HIV/AIDS Among Female Sex Workers in China: Epidemiology and Recent Prevention Strategies

Abstract

Purpose of Review

This review aims to describe the epidemiology of HIV among female sex workers (FSWs) in China over the past decade, to summarize current gaps in knowledge regarding risk factors, and to identify new directions for HIV prevention strategies.

Recent Findings

We summarized national and regional levels of HIV prevalence among FSWs based on reported rates in the literature from 2008 to 2018. Studies identified cases of HIV infection among FSWs in all but one province during this time period, and demonstrated a sporadic pattern in most provinces, with a low overall national HIV prevalence below 1%. However, in Yunnan and Guanxi Zhuang Autonomous Regions, the median-reported prevalence rates were close to or slightly above 1%. National prevention programs have widely promoted male condoms as a primary and practical measure for HIV prevention, but studies evaluating condom use practices among FSWs demonstrated wide variability. A rise in illicit use of synthetic drugs and changing sexual practices in the setting of sex work (e.g., anal sex) may represent newer risk factors for HIV transmission among FSWs; however, more data are needed to better characterize these trends. Limited studies have examined the feasibility and efficacy of innovative prevention tools (e.g., female condoms) or strategies (e.g., pre-exposure prophylaxis, PrEP) to prevent HIV among FSWs.

Summary

We call for a more comprehensive understanding of current trends in HIV risk among FSWs, as well as more research focuses on innovative strategies to reduce the spread of HIV in this vulnerable population.

Introduction

Estimates from the National Center for AIDS/STD Control and Prevention of the China Center for Disease Control and Prevention (NCAIDS, China CDC) indicate that over 842,000 people live with HIV/AIDS in mainland China [1•]. The HIV/AIDS epidemic in China originated among injection drug users (IDUs), but sexual transmission has gradually become the dominant route of HIV infection. In recent decades, more than two-thirds of newly reported cases of HIV/AIDS are believed to have been acquired through heterosexual transmission, comprising 68% of cases in 2012 and 67% in 2016 [2, 3•]. The HIVepidemic in China is primarily concentrated in key populations, and among these, female sex workers (FSWs) have received attention because they are believed to represent an important “bridge population” driving HIV infections in the general population through sexual transmission [4, 5].

Since the 1950s, both the commercial sex industry and sex work itself have been considered morally unacceptable and are prohibited legally in mainland China. Sex work was considered largely eliminated in the 1950s–1970s. However, the commercial sex industry re-emerged following the economic reforms and open policies of the 1980s and has flourished alongside China’s rapid economic growth, diversified attitudes toward sex, and rapidly changing lifestyles in society [6, 7], though it remains illegal [8]. In 2007, there were an estimated 1.8–3.8 million female sex workers (FSWs) in mainland China [9].

To monitor the HIV epidemic in mainland China, HIV sentinel surveillance systems were established in 1995 among several key populations [10, 11] including drug users, FSWs, and men who have sex with men (MSM). This system consists of national and provincial surveillance sites and is operated by the China CDC system at the national, provincial, and county levels. Since 2010, the sentinel system has included more than 1888 surveillance sites within 300 cities across 31 provinces; among them, 520 sites exist for FSWs [12••]. Annual cross-sectional surveys are conducted at each sentinel site to measure the prevalence of HIV and syphilis infection, as well as high-risk behaviors among target subpopulations. FSWs surveyed, who test negative for HIV with the first ELISA-based screening test, are considered HIV-negative. If a FSW’s first ELISA is positive, she either is re-tested with a second ELISA test kit from a different manufacturer (in areas of high HIV prevalence such as Yunnan and Guangxi), or is notified to receive a confirmatory western blot test (in most areas of low prevalence). Only FSWs with ≥ 1 positive ELISA result in HIV prevalent areas and those with a positive confirmatory test are counted when estimating HIV prevalence. However, surveillance survey sampling for FSWs has been highly dependent on identifying sex work venues and building rapport with local bosses and FSWs. This method of convenience sampling has tended to oversample FSWs working in low-tier venues—defined as venues where FSWs charges a fee of less than 50 Chinese yuan (approximately 7–8 US dollars) per client—including hair salons/barber shops, massage parlors, foot-bathing shops, roadside shops and restaurants, guesthouses, streets, and other public outdoor areas. FSWs working in higher-tier venues such as karaoke bars, hotels, or nightclubs, where clients are primarily higher in socioeconomic status, are believed to be much less represented in these surveillance surveys [13].

