-identified money boys (20.8% versus
12.8%; p #0.2). The results were not statistically
significant, possibly due to the small sample size.
Among all participants who experienced sexual
violence, 76.1% reported that a client had been the
perpetrator. Nearly 6% of gay-identified participants
indicated that they had been forced to have sex with

an adult before the age of ten, while none of the nongay-identified money boys indicated that they had
experienced such abuse.
HIV/STD testing and prevalence
Table III also summarizes health services utilization,
including HIV and STI testing. The data revealed no
significant differences. HIV testing was more common than STI testing. Among all participants,
56.5% had been tested for HIV and 30.1% had
been tested for STIs. Only 1.7% indicated that they
were HIV-positive, though a majority (58.2%) did
not know their HIV status. STI infections were more
common (8.8%) but three-quarters of participants
(75.3%) did not know if they had an STI. Only
about half (49.8%) knew that HIV testing was freeof-charge. Gay-identified participants were more
likely to believe that it was possible to get a free
physical examination annually (24% versus 14.9%).

Chinese money boys and HIV risk

177

Table III. Depression, episodes of violence and health services access for gay versus non-gay-identified money boys, Shanghai, China, 2006.

All money boys

%

Gay-identified
money boys
%

Non-gay-identified
money boys
%

28.03
32.64
39.33

28.65
34.38
36.98

25.53
25.53
48.94

2. Have you been forced to have sex with an adult before 10 years old?
Yes
4.62
No
95.38

5.76
94.24

3. Have you ever experienced sexual violence?

Yes
No

19.25
80.75

20.83
79.17

12.77
87.23

0.2086

4. If so, this person was:

Main partner
Client
Other

6.52
76.09
17.39

7.50
75.00
17.50

0.00
83.33
16.67

0.7789

1. Ever tested for HIV

Yes
No
Dont know

56.49
43.51
0.00

55.73
44.27
0.00

59.57
40.43
0.00

0.6337

2. Ever tested for STI

Yes
No
Dont know

30.13
66.95
2.93

29.17
68.23
2.60

34.04
61.70
4.26

0.6409

3. Do you think HIV testing is free?

Yes
No
Dont know

49.79
11.72
38.49

51.56
11.98
36.46

42.55
10.64
46.81

0.4223

4. Do you know where to take HIV testing?

No
Hospital
Private clinic
Public health service
Other

23.01
24.27
0.42
51.46
0.84

23.96
26.04
0.52
48.96
0.52

19.15
17.02
0.00
61.70
2.13

0.3736

5. Can you get free annual physical examination?

Yes
No
Dont know

22.18
67.78
10.04

23.96
65.10
10.94

14.89
78.72
6.38

0.2007

Downloaded by [The University of British Columbia] at 14:09 28 February 2016

1. CESD
High probability of depression (CESD"22)
Moderate probability of depression (CESD#16!22)
Low probability of depression (CESDB16)

Multi-variable analysis
Five hypothesized correlates of behaviour and health
risks were tested simultaneously in the multivariable
logistic models. For a co-variate to be selected, it
must have been: (1) a variable that corresponded to
our hypothesis (gay versus non-gay money boys) or
one of the tripartite model domains (year away from
home); (2) a variable that showed a statistically
significant association with at least one of the
behaviour or health indices; and (3) a variable that
was not highly correlated with other co-variables in
the model (education was found to be highly
correlated with HIV knowledge, so education was
not in the model). Co-variates selected in the model
are age, current income, years away from home, and
sexual orientation.

0
100

P-value

0.3000

0.0921

Table IV demonstrates the results from the multivariable models. HIV testing was independently
predicted by HIV knowledge score (a scale of
0!12, sum of correct answers for the 12 questions
concerning HIV transmission). Each point increase
in the score was associated with a 30% increase in
receiving HIV testing. The sexual risk behaviour
index*a measure of the possibility of not using
a condom in ten different settings (e.g. with a
customer)*was associated with sexual orientation.
A money boy with a score of more then seven (the
median score) was categorized as having higher HIV
sexual risks. Non-gay-identified money boys were
80% less likely to be in the high-risk category. Drug
use was independently associated with income.
Those with incomes higher than 3000 Yuan per
month were four times more likely to have used drug

178

F. Y. Wong et al.

Downloaded by [The University of British Columbia] at 14:09 28 February 2016

Table IV. Adjusted ORs for risk KABB, depression and HIV risk behaviours by sexual orientation of money boys, Shanghai, China, 2006.

