Secondary Logo

Journal Logo

Original Study

Concurrent Sexual Partnerships Among Men Who Have Sex With Men in Shenzhen, China

Ha, Toan H. MD, DrPH*; Liu, Hongjie PhD*; Liu, Hui MD; Cai, Yumao MS; Feng, Tiejian MD

Author Information
Sexually Transmitted Diseases: August 2010 - Volume 37 - Issue 8 - p 506-511
doi: 10.1097/OLQ.0b013e3181d707c9
  • Free

The rates of HIV and sexually transmitted diseases (STDs) among men who have sex with men (MSM) are dramatically increasing in China in recent years.1–4 Sexual networks may contribute to the fast HIV/STD transmission among Chinese MSM as the role of sexual networks in determining the spread of HIV/STDs has been demonstrated in several mathematical models5–8 and empirical studies.9–11 Sexual network is defined by a set of people who have sexual interrelationship.12 The sexual network approach in studying sexual risk differs from the traditional epidemiologic approach by focusing on the dynamic changes of sexual partnerships and the network linkage in a group.13 An individual's risk is not only determined by one's own behavior, but one's partner's.14

One important aspect of the sexual network is concurrent sexual partnership. Concurrent sexual partnership is conceptualized as partnerships overlapping in time.15 According to Morris & Kretzschmar, concurrent partnerships not only exponentially increase the number of infected persons, but accelerate the growth rate of the HIV epidemic throughout the initial phase as well.7 The presence of concurrent partnerships may make the speed of infection transmission through a sexual network much faster than sequential monogamy partnerships.16 By using a data-driven network simulation model, Morris et al., revealed that concurrent partnerships affected the infection spread by reducing the average time to the secondary transmission. Compared to the time among those who engage in monogamous partnerships, the time interval between acquiring an infection and passing on the infection to other by an individual with concurrent partners is shortened by approximately one-third.14 Concurrent partnerships have been identified as an important factor associated with the spread of Chlamydia infection in Colorado10 and has been implied as one of the factors contributing to the high prevalence of HIV infection in Africa.17 A recent study in the United States documented that MSM who reported completely concurrent partnerships had a greater number of sexual episodes and unprotected sexual episodes per partnership than those whose partnerships were not completely concurrent did. The higher number of unprotected sex within completely overlapping partnerships might lead to higher possibilities of HIV transmission among MSM population.18 Previous research in United States has documented a number of factors associated with the possession of concurrent partnerships, such as younger age,19 race,14 gender,20–22 being single,19 and early sexual initiation.23 However, few empirical studies have examined HIV-related risk and factors that are associated with the possession of concurrent partnerships in China.

Homosexuality is viewed with disapproval in traditional Chinese society.24 The same-sex relationship is considered immoral or abnormal.25 However, with China's economy opened to the world in late 1970s, and the government's triple-no policy on homosexuality (characterized as no explicit law against homosexuality, no governmental regulations protecting MSM from discrimination, and no law promoting MSM' rights), MSM communities have increased rapidly in both urban and rural areas in recent decades.26,27 Information about gay community activities is widely available in a large number of websites in China.28 An estimated 3.6 to 7.1 million MSM were living in China in 2005.29 Similar to other countries,30,31 Chinese MSM includes those who have only male partners and those who have both male and female partners. Given the culturally favorable procreation and social order, homosexuality continues to be widely stigmatized in China.24,32 Sons are expected to get married, have children and pass on the family name.32 The cultural imperative of familial responsibility, rather than individual rights, ensures that marriage and procreation are not just personal choices, but family ones.29 Because of the societal and family pressure, a substantial number of Chinese MSM gets married with women while continuing to have a secret homosexual relationship.33 Given that male partnerships are kept secret, MSM may choose not to use condoms with female sexual partners to avoid the suspicion of other secret concurrent male partnerships.34 This risky sexual behavior along with concurrent sexual practice may make bisexual MSM as a bridge group for the spread of HIV to the general population in China. In addition, a proportion of Chinese MSM engages in commercial sex.35 The variety of sexual partners, coupled with the high percentage of unprotected sex,36,37 makes Chinese MSM highly vulnerable to HIV infections, which may facilitate the expansion of the HIV epidemic among this population and beyond.