In the present review, we summarize the epidemiologic patterns of HIV infection among FSWs in China from 2008 to 2018 as reported in the English and Chinese literature, describe relevant risk factors for HIV among FSWs in China, highlight findings from recent studies examining HIV prevention strategies targeting FSWs, and provide recommendations for future work.

Epidemiology of HIV Among FSWs in China

To understand the epidemiology of HIV among FSWs, we identified publications measuring HIV prevalence (including studies measuring incidence that specifically reported prevalence data) in peer-reviewed journals published in English and Chinese. We searched PubMed, China National Knowledge Infrastructure (CNKI), and Wan Fang database; the latter two are well-established databases for academic publications in mainland China. Search terms included (“HIV” OR human immunodeficiency virus OR “AIDS” OR acquired immunodeficiency syndrome OR HIV/AIDS) AND (“female sex workers” OR “commercial sex workers” OR “women who sell sex” OR “FSW” or “CSW”) AND (China OR Chinese). The search was limited to studies carried out from January 1, 2008 to December 31, 2018.

The titles and abstracts of articles were screened to determine if the articles met the following criteria: (1) the study population was comprised of FSWs; (2) the study was conducted in mainland China. Studies were excluded if they (1) described a dataset that overlapped with another included publication, (2) did not report data regarding HIV prevalence, (3) had a sample size less than 400, and (4) were a systematic review/meta-analysis, conference report, or case study. The flow diagram for the identification and selection of articles for inclusion is shown in Supplement Material Fig. 1.

Prevalence of HIV Among FSWs in China

Of 6155 abstracts screened, a total of 608 full-length articles were reviewed, and 328 publications met criteria for inclusion. Studies included sites from all 31 provinces in mainland China. A majority of the study sites were from the East (111,33.84%), South central (88,26.83%), and Southwest China (64,19.51%). Nineteen (5.79%) publications represented municipalities such as Beijing and Shanghai. Almost all studies utilized convenience sampling or respondent-driven sampling (RDS), with 255 (77.74%) publications based on sentinel surveillance data. The sample included 306 Chinese-language articles and 22 English-language articles.

Cui and colleagues from the National Centre for AIDS/STD Control and Prevention (NCAIDS) analyzed sentinel surveillance data from 2010 to 2014, and found that the national HIV prevalence remained stable over the time period between 0.17–0.24%. Looking more granularly at the numbers, the authors surveyed all 499 FSWs surveillance sites at 2014 and reported that prevalence at 473 sites was < 1%; however, among 26 sites, the prevalence ranged from 1 to 5% [12••]. Wang et al. conducted a series of comprehensive nationwide cross-sectional surveys among 827,079 FSWs from 2008 to 2012. In their study, the national prevalence of HIV declined from 0.6% in 2008 to 0.3% in 2012. However, among FSWs working in low-tier venues (28.1–41.5% of overall FSWs), higher HIV prevalence was observed compared with those working in high-tier venues (0.8% vs 0.4% in 2008; 0.6% vs 0.1% in 2012) [14].

Table 1 demonstrates the distribution of publications identified from the past decade reporting HIV prevalence among FSWs, organized by province and region. HIV infection was identified among FSWs from all but one province (Hebei province in the Northern region) in the past decade. Cases were sporadic in most provinces with prevalences reported below 1%. However, in Yunnan province in the Southwest and Guangxi Zhuang Autonomous Region in the South central regions, the overall HIV prevalence rates were higher.

Table 1.

