Methods used to find clients and risky sexual behavior among female sex workers in China: a cross-sectional study

Abstract

Background

Female sex workers (FSWs) are high-incidence population of sexually transmitted infections (STIs), including human immunodeficiency virus (HIV). The methods used to find clients for commercial sex have diversified in recent years. As a result, it is crucial to analyze the correlation between methods used to find clients and risky sexual behavior and explore changes in risky sexual behavior among FSWs over time.

Methods

HIV sentinel surveillance data were collected from Zhejiang Province, China, between 2016 and 2021. In total, 4,798 FSWs were included in our analysis. We collected information related to socio-demographic variables, HIV knowledge, methods used to find clients, and condom use. Descriptive statistics were used to analyze the socio-demographic characteristics of FSWs. Univariate and sequential logistic regression models were employed to determine the influence of methods used to find clients on risky sexual behavior after controlling for other factors.

Results

The average age of the 4,798 FSWs was 33.0 ± 7.4 years. The majority (70.6%) who find clients online reported always using condoms over the past month. The proportion of FSWs engaging in risky sexual behavior, defined as not using condoms at the last sexual encounter or not using them consistently over the past month, was lowest among those who find clients online, whereas the risk was highest among those who worked at a fixed site (p < 0.001). Sequential logistic regression demonstrated that compared to FSWs who find clients at a fixed site, those finding clients online (OR = 0.6, p < 0.001) were significantly associated with risky sexual behavior.

Conclusion

FSWs had a high knowledge regarding acquired immunodeficiency syndrome (AIDS), but with greater likelihood of risky sexual behavior, particularly among FSWs who find customers at a fixed site. The risk of engaging in condomless sex among FSWs who find client at fixed sites was higher than those online. Therefore, it is essential to explore health interventions that improve condom use among FSWs.

Background

According to the 2019 Global Burden of Disease study, unsafe sex was the 12th Level 2 risk factor for attributable deaths among females and the 17th among males [1]. In recent years, more than 95% of newly diagnosed patients with HIV were infected through sexual transmission, with heterosexual transmission accounting for about 70% of cases, suggesting that risky sexual behavior is the main cause of the HIV epidemic [2]. Risky sexual behaviors, such as sex with multiple partners, condomless sex, and sex under substance abuse, may not only lead to sexually transmitted infections (STIs), but also cause unintended pregnancy, sexual and psychological violence and poor mental health, self-denial, and impose a substantial public health burden [3].

Commercial sex already existed in ancient civilizations. For example, brothels in ancient Rome were legal and regulated. In the Middle Ages, Christian Ethics had a significant impact on commercial sex, with sexual behavior strictly confined to marriage. During the Renaissance, the rise of humanism led to a relaxation of attitudes toward sex. The Industrial Revolution caused a massive influx of people into cities, leading to the rapid growth of the sex industry. The sexual liberation movement of the 1960s and 1970s had a profound impact on commercial sex, and the sex industry gradually became more publicized. In the 21st century, the widespread use of the internet has dramatically transformed the sex industry, with cyberpornography and online sex becoming the common [4, 5]. Female sex workers (FSWs) are a population who are at heightened risk of HIV infection secondary to biological, behavioural, and structural risk factors [6]. An estimated 15% of HIV transmission in the general female adult population were attributed to (unsafe) female sex work [7]. A study from Thailand corroborated that sexual partners of FSWs exhibited a high prevalence of HIV and could serve as a bridge for HIV transmission to other populations [8]. Notably, commercial sex transmission accounted for 53% of male HIV cases in China [9]. Although the rate of HIV prevalence among FSWs (<1% in China [10]) is lower compared to other key populations such as men who have sex with men (MSM) and People who injects drugs (PWID), sentinel surveillance data revealed a significant rise in HIV prevalence among sex workers [11]. Therefore, interventions targeting behavioural and structural-level risk factors for HIV among FSWs have become an important step in increasing protective behaviors and disease control and prevention.