Limited studies has documented that the proportion of Chinese MSM who had concurrent sexual partners with males only was 33%, and between 15% to 17% with both males and females.4,34 However, few empirical studies have compared HIV risk behaviors between MSM with and without concurrent partnerships in China. The objective of this study was to examine and compare HIV knowledge, perceived risk of HIV infection and consistent condom use between MSM with and without concurrent sexual partnerships, and between MSM with only male concurrent partners and MSM with both male and female concurrent partners.


Study Site and Subjects

The study site has been previously described.35 A cross-sectional study was conducted among MSM between November and December 2007 in Shenzhen, China. Shenzhen, the first special economic region in China, is located along the southern coast of China, bordering Hong Kong and Guangzhou. It is estimated that there are 60,000 MSM in the city.38 A man was eligible for this study if he (1) was between 18 to 45 years old; (2) reported having engaged in anal intercourse with one or more men in the past year; and (3) had lived in Shenzhen for more than 3 months at the time of the interview.

Respondent-Driven Sampling

Respondent-driven sampling (RDS) approach was used to recruit MSM.39 A group of 12 MSM was initially selected and served as “seeds.” MSM patrons who worked in MSM congregation venues helped the researchers to select the initial seeds. The seeds were diverse in age (18–30 and 31–45 years old), MSM congregation venues (sauna, bar, and public park), or having sex with other MSM (commercial or noncommercial sex) in the past 6 month. These 12 seeds received an explanation of the study purpose and 3 coupons to recruit MSM peers from their network ties. The coupon included project name, address and phone number of the interview site, office house, and a bus-route map to the site. The unique serial number on each coupon linked each subject to his recruiters. All new recruits in the subsequent waves (wave 1, 2, 3, 4 and 5) received an anonymous interview and offered three coupons as were the seeds. We obtained the expected sample size of 351 eligible MSM after 4 to 5 wave of RDS recruitment. Eligible subjects received a face-to-face anonymous interview in a private setting. All interviewers underwent an intensive training in interviewing techniques, developing rapport and ensuring confidentiality. A pilot test of interviewers' skills, and RDS handling procedures was performed with 10 MSM.


HIV/AIDS Knowledge.

Knowledge regarding HIV transmission and prevention was measured by 11 true/false/unsure questions. Participants were asked to indicate what pathways they thought a person could be infected with HIV (e.g., having sex with a HIV/AIDS patient without using a condom). Each correct answer was given one point, with a total score ranging from 0 to 11 points. Based on its distribution, HIV knowledge was categorized into low (0–6), medium (7–9), and high (10–11) in the data analysis.

Perceived Risk of HIV Infection.

Perceived risk of HIV infection was measured using one question asking participants' perception about their possibility of acquiring HIV. The response format was a 5-point scale ranging from at no risk at all to at extremely high risk. In the data analysis, the variable was dichotomized into “having no risk” (no risk at all) and “having risk” (some possibility, large possibility, and I am sure that I will be infected).

Condom Use.

Condom use was measured by asking respondents about condom use with each type of sexual partners during the last 6 months. Consistent condom use was defined as using condoms in every sexual act, including vaginal and/or anal sex in the past 6 months and inconsistent condom use were defined as not using condoms for every sexual act in the past 6 months.

Concurrent Sexual Partnerships.

Concurrent sexual partnerships were assessed by comparing dates of first and last sexual intercourse with sexual partners in the past 6 months. Overlapping of these dates was categorized as “having concurrent partnerships.”

Data Analysis

Data were entered independently by 2 research associates into the database using EpiData software 2.1b (EpiData Association, Odense, Denmark). We used Respondent Driven Sampling Analysis Tool (RDSAT, version 5.6, RDS Incorporated, Ithaca, NY) to estimate proportions of variables of interest (e.g., concurrent sexual partnerships; male concurrent sexual partners, and both male and female concurrent sexual partners). According to the RDS methodology, there are several potential biases that need to be taken into account when analyzing RDS data: recruitment patterns across groups (groups more effectively recruited would be overrepresented in the sample), network size (those with larger network sizes would be oversampled as more recruitment paths lead to their members), and homophily (groups that recruit only within their own group would be overrepresented).40–42 The RDSAT statistical program is designed to adjust for these biases.