HIV testing

Sexual risk behaviours

(no condom use in
multiple settings)

Any drug use

Sexual orientation
Gay
Non-gay

Ref
1.1 (0.5!2.1)

Ref
0.2 (0.1!0.5)

Ref
.5 (0.5!4.2)

HIV knowledge

1.3 (1.1!1.4)

1 (0.9!1.1)

Depression (CESDB16
versus CESD"#16)
Alcohol dependence

Ref
0.6 (0.3!1.3)

Ref
0.9 (0.5!1.9)

0.9 (0.8!1.1)

0.9 (0.8!1)

1.1 (1!1.3)

Age

1 (0.9!1.1)

1 (0.9!1.1)

1 (1!1.1)

1 (1!1.1)

1 (0.9!1.1)

Current income
B1000 Yuan/month
1000!2999.99 Yuan/month
3000"Yuan/month

Ref
0.6 (0.3!1.3)
0.9 (0.4!1.8)

Ref
1.3 (0.7!2.7)
0.8 (0.4!1.7)

Ref
1.8 (0.5!6.8)
4.1 (1.1!15.2)

Ref
1 (0.5!2.1)
1.1 (0.5!2.2)

Ref
0.6 (0.3!1.3)
0.8 (0.3!1.7)

Years away from home

B5 years
6!9
10"

Ref
1.3 (0.7!2.5)
0.6 (0.2!1.8)

Ref
1.1 (0.6!2.0)
1.0 (0.3!2.6)

Ref
2.6 (1.0!7.1)
3.5 (0.8!15.1)

Ref
1.2 (0.6!2.2)
0.6 (0.2!1.6)

Ref
1.1 (0.6!2.0)
0.9 (0.3!2.7)

(including both hard and soft drugs) compared to

those whose income was less than 1000 Yuan per
month. Alcohol dependence, as measured by the
CAGE scale in our survey, was associated with HIV
knowledge and education. Depression measured by
CES-D was not associated with any co-variate in the
model other than age. We found depression was
predicted by being forced to have sex by an adult
before age ten and having experienced sexual
violence. Those who had higher CES-D scores
were also more likely to report stress and dissatisfaction with life in Shanghai.
Discussion
This is the first quantitative study to systematically
examine the influence of a multitude of factors on
risk-taking behaviours in this invisible and high-risk
population of migrant male sex workers in a large
Chinese city guided by Chng et al.s (2003) tripartite
migration model. The mean age of money boys in
our study was 25 years. Like many of their peers in
China, they grew up in an age of dramatic economic
and social change. The open door policy that began
in 1986 removed long-standing taboos associated
with the free market, migration and commercial sex.
The driving force for migration from rural areas to
nearby major cities like Shanghai was primarily
economic.
Among the money boy participants, 12% of gayidentified money boys and 4% of non-gay-identified
money boys said they were married. This is consistent with findings from other studies of MSM in
China, India and other Asian countries (Go et al.,
2004; Kumar & Ross, 1991) that many MSM are
hidden from the mainstream society, leading a
normal life and fulfilling family obligations due to
social pressure. A prior qualitative study (He et al.,

2007) revealed that many money boys have families

in their rural hometowns whom they support financially and visit regularly in return for gaining some
degree of freedom in the city.
Our findings show that money boys had a high
prevalence of sexual risk behaviours (such as unprotected oral and anal sex), depression, alcohol
dependence, exposure to sexual violence, relatively
limited access to HIV testing and limited knowledge
of HIV/AIDS. Compared to the non-gay-identified
money boys, the gay-identified money boys were
more likely to engage in oral or anal sex, less likely to
use condoms in different settings, more likely to have
been forced to have sex by an adult before the age of
ten and more likely to experience sexual violence
from either their main partner or a client. Non-gayidentified money boys tended to be younger, more
mobile and more likely to use illegal drugs.
Other factors affecting HIV-related behavioural
and health-risk factors (HIV testing, sexual risk,
drug use, alcohol dependence and depression) were:
knowledge of HIV, age, current income and years
away from home. For example, HIV knowledge was
highly correlated with level of education. Those
participants who were more knowledgeable about
HIV were more likely to receive HIV testing, which
confirms the importance of HIV/STD-prevention
education for this high-risk group.
HIV screening tests are free throughout China
(Shao, 2006). Those diagnosed with HIV/ AIDS are
now able to receive free antiretroviral drugs per
national guidelines (Shao, 2006). Because the majority of reported HIV cases in Shanghai are among
migrants, the central government has given special
permission for HIV-infected migrants in Shanghai to
receive free antiretroviral treatment in Shanghai
without having to return to their hometowns
(Shao, 2006). Despite the existence of free HIV

