Heterosexual contact has become the predominant mode of HIV transmission in China, indicating that 46.5% of the estimated 740,000 people living with HIV and AIDS (PLWA) in 2011 were due to transmission through heterosexual contacts [1]. Female sex workers (FSWs) are one of the key populations to facilitate increase of the HIV epidemic and likely to determine how fast the HIV epidemic will spread from high risk groups to the general population [2]. In the national HIV/AIDS surveillance system in China, periodic prevalence surveys of syphilis infection among FSWs have been integrated but current Neisseria gonorrhoeae (NG) and Chlamydia trachomatis (CT) surveillance activities concentrate mainly on passive case reporting. Similarly, there have been many studies on prevalences of HIV and syphilis infections in this population [3–5] but few on other sexually transmitted infections (STIs) including NG and CT [2, 6].These two infections can not only cause serious health complications [7] but act as facilitating factors in transmission of HIV [8]. In the study described in this article, we sought to determine the frequency of HIV, syphilis, NG and CT infections in a sample of FSWs recruited from 4 provinces in China, and to identify significant predictors associated with CT and NG infections.

To our knowledge, this study was a cross-sectional survey with biggest sample size of FSWs recruited from different categories of sex venues in multiple provinces in China. Our study finding that there was higher prevalence of chlamydial infection in LV-based FSWs is in accord with previous studies [2], and high prevalences of syphilis and HIV infections among this population and differences in these two infections between FSWs recruited from different categories of sex work venues were reported elsewhere [9, 10]. In China, FSWs who solicit on streets or other outdoor places (freelance FSWs) and conduct commercial sex in rental houses are a special segment of the FSW population with more vulnerable nature in terms of socio-demographic characteristics, organization of sex work, employment and economic status, and relationship with clients [11, 12]. Previous studies have shown that the FSWs working in low-class sex venues tended to be older in age, have high turnovers of sexual clients due to a low pay, but infrequently used condoms due to extra pay for unsafe sex [13]. The freelance FSWs either independently solicit clients on streets or find clients through nearby construction sites and factories [14]. They usually earn far less than what FSWs in higher-class sex venues earn but engage in riskier behaviors when having sex with both commercial and regular non-paying partners [12–14] and are highly stigmatized and marginalized by society. In addition, these FSWs may have poorer access to health information and health care services. High prevalence of NG or CT infection was found among FSWs, particuarly LV-based FSWs in the current study. Considering the synergistic effect of STIs on HIV infectivity and susceptibility [8], effective control of NG and CT in this population is not only important for avoiding the complications caused by these infections but also for preventing the transmission of HIV infection. A high co-infection of CT among FSWs infected with NG further supports the presumptive treatment of the patients with NG for CT.

Regarding sexual behaviors, self-reported condom use at last commercial sex was reported by 74.0% and consistent use of a condom during the previous month was reported by 52.0%. Unexpected was that the rates of condom use were not associated with NG or CT infections. This may be due, in part, to an upward bias in self-reported condom use rates [15]. Older age, which was usually associated with longer duration of working in sex trade, has also been identified as a risk factor for syphilis in previous studies among FSWs in China and other countries [16, 17]. However, younger age is a significant risk factor for CT infection in the current study. A possible explanation is that young FSWs have more sexual activity, and less knowledge and experience with STI prevention. As well, younger women may have increased susceptibility to some STIs due to cervical ectopy following sexual initiation [18, 19] or because of less likelihood of acquired protective immunity from previous STI exposure [20]. Previous study reported that the probability of incident chlamydial infection was inversely related to duration of prostitution [21]. Although the prevalence of CT infection among FSWs in low-class sex venues with higher risk behaviors was higher than that in middle- or high-class ones, the difference in the prevalence rates among LV-, MV- and HV-based FSWs was not as much as that for syphilis [9]. Previous study reported no difference in CT prevalence between establishment- and street-based FSWs [2]. Therefore, the likelihood of CT infection among FSWs may be not only related to the current risk behaviors but also the previous exposure to the infection which may be relevant to stimulation of protective immunity. In light of these findings and hypothesis further studies is warranted.

As FSWs are one of the important populations to drive the STI epidemic and probably the bridge population for the heterosexual transmission of STIs, the findings of this study have a number of important implications. First, this study shows substantial prevalences of NG and CT infections among FSWs and much higher prevalences of the infections among FSWs at low-class sex venues. The patterns of infections have served as a call for action to draw further attention to control the critical disease burden of CT and NG among this population and suggested the importance of including NG and CT prevalence surveys in the current surveillance program and prioritizing the FSWs at low-tier venues in the current program for intervention. Second, this study indicates the association of young age with CT infection, which further supports the concept of protective immunity to the recurrent infections although more studies are needed. Based on these findings, it is important to consider the potential protective immunity related to age and duration of prostitution when we use CT or NC prevalence as a proxy to indicate risk behaviours among high-risk groups. However, further biological and epidemiological studies are needed.

Several limitations of this study should be considered in the interpretation of the results. First, study participants were not randomly recruited, so they may not be an accurate representation of the target study population. Second, as prostitution is illegal in China, some FSWs (ranging from 10% to 40% and varying between different types of venues) refused participating in the study, resulting in potential selection bias. Third, our results may also be affected by a self-reporting bias, particularly those related to sexual behaviours.

The high NG and CT prevalence rates found among FSWs in China, especially among LV-based FSWs, suggest that it is essential that comprehensive prevention and control programs including not only behavioral interventions but also screening and medical care are needed to meet the needs of this high-risk population.