As the current version of RDSAT does not provide options for multivariate analyses, the multivariate analyses were performed using SAS 9.2 software (SAS Institute, Cary, NC). We used logistic regression models to explore the relationships among concurrent relationships, HIV/AIDS knowledge, perceived risk of HIV infection, and condom use, controlling for social-demographic variables (age, marriage, occupation, education, and number of sexual partners).


Sample Characteristics and RDS-Adjusted Proportion of Sexual Behaviors

A total of 351 MSM were recruited and interviewed. The mean age was 27 years old (range: 18–44 years old, standard deviation = 6.3). About two-thirds of respondents (65%) received a high-school or above education and 38% worked in the entertainment venues such as bar, hair salon, bathhouse, karaoke bar, and dancing hall. A majority of the respondents (78%) were single.

As shown in Table 1, 49% of respondents reported having concurrent sexual partnerships during the past 6 months. Among those who reported having concurrent sexual partnerships, 62% reported having only male partners and 38% had both male and female partners. Among MSM who had concurrent partnerships with both male and female partners, 38% reported having concurrent partnerships with either male or female commercial sex workers. Among those without concurrent sexual partnerships, 93% had sex with only male partners and 7% had both sequential monogamy male and female partners. For those who had sequential partnerships with both male and female partners, only 15% reported having sex with either male or female commercial sex workers.

Proportion Estimates for Sexual Behaviors Among MSM

Length of Sexual Partnerships

The total number of sexual partners reported by MSM in the past 6 months was 2,382. Each MSM reported an average of 6.8 sexual partners. For all MSM, the average length of male partnerships were 10.2 months (range: 0.5–145.1); whereas the mean length of female partnerships was 43.6 months (range: 1–259.8). Among those with concurrent sexual partnerships, the average length of male partnership were 9.9 months (range: 0.5 –145.1), and the mean length of female partnership was 42.6 months (range: 1–259.8).

Characteristics of MSM With and Without Concurrent Sexual Partnerships

Table 2 presents the RDS estimated proportions by demographic characteristics, HIV/AIDS knowledge, perceived risk of HIV infection, and condom use among MSM with and without concurrent partnerships. MSM with concurrent partnerships (71% and 95% CI: 60.2–79.2) were more likely to work in entertainment venues than those without concurrent partnerships (43% and 95% CI: 23.7–44.9; P < 0.05). Although the differences were not statistically significant, compared to MSM without concurrent partnerships, MSM with concurrent partnerships tended to report (1) poorer HIV/AIDS related knowledge (32% and 95% CI: 28.8–51.5 vs. 55% and 95% CI: 44.5–73.1; P > 0.05), and (2) a higher proportion of inconsistent condom use (42% and 95% CI: 34.7–55.2 vs. 30% and 95% CI: 22.5 – 46.4; P > 0.05).

Comparisons Between MSM With and Without Concurrent Sexual Partners

Characteristics of MSM With Only Male Concurrent Partners and MSM With Male and Female Concurrent Partners

As shown in Table 3, compared to MSM with only male concurrent partners, MSM with both male and female concurrent were more likely to be older (35% and 95% CI: 14.4–44.4 vs. 80% and 95% CI: 50.9–97.7; P < 0.05). While 24% (95% CI: 17.1–48.3) of MSM with only male concurrent partners possessed a higher level of HIV knowledge, only 16% (95% CI: 14.5–78.1) of MSM with both male and female concurrent partners reported a high level of HIV knowledge. However, the difference was not statistically significant. The proportions of consistent condom use were similar between the 2 groups (52% and 95% CI: 29.8–58.3 vs. 57% and 95% CI: 39.8–96.8; P > 0.05).

Comparisons Between MSM With Only Male Concurrent Partners and MSM With Both Male and Female Concurrent Partners

Multivariate Analysis of Factors Associated With Concurrent Sexual Partnerships

Results from logistic regression analysis documented that MSM who had concurrent sexual partners had a lower level of HIV/AIDS knowledge than those without concurrent partners. No statistically significant differences were found in perceived risk of HIV infection and consistent condom use.