Prevalence of HIV among FSWs in China, by province from 2008 to 2018

Region*	Province	n of articles	No. of reference	HIV prevalence (%)
				Median	Q25, Q75	Range
East	Anhui	5	1–5	0.07	0.00, 0.25	0.00, 0.25
	Fujian	6	6–11	0.00	0.00, 0.00	0.00, 2.87
	Jiangsu	33	12–44	0.00	0.00, 0.19	0.00, 2.07
	Jiangxi	9	45–53	0.00	0.00, 0.37	0.00, 1.20
	Shandong	15	54–68	0.00	0.00, 0.00	0.00, 0.63
	Shanghai	15	69–83	0.00	0.00, 0.12	0.00, 0.22
	Zhejiang	28	84–111	0.04	0.00, 0.24	0.00, 0.48
North	Beijing	4	112–115	0.00	0.00, 0.04	0.00, 0.23
	Hebei	5	116–120	0.00	0.00	0
	Neimenggu	2	121–122	0.13	0.00, 0.25	0.00, 0.25
	Shanxi	6	123–128	0.00	0.00, 0.00	0.00, 0.25
	Tianjin	2	129–130	0.10	0.06, 0.15	0.06, 0.15
Northeast	Heilongjiang	4	131–134	0.00	0.00, 0.01	0.00, 0.01
	Jilin	5	135–139	0.00	0.00, 0.00	0.00, 0.30
	Liaoning	5	140–144	0.00	0.00, 0.00	0.00, 0.04
Northwest	Ningxia	2	145–146	0.04	0.00, 0.11	0.00, 0.25
	Qinghai	2	147–148	0.00	0.00, 0.12	0.00, 0.25
	Shanxi	6	149–154	0.00	0.00, 0.21	0.00, 0.74
	Xinjiang	9	155–163	0.50	0.25, 1.00	0.00, 1.95
	Gansu	10	164–173	0.00	0.00, 0.00	0.00, 0.25
South central	Guangdong	18	174–191	0.25	0.00, 0.47	0.00, 3.09
	Guangxi	51	192–242	1.08	0.64, 1.80	0.13, 6.78
	Hainan	3	243–245	0.09	0.03, 0.16	0.00, 0.21
	Henan	3	246–248	0.23	0.21, 0.25	0.21, 0.25
	Hubei	9	249–257	0.25	0.00, 0.62	0.00, 1.00
	Hunan	4	258–261	0.16	0.03, 0.93	0.00, 1.60
Southwest	Chongqing	3	262–264	0.11	0.00, 0.25	0.00, 0.50
	Guizhou	4	265–268	0.50	0.32, 1.00	0.14, 1.50
	Sichuan	26	269–294	0.84	0.42, 1.37	0.00, 2.73
	Tibet	2	295–296	0.02	0.00, 0.05	0.00, 0.05
	Yunnan	29	297–325	1.50	0.51, 2.75	0.13, 10.12

^*The 31 provinces/metropolitans were categorized into six geographical areas based upon Chinese conventional administrative regions as below: East (including 7 provinces/municipalities), Northeast (including 3 provinces), North (including 5 provinces/municipalities), South central (6 provinces), Northwest (5 provinces), and Southwest (5 provinces/municipalities)

Regions in the Southwest and South central China included provinces traditionally associated with drug abuse (e.g., Yunnan and Sichuan in the Southwest) or those neighboring the “Golden Triangle” and on the drug trafficking route to Hong Kong (e.g., Yunnan and Guangxi). The HIV epidemic originated in the late 1980s–1990s among drug users in these regions, and has since become steadily established among FSWs [15, 16]. In our literature search, we identified 51 publications reporting HIV prevalence rates among FSWs from 2008 to 2018 in Guangxi ranging from 0.13 to 6.78%, with a median and interquartile range (IQR) of 1.08% (0.64%, 1.80%). Twenty-nine articles measuring HIV prevalence data among FSWs in Yunnan reported a range from 0.13 to 10.12% (1.50% (0.51%, 2.75%)). While substantially fewer publications were identified from most regions of low prevalence, we did find a large number of publications from the economically developed Eastern region (including 7 provinces and municipalities such as Shanghai). These publications indicated very low prevalence of HIV infection identified through sentinel surveillance, although relatively high prevalences of syphilis were observed (range 0–10.12%) among FSWs in these studies.