Recent studies showed the significance of characterizing FSWs practice according to workplace [12]. FSWs with unstable workplaces (street, vehicle, gas station, etc.) had lower condom use self-efficacy scores and were more subject to STIs and have a higher prevalence of HIV [13]. The physical environment and location of workplace may have an influence on sex behaviors and risk of HIV. FSWs mostly worked in fixed sites Previously. In recent years, the sex industry has been augmented by the widespread availability of the internet. Online communication platforms have the potential to facilitate commercial sex, which provide FSWs with more agency over their work, a wider variety of opportunities, convenience and safety [14]. A significant proportion of FSWs engaged in finding clients online, this behavior was positively correlated with both STIs and reproductive health outcomes [15, 16]. This shift in client acquisition strategies has created new challenges for HIV surveillance and intervention among FSWs. Research on the relationship of these new methods to find clients on risky sexual behavior is limited. Therefore, our study aims to quantitatively analyze the correlation between different methods of finding clients and risky sexual behavior. We also aim to explore changes in risky sexual behavior among FSWs since 2016. The findings from this study may provide evidence for effective interventions targeting risky sexual behavior, thereby reducing the risk of HIV and other STIs among FSWs.

Methods

Participants

Data were collected from HIV sentinel surveillance, which has been operational in China since 2010. This surveillance system continuously gathers information on HIV status, behavioral characteristics, and relevant details among specific populations at fixed locations and times. The aim is to analyze local HIV epidemic trends and evaluate the effectiveness of HIV prevention and control measures. From 2016 to 2021, investigations were conducted annually from April to July at the workplaces of FSWs in Hangzhou, Zhejiang province. These workplaces were typically categorized into low, medium, and high-grade locations based on the type of site and the prices of sex trades. High-grade places included sauna centers, nightclubs, dance halls, bars, and hotels, where the price per sex trade exceeded RMB 200 (equivalent to 28 dollars). Medium-grade places encompassed inns, hair salons, and foot massage shops, with prices ranging from RMB 50 to 200 per sex trade. Low-grade places comprised roadside stores, small restaurants, and streets, with prices below RMB 50 (equivalent to 7 dollars) per sex trade. The survey was conducted by professionals from the local health supervision institution and community Healthcare centers with the former’s job responsibility being to regularly supervise entertainment in the district. Convenience sampling was employed to collect data from FSWs aged 18 to 65 years in the district. Trained investigators conducted face-to-face questionnaire surveys in independent spaces, adhering strictly to ethical principles of consent, respect, and non-judgment. A total of 4,798 FSWs completed the questionnaire and were included in our analysis.

Data collection and measurement

The following information was collected: socio-demographics, HIV knowledge, strategies to find clients, and condom use (always using, sometimes using, and never using). The socio-demographic variables included age, ethnicity, marital status, educational level, workplace grade, and previous workplace (the previous city of sex work). Their methods of client acquisition included fixed site, call-out and online. “Fixed sites” include sauna centers, nightclubs, dance halls, bars, and hotels, etc. “Call-out” is mainly made through telephone appointment for on-site service. “Online” mainly attracts customers through Instant Messenger (IM), dating application (App), and pornographic websites, etc. Risky sexual behavior was defined as not using condoms at the last sexual encounter or not using condoms consistently over the past month. HIV knowledge was measured using the 8-item HIV Knowledge Questionnaire, which has been widely applied in HIV-related surveys in China and has been proven to have a good validity [17]. Responses were recorded as “true,” “false,” or “don’t know.” A score of 1 was assigned if the answer was correct and 0 if the answer was wrong or “don’t know.” HIV knowledge was determined based on total scores, with a higher score indicating a higher HIV knowledge level.

Statistical analysis

Descriptive statistics were employed to characterize the socio-demographic attributes of FSWs, their methods of client acquisition, HIV knowledge scores, and risky sexual behaviors, presented using frequency, percentage, and mean ± standard deviation (SD). Both univariate and sequential logistic regression models were used to determine the influence of methods of finding clients on risky sexual behavior, which was measured by condom use, after controlling for other factors such as HIV knowledge score, investigation year (from 2016 to 2021), workplace grade, previous workplace, age, marital status, and education level. An odds ratio (OR) greater than 1 indicated that the method of finding clients increased the likelihood of engaging in risky sexual behavior, while an OR less than 1 indicated a decreased likelihood. All data analyses were conducted using SPSS version 25.0 (IBM, Armonk, NY, USA), and variables with a P-value less than 0.05 were considered statistically significant.