This study documents that about half of MSM reported having had concurrent sexual partners among MSM. Among MSM who reported concurrent partnerships, 38% were engaged in concurrent partnerships with males and females. These figures are higher than that of other studies conducted in China4,34 and in the United States.43 As demonstrated by Choi et al., while relationships with females tended to be longer, the duration of relationships with male partners were of shorter. The longer duration of female partnerships could increase the chance for those relationships to be concurrent with other partnerships.34 Further, a substantial proportion of MSM who had both male and female partners did not consistently use condoms.4,37 Taken together, our results provide evidence that MSM may serve as a bridge for HIV transmission to the general population. A similar pattern has been reported in a study among men in Thailand, in which a group of men reported using condoms inconsistently while having sex with both marital and commercial sex partners.9 Given a large number of MSM engaging in concurrent partnerships and the increased prevalence of HIV among MSM,2 the spread of HIV to the general population could be accelerated unless effective HIV interventions for MSM are implemented in China.

MSM with concurrent partnerships reported a high level of inconsistent condom use and a high level of perceived risk of HIV infection. This finding indicates that, despite having awareness of the risk of HIV infection, MSM with concurrent partnerships did not use condoms consistently when engaging in sexual activities. The possible explanation could be due to the barriers of surrounding environments that prevent MSM from using condoms. As argued by Weinstein and Nocolich, people may continue to engage in high risk behaviors while acknowledging the danger of acquiring HIV. These individuals fail to reduce their risk not because they do not want to adopt the preventive behaviors, but because they may face obstacles that inhibit the protective actions, for instance, stigma attached to condom use, negative attitudes toward condom use, or partners do not endorse safe-sex practices.37,44

Another important finding of this study is that sexual concurrency was not significantly associated with consistent condom use. Similar findings have been reported by other studies.43,45 This finding implies, as supported by an argument made by Reihman et al., the lack of the association between concurrent partnerships and condom use indicates that individuals who engage in sexual activities outside their steady relationships may not act to protect their partners from contracting HIV/STDs.43 The possible explanation to this finding is that MSM with concurrent partnerships may not want to use condoms in order to demonstrate their commitment to a faithful relationship. For example, a study among MSM in Beijing, China found that 61% of the respondents cited a mutually faithful sexual relationship as the reason for not using condoms.37 Similarly, greater relationship commitment was found to be negatively associated consistent condom use among young American adult.43 Given that homosexuality is largely unaccepted and widely stigmatized by Chinese society,32,36 MSM may choose not to use condoms to avoid the suspicion of their secret concurrent male partnerships when they have sex with their wives or regular male partners.34

There are several limitations in our study. Given the nature of the cross-sectional design, our findings should not be interpreted beyond associations, and the role of causality may not be inferred. Generalization of the findings to other MSM populations cannot be made as subjects were recruited from one city which is not representative of all cities in China. Information about sexual behaviors were self-reported, as such may suffer from reporting bias. Because the current version of RDSAT does not provide options for RDS multivariate analyses, we used the traditional multivariate analytical approach to examine factors associated with sexual concurrency. Given that the traditional multivariate analysis could not control for biases (homophily, differential network sizes and recruitment patterns) in the RDS recruitment process, the results may be suffered from the biased estimates, for instance, the 95% CI presented in Table 4 may be narrower than their actual 95% CI. Currently, there is no consensus on the appropriate statistical methods for using RDS data in multiple regression analyses.46

Multiple Logistic Regression Analysis of Correlates of Concurrent Partnerships

Our study added to the literature by differentiating characteristics of MSM with and without concurrent partnerships, MSM with only male concurrent partners and MSM with male and female concurrent partners among Chinese MSM population. The differences between these MSM groups should be addressed in the HIV intervention programs in China and other countries with similar settings. The findings also underscore the fact that HIV behavioral interventions need to target not only individual risk behaviors but also take into consideration sexual contextual factors resulting in the concurrent partnerships in China.