We further stratified the studies into two time periods in order to explore temporal variations in HIV prevalence among FSWs in different regions of mainland China. Due to the notable paucity of studies reporting HIV prevalence published after 2015 (3 studies available), we chose to focus on the periods from 2008 to 2011 (246 articles) and 2012 to 2015 (181 articles). Figure 1 demonstrates the median and IQR of the reported HIV prevalences among FSWs by province during each 4-year period, as well as during the overall time period from 2008 to 2015. In four provinces (Inner Mongolia, Qinghai, Tibet, Henan), no data have been published regarding HIV prevalence among FSWs since 2012. The three studies published after 2015 were from Jiangsu (2) and Sichuan (1) provinces and the reported HIV prevalences were within the range reported in earlier studies.

Fig. 1.

Median HIV Prevalence among FSWs in China,by Province

Although no longitudinal cohort studies measuring HIV incidence among FSWs during this time period were identified from our review of the literature, a few studies in Yunnan province have measured recently infected cases using new laboratory techniques from cross-sectional surveys among FSWs. For instance, Yang et al. tested blood samples from 454 HIV-positive FSWs identified among 33,829 women in 32 FSWs surveillance sites in Yunnan province from 2008 to 2010, and estimated proportion of newly infected HIV cases in the study sample using a BED-CEIA (BED-capture enzyme immunoassay) method, which detects seroconversions within the last 6 months based on measurement of the proportion of HIV-1-specific IgG to total IgG. The authors found that while overall HIV prevalence was stable between 1.95–2.22% during the study period, the proportion of recently infected FSWs in the sample was 0.40% in 2008 and decreased to 0.31% in 2010 [17•]. Mathematical modeling is another method used to estimate HIV incidence from serial cross-sectional prevalence data under certain assumptions [18, 19]. Cui et al. used this approach to estimate HIV incidence rates among FSWs between 2010 and 2014, based upon national HIV prevalence data collected from all 499 FSWs sentinel surveillance sites in China, and found very low annual HIV incidence rates (0.03% in 2011, 0.06% in 2012, < 0.001% in 2013, and 0.01% in 2014). However, the authors cautioned that the sentinel surveillance prevalence data might be subject to selection bias not fully adjusted for in the model. Furthermore, most past studies applied this methodology in settings where the HIV epidemic is generalized such as sub-Saharan Africa, and the low FSWs prevalence rates might influence reliability of the incidence estimates [12••].

Behavioral Characteristics and Risk Factors

Since most studies on HIV prevalence report data collected through the sentinel surveillance system, knowledge regarding correlates related to HIV infection is also primarily based on risk factors measured in sentinel surveillance surveys such as age, years working in the sex industry, types of venues, frequency of engaging in commercial sex, and frequency of condom use. Most studies found that HIV infection among FSWs was associated with older age, longer years working in the sex industry, lower education level, and working in a low-tier venue (vs middle or high-tier) [20–27].

Condom Use Behaviors Among FSWs

The male condom was initially promoted for family planning purposes in mainland China and family planning services have historically been closely involved with producing and dispensing condoms. Along with increased awareness and programs promoting condoms for HIV/STI (sexually transmitted infections) prevention, uptake and use of condoms among FSWs and their clients have gradually increased since the 1990s (the 100% condom use program). During this same period, condom quality, accessibility, and diversified designs have also improved enormously in China. In addition to widespread availability through commercial systems, free condoms are also provided to key populations such as FSWs as part of many HIV/AIDS prevention programs through the China CDC or other organizations including the UN or international NGOs [28–30].

One of the aims of sentinel surveillance for FSWs has been to understand and quantify condom use in this population. Accordingly, the standardized questionnaire includes two questions dedicated to measuring condom use among FSWs: (1) “Did you use a condom in the last sex act with your client (yes/no)” and 2) “How often did you use condoms in last month with your clients (never, sometimes, always and do not know)” [31]. The two questions are asked with regard to clients as well as non-client steady partners, respectively. A few independent studies have estimated overall level of condom use among populations of FSWs by measuring the proportion of condom-protected sexual acts compared with total sexual acts over the last 30 days [32]. Irrespective of the indicator employed, the proportion of FSWs reporting “always used condoms with commercial sexual partners in last month” was highly variable (14.8–100.00%) among studies published over the last decade [33, 34].