Results

Table 1 displays the socio-demographic characteristics of 4,798 FSWs. The average age of FSWs who find clients via fixed sites was 33.6 ± 7.5 years, via call-out was 32.9 ± 8.1 years, and via online methods was 30.0 ± 6.1 years. FSWs who find clients online had a higher proportion of working in low-grade places compared to those who find clients at fix sites and though call- out (24.0% vs. 14.9% vs. 14.2%) (p < 0.001). Regarding marital status, a significant proportion of FSWs at fixed sites were married (48.1%), while most of FSWs finding clients online were unmarried (56.5%). Additionally, 42.8% (n = 329) of FSWs who find clients online had a senior high school education or higher.

Table 1.

Characteristics of female sex workers in different methods to find clients between 2016 and 2021

	Total participants (n = 4798,%)	Fixed site (n = 3712,%)	Call-out (n = 318,%)	Online (n = 768,%)	P value
Age (mean ± SD)	33.0 ± 7.4	33.6 ± 7.5	32.9 ± 8.1	30.0 ± 6.1	< 0.001
Workplace grade					< 0.001
Low	783 (16.3)	554 (14.9)	45 (14.2)	184 (24.0)
Medium	3491 (72.8)	2736 (73.7)	257 (80.8)	498 (64.8)
High	524 (10.9)	422 (11.4)	16 (5.0)	86 (11.2)
Previous workplace					0.001
None	2069 (43.1)	1577 (43.5)	147 (46.2)	345 (44.9)
Other provinces	1201 (25.0)	979 (26.4)	68 (21.4)	154 (20.1)
Cities outside the province	1170 (24.4)	896 (24.1)	68 (21.4)	206 (26.8)
Local	358 (7.5)	260 (7.0)	35 (11.0)	63 (8.2)
Marital status					< 0.001
Unmarried	2365 (49.3)	1757 (47.3)	174 (54.7)	434 (56.5)
Married	2230 (46.5)	1786 (48.1)	124 (39.0)	320 (41.7)
Cohabit	71 (1.5)	60 (1.6)	3 (0.9)	8 (1.0)
Divorced/widowed	132 (2.8)	109 (2.9)	17 (5.3)	6 (0.8)
Ethnicity					0.372
Han	4743 (98.9)	3666 (98.8)	314 (98.7)	763 (99.3)
Others	55 (1.1)	46 (1.2)	4 (1.3)	5 (0.7)
Education					< 0.001
Illiterate	25 (0.5)	19 (0.5)	3 (0.9)	3 (0.4)
Primary school	451 (9.4)	381 (10.3)	27 (8.5)	43 (5.6)
Junior high school	2450 (51.1)	1900 (51.2)	157 (49.4)	393 (51.2)
Senior high school	1626 (33.9)	1250 (33.7)	107 (33.6)	269 (35.0)
Junior college or above	246 (5.1)	162 (4.4)	24 (7.5)	60 (7.8)

Figure 1 illustrates condom use in sexual behavior among FSWs. The proportion of FSWs using condom at the last sexual encounter were high among those who find clients via call-out and online, at 95.8% and 95.8% respectively. The majority (70.6%) who find clients online reported always using condoms over the past month. The proportion of FSWs engaging in risky sexual behavior, defined as not using condoms at the last sexual encounter or not using them consistently over the past month, was lowest among those who find clients online.

Fig. 1.

Condom use in sexual intercourse among female sex workers

Figure 2 indicates that the average HIV knowledge scores among FSWs were concentrated between 7 and 8 points. Specifically, the scores for those working at fixed sites increased from 6.9 in 2018 to 7.4 in 2021. Figures 3 and 4 reveal that the risk of condomless sex was highest among FSWs at fixed sites from 2016 to 2021 (98.8%,95.5%,91.2%,78.4%,89.8%,58.8% in last sex and 97.2%,79.6%,93.7%, 76.0%,80.0%,47.8% in last month). However, the rate of condomless sex showed a decreasing trend from 2016 (98.8% in last sex and 97.2% in last month) to 2021 (58.8% in last sex and 47.8% in last month).

Fig. 2.

Average HIV knowledge score among female sex workers between 2016 and 2021

Fig. 3.

Condomless sex at the last sex among female sex workers between 2016 and 2021

Fig. 4.

