Female sex workers (FSWs) are a key population at risk for HIV infection in India, with a prevalence that is 20 times higher than the general population [1]. Beginning in 2003, large scale, targeted HIV prevention programs, including the Avahan program funded by the Bill & Melinda Gates Foundation, were implemented among FSWs and other key populations in high prevalence states, and resulted in significant increases in the uptake of HIV prevention services and safer sexual behaviors, and reductions in the number of sexually transmitted infections (STIs) [2, 3]. According to an UNAIDS report, by the year 2010, India saw a drop of 56% in the incidence of new HIV infections [4]. Despite these important gains in HIV prevention, HIV prevalence continues to be relatively high among FSWs in India at about 4.9% [1]. In some southern states, up to 15% of FSWs are reported to be living with HIV [5]. Further gains in prevention are likely to depend on a more nuanced understanding of the factors contributing to the persistence of high levels of risk among FSWs.

To date, HIV prevention research and programs have mainly focused on FSWs’ sexual practices vis-à-vis commercial partners, and their practices within the context of intimate relationships have been less consistently addressed. It is well known that women in India have limited decision-making agency, especially in the economic and sexual realms, in the context of their intimate relationships [6–9]. A large body of evidence has highlighted the ways in which entrenched gender inequalities compromise women’s access to and use of critical reproductive and sexual health resources [6, 9–11]. Yet, relatively few studies have explored FSWs’ sexual risks and their access to HIV prevention services in context of their relationships with non-paying partners [12–15]. Data from targeted interventions in the high prevalence states of southern India from 1995 to 2008 indicate that FSWs’ consistent use of condoms with paying clients significantly increased from 58.6% to 83.7% (p < 0.001) [16], although condom use with non-paying partners remained a low 10% [12]. However, studies have also found that HIV and STI prevalence was lower among married FSWs compared with widowed and unmarried FSWs [17–19].

The available evidence suggests that several factors are likely to shape FSWs’ HIV risks in the context of relationships with non-paying or intimate partners. For example, married FSWs are more likely to solicit clients on the street or at non-brothel-based venues whereas single FSWs are more likely to be brothel- or home-based [17]. When street-based, women may not identify themselves as sex workers, making it difficult to identify and engage with them on HIV prevention and heightening their risk of HIV infection [18]. Of the 80% of FSWs reporting indebtedness, debt has been found to be more common among FSWs who have an intimate partner; indebtedness is also associated with an increased likelihood of experiencing violence and STI symptoms [19]. Moreover, intimate partner violence (IPV) is widely prevalent among FSWs, although prevalence estimates vary due to differences in definition and reference period. In one study in Chennai, nearly all (98%) surveyed FSWs reported IPV [20] while about a third (35%) reported IPV in another study in Goa [21]. FSWs who face violence from clients or non-paying partners are less likely to use condoms consistently even with clients or utilize HIV prevention program services [22]. However, there is limited literature on the influence of these factors on the uptake of HIV prevention services by FSWs reporting a non-paying partner, including those with whom they are cohabiting. The aim of this paper is to explore the association between FSWs’ non-paying partner status, including cohabitation, and their HIV risk behaviors, HIV prevention program exposure, participation in groups/events and health care service utilization in three states in India.

In this study we explored the association between FSWs’ non-paying partner status and their HIV risk behaviors, HIV prevention program exposure, participation in groups/events and health care service utilization in three high priority states in India. We observed no differences in the HIV risk behaviors between FSWs reporting a non-paying partner and those with no non-paying partners. However, FSWs with a non-paying partner were more likely to be exposed to HIV prevention program services, participate in groups/events and utilize services compared to those reporting no non-paying partner, despite having higher vulnerability scores. Our analysis shows that there is a strong association between the vulnerability of FSWs and their HIV prevention services uptake indicating that the HIV prevention program was successful in reaching the most vulnerable section of this key population at risk.

Recent research has indicated that even though FSWs with non-paying partners had higher vulnerability, the HIV prevalence among them is 11% as compared to 18% among those with no non-paying partner [27]. This reflects the reach of HIV prevention programs to FSWs with non-paying partner, which is further supported by findings of our study. Our analysis shows that FSWs having a non-cohabiting non-paying partner were most likely to avail HIV prevention services, followed by FSWs reporting a cohabiting non-paying partner and then by those reporting a cohabiting husband. One of the reasons for higher service utilization among FSWs with non-paying partners could be that they may be less mobile as compared to those with non-paying partners. Lesser mobility of FSWs could make them more accessible to HIV prevention programs. However, more research is required to understand the motivation behind FSWs having non-paying partners accessing prevention services. Further investigation of reasons underlying this observation as to what factors contribute to the higher uptake of HIV prevention services despite being more vulnerable – such as differences in perceptions of HIV/STI risk and motivation to stay healthy – should be explored using qualitative research.