Condom use depends not only on individual-level factors, but also on contextual factors as well. Across studies, individual-level FSWs characteristics were most commonly observed to influence condom use. For example, older age, lower education level, working in low tier venues, and lower HIV-related knowledge were all factors commonly associated with inconsistent or lack of condom use during sexual encounters [32]. But contextual factors may also influence condom use patterns. For instance, in a study among 445 FSWs in four sites in Hainan and Guangxi (two rural towns in Hainan, a small urban center in Guangxi and a mid-sized city in Hainan), the size of sex industry establishments was found to vary within and across the different types of sites. The mean proportions of condom use during sexual encounters in the last 30 days were 77.00%, 53.00%, 68.00%, and 53.00%, respectively (median proportions 59–100%). Twenty-five to 54.8% of FSWs across the four towns reported always using condoms during sex in the last 30 days (54.79%, 35.89%, 47.37%, and 25.35%, respectively), while 8.21–32.05% of FSWs reported never using condoms (8.21%, 32.05%, 19.74%, and 29.58%, respectively) [35]. The variation in condom use between towns suggests the influence of differences in local HIV prevalence and types of sex venues, but also differences in quality and quantity of local condom education and promotion.

In addition to clients, FSWs often have other sexual partners such as steady partners (spouse or boyfriends) or non-paying casual partners. Condom use in the setting of these types of relationships is much lower than that with commercial partners. A cross-sectional study of willingness and actual use of condoms among 700 FSWs in Guangxi showed lower intention to use condoms consistently with fixed partners (24.10%) compared with temporary partners (80.29%), and lower actual frequency of condom use in the last three sex acts with fixed partners (33.39%), compared with temporary partners (71.32%) [36•]. A study among FSWs in 3 rural towns in Hainan and Guangxi reported similar findings, with 68.10–77.10% of FSWs reporting universal condom use with clients in the last 30 days, compared with 24.50–40.50% reporting universal condom use with their steady partners [37].

Despite over two decades of extensive efforts to promote condom use among FSWs, barriers to condom use, in particular those arising from clients (e.g., the condom felt uncomfortable during sex), remain commonly cited in the literature over the past 10 years. A study of 324 FSWs from three suburbs and two downtown areas in Shanghai in 2010 found that 51.23% of women reported experiencing condom breakage or slippage during sex. One-third of FSWs reported having had sex without a condom because clients paid more money or they looked clean [38]. Another study in a respondent-driven sample of 320 FSWs in Guangdong province found that only 60.2% of FSWs used condoms correctly. Examples of incorrect condom use included not wearing a condom until ejaculation, condom rupture, and not using the condom during the entire sex act [39].

Other High-Risk Behaviors

Illicit drug use is another important risk factor for HIV/STI among FSWs. While use of traditional drugs such as heroin still exists in some areas, newer synthetic drugs such as methamphetamine and ketamine have gradually replaced heroin and become the drugs of choice in some entertainment venues. Serial cross-sectional surveys from 2006 to 2013 among FSWs in Qingdao, a coastal city in eastern China, showed that 21.75–55.50% of participants reported having used illicit drugs each year, of which more than 95% used methamphetamine orally and only a small percentage injected heroin between 2006 and 2008 [40]. HIV prevalence was different between FSWs who used drugs and who not. Another study conducted among 973 FSWs in the tourist city of Sanya in Hainan province from 2009 to 2010 showed that 5% of FSWs interviewed reported using illicit drugs in the previous year, with FSWs in higher-tier venues reporting higher rates of drug use (high-tier 7.95%; middle-tier 5.06%; low-tier 2.00%) [41].

