Introduction

Female sex workers’ (FSWs) sexual and reproductive health needs are complicated given their vulnerability to sexually transmitted infections (STIs) and unintended pregnancies [1]. However, due to the sensitive nature of sex work, they remain underrepresented in fertility-related research and their fertility needs are often invisible or overshadowed by their ‘risky’ occupation [1]. Several studies have reported high HIV and STIs prevalence, physical and sexual violence, and limited access to healthcare services among FSWs; all of which have adverse effects on their pregnancy outcomes [1–4]. Not often recognized as mothers, health risks associated with pregnancy outcomes in FSWs remain overlooked [2, 5].

Recent research, however, suggests that despite complex reasons for pregnancy among FSWs, most desire to have or have children [3, 5, 6]. While some FSWs may consider having children as a way to solidify their relationship with a sex partner or leave sex work, others may be forced into sex work to provide for their children [1, 5, 7]. On the other hand, unintended pregnancies–as an occupational hazard–are very frequent among FSWs across different settings and often lead to abortion [1, 2]. Evidence suggests that many abortions among this population are unsafe, and most FSWs suffer from its complications [1, 8]. Unsafe practices of abortion among FSWs are mainly driven by the criminalization of sex work and abortion [1].

In the context of Iran, similar to several other settings, sex work remains criminalized and highly stigmatized [2, 4, 9]. Additionally, induced abortion is illegal, except for medical indications that are life-threatening to mother or diagnoses of congenital fetal defects (e.g., Down syndrome) [10, 11]. Despite the criminalization of induced abortion, unsafe abortions do take place underground and studies in Iran suggest an annual rate of around eight abortions per 1,000 women of reproductive ages [10, 11]. Iranian health policy makers have recently scaled up health care services, harm reduction efforts and educational interventions catered towards FSWs. However, criminalization of abortion and female sex work have created barriers in the recognition of important contraceptive needs of FSWs in Iran. Given the considerable burden of HIV among Iranian FSWs (i.e., Point prevalence of 4.5%) [12] and the intersection of HIV and reproductive health outcomes among FSWs through mother-to-child transmission of the virus, it is essential to inform policies aimed at improving FSWs’ reproductive health outcomes by characterizing abortion practices among Iranian FSWs. Using data from the first national survey of FSWs, we try to describe characteristics of lifetime abortion among FSWs in Iran.

Materials and Methods

Results

The characteristics of FSWs with and without a history of lifetime abortion are presented in Table 1. Overall, a total of 305 (35.3%, 95% CI: 32.1–38.6) reported a history of lifetime abortion. Assuming that FSWs in our study have had their first pregnancy at the age of 15, considering the mean age of 32, the average number of reproductive years was estimated to be roughly 17 years. Given the total reported number of 305 lifetime abortions, the annual rates of abortion among FSWs were estimated to be around 20.7 per 1000 FSWs [1000 × 305/ (32–15) × 863] = 20.7. Around 31.2% of FSWs (n = 251) reported no usual contraceptive use; however, barrier method, non-barrier modern contraception methods, and dual protection were reported by 32.6% (n = 262), 23.6% (n = 190), and 12.5% (n = 152) of the participants, respectively.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Lifetime self-reported abortion in female sex workers with different socio-demographics, sexual behaviour, and injection risk behaviours in Iran (2010).

https://doi.org/10.1371/journal.pone.0166042.t001

Higher prevalence of lifetime abortion was reported among FSWs who were older (42.8% vs. 26.9%; P-value = 0.001), had ever got married (37.6% vs. 25.5%; P-value = 0.004), had children (40.3% vs. 32.7%; P-value = 0.026), had been involved in sex work for longer (40.2% vs. 30.7%; P-value = 0.003), and were younger at first sex (37.5% vs. 29.0%; P-value = 0.021). Lifetime abortion was also higher among FSWs that had ever worked in brothels (43.3% vs. 31.2%; P-value<0.001), practiced group sex within past week (52.9% vs. 33.6%; P-value<0.001), ever travelled for sex work (45.2% vs. 30.0%), experienced sexual violence in the past month (46.0% vs. 26.9%; P-value = 0.001), reported STIs in the last year (42.4% vs. 27.6%; P-value<0.001), accessed family planning services (42.0% vs. 27.7%; P-value<0.001), self-perceived risk of HIV (40.4% vs. 31.3%; P-value = 0.006), and ever tested for HIV (40.3% vs. 30.9%; P-value = 0.004). Lastly, FSWs who had ever consumed alcohol (39.5% vs. 30.8%; P-value = 0.008), ever used drugs (37.6% vs. 29.8%; P-value = 0.031), and ever injected drugs (48.8% vs. 33.1%; P-value = 0.001) were more likely to have a history of lifetime abortion.