Downloaded by [The University of British Columbia] at 14:09 28 February 2016

Chinese money boys and HIV risk

testing, only half of participants had ever been tested
for HIV. The testing rates are higher than for those
of general MSM reported in Hong Kong (Lau et al.,
2003, 2004a; 2004b; Lau, Siah, & Tsui, 2002;
Lau & Wong, 2002), probably reflecting their highrisk professional activities. An HIV prevalence of
3.1% (based on blood tests) was observed in MSM
in Beijing (Choi et al., 2004) so the self-reported
HIV-positive rate (1.6% in gay-identified money
boys and 2.1% in non-gay-identified money boys)
and STI infection rate (8.3% and 10.6% in gayidentified and non-gay-identified money boys) might
be underestimated in our study. Participants knowledge and attitudes towards HIV/ AIDS were also
unfavorable to HIV prevention.
Depression prevalence was high among money
boys in Shanghai, with no revealed differences based
on sexual orientation, HIV knowledge, income or
level of acculturation. Instead, depression was associated with stress, dissatisfaction with life and earlier
or current exposure to sexual violence. Sexual abuse
and violence were disclosed by some money boys,
especially by those who were gay-identified. Approximately 60% of money boys had CES-D scores
above the threshold score of 16 and 28% reached the
high probability of depression level. Future studies
targeting mental health services among this population are needed.
In sum, gay- and non-gay-identified money boys
in Shanghai face multiple layers of HIV risks.
Chinas conservative social culture, especially strong
in rural areas, prevents these money boys from
gaining access to knowledge about HIV and receiving HIV/STI testing, which increases the risk of
transmission. In addition, the economic and social
opportunities presented to these migrant money
boys by Chinas growing role in the global economy
may inadvertently promote risk behaviours. Systematic and larger-scale studies that target money
boys and employ Chng et al.s (2003) tripartite
model deserve critical examination. The public
health implications of HIV infection among migrant
money boys might be disastrous, as this population
carries the potential to act as a bridge population for
the rapid spread of HIV and STIs to their wives,
other MSM and male prostitutes and the general
population.
Limitations of the study
This study has several limitations. First, a crosssectional and correlational design was used, so
causation could not be inferred. Second, the use of
a snowball-type sampling or a truncated RDS
methodology may restrict generalizability. Third,
we were not able to measure pre- and post-migration
normative attitudes, beliefs and behaviours, thus

179

validation of Chng et al.s (2003) model is not

possible; although overall the model is heuristic
in predicting HIV and sexual risks. Finally, selfreported measures may underestimate HIV risk
behaviours and substance use/abuse. Nonetheless,
estimates of unprotected anal intercourse observed
appear comparable to the eight English languagebased studies.
Acknowledgements
The study was supported by a United States
National Institute on Drug Abuse grant
(R01DA15623-04S1) awarded to the first author.
Notes
1 The Choi 2002 and Choi 2003 articles use data from the same
parent study.
2 The instrument has undergone standard translation and backtranslation procedures. Both the Chinese and English versions
are available from the first author upon request.