Studies also found that drug use correlated with lower condom use and higher HIV prevalence. Zhang et al. conducted a survey among low-tier venues FSWs in Guangxi, which showed that 2.4% of the participants reported drug abuse, and HIV infection rate among drug users was as high as 6.42%, significantly higher than that of non-drug users (non-DU) (OR = 5.32, p< 0.001) [42]. Another study analyzed surveillance data from 2010 among FSWs in Guangxi and found that compared with non-DU, inject drug users (IDU) were more likely to report not using a condom during the last sex act (aOR = 3.25, 95% CI = 1.65, 6.40), and inconsistent condom use in the past month (aOR = 4.88, 95% CI = 2.66, 8.96) [43].

Very few studies have examined high-risk sexual behaviors among FSWs such as anal intercourse. A prior study conducted in 2004 among 454 FSWs in Guangxi reported that 1.98% of the women practiced anal sex with clients, 6.70% had sex during menstruation, and 29.40% used alcohol before sex [44]. By contrast, in 2011, Gao et al. found that 28.16% (158/561) of FSWs working in low-tier venues in Kunming, the capital city of Yunnan province, provided anal sex as part of their services [45]. These data underscore the pressing need for more research examining changes in patterns of sexual behaviors in the sex industry in China, particularly the increased adoption of high-risk behaviors among newer generations of FSWs.

Vaginal douching, i.e., washing or flushing the vagina with water or liquids with or without other substances mixed (e.g., salt, toothpaste or soap, or commercial douching products), is a common “hygienic practice” among FSWs in China. The practice has raised concerns because it may affect biological susceptibility to HIV/STIs by altering the vaginal flora and normal pH, or discourage condom use. However, the few studies focused on vaginal douching among FSWs in China have produced inconsistent results. For example, one study carried out in three towns in Guangxi in 2004 reported that 64.7% of FSWs had ever practiced douching, and that vaginal douching was significantly associated with decreased life-time use of condoms with clients (OR = 0.31) and with regular partners (OR = 0.22). Vaginal douching was also associated with an increased history of self-reported STI (OR = 1.95) but not necessarily with laboratory confirmed current STI [44]. Chen et al. explored the association between vaginal douching and condom use among 905 FSWs from 5 rural or small urban towns in Guangxi province and found highly variable rates of douching with 38.41–84.21% of women reporting having douched in the last 30 days, and 18.90–49.10% women reporting douching at least once a day. Compared with the women who had never used condoms in the past 30 days, women using condoms consistently were more likely to have occasional douche (aOR = 1.61, 95% CI 1.09∼7.76) or frequent douche (aOR = 2.80, 95% CI 1.09∼6.26) [46]. Recent data regarding vaginal douching practices among FSWs in China are lacking and more data are needed to better understand how douching practices may or may not contribute to HIV risk in this population.

FSWs Prevention Programs and Research

The China CDC is a nationwide system focused on addressing disease prevention and control at the national, provincial, and county levels. The system provides HIV-related education and prevention programs for the general public as well as key subpopulations, such as condom promotion, free condom distribution, and other routine health services. Since 2005, the CDC system has offered free HIV screening, counseling, referral of infected individuals to HIV treatment centers, and access to free antiviral treatment through the National Free AIDS Treatment Program (NFATP). Through provision of technical and financial support, the China CDC has led a national demonstration program for HIV/AIDS since 2003 to strengthen capacity for HIV/AIDS prevention among key populations at the county or urban district level. Currently, this demonstration program is in its third round, and encompasses 241 demonstration areas across all 31 provinces.