The results of the bivariable and multivariable analyses are presented in Table 2. In the multivariable logistic regression model, being older than 34 years (AOR = 1.74, 95% CI: 1.02, 2.96), recent group sex (AOR = 1.92, 95% CI: 1.10, 3.35), history of travel for sex work (AOR = 1.55, 95% CI: 1.09, 2.20), history of sexual violence (AOR = 1.77, 95% CI: 1.25, 2.50), reported STIs in last year (AOR = 1.53, 95% CI: 1.09, 2.14), and accessing family planning services (AOR = 1.76, 95% CI: 1.24, 2.49) remained independently significantly associated with history of lifetime abortion.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Correlates of having a recent HIV test results among female sex workers in the first national bio-behavioural surveillance survey in Iran (2010).

https://doi.org/10.1371/journal.pone.0166042.t002

Discussion

Our study, the first of its kind in Iran, suggests that over 70% of FSWs in our national study were usually on contraceptives, and more than one-third had a history of lifetime abortion. Lifetime abortion was significantly associated with being older, as well as history of group sex within last week, travel for sex, sexual violence in last month, STIs in last year, and accessing family planning services. Our findings indicate that similar to a variety of international settings which have reported pregnancy termination prevalence ranging from 11.7% in Swaziland to over 80% in Cote d'Ivoire [14–20], FSWs’ reproductive health need are still unmet in Iran.

Moreover, FSWs’ contraceptive practices are comparable with those in women of reproductive age in the general population. While annual abortion rates among the general population are estimated 7.5 per 1000 women [10, 11], our findings suggest an estimate of about three times higher among FSWs. Although over two-third of Iranian FSWs and the general population (70.5% vs. 74.5%) have reported using a method of contraception, condom use practices of FSWs exceed that in the general population (30.6% vs. 11.4%). Conversely, women of the general population have reported higher practices of non-barrier modern contraception methods compared to FSWs (34.8% vs. 22.2%) [21]. Such differences between FSWs and women of the general population in Iran could be due to FSWs’ overstatement of condom use as well as higher fees associated with non-barrier modern contraception methods, active condom promotion efforts (e.g., free condom dispensation) catered towards FSWs receiving services at centers for vulnerable women, and higher self-perceived risk of FSWs for acquiring HIV/STIs [22–25].

Studies among FSWs have associated non-barrier modern contraception methods with reduced condom use, reflecting on a perceived decreased need for using barrier methods (i.e. condom) [18, 26, 27]. Among our participants, rates of condom use with last paying and non-paying partners were around 60% which is comparable with studies in various international settings [14–20]. Moreover, dual protection was reported among around 12.5% of our participants which is similar to studies elsewhere (e.g., 12% in Russia [18], 38% in Kenya [25]). Despite relatively high knowledge of HIV among Iranian FSWs and rising rates of condom use [12, 28], their inconsistent condom use could be influenced by barriers to condom use such as limited condom negotiation skills. This might urge FSWs to forgo condom use when on non-barrier modern contraception methods, despite their acknowledgment of the primacy of HIV risks associated with sex work [18]. Therefore, regardless of FSWs’ use of a non-barrier modern contraception method, the importance of consistent and correct condom use with all their sexual partners (paying and non-paying) should be further emphasized to prevent HIV/STIs transmission within their sexual networks. While promoting dual protection to FSWs may raise concerns around FSWs lower tendency towards using condoms while on other contraceptive methods [29], there is a great need for exploring alternative non-daily use or non-coitally dependent contraceptive methods (e.g., injectables and implants) on top of condoms among Iranian FSWs to reduce the risk of unintended pregnancies when they cannot follow correct and consistent condom use practices [25]. Nonetheless, promoting condom use with non-paying partners remains challenging in this sub-population as it may be interpreted as a sign of mistrust by their partners [20, 30, 31]. Despite the difficulties in accessing FSWs’ clients for health promoters in Iran, it is critical to find strategies (e.g., phone-based or web-based surveys) to direct prevention activities at paying and non-paying partners of FSWs.