References
amFAR (2006). MSM and HIV/AIDS risk in Asia: What is fueling
the epidemic among MSM and how can it be stopped?
Bangkok, Thailand: AmFAR. Available at: www.amfar.org/
binary-data/AMFAR_PUBLICATION/download_file/47.pdf
Anderson, A.F., Qingsi, Z., Hua, X., & Jianfeng, B. (2003).
Chinas floating population and the potential for HIV transmission: A social-behavioural perspective. AIDS Care, 15, 177!
185.
China National Bureau of Statistics (2002). Characteristics of
Chinese rural migrants: 2000. Beijing, China: China National
Bureau of Statistics.
Chinese State Council (1998). Chinese national medium- and
long-term strategic plan for HIV/AIDS prevention and control
(1998!2010). Available at: www.casy.org/Chindoc/lgAIDS
plan_98.htm (accessed 19 November, 1998).
Chng, C.L., Wong, F.Y., Park, R.J., Edberg, M.C., & Lai, D.S.
(2003). A model for understanding sexual health among Asian
American/Pacific Islander men who have sex with men in the
US. AIDS Education and Prevention, 15(Suppl. A), S21!S38.
Choi, K.H., Diehl, E., Guo, Y., Qu, S., & Mandel, J. (2002). High
HIV risk but inadequate prevention services for men in China
who have sex with men: An ethnographic study. AIDS and
Behavior, 6, 255!266.
Choi, K.H., Gibson, D.R., Han, L., & Guo, Y. (2004). High levels
of unprotected sex with men and women among men who have
sex with men: A potential bridge of HIV transmission in
Beijing, China. AIDS Education and Prevention, 16, 19!30.
Choi, K.H., Liu, H., Guo, Y., Han, L., Mandel, J.S., &
Rutherford, G.W. (2003). Emerging HIV-1 epidemic in China
in men who have sex with men. Lancet, 361, 2125!2126.
Cohen, M.S., Henderson, G.E., Aiello, P., & Zheng, H. (1996).
Successful eradication of sexually transmitted diseases in the
Peoples Republic of China: Implications for the 21st century.
Journal of Infectious Diseases, 174(Suppl. 2), S223!S229.
Go, V.F., Srikrishnan, A.K., Sivaram, S., Murugavel, G.K., Galai,
N., Johnson, S.C., et al. (2004). High HIV prevalence and risk
behaviors in men who have sex with men in Chennai, India.
Journal of Acquired Immune Deficiency Syndromes, 35, 314!319.
He, N., Detels, R., Zhu, J., Jiang, Q., Chen, Z., Fang, Y., et al.
(2005). Characteristics and sexually transmitted diseases of

Downloaded by [The University of British Columbia] at 14:09 28 February 2016

180

F. Y. Wong et al.

male rural migrants in a metropolitan area of Eastern China.

Sexually Transmitted Diseases, 32, 286!292.
He, N., Wong, F.Y., Huang, Z.J., Thompson, E.E., Fu, C., &
Ding, Y. (2007). Substance use and HIV risks among two types
of male migrants in Shang Hai, China. AIDS Care, 19, 109!
115.
He, Q., Wang, Y., Lin, P., Liu, Y., Yang, F., Fu, X., et al. (2006).
Potential bridges for HIV infection to men who have sex with
men in Guangzhou, China. AIDS Behaviour, 10(Suppl. 4), 17!
23.
Hong, Y., Stanton, B., Li, X., Yang, H., Lin, D., Fang, X., et al.
(2006). Rural-to-urban migrants and the HIV epidemic in
China. AIDS Behaviour, 10, 421!430.
Hu, Z., Liu, H., Li, X., Stanton, B., & Chen, X. (2006). HIVrelated sexual behaviour among migrants and non-migrants in a
rural area of China: Role of rural-to-urban migration. Public
Health, 120, 339!345.
Huang, R., & Yang, L. (2000). Studies on basic structure of the
floating population in China. Northwest Population, 4, 47!51.
Kumar, B., & Ross, M.W. (1991). Sexual behaviour and HIV
infection risks in Indian homosexual men: A cross-cultural
comparison. International Journal of STD and AIDS, 2, 442!
444.
Lau, J.T., Kim, J.H., Lau, M., & Tsui, H.Y. (2003). Prevalence
and risk behaviors of Chinese men who seek same-sex partners
via the internet in Hong Kong. AIDS Education and Prevention,
15, 516!528.
Lau, J.T., Kim, J.H., Lau, M., & Tsui, H.Y. (2004a). HIV related
behaviours and attitudes among Chinese men who have sex
with men in Hong Kong: A population based study. Sexually
Transmitted Infections, 80, 459!465.
Lau, J.T., Kim, J.H., Lau, M., & Tsui, H.Y. (2004b). Prevalence
and risk behaviors of Hong Kong males who seek cross-border
same-sex partners in mainland China. Sexually Transmitted
Diseases, 31, 568!574.
Lau, J.T., Siah, P.C., & Tsui, H.Y. (2002). A study of the STD/
AIDS-related attitudes and behaviors of men who have sex with
men in Hong Kong. Archives of Sexual Behaviour, 31, 367!373.
Lau, J.T., & Wong, W.S. (2002). HIV antibody testing among
male commercial sex networkers, men who have sex with men
and the lower-risk male general population in Hong Kong.
AIDS Care, 14, 55!61.
Li, X., Fang, X., Lin, D., Mao, R., Wang, J., Cottrell, L., et al.
(2004). HIV/STD risk behaviors and perceptions among ruralto-urban migrants in China. AIDS Education and Prevention, 16,
538!556.
Lin, D., Fang, X., Li, X., Xu, J., & Liu, H. (2005). HIV/AIDS
prevention and intervention in Chinese migrants. China Journal
of AIDS/STD, 11, 158!160.
Liu, H., Xie, J., Yu, W., Song, W., Gao, Z., Ma, Z., et al. (1998). A
study of sexual behavior among rural residents of China.
Journal Acquired Immune Deficiency Syndromes and Human
Retrovirology, 19, 80!88.
Ma, X., Ge, J., Zhang, J.Y., & Wen, W. (2004). [Survey of
behaviors and knowledge about HIV/AIDS among intravenous
drug users at a city in Sichuan Province]. Sichuan Da Xue Xue
Bao Yi Xue Ban, 35, 376!378.
Mayfield, D., McLead, G., & Hall, P. (1974). The CAGE
questionnaire validation of a new alcoholism screening instrument. American Journal of Psychiatry, 131, 1121!1123.