Condomless sex in the last month among female sex workers between 2016 and 2021

Table 2 demonstrated that FSWs who find clients online are significantly less likely to engage in risky sexual behavior compared to those who find clients via fixed sites (OR = 0.6, p < 0.001). Moreover, those with lower HIV knowledge scores (0–2) (OR = 3.0, p = 0.006) and worked at low-grade places (OR = 1.7, p < 0.001) had a higher likelihood of engaging in risky sexual behavior compared to those at high-grade places. Furthermore, FSWs with primary school education were more likely to engage in risky sexual behavior (OR = 4.2, p < 0.001).

Table 2.

Univariate logistic regression and sequential logistic regression analyses for predicting risky sexual behavior in the previous month among female sex workers according to the methods used to find clients between 2016 and 2021

Items	Percentage (%)	Model 1			Model 2
		COR	95% CI	P value	AOR	95% CI	P value
Methods used to find clients
Online	16.0	0.5	0.5–0.6	< 0.001	0.6	0.5–0.7	< 0.001
Call-out	6.6	0.7	0.5–0.9	0.001	0.9	0.7–1.1	0.33
Fixed site	77.4	1	-	-	1	-	-
HIV knowledge score
0–2	0.6	4.1	1.9–8.9	< 0.001	3.0	1.4–6.7	0.006
3–4	2.6	5.2	3.6–7.7	< 0.001	3.3	2.2–4.8	< 0.001
5–6	14.6	3.3	2.8–3.9	< 0.001	2.5	2.1–3.0	< 0.001
7–8	82.2	1	-	-	1	-	-
Year
2016	16.6	2.6	2.1–3.3	< 0.001	2.3	1.8–2.9	< 0.001
2017	16.7	5.3	4.2–6.6	< 0.001	4.4	3.4–5.6	< 0.001
2018	16.7	5.5	4.4–6.9	< 0.001	4.8	3.8–6.1	< 0.001
2019	16.7	1.8	1.5–2.3	< 0.001	2.0	1.5–2.5	< 0.001
2020	16.7	2.7	2.2–3.4	< 0.001	2.9	2.3–3.7	< 0.001
2021	16.7	1	-	-	1	-	-
Workplace grade
Low	16.3	2.4	1.9–3.0	< 0.001	1.7	1.3–2.2	< 0.001
Medium	72.8	1.1	0.9–1.3	0.34	0.98	0.8–1.2	0.82
High	10.9	1	-	-	1	-	-
Previous workplace
None	7.5	0.8	0.6–1.0	0.10	1.0	0.7–1.3	0.78
Other provinces	43.1	1.9	1.6–2.2	< 0.001	1.8	1.5–2.1	< 0.001
Cities outside the province	25.0	1.9	1.6–2.2	< 0.001	1.9	1.5–2.2	< 0.001
Local	24.4	1	-	-	1	-	-
Age (mean ± SD)		1.01	1.006–1.02	0.001	1.004	0.99–1.01	0.43
Marital status
Divorced/widowed	49.3	1.8	1.3–2.5	0.001	1.4	0.9-2.0	0.12
Married	46.5	0.9	0.8–1.04	0.23	0.8	0.7–0.99	0.04
Cohabit	1.5	0.8	0.5–1.3	0.43	0.7	0.4–1.2	0.22
Unmarried	2.8	1	-	-	1	-	-
Education
Illiterate	0.5	4.4	1.9–10.2	< 0.001	3.1	1.3–7.5	0.01
Primary school	9.4	6.4	4.4–9.2	< 0.001	4.2	2.8–6.2	< 0.001
Junior high school	51.1	3.7	2.7–5.2	< 0.001	2.5	1.8–3.5	< 0.001
Senior high school	33.9	1.9	1.4–2.7	< 0.001	1.5	1.04–2.1	0.03
Junior college or above	5.1	1	-	-	1	-	-

Notes: COR: crude odds ratio; AOR: adjusted odds ratio; Model 1: Univariate logistic regression; Model 2: after controlling for HIV knowledge score, year, workplace grade, previous workplace, age, marital status, and education

Discussion

Our results showed a correlation between the method FSWs find clients and their engagement in risky sexual behaviors. In contrast to the findings by Wang et al. [18], our study revealed that the risk of engaging in condomless sex among FSWs online was 0.6 times lower than among those who find clients at fixed sites. This discrepancy may be attributed to differences in surveyed areas and demographic groups. FSWs at fixed sites are more likely to engage in repeated sexual encounters with the same client, potentially leading to a decrease in vigilance over time. Moreover, they may choose to forgo condom use in order to retain clients. In fixed sites, clients may hold the misconception that FSWs are relatively healthy, leading them to take chances and opt not to use condoms. Conversely, clients engaging with sex workers online may be more inclined to prioritize safety by using condoms, possibly due to the perceived casual nature of online encounters [19].