HIV testing has traditionally been considered the entry point into the system for HIV prevention and treatment, and has been one service widely promoted for HIV prevention in China. Reported uptake of HIV testing has varied among FSWs. A study conducted in Beijing from 2010 to 2013 among 481 FSWs showed that only 11.02% of participants had been tested for HIV in the last year [47]. Another study in Liuzhou, a city in the Guangxi Autonomous Region, reported that the HIV testing rate was 45.28% among 2800 FSWs, and women receiving HIV testing also reported a higher proportion of condom use compared with women who did not receive HIV testing [48]. With regard to follow-up after receiving a diagnosis of HIV, in a cohort study conducted in Yunnan between 2010 and 2012, 99 HIV-infected FSWs were enrolled at the time of HIV diagnosis and followed up at 3, 6, 9, and 12 months. A significant proportion of FSWs remained involved in commercial sex at follow-up (45.00%, 48.75%, 58.67%, and 53.33%, respectively); however, the proportion of antiretroviral therapy (ART) was low and steady (p = 0.663) at baseline and follow-up (25.3%, 23.8%, 26.3%, 30.7%, and 20.0%, respectively) [49]. Similarly, Tan et al. studied behavioral change among 101 FSWs recently diagnosed with HIV recruited from low-tier venues in Guangxi and found that 52.47% (53/101) of FSWs still engaged in prostitution, and only 24.75% of women reported taking ART; but the proportion of women using condoms during commercial sex within the last month increased significantly following HIV diagnosis (71.69%, 38/53 vs 49.05%, 26/53; p < 0.05) [50]. These findings underscore the critical point that for FSWs, HIV testing and diagnosis do not automatically lead to prevention or treatment-seeking behaviors as expected. Possible contributing factors include low levels of education among FSWs, economic pressure to support family, and lack of other work skills [50, 51].

In order to promote a barrier method that women can initiate, and an option for dual barrier protection, several female condoms (FCs) with different designs and materials have been developed and produced globally, including two products approved and registered in China. Nie and Liao et al. conducted an intervention study among 445 FSWs from 2008 to 2012 in four towns in Hainan and Guangxi to evaluate the acceptability and adoption of an FC produced by the Female Health Company. The 1-year intervention—which included an educational component, training for FSWs regarding proper FC use and trouble-shooting potential difficulties, and distribution of free FCs along with male condoms—was delivered via outreach to local sex venues by the health workers. Serial surveys were administered at baseline, 6 months and 12 months. At 12 months, 19.7%, 28.4%, and 50.3% (with one town missing data at 12 months) of participants reported using FCs at least once. Among those using FCs, 50% used FCs more than once, but only 6% reported using FCs regularly in the last 6 months. Compared with baseline, the overall proportion of women reporting 100% condom use in the last 30 days was relatively stable (near 40%), but the proportion of women reporting never using condoms declined in all towns, although the decline was modest in two of the four towns. Some FSWs considered FCs to be a backup protection method when male condoms could not be used [35, 52].

Another vaginal barrier product with a different design, the Woman’s Condom (brand O’Lavie, registered and produced in Shanghai), has also been used in a few intervention studies for FSWs with differing follow-up periods of 1 month [53] or 1 year [54]. The results also demonstrated that when FCs are provided for free as part of an intervention, FSWs are willing to consider them as a barrier protection method. Zhou et al. carried out a brief educational intervention and provided free FCs to 312 low-fee FSWs in three cities in central and Southern China. At the 1-month follow up assessment, 87.5% of the low-fee FSWs had used at least one FC, and 39.4% reported using more than two FCs during the course of the month. FCs were more commonly used among FSWs who were less educated or charged the lowest fees, and with FSWs’ regular clients [53]. Zheng & Chen et al. conducted a 1-year intervention study, using a similar protocol to Nie and colleagues, among 419 FSWs in three towns of Guangxi between 2013 and 2014. The project provided 3800 O’Lavie FCs and 23,000 male condoms during the course of two 6-month intervention periods. At the 12-month follow-up, more than half of the participants (55.3%) reported having used at least one FC in comparison with 13.3% at baseline. The uptake of FC was different across study sites, and FSWs working in boarding houses, who were freelance FSWs without proprietors, reported 5 times higher odds of using FCs more than once than FSWs in sex-work venues operated by bosses (OR = 5.27, 95% CI = 2.18–12.76). The uptake of FCs was not associated with frequency of male condom use. [54]

Taken together, the overall uptake of vaginal barrier products still remains lower than that of male condoms in the context of the commercial sex industry in China; however, the incorporation of free FCs into male condom interventions may provide an important, woman-initiated, incremental benefit toward reducing unprotected sex among FSWs in China, thereby promoting safe contraception and controlling HIV/STI transmission [55]. Possible reasons for the lower uptake include the shape and size of the FC, and discomfort after placing it into vagina [55]. Factors impacting FC use may include intervention-related variables, as well as contextual characteristics such as study site, type of sex work establishments, and type of sexual partners [35, 53, 54]. Additional research is necessary to clarify the factors affecting FC use, and to develop a viable, comprehensive promotion model to increase uptake.

