Secondary Logo

Journal Logo


Client-Perpetuated Violence and Condom Failure Among Female Sex Workers in Southwestern China

Choi, Susanne Y.P. PhD*; Chen, K L. MD; Jiang, Z Q. MD

Author Information
doi: 10.1097/OLQ.0b013e31815407c3
  • Free

IN 2002, OVER 830,000 cases of sexually transmitted infections (STIs) were reported nationwide in China,1 whereas the actual number was probably 10 times higher.2 Female sex workers (FSWs) in China are particularly at risk of being infected by STIs. Commercial sex is illegal in China. Sellers and buyers of sexual services may be detained for up to 4 years in reeducation centers or labor camps.3 Establishments selling sex operate in an underground economy and are closed down by police raids in periodic “strike-hard” antiprostitution campaigns.4 A review of studies on FSWs conducted in China in 2005 and 2006 suggested that rates of consistent condom use in this population are low, ranging from 15% to 58%.3 Within these contexts of state criminalization and low rates of consistent condom use, infection rates of STIs among FSWs are considerably higher compared with other population groups. Data from the national sexually transmitted disease surveillance system and sentinel site network showed that in the nation as a whole, the reported total rate of cases of syphilis was 5.7 cases per 100,000 persons in 2005.5 A study amongst FSWs in southwestern China, however, found that the prevalence rates of syphilis and human immunodeficiency virus (HIV) were 15.7% and 0.6%, respectively.6 Figures on other STIs among FSWs are equally alarming. A study of FSWs in Yunnan province found that 24.6% were coinfected with Chlamydia trachomatis and Neisseria gonorrhoeae.7 Another study in the same region found that the prevalence rates of C. trachomatis, Trichomonas vaginalis, and N. gonorrhoeae were 58.6%, 43.2% and 37.8%, respectively.8 A study of FSWs in an unspecified Chinese county found that self-reported history of STIs and current STIs was 19.4% and 41.5%, respectively.9

The vulnerability of FSWs to STIs suggests the importance of examining sexual risk behavior, including the inconsistent use of condoms and condom failure (condom slippage and condom breakage). Factors associated with inconsistent condom use among FSWs in China have received considerable attention,3,9–14 yet very little is known about factors associated with condom failure among this population in China. Studies elsewhere have found that condom slippage and breakage are significantly associated with STIs.15,16 This study sought to examine the prevalence of condom failure among FSWs and the factors associated with it.

So far studies on condom failure have focused on condom knowledge and condom efficacy.16 Yet qualitative studies among FSWs have suggested that condom use is not only about knowledge and attitudes but also depends on negotiation between FSWs and their clients.17,18 When sex workers can assert control over the interaction with clients, they are more likely to ensure proper use of condoms.19 Yet the control of sex workers over the interaction with clients varies between street-based and indoor sex workers and is affected by factors such as drug use, occupational code of conduct, and violence. Previous studies have suggested that drug use may affect the professional judgment of sex workers and may increase the risk of condom failure.20 Other studies show that an occupational code of conduct that incorporates condom use as a regular practice enhances the assertiveness of sex workers over clients and increases client compliance with consistent and proper use of condoms.19 On the other hand, violence and its threat undermine the ability of sex workers to negotiate condom use with clients and ensure the correct use of condoms.21 The use of force in intercourse may also increase condom slippage and breakage. In our research, we examined the relationship between condom failure (condom slippage and breakage) with sociodemographic characteristics (age, ethnicity, education, relationship status, drug use, and venue of sex work), acquired immune deficiency syndrome and condom knowledge, condom efficacy, control of sex workers over condom use, workplace condom use norms, and client-perpetuated violence.

Materials and Methods

Sampling and Recruitment

The research site was a medium-sized city situated in southwestern China. A study of 379 intravenous drug users conducted in the area in 2002 found that 11.3% were HIV-positive.8 Another community survey of 200 FSWs conducted in the area in 2005 showed that the prevalence rates of syphilis and HIV were 8% and 0.5%, respectively.22 The research team (including the authors and 9 female outreach workers) identified 171 locations where sexual services were exchanged for money in July 2004. The research team returned to visit all of these 171 venues in December 2005 to conduct this study. However, only 114 venues, including 60 beauty/ hair salons, 28 karaoke bars, 10 massage parlors, 11 sauna bath, 1 night club, and 4 hostels were open during this return visit. In these venues, we contacted 426 women. We invited all of the women who had described themselves as persons who had exchanged money for sex in the past 3 months or who were confirmed as such by the owners/managers of these venues, to participate in the research. As a result of our invitation, 176 women agreed to participate in the study. The response rate was 41%. We recruited a further 24 women through the local methadone center, and through visits to 3 street locations where streetwalkers congregated.