History of travel for sex, group sex, and reported sexual violence were significantly associated with higher likelihood of lifetime abortion among FSWs. While our understanding of the mobility of FSWs in Iran is limited, this association is also reported in Afghanistan [22] and may reflect the importance of future mixed-methods research on the dynamics of selling sex outside the city of residence. Traveling for sex work purposes is associated with lower availability of contraception methods, higher risky sexual practices (e.g., violent sexual practices, anal sex), limited condom negotiation ability, and involvement in risky environments which often lead to harmful behaviours [22, 32, 33]. In particular, group sex -an indicator of high-risk behaviours- could be a frequent practice in such risky environments and has been associated with injection and non-injection drug use, selling sex, sex under the influence of drugs and alcohol, sexual mixing of HIV/STIs-infected people with non-infected individuals, and using sex toys and condoms with several sex partners [34, 35]. Our findings indicate the need to highlight the significance of contraceptive use and safe sex negotiation for FSWs, particularly among those who report traveling for sex or engaging in group sex events.

As expected, older age was significantly associated with a history of lifetime abortion as older FSWs’ desire for bearing children may decrease as they get older. Our finding, however, should not overshadow the importance of providing early access to contraceptives among young and adolescent FSWs as about one-fourth of 18-24-year-old FSWs in our sample reported a lifetime abortion. Indeed, studies suggest that given the illegal nature of abortion in several settings (e.g., Iran), the profound stigma associated with premarital pregnancy and sexual activity, limited sexual health education for youth and adolescents, and negative attitudes of healthcare providers towards teenage pregnancy, young women and FSWs’ abortion practices are highly under-reported and often carried out within unauthorized and underground health facilities [36–38].

Accessing family planning services—one of the main reasons for FSWs’ referrals to harm reduction services [24]–and self-reported STIs in the previous year—an indicator of high-risk behaviour—were significantly associated with higher lifetime abortion prevalence. This could be partly explained by the reverse causality effect of prior lifetime abortion experiences on current contraceptive use as well as the availability of STIs counseling services within these centers. While it is encouraging that despite Iran’s recent shift in population control policies, family planning services targeting FSWs are not significantly affected and are still freely available [39, 40], around 40% of our participants had not received family planning services (e.g., STIs counselling, free condoms). This calls for further research on barriers to accessing the available services.

We acknowledge the limitations of our study that are common to studies conducted on such sensitive topics among FSWs. Given that these data were collected in 2010, our findings may not give a full picture of current contraceptive and abortion practices of Iranian FSWs; however, we were not able to pursue publication earlier due to the sensitivities around the topic. Given the lack data on Iranian FSWs’ abortion practices, we believe our findings of the only national survey of FSWs to date, will still have crucial implications for both research and policy. Moreover, our cross-sectional study design limits causal inferences and the voluntary nature of participation and self-reported data do not rule out possible selection or social desirability biases. An important limitation of the study was the potential for time lag bias as the timing of abortion was not captured in the survey. However, given the difficulties around changing behaviour in any population including FSWs, those with prior abortions are likely to be exposed to present or future high risks of abortions. Our non-random sample may also limit the generalizability of our findings to high socioeconomic FSWs and those who do/cannot use the facilities serving FSWs (e.g., outreach FSWs). They might be likely to have different abortion experiences and seek abortion services in the private sector (e.g., high socio-economic FSWs) or underground (e.g., outreach FSWs). Nonetheless, whereas accessing a very representative sample of FSWs in Iran continues to be challenging, efforts were made to reduce potential biases by recruiting a large national sample size as well as engaging local organizations and staff as well as experienced interviewers.