We describe the behavioral characteristics and sexually transmitted disease (STD) prevalence of Chinese men who have sex with men (MSM) (n = 41) from a national probability sample of men (n = 1861). Most MSM were partnered with females (97%) and had a low rate of consistent condom use (7%). More MSM than heterosexual men self-reported a prior STD and risky sexual behaviors. MSM may act as a bridge for HIV transmission to female partners. Targeted interventions may help prevent a generalized HIV epidemic in China.
In some parts of China, HIV prevalence has exceeded 1% of the general population.1,2 Men who have sex with men (MSM) accounted for 7.3% of HIV infections3 and 11% of new HIV infections in 2007.4 The proportion of HIV infections among MSM is likely to grow, as studies have documented increasing HIV prevalence and high prevalence of risky sexual behaviors and of sexually transmitted diseases (STDs) among this population.5–12
MSM may serve as a bridge population for transmitting HIV to the general population. Empirical data have shown that a large proportion of Chinese MSM are either married or have female sex partners.5,13 High-risk sexual behaviors with both male and female partners among MSM are also common. Surveys have documented extremely high rates of inconsistent condom use with female partners.13–15 We report the sociodemographic and sexual behavioral characteristics and the STD prevalence of a national probability sample of Chinese heterosexual men and men who had homosexual experiences.
A Chinese household survey (not including Hong Kong and Tibet) of adults aged 20 to 64 years was implemented between 1999 and 2000 with probabilistic sampling, as part of the Chinese Health and Family Life Survey. The procedures in this study were previously reported in detail elsewhere.16 Of the 5000 individuals sampled nationally, 3813 completed the computerized interview and 3426 (participation rate = 69%) also provided a urine sample.
Adult MSM (n = 41) and heterosexual men (n = 1820) were identified through participants' reply to the question, “Have you ever had sex with another man?” Sociodemographic characteristics, including place of residence, education, employment, marital status, and monthly income, were also recorded. Condom use with spouse or primary sex partner, sexual encounters with nonprimary partners, and whether participants ever gave or received money, gifts, or valuables for sex were assessed. Finally, participants self-reported lifetime history of any STD. Urine specimens were collected and tested for gonorrhea and chlamydia by standard laboratory methods.
Of 1861 male participants who answered the question concerning MSM behavior, 41 (2.2%) reported ever having had sex with men (Table 1). MSM appeared to be younger than heterosexuals (42% vs 27% were younger than 30 years old), although the difference was statistically nonsignificant. A majority of MSM (85%) did not attend any form of college, but they were significantly more likely than were heterosexual men to have attained more than a primary school education (98% vs 84%; P = .041). Heterosexual men were more likely to be married than were MSM (85% vs 68%; P = .027), but it is important to recognize that most MSM were currently married.
Sociodemographic Characteristics and Sexual Behaviors of Male Study Participants: Chinese Health and Family Life Survey, 1999–2000
|No./Total Sample||%||No./Total Sample||%||No./Total Sample||%||Pa|
|Primary school or less||299/1861||16.1||1/41||2.4||298/1820||16.4|
|Current location of residence||.116b|
|Monthly income, yuan||.911b|
|Gender of spouse/primary sex partner||.021b|
|Last time had sex with spouse/primary sex partner||.262b|
|Within last 2 wk||1198/1645||72.8||23/34||67.6||1175/1611||72.9|
|Within last mo||202/1645||12.3||5/34||14.7||197/1611||12.2|
|Within last y||161/1645||9.8||2/34||5.9||159/1611||9.9|
|1 y ago or more||84/1645||5.1||4/34||11.8||80/1611||5.0|
|Frequency of sex with spouse/primary sex partner||.730b|
|More than once per wk||892/1561||57.1||17/30||56.7||875/1531||57.2|
|2–3 times per mo||443/1561||28.4||10/30||33.3||433/1531||28.3|
|Once per mo or less||226/1561||14.5||3/30||10.0||223/1531||14.6|
|Condom use with spouse/primary sex partner||.367b|
|Had sex with others while with spouse/primary sex partner||.047|
|Received money, gifts, or valuables for sex in past y||.128b|
|Gave money, gifts, or valuables for sex in past y||.031|
Note. MSM = men who have sex with men. Sum of total sample varies because of missing values.
aP value obtained from the Pearson χ2 test for categorical variables unless otherwise indicated.
bP value obtained from the Fisher exact test
cJunior college or above.
Of the 34 MSM who reported having a spouse or primary sex partner (hereafter called “partner”), the partner was female in all but 1 case (Table 1). Most MSM (82%) reported having had sex with their partner within the past month, and 90% reported having had sex with their partner at least 2 to 3 times per month on average. Consistent condom use with female partners was low for both MSM and heterosexuals (7% and 6%, respectively); 53% of MSM had never used condoms with their partner in the past year. MSM were significantly more likely than heterosexuals to have had sex outside of their primary partnership (32% vs 19%; P = .047) and to have given money, gifts, or valuables for sex in the past year (17% vs 8%; P = .031).