In total 200 women participated in the study. They each completed a structured questionnaire administered face-to-face by a female outreach worker. All the female outreach workers attended a 2-day training workshop on interviewing techniques organized by the first author. To ensure confidentiality, all the interviews were conducted in private, mostly in the rooms of women at the venues where they worked or in the consultation rooms of the city's Center for Sexually Transmitted Diseases. The questionnaire was anonymous. Informed consent was obtained from all the women who participated in the study. They were told that (a) the study was conducted by an overseas university not connected with the Chinese government, (b) the study was to understand the occupational risk they experienced at work for HIV prevention, (c) participation was entirely voluntary, and (d) that they could terminate the interview at any time. A small gift in the form of 2 packs of condoms and a health voucher (worth ¥100 and redeemable to cover the consultation fee of one visit to physicians at the city's Center for Sexually Transmitted Diseases for STI treatment) was given to participants as a token of gratitude at the end of the interview. The study was reviewed and approved by the Institutional Review Board of the first author's University.


Condom Failure.

Information on condom failure was collected by 2 questions asking the frequency of condom slippage and condom breakage during intercourse with clients in the past 3 months (Did you experience any condom slippage/breakage when you were using condoms with clients in the past 3 months?). Each question had 4 responses: “many times,” “a few times,” “once or twice,” and “never.” Because the majority (97% of respondents) answered “once or twice” or “never” to these 2 questions, answers were grouped into 2 categories: (a) either condom slippage or condom breakage (scored 1) and (b) never experienced slippage or breakage (scored 0).

Sexual Health Knowledge.

Sexual health knowledge was examined by 10 true/false/unsure questions asking about the routes of HIV transmission (HIV infection can be detected from appearance, HIV can be transmitted through dining together, shaking hands, by having unprotected sex, by sharing injection equipment, and by mother-to-child transmission); the proper use of condoms (a condom can be reused after washing, it is appropriate to wear condoms only immediately before ejaculation); and knowledge related to the transmission of STIs (washing the vagina after each intercourse is the best way to prevent STIs, all the STIs can be detected from examining the sexual organs). One point was given for a correct answer, with a possible score ranging from 0 to 10 (α value 0.52).

Perceived HIV Risk.

Respondents were also asked about their perceptions about the likelihood of HIV infection from unprotected sex with clients: “Do you think that you are likely to be infected with HIV if you have unprotected sex with clients?” (Very unlikely, unlikely, likely, and very likely).

Condom Efficacy.

Four items were used to measure condom efficacy: (a) If I try my best to explain, I can always persuade clients to use condoms. (b) Persuading clients to use condoms is easy for me. (c) I am always very firm when rejecting clients who refuse to use condoms. (d) I can always find a way to handle troublesome clients. Response choices for all the 4 items were: definitely true (scored 3), true (scored 2), not true (scored 1), and definitely untrue (scored 0). The scores of these 4 items were summed, with a possible score ranging from 0 to 12 (α value 0.62).

Control Over Condom Use.

One item was used to measure control over condom use at work: “Do you always put on the condom for clients in the past 3 months?” (Every time, most of the time, seldom, or never).

Workplace Norm of Condom Use.

One item was used to measure workplace norm of condom use: “Do sex workers working in your venue/area use condoms with clients?” (All of them, some of them, few of them, or none of them).


Thirteen questions were used to measure client-perpetuated violence in the past year: (a) insulted or sworn at by clients; (b) threatened; (c) threw things at, pushed/shoved, grabbed, or slapped; (d) kicked, bitten, or punched; (e) threatened with weapons (knife/gun); (f) choked; (g) attacked with weapons (knife/gun); (h) clients used threats to make me provide oral sex; (i) clients used force to make me provide oral sex; (j) clients used threats to make me have anal sex; (k) clients used force to make me have anal sex; (l) clients used threats to make me have sex when I wanted to terminate the transaction; (m) clients used force to make me have sex when I wanted to terminate the transaction. Each item had 4 response choices: often (scored 3), sometimes (scored 2), seldom (scored 1), and never (scored 0). Scores of all the 13 items were summed, with a possible score ranging from 0 to 39 (α value 0.88).