A few studies have started to examine the potential application of novel strategies such as pre-exposure prophylaxis (PrEP) and harnessing the capabilities of the internet or mobile phone applications for HIV prevention among FSWs in China. Peng et al. surveyed 1611 FSWs across four provinces of China (Chongqing, Guangxi, Xinjiang, and Sichuan) from 2009 to 2010 and explored FSWs’ willingness to take PrEP. The authors found that 69% of FSWs reported intent to use PrEP, and that Han ethnicity, urban residence, high HIV-related knowledge, a history of STIs, and use of prophylactic medication for STIs were independently associated with higher acceptability of PrEP. Barriers to acceptability included economic status and concerns regarding the side effects of the PrEP drugs [56]. Another study in Guangxi focused on the acceptability of PrEP and willingness to participate in a clinical trial for PrEP. Among 405 participants, 15.1% had heard of PrEP. If PrEP was deemed to be effective, safe, and provided for free, 85.9% reported that they would accept it, and 54.3% of those who accepted PrEP said that they would participate in future clinical trials. The study identified the following factors as potentially associated with increased acceptability of PrEP: having a higher income, poor relationships with family members, increased knowledge regarding HIV/AIDS, lack of recognition of HIV risk from unfamiliar clients, lack of requirement for mandatory condom use in their work setting, consistent use of condoms, and use of prophylactic treatments to prevent STI [57]. However, a cross-sectional study published in 2019 of 1466 FSWs across China found that awareness of PrEP (10.2%) and willingness to use PrEP (35.5%) both remained low, demonstrating a significant knowledge gap regarding PrEP in this population [58]. Additional studies among key populations such as FSWs will provide important insights as the feasibility and optimal strategy for integration of PrEP in the national HIV prevention programs.

As a report on development of internet in China by the Cyberspace Administration of China, the population of Internet and mobile technology users in China has reached 0.77 billion up to 2017 [59], which has tremendous implications for both HIV transmission and HIV prevention strategies. However, much fewer studies among FSWs were found in these aspects. One cross-sectional survey recruited 1022 FSWs, representing nine types of commercial sex venues at different levels in two tourist cities (Beihai and Guilin) of Guangxi in 2009. The study found that among FSWs who use the internet, two-thirds were willing to participate in the online HIV/STI prevention program. Internet usage among FSWs was associated with younger age, higher level of education, higher income, and participation in internet dating [60]. Another study conducted in Beijing assessed associations between levels of mass media exposure (via television, street posters, radio, and the Internet) of HIV/AIDS prevention information and HIV/AIDS knowledge as well as condom use–related attitudes, beliefs, and behaviors. The study showed that among 359 FSWs, exposure to HIV/AIDS prevention information delivered by radio, street posters, and the internet was associated with higher levels of communication about HIV or condom use with sexual partners [61]. These studies demonstrate the important role that different forms of media, particularly the internet, can play in HIV prevention. Future studies are needed to explore innovative approaches integrating the internet and mobile applications into new or existing HIV prevention programs targeting FSWs.

Conclusion

HIV prevalence among FSWs is generally low in mainland China, but sporadic infection has been reported among FSWs in all but one province. Higher prevalence rates were found among FSWs in traditional centers of the HIV epidemic such as South central and Southwest China. Condom education and promotion are still the cornerstone of prevention programs among FSWs. However, in light of changing behavioral patterns in the Chinese sex industry, more comprehensive prevention strategies may be necessary to maintain the current low levels of HIV prevalence in this population. Such strategies should adopt a multidimensional approach to address risk factors such increasing drug use and risky sexual behaviors, to incorporate other health needs such as STIs and reproductive health, and to integrate new technologies. Finally, a greater focus on implementation science research is necessary to ensure the successful integration and uptake of these strategies among FSWs in the context of a rapidly changing society.