From the perspective of public health, the Government can set up a free or low-cost medical check-up program specifically for FSWs at fixed sites, focusing on testing for STIs such as AIDS, syphilis and gonorrhea. At the same time, condoms and other protective equipment could be distributed free of charge through the CDC and public interest organizations, and guidance is also provided on their proper use. Regularly, medical professionals are organized to visit workplaces to conduct health-knowledge seminars for FSWs, helping them to understand their own health conditions and enhance their awareness of self-protection.

Although the proportion of risky sexual behavior was highest among FSWs at fixed sites, there was a fluctuating decline in this behavior from 2016 to 2021. The National Health Commission has expanded STI intervention coverage through various strategies, including regular monitoring, health education, condom promotion, mobilization testing, and involving “mommy” figures and business owners as peer educators. Additionally, adopting a bidirectional intervention mode for clients has been implemented. These multifaceted interventions are likely contributing factors to the observed decrease in the rate of risky sexual behavior among FSWs.

Our survey revealed that HIV knowledge scores were concentrated between 7 and 8 points, indicating a high awareness of AIDS among participants. However, this coexisted with a high risk of engaging in condomless sex, reflecting the phenomenon of “separation of knowledge and behavior” in HIV prevention and intervention [19]. Factors influencing condom use include client demands for sexual pleasure without condoms, the cognitive and communication skills of FSWs, peer pressure, and environmental influences. Additionally, structural factors such as the commercial sex environment, resource access, poverty, stigma, legal frameworks, and media influence all play roles in the negotiation process surrounding condom use [20]. Increasing FSWs’ ability to initiate discussion of condom use with their partners, refusing uncooperative clients and self-protection in violent situations are effective interventions among FSWs. They should be provided with information on risk reduction, build negotiation and self-protection skills [21, 22].

Conclusion

Our study demonstrated heightened levels of risky sexual behavior among FSWs who find clients at fixed sites. Considering the phenomenon of “separation of knowledge and behavior,” there is an urgent need to provide support aimed at raising awareness of the risks associated with sexual behaviors and improving adherence to condom use. Implementing online and offline interventions could be highly cost-effective strategies for preventing HIV transmission in this population.

Limitations

Our study has several limitations that should be acknowledged. First, the frequency of condom use monitored in this study may not fully reflect the long-term situation, and there is a potential for recall bias among participants. Additionally, the study only investigated the frequency of condom use without considering whether condoms were used consistently throughout the entire sexual encounter or if they were used correctly. Second, considering social desirability of sexual behaviors, the participants may response the related questions with answers that meet social expectations. Third, participants may not have been consistent across the study years, which could potentially influence the results. Moreover, considering the significant number of FSWs and regional differences within China, our study findings may not be representative of the entire population of FSWs in the country.

Author contributions

ZL and JX conducted the literature review and data analysis and drafted the paper. XZ, JJ, and KX contributed to the interpretation of the data and critically revised the manuscript. All authors approved the final versions for submission.

Funding

This work was supported by the China Medical Board (Project No.20–391), Fundamental Research Funds for the Central Universities, and Zhejiang Province Medical and Health Technology Plan Project (project number 2022KY1041).

Data availability

All of the main data have been included in the results. Additional materials with details may be obtained from the corresponding author.

Declarations

Ethics approval and consent to participate

The study protocol and consent procedure were approved by Medical Ethics Committee of Public health school of Zhejiang University (ZGL202306-9) and adhered to the principles delineated in the Declaration of Helsinki. Participation was entirely voluntary and anonymous and based on written informed consent, and participants had the right to withdraw from the study at any time. To ensure participants’ privacy, all collected data were deidentified, and no personally identifiable information was collected or stored. The confidentiality of individuals was properly protected in the management of the investigation and the processing of data.

Consent for publication

All authors have agreed with the content and approved the submission of the manuscript.

Competing interests

The authors declare no competing interests.