Sociodemographic Characteristics.

The following sociodemographic information was collected from the respondents: age, ethnicity, educational level, drug use, relationship status, and venue of work.


One of the research assistants entered the data into a computerized database using SPSS [SPSS for Windows version 12.0 (SPSS, Chicago, IL)] whereas another 2 checked the reliability of the data input. Bivariate and multivariable regression analysis was performed to estimate the crude odds ratios and the adjusted odds ratios (aOR) and their 95% confidence intervals (CIs) of factors associated with condom failure. A P value ≤0.05 was considered statistically significant.


Sociodemographic Profile of the Study Population

The mean age of the sex workers who were interviewed was 25 [standard deviation (SD) 4.8]. The majority of women interviewed were Han (85.5%), which is the majority ethnic group in China, and 14.5% were from the Yi minority. Over half of the women interviewed had secondary school education (56.3%) followed by primary or no schooling (35.7%). Less than 10% (8%) had tertiary education. Less than half of the women interviewed (42.6%) had a primary sex partner. Ten percent of the women interviewed were street-based sex workers. The majority (89.5%) were indoor sex workers. Slightly more than 10% (11.1%) of the women were drug users.

Condom Failure

More than one-third of the women (36.2%) reported condom slippage with clients in the past 3 months. A roughly equal percentage of women (34%) reported condom breakage with clients in the past 3 months. The overall rate of condom failure (either slippage or breakage) in the past 3 months was 48%.

Sexual Health Knowledge, Perceived HIV Risk, and Condom Efficacy

With regard to sexual health knowledge, the mean number of corrected answers recorded was 8.5 of a total of 10 (SD 1.7). Most respondents had a reasonably good understanding about the routes of HIV transmission. However, some incorrect beliefs persisted, for instance, one-third of the respondents (32.5%) thought that washing the vagina after sex was the best way to protect against STIs (including HIV). Twenty-three percent thought that all STIs could be identified by checking genitals.

Turning to perceived HIV risk, only 4% of women interviewed thought that they were unlikely to be infected with HIV if they had unprotected sex with clients. The mean condom efficacy score was 7.47 of a total of 12 (SD 1.55).

Control Over Condom Use and Work Place Norm of Condom Use

Slightly more than half of the respondents (51.5%) always put the condom on their clients. About workplace norms concerning condom use, 78% (78.4%) reported that all of their coworkers used condoms with clients.


Over two-thirds of the respondents (68.4%) reported that they had experienced some form of abuse from clients in the previous year. Among them, 63.5% reported verbal abuse, 32% reported physical violence (26.5% were pelted with objects/pushed/shoved/grabbed/slapped, 15.6% were hit with fists or bitten, 8.6% were threatened with a weapon, 10.1% were choked, 6.5% were attacked with a weapon.) With regard to sexual violence, 33.5% were threatened into providing oral sex; 20.5% were forced to have oral sex; 8.5% were threatened into providing anal sex; 7% were forced to have anal sex; 33% were threatened into having sex when they wanted to terminate the transaction; and 23% were forced to have sex when they wanted to terminate the transaction. The mean violence victimization score was 3.85 (SD 4.83).

Factors Associated With Condom Failure

Table 1 presents the results of bivariate and multivariable logistic regression analysis. After adjusting for all other factors in Table 1, condom failure was found to be significantly associated with drug use, workplace norm of condom use, and violence victimization. Compared with nondrug-using FSWs, drug users were more likely to have condom failure (aOR = 4.01; 95% CI: 0.99–16.25; P = 0.05). Compared with sex workers who reported that not all of their coworkers used condoms with clients, those who reported that all their coworkers used condoms with clients were less likely to have condom failure (aOR = 0.39; 95% CI: 0.16–0.98; P = 0.04). Finally, compared with sex workers with “low” violence victimization scores, those with a “high” victimization scores were more likely to experience condom failure (aOR = 2.30; 95% CI: 1.06–4.98; P = 0.03).