1. Jiang J, Cao N, Zhang J, et al. High prevalence of sexually transmitted diseases among men who have sex with men in Jiangsu province, China. Sex Transm Dis 2006; 33:118–223.
2. Wang L, Wang N, Wang L, et al. The 2007 estimates for people at risk for and living with HIV in China: Progress and challenges. J Acquir Immun Defic Syndr 2009; 50:414–418.
3. Hong F, Zhou H, Cai Y, et al. Prevalence of syphilis and HIV infections among men who have sex with men from different settings in Shenzhen, China: Implications for HIV/STD surveillance. Sex Transm Infect 2009; 85:42–44.
4. Choi K, Ning Z, Gregorich ES, et al. The influence of social and sexual networks in the spread of HIV and syphilis among men who have sex with men in Shanghai, China. J Acquir Immun Defic Syndr 2007; 45:77–84.
5. Watts CH, May RM. The influence of concurrent partnerships on the dynamics of HIV/AIDS. Math Biosci 1992; 108:89–104.
6. Kretzschmar M, Morris M. Measures of concurrency in networks and the spread of infectious disease. Math Biosci 1996; 133:165–195.
7. Morris M, Kretzschmar M. Concurrent partnerships and the spread of HIV. AIDS 1997; 11:641–648.
8. Potterat JJ, Rothenberg RB, Muth SQ. Network structural dynamics and infectious disease propagation. Int J STD AIDS 1999; 10:182–185.
9. Morris M, Podhisita C, Wawer JM, et al. Bridge population in the spread of HIV/AIDS in Thailand. AIDS 1996; 10:1265–1271.
10. Potterat JJ, Zimmerman-Rogers H, Muth SQ, et al. Chlamydia transmission: Concurrency, reproduction number, and the epidemic trajectory. Am J Epidemiol 1999; 150:1331–1339.
11. Aral SO, Hughes JP, Stoner B, et al. Sexual mixing patterns in the spread of gonococcal and chlamydial infections. Am J Public Health 1999; 89:825–833.
12. Scott J. Social Network Analysis: A Handbook, 1st ed. London, CA: Sage Publications, 1991.
13. Doherty AI, Padian SN, Marlow C, et al. Determinants and consequences of sexual networks as they affect the spread of sexually transmitted infections. J Infect Dis 2005; 191(suppl 1):S42–S54.
14. Morris M, Kurth EA, Hamilton TD, et al. Concurrent partnerships and HIV prevalence disparities by race: Linking science and public health practice. Am J Public Health 2009; 99:1023–1031.
15. Adimora A, Schoenbach JV. Social context, sexual networks, and racial disparities in rates of sexually transmitted infections. J Infect Dis 2005; 191(suppl 1):S115–S122.
16. Morris M, Kretzschmar M. Concurrent partnerships and transmission dynamics in networks. Soc Networks 1995; 17:299–318.
17. Halperin TD, Epstein H. Concurrent sexual partnerships help to explain Africa's high HIV prevalence: Implications for prevention. Lancet 2004; 364:4–6.
18. Bohl DD, Raymond HF, Armold M, et al. Concurrent sexual partnerships and racial disparities in HIV infection among men who have sex with men. Sex Transm Infect 2009; 85:367–369.
19. Adimora AA, Schoenbach JV, Martinson F, et al. Concurrent sexual partnerships among African Americans in the rural south. Ann Epidemiol 2004; 14:155–160.
20. Adimora AA, Schoenbach JV, Doherty AI. Concurrent sexual partnerships among men in the United States. Am J Public Health 2007; 197:2230–2237.
21. Adimora AA, Schoenbach VJ, Martinson FE, et al. Concurrent partnerships among rural African Americans with recently reported heterosexually transmitted HIV infection. J Acquir Immun Defic Syndr 2003; 34:423.
22. Ford K, Sohn W, Lepkowski J. Am adolescents: Sexual mixing patterns, bridge partners, and concurrency. Sex Transm Dis 2002;19:13–19.
23. Manhart EL, Aral SO, Holmes KK, et al. Sex partner concurrency: Measurement, prevalence, and correlates among urban 18–39-year-olds. Sex Transm Dis 2002; 29:133–143.
24. Liu H, Yang H, Li X, et al. Men who have sex with men and human immunodeficiency virus/sexually transmitted disease control in China. Sex Transm Dis 2006; 33:68–76.