McCoy, V.H., & Yu, Z. (1999). The effect of migration patterns

on exposure to HIV prevention in a migrant community.
Population Research and Policy Review, 18, 155!169.
Merli, G. (2006). Unbalanced sex ratios, migration and spread of
HIV in China. Paper presented at the International Symposium
on Urbanization, Gender and Public Health, Shanghai, December 14!15.
Ministry of Health, UNAIDS, & WHO (2006). 2005 Update on the
HIV/AIDS epidemic and response in China. Beijing: National
Center for AIDS/STD Prevention and Control, China CDC.
Qi, S., Wang, Q., Zhang, G., et al. (2006). Urine screening of
chlamydial and gonococcal infections and associated risk
factors in male sex workers. China Journal of Leprosy and Skin
Diseases, 22, 819!821.
Qian, H.Z., Vermund, S.H., & Wang, N. (2005). Risk of HIV/
AIDS in China: Subpopulations of special importance. Sexually
Transmitted Infections, 81, 442!447.
Qiao, X., Guo, X., Zhang, L., et al. (2000). HIV infection among
migrants in Shanxi Province. Chinese Journal of STD and HIV
Prevention and Treatment, 6, 265!267.
Radloff, L.S. (1977). A self-report depression scale for research in
the general population. Applied Psychological Measurement, 1,
385!401.
Shao, Y. (2006). AIDS epidemic at age 25 and control efforts in
China. Retrovirology, 3, 87.
Thomas, J. (1998). HIV/AIDS in China: Migrant population,
drug injection responsible for increased transmission. AIDS
Link, 49, 12!14.
UNAIDS (2004). A joint assessment of HIV/AIDS prevention,
treatment and care in China. Paper presented at the UNTG on
HIV/AIDS, Beijing, China.
UNAIDS (2006). Global AIDS epidemic continues to grow.
Available at: http://data.unaids.org/pub/PressRelease/2006/
061121_2006_EPI_Update_en.pdf
Wang, Q., & Ross, M.W. (2002). Differences between chat room
and e-mail sampling approaches in Chinese men who have sex
with men. AIDS Education and Prevention, 14, 361!366.
Wu, Z. (2006). Personal communication.
Xia, G. (2006). Advancement and reformation of Chinas HIV/
AIDS social policies. Paper presented at the International
Symposium on Urbanization, Gender and Public Health,
Shanghai, December 14!15.
Xu, C., Wu, Z., & Zhang, Y. (1998). A study on HIV prevalence
among rural-to-urban migrants. Paper presented at the in the
proceedings of the selected studies presented on Symposium on
AIDS Prevention and Control in China, Beijing, China.
Zhang, B., Liu, D., Li, X., & Hu, T. (2000). A survey of men who
have sex with men: Mainland China. American Journal of Public
Health, 90, 1949!1950.
Zhang, B.C., & Chu, Q.S. (2005). MSM and HIV/AIDS in China.
Cell Research, 15, 858!864.
Zhang, L. (2001). Strangers in the city: Reconfigurations of space,
power and social networks within Chinas floating population.
Stanford: Stanford University Press.
Zheng, L., Zhu, J., Tian, P., Chen, Y., Song, Y., & Chen, Y.S.
(2000). Sexual behaviors and associated factors among unmarried female migratory workers in Guangzhou. Chinese
Journal of Family Planning, 60, 162!164.
Zhuang, T., Mao, X., Liu, G., et al. (2005). An HIV/AIDS related
survey of 86 men who have sex with men. China Journal of
AIDS and STDs, 11, 329!332.