Factors Associated With Condom Failure


This study examined the factors associated with condom failure among FSWs in an area within which the prevalence of HIV was relatively high (>10% among IDUs). Condom failure has received little attention compared with consistent condom use. Our study, however, suggests that it might be important to look into the issue of condom failure because of the high rates of condom slippage and breakage (over 30%) among FSWs.

The significant association between drug use and condom failure is particularly worrying. Although HIV infection rate remains low (<0.1% among adults) in the general population in China, even after 2 decades of the epidemic among IDUs there remains the risk of HIV transmission from IDUs to the general population through commercial sex.12 Studies in Yunnan and Guangxi found that women constitute 16% to 25% of all drug users in detoxification centers.23To support their drug habit, most female drug users (52%–98%) reported having exchanged sex for money.24–26 Furthermore, compared with male drug users, female drug users are more likely to have a primary sex partner who also uses drugs (10.3% vs. 25%) and to share injection equipment with their drug-using male sex partners (8.6% vs. 13.1%).26 Putting these data together, drug-using FSWs may constitute a potential bridge population for HIV transmission between the core IDU population with other populations. More intervention work is therefore urgently needed to help reduce the rate of condom failure amongst drug-using sex workers. Existing HIV prevention projects among FSWs are mostly based on the health belief model and focus on increasing HIV knowledge. Yet the provision of general HIV knowledge, e.g., the routes of HIV transmission, may be inadequate amongst this population group. Rather, health education may need to be more specific, focusing on educating drug-using FSWs about condom negotiation skills and techniques for the proper use of condoms.

A workplace norm of condom use (all vs. not all sex workers used condoms with clients) was negatively associated with condom failure. Several possible factors might contribute to this negative relationship. An ethnographic study of FSWs in the field site showed that FSWs in the same establishment (or location in the case of street-based FSWs) often share food and accommodation and spend their waiting time chatting, playing mahjong, knitting, and watching TV together.27 A sense of community is developed through these activities. Information about clients and health tips is circulated through these informal social networks, which also act as channels of mutual help and agents of socialization and social control. This means, first, that FSWs working in establishments with an established norm of condom use may have better access to knowledge about proper condom use and may receive training on condom use skills from their coworkers. Second, they may also be subjected to greater peer pressure to enforce consistent and correct use of condoms. For example, in establishments where all the FSWs use condoms with clients, noncondom sex with clients may be regarded as a lapse of the occupational code of conduct and hence be condemned, criticized, or ridiculed. Likewise, incidences of condom failure may be discussed in these networks and solutions exchanged. Finally, clients visiting establishments with an established condom use norm may feel a greater need to be compliant with the discourse of safe sex. Studies in several Western countries have shown that client compliance with the discourse of safe sex is an important factor contributing to the relatively high rates of consistent condom use in these countries.19,28,29 In an intervention research project in the Dominican Republic, Kerrigan and colleagues showed that the combination of solidarity-building activities in establishments and government regulation mandating 100% condom use was an effective strategy in enhancing consistent condom use and the ability of FSWs to reject unsafe sex.30 However, government regulation of commercial sex establishments to promote safe sex presume an official acceptance of commercial sex as a legal activity. Within the present context of criminalization of sex work in China, it is unlikely that the Chinese government will support such a strategy. Without supportive policy initiatives, normative changes at establishments that aim at strengthening collective commitment to support consistent and correct condom use between FSWs and their clients become a very significant community level intervention strategy.12,14 A study of 454 establishment-based FSWs found that perceived gate-keeper support was significantly associated with condom use during the previous 3 sexual encounters, condom communication, and condom use intention.12 In addition to gate-keepers, results of this study suggested that the role of coworkers (e.g., their condom use behavior and peer norm) in influencing condom behavior (use and failure) may be equally important and thus need to be more thoroughly examined in the future.

The significant and positive association between vulnerability to client-perpetuated violence and condom failure is alarming. Violence in itself is a serious problem as it jeopardizes the well-being of FSWs. In Western countries, client-perpetuated violence is mainly a problem for street-based FSWs.21,31 Our data, however, showed that around 60.9% of establishment-based FSWs in this study experienced verbal abuse, 32% suffered physical violence, and 48.5% experienced sexual violence perpetuated by clients. These figures suggested that client-perpetuated violence against FSWs is a pandemic that affects FSWs from different settings in China.