Two percent of men in our sample self-reported ever having had an STD (Table 2). MSM were significantly more likely than heterosexuals to report ever having had an STD (11% vs 2%; P = .001). There was a trend difference in the cross-sectional prevalence of gonorrhea- or chlamydia-positive test results between MSM and heterosexuals (8% vs 3%; P = .074).
Prevalence of Sexually Transmitted Diseases (STDs) Among Male Study Participants as Determined by Self-Reports and Urine Test Results: Chinese Health and Family Life Survey, 1999–2000
|Lifetime STD||No./Total Sample||% (95% CI)||No./Total Sample||% (95% CI)||No./Total Sample||% (95% CI)||Pa|
|Yes||34/1752||1.9 (1.3, 2.7)||4/38||10.5 (2.9, 24.8)||30/1714||1.8 (1.2, 2.5)|
|No||1643/1752||93.8 (92.5, 94.9)||30/38||78.9 (62.7, 90.4)||1613/1714||94.1 (92.2, 95.2)|
|Do not know||75/1752||4.3 (3.4, 5.3)||4/38||10.5 (2.9, 24.8)||71/1714||4.1 (3.2, 5.2)|
|Positive||44/1685||2.6 (1.9, 3.5)||3/38||7.9 (1.7, 21.4)||41/1647||2.5 (1.8, 3.4)|
|Negative||1641/1685||97.4 (96.5, 98.1)||35/38||92.1 (78.6, 98.3)||1606/1647||97.5 (96.6, 98.2)|
|Positive||2/1685||0.1 (0.01, 0.43)||0/38||0 (0, 9.3)c||2/1647||0.1 (0.01, 0.44)|
|Negative||1683/1685||99.9 (99.6, 100)||38/38||100.0 (90.7, 100)c||1645/1647||99.9 (99.6, 100)|
|Positive||42/1685||2.5 (1.8, 3.4)||3/38||7.9 (1.7, 21.4)||39/1647||2.4 (1.7, 3.2)|
|Negative||1643/1685||97.5 (96.6, 98.2)||35/38||92.1 (78.6, 98.3)||1608/1647||97.6 (96.8, 98.3)|
Note. CI = confidence interval; MSM = men who have sex with men. Participants who knew the term “STD” self-reported lifetime history of any STD.
aP value obtained from the Fisher exact test.
bDetermined by a urine test.
cOne-sided 97.5% confidence interval.
In this population-based Chinese probability sample, an estimated 2.2% (95% confidence interval [CI] = 1.6, 3.0) of adult males aged 20 to 64 years were MSM. Even this modest prevalence rate produces a conservative estimate of 7 000 000 MSM in China. We found that these men had regular sex and inconsistent condom use with their female partners. Compared with heterosexual men, MSM had higher rates of STDs and engaged in more risky sexual behaviors. As most MSM were married to females, their female partners could be exposed to STDs and at elevated risk for HIV infection. The concern that Chinese MSM may be a potential bridge for HIV transmission to the general population is therefore valid and deserves more focused investigation.
This study had a small sample size of MSM (n = 41), thereby limiting our power to detect statistically significant relationships. We were also not able to examine variables related to sex partners and behaviors in more detail. We did not have MSM-specific questions relating to sexual behavior (e.g., unprotected anal intercourse) and other contextual issues (e.g., homophobia, stigma). Some MSM may have chosen not to participate in the survey or may have denied MSM behavior because of societal stigma. Our findings may therefore be an underestimate of STDs and sexual risk behaviors among Chinese MSM.
Other methods to reach representative samples of MSM, such as time-location sampling or respondent-driven sampling, may be better positioned to enhance our understanding of the risk behaviors of MSM. Additionally, STD and HIV behavioral surveillance needs to be regularly conducted to monitor trends over time. Whereas interventions in the West have been developed in the almost exclusive context of MSM-only behavior, interventions in China will need to take into account the unique context of bisexuality evident among Chinese MSM. Qualitative research aimed at understanding these sociocultural contexts is therefore essential for the development of surveys and interventions.
Finally, it is of note that the prevalence of HIV infection in China is still relatively low. We therefore have a rare opportunity in HIV prevention work: time to act. The HIV epidemic among MSM in Thailand17,18 strongly suggests that this window of opportunity to prevent a widespread HIV epidemic among MSM and, potentially, their female partners will not last. The time to act is now.
We thank Mark S. Friedman, Michael P. Marshal, and H. Fisher Raymond for their invaluable feedback on this article. We also thank the original investigators of this study for making the data publicly accessible. Finally, we thank the Chinese participants for their time and effort.
Human Participant Protection
The assessment protocol was reviewed and approved by the institutional review boards of the University of Chicago, Renmin University, and Peking Union Medical College. The data analysis portion of the study was considered nonhuman subjects activity (publicly accessible data) and thereby did not need institutional review board approval by the University of Pittsburgh and Georgetown University.