25. Liu JX, Choi K. Experiences of social discrimination among men who have sex with men in Shanghai, China. AIDS Behav 2006; 10:S25–S33.
26. Chen P. Homosexuality, sexually transmitted diseases and health education. Chinese J Dermatol Venereal Dis 2003; 17:62–63.
27. Elegant S. Postcard: Beijing. Time 2008; 171:8–9.
28. Wang QR, Ross MW. Difference between chat room and e-mail sampling approaches in Chinese men who have sex with men. AIDS Educ Prev 2002; 14:361–366.
29. Liu H, Feng T, Liu H, et al. Egocentric networks of Chinese men who have sex with men: Network components, condom use norms and safer sex. AIDS Patient Care STDs 2009; 23:885–893.
30. Wold CM, Seage GR, Lenderking WR, et al. Unsafe sex among men who have sex with men and women. Paper presented at: HIV Infection Women Conference; February 22–24, 1995; Washington, DC. Available at: Accessed May 10, 2009.
31. Flores SA, Bakeman R, Millett GA, et al. HIV risk among bisexually and homosexually active racially diverse young men. Sex Transm Dis 2009; 36:325–329.
32. Choi K, Liu XJ. HIV/AIDS in Asia: HIV and men who have sex with men in China: The potential for a rapid increase in infection. J AIDS Res 2006; 8:1–6.
33. Zhang B, Li X, Shi T. A primary estimation of the number of population and HIV prevalence in homosexual and bisexual men in China. J China AIDS/STD Prevent Control 2002; 8:197.
34. Choi KH, Hudes ES, Steward WT. Social discrimination, concurrent sexual partnerships, and HIV risk among men who have sex with men in Shanghai, China. AIDS Behav 2008; 12(suppl 4):S71–S77.
35. Liu H, Liu H, Cai Y, et al. Money boys, HIV risks, and the associations between norms and safer sex: A respondent-driven sampling study in Shenzhen, China. AIDS Behav 2009; 13:652–665.
36. Li X, Shi W, Li D, et al. Predictors of unprotected sex among men who have sex with men in Beijing, China SE Asian. J Trop Med Public Health 2008; 39:99–108.
37. Choi K, Gibson RD, Han L, et al. 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 Educ Prev 2004; 16:19–30.
38. Feng T, Cai W, Tan J, et al. Using capture-recapture method to estimate size of MSM in Shenzhen. Paper presented at: China HIV/AIDS Surveillance Annual Report; 2005; Haebin, China.
39. Heckathorn DD. Respondent-driven sampling: A new approach to the study of hidden populations. Soc Probl 1997; 44:174–199.
40. Abdul-Quader SA, Heckathorn DD, Sabin K, et al. Implementation and analysis of respondent driven sampling: Lessons learned from the field. J Urban Health 2006; 83(suppl 6):S1–S5.
41. Heckathorn D. Respondent Driven Sampling. Available at: Accessed December 4, 2009.
42. Lansky A, Abdul-Quader AS, Cribbin M, et al. Developing an HIV behavioral surveillance system for injecting drug users: The national HIV behavioral surveillance system. Public Health Rep 2007; 122(suppl 1):S48–S55.
43. Riehman KS, Wechsberg WM, Francis SA, et al. Discordance in monogamy beliefs, sexual concurrency, and condom use among young adult substance-involved couples: Implications for risk of sexually transmitted infections. Sex Transm Dis 2006; 33:677–682.
44. Weinstein N, Nicolich M. Correct and incorrect interpretations of correlations between risk perceptions and risk behaviors. Health Psychol 1993; 12:235–245.
45. Lichtenstein B, Desmond RA, Schwebke JR. Partnership concurrency status and condom use among women diagnosed with Trichomonas vaginalis. Womens Health Issues 2008; 18:369–374.
46. Johnston GL, Malekinejad M, Kendall C, et al. Implementation challenges to using respondent-driven sampling methodology for HIV biological and behavioral surveillance: Field experiences in international settings. AIDS Behav 2008; 12(suppl 1):S131–S141.
© Copyright 2010 American Sexually Transmitted Diseases Association