The association between violence and condom failure suggests the intersection between violence and HIV risk exposure. Our findings do not explicate the nature of this association or the causal pathways of influence on condom failure. Two links, however, may be possible and provide a challenge for future research. First, physical and sexual violence may directly cause condom slippage or breakage. Second, violence may be a sign that FSWs have failed to establish or maintain client compliance throughout the sexual encounter, with condom failure a consequence of this loss of control on the part of FSWs. In this scenario, violence and condom failure are both the result of the failure of FSWs to gain control in the encounter.

Limitations of the Study

This is a small-scale study examining the factors associated with condom failure. Findings may not be representative of the larger community of FSWs outside the study site and FSWs working for higher-end settings, e.g., escort agencies. The response rate of 41% among establishment-based FSWs was not high. We refrained from exerting pressure on managers/owners of establishments and FSWs because we believed that only when FSWs were genuinely interested in our research would they share their views with us on these sensitive topics. The second and third authors who have been conducting research and outreach work in the field site since 1996 felt that a considerable number of those who declined to participate in the study were newcomers to the city. It was therefore likely that our study excluded FSWs with a shorter duration of stay at the site or a shorter duration of work experience in the industry. Turning to measurement limitations, the α value of the health knowledge scale was relatively low, suggesting the need for refining the scale to improve its reliability.


Despite these limitations, the findings of this study contribute to a growing literature that examines the intersection between victimization caused by violence and HIV risk exposure.31–33 The findings highlight the prevalence of condom failure among FSWs and its intersection with drug use, workplace condom use norms, and violence. Further research is needed to scrutinize these associations more closely. Meanwhile health education among FSWs needs to focus more attention on condom education, with specific attention being paid to educating drug-using FSWs about the correct ways of using condoms. The provision of sex workers with communication training to strengthen their skills to negotiate condom use with reluctant clients is also essential. Condoms of different sizes and reliable quality should be distributed to FSWs. The peer network of FSWs should be mobilized to enforce the norm for condom use at the workplace, e.g., through the training of peer educators. In the long term, the national policy of criminalizing sex work may need to be addressed. Outright criminalization limits the state's capacity to use innovative policy initiatives to promote safe sex among workers in the sex industry. Fear of being arrested by the police often inhibits FSWs from seeking police help in order to deal with violence.3,4,26,27 This encourages violent clients who are sure of impunity. The criminalization of sex work thus inadvertently increases the vulnerability of FSWs to client-perpetuated violence and HIV/STI risk exposure.


1. Chinese Center for Disease Control and Prevention. Information on HIV/STDs Surveillance. Beijing, China: Chinese Center for Disease Control and Prevention, 2003.
2. He N, Detels R, Zhu J, et al. Characteristics and sexually transmitted diseases of male rural migrants in a metropolitan area of eastern china. Sex Transm Dis 2005; 32:286–292.
3. Choi SYP, Holroyd E. The influence of power, poverty and agency in the negotiation of condom use for female sex workers in mainland China. Cult Health Sex 2007; 9:1–15.
4. Choi SYP, David R. Law enforcement, public health, and HIV/AIDS in China. In: Harris P, Siplon P, eds. The Global Politics of AIDS 2007. London, UK: Lynne Rienner Publishers,137–154.
5. Chen ZQ, Zhang GC, Gong XD, et al. Syphilis in China: Results of a national surveillance programme. Lancet 2007; 369:132–138.
6. Chen XS, Yin YP, Liang GJ, et al. Co-infection with genital gonorrhoea and genital chlamydia in female sex workers in Yunnan, China. Int J STD AIDS 2006; 17:329–332.
7. Chen XS, Yin YP, Liang GJ, et al. Sexually transmitted infections among female sex workers in Yunnan, China. AIDS Patient Care STDS 2005; 19:853–860.
8. Ruan Y, Cao X, Qian HZ, et al. Syphilis among female sex workers in southwestern China: Potential for HIV transmission. Sex Transm Dis 2006; 33:724–725.
9. Wang B, Li X, Stanton B, et al. Vaginal douching, condom use, and sexually transmitted infections among Chinese female sex workers. Sex Transm Dis 2005; 32:696–702.
10. Lau JTF, Tsui HY, Siah PC, et al. A study on female sex workers in southern China (Shenzhen): HIV-related knowledge, condom use and STD history. AIDS Care 2002; 14:219–233.
11. Qu S, Liu W, Choi KH, et al. The potential for rapid sexual transmission of HIV in China: Sexually transmitted diseases and condom failure highly prevalent among female sex workers. AIDS Behav 2002; 6:267–274.
12. Yang H, Li X, Stanton B, et al. Condom use among female sex workers in China: Role of gatekeepers. Sex Transm Dis 2005; 32:572–580.
13. Hesketh T, Zhang J, Qiang DJ. HIV knowledge and risk behavior of female sex workers in Yunnan Province, China: Potential as bridging groups to the general population AIDS Care 2005; 17:958–966.
14. Huang Y, Henderson GE, Pan S, et al. HIV/AIDS risk among brothel-based female sex workers in China: Assessing the terms, content, and the knowledge of sex work. Sex Transm Dis 2004; 31:695–700.
15. Crosby RA, DiClemente RJ, Wingood GM, et al. Condom failure among adolescents: Implications for STD prevention. J Adolesc Health 2005; 36:534–536.
16. Crosby RA, Yarber WL, Sanders SA, et al. Men with broken condoms: Who and why? Sex Transm Infect 2007; 83:71–75.
17. Campbell C. Selling sex in the time of AIDS: The psycho-social context of condom use by sex workers on a Southern African mine. Soc Sci Med 2000; 50:479–494.
18. Wojcicki JM, Malala J. Condom use, power and HIV/AIDS risk: Sex-workers bargain for survival in Hillbrow/Joubert Park/Berea, Johannesburg. Soc Sci Med 2001; 53:99–121.
19. Sanders T. A Continuum of risk? The management of health, physical and emotional risks by female sex workers. Sociol Health Illn 2004; 26:557–574.
20. Green A, Day S, Ward H. Crack cocaine and prostitution in London in the 1990s. Sociol Health Illn 2000; 22:27–39.
21. Barnard MA. Violence and vulnerability: Conditions of work for street working prostitutes. Sociol Health Illn 1993; 15:683–705.
22. Choi SYP. Violence against sex workers in mainland China and its association with STD infection. Presented at: The XIV International AIDS Conference; August 2006; Toronto, Canada.
23. Reid G, Costigan G. Revisiting the “Hidden Epidemic”: A Situation Assessment of Drug Use in Asia in the Context of HIV/AIDS. Australia: The Centre for Harm Reduction, 2002.
24. Li X, Shanton B, Zhou Y. Injection drug use and unprotected sex among institutionalized drug users in China. J Drug Issues 2000; 30:633–674.
25. Wang Q, Lin G. Sex exchange and HIV-related risk behaviors among female heroin users in China. J Drug Issues 2003; 33:119–132.
26. Choi SYP, Cheung YW, Chen K. Gender and HIV risk behavior among intravenous drug users in Sichuan Province, China. Soc Sci Med 2006; 62:1672–1684.
27. Choi SYP. Condom management among female sex workers in China: Structure, identity and strategy. Presented at: The American Sociological Association Annual Meeting; August 2006; Montreal, Canada.
28. Plumridge EW, Chetwynd SJ, Reed A. Control and condoms in commercial sex: Client perspectives. Sociol Health Illn 1997; 19:228–243.
29. Whittaker D, Hart G. Research note: Managing risks: The social organization of indoor sex work. Sociol Health Illn 1996; 18:399–414.
30. Kerrigan D, Moreno L, Rosario S, et al. Community Approaches and Government Policy Reduce HIV Risk in the Dominican Republic. Horizons Research Summary. Washington, DC: Population Council, 2004.
31. El-Bassel N, Witte SS, Wada T, et al. Correlates of partner violence among female street-based sex workers: Substance abuse, history of childhood abuse, and HIV risks. AIDS Patient Care STDS 2001; 15:41–51.
32. Nemoto T, Iwamoto M, Wong S, et al. Social factors related to risk for violence and sexually transmitted infections/HIV among Asian massage parlor workers in San Francisco. AIDS Behav 2004; 8:475–483.
33. Wechsberg WM, Luseno WK, Lam WK. Violence against substance-abusing South African sex workers: Intersection with culture and HIV risk. AIDS Care 2005; 17:S55–S64.
© Copyright 2008 American Sexually Transmitted Diseases Association