While FSWs in non-paying partner relationships avail HIV prevention program services, inconsistent condom use with these partners is almost universal. Besides the risk of HIV infection associated with inconsistent condom use, it could also result in unwanted pregnancies and abortions [28, 29]. Data on factors that can help explicate the observed associations such as reasons for condom non-use with non-paying partners as well as the family planning and abortion experiences of FSWs availing HIV prevention services, which would have been useful to explore the sexual and reproductive health concerns of FSWs, were not available in our dataset. A study in central India revealed a 70% likelihood of the unwanted and unintended pregnancies among HIV positive women [28] while a survey conducted among FSWs in Goa, India revealed that while 45% of FSWs did not use any method of contraception, 26% had experienced an abortion [29]. The study also indicated that exposure to HIV prevention programs was not significantly associated with contraceptive use [29].

A possible explanation for the lack of condom use is that FSWs lack condom negotiation skills with non-paying partners. Research from south India found that FSWs whose non-paying partners were aware of their sex work practices were more likely to use condoms consistently [30]. Also consistency of condom use was reduced by 18% per year of increase in the duration of the non-paying partner relationship [30]. A study conducted among sex workers and their non-paying partners in Vietnam pointed out that couples who were comfortable in speaking to each other about HIV risk were also more likely to report more consistent condom use within such relationships, hence emphasizing the need to strengthen communication skills among FSWs in non-paying partner relationships [31].

Community mobilization efforts in India have shown some promise in influencing FSWs’ risks in the context of intimate relationships. For example, the Frontiers Prevention Program conducted in Andhra Pradesh found that after four years of intense community mobilization accompanied by access to HIV prevention services, FSWs were more likely to use condoms with regular partners [32]. In contrast, in Karnataka a state-wide survey among a stratified sample of 1,512 FSWs found that women who are part of a sex-worker collective were more likely to report condom use with paying clients but were no more likely to use condoms with non-commercial sexual partners as compared to those who were not a part of a collective [33].

Future HIV prevention interventions in India should focus on addressing relationship factors such as risk communication and condom negotiation. For example, in South Africa, a randomized community trial of a gender-focused intervention (the Women’s Health CoOp) with FSWs increased their condom negotiation skills which in turn significantly increased condom use with their primary sexual partners at 6-month follow-up (52% vs. 37%, p < 0.05) [34]. Related outcomes included improved communication with the main partner about HIV risk, improved self-efficacy and sexual assertiveness and decreased number of experiences of sexual coercion [34].

Our analysis confirms literature on the vulnerability of FSWs reporting a non-paying partner [18, 19]. FSWs reporting a non-cohabiting non-paying partner had the highest vulnerability scores followed by FSWs reporting a cohabiting non-paying partner and a cohabiting husband. Scores were lowest among FSWs having no non-paying partner. Research among FSWs in southern India has found that being in a relationship can exacerbate their indebtedness as they have limited control over financial resources in the context of intimate relationships [19]. Furthermore, FSWs reporting a non-paying partner were more likely to be street-based. Street-based FSWs have been found to be at a significantly higher risk of HIV infection as compared to brothel-based workers because they tend to operate independently and have less access to institutional and social support [18]. The fact that three-fourths of FSWs with a non-paying partner reported being in debt and that a majority of them are engaged in street-based sex work and were older in age (data not shown) points to the need for interventions that address their financial and sex work-related risks. Interestingly, unlike previous studies, we found no significant differences in experience of violence by FSWs’ partner status. Further investigation into FSWs’ perceptions and understanding of violence may shed light on this finding.

The limitations of this study should be noted in interpreting the findings. First, data were self-reported by participants and hence subject to recall and social desirability biases. Second, being cross-sectional data we cannot establish causal relationships between partner status and HIV prevention services uptake, highlighting the need for longitudinal studies. Third, the vulnerability score was created based on only three measures of vulnerability. There are many vulnerability factors among FSWs which could have been incorporated in creating the vulnerability index. However, due to unavailability of additional information in the data set, we were limited with the choice of selection of variable to create the index. The question on debt does not measure extent of indebtedness and the question on violence is subjective. Unfortunately, we are limited by the questions that were posed in the Integrated Behavioural and Biological Assessment (IBBA) survey. Furthermore, experience of violence is typically measured using questions such as in the IBBA and is a limitation of all research on women’s experience of violence. Further research, including qualitative investigations, to consider the nuances of these issues in assessing FSWs vulnerability and its relationship with outcomes are warranted. Despite these limitations, our analyses show a strong association between vulnerability of FSWs and HIV prevention service uptake indicating that the program was able to reach out to the most vulnerable FSWs.

The fourth phase of national AIDS control program in India has been launched with the vision that by the year 2020 there will be a significant reduction in new HIV infections building on the already successful targeted interventions [35]. With declining incidence rates of new HIV infections in India and inconsistent condom use with non-paying partners among FSWs being almost universal, the next generation of HIV prevention interventions among FSWs should emphasize the need for universal condom use in efforts to achieve zero new infections in India. Our findings suggest that strengthening the currently existing HIV prevention efforts should include components focused on enhancing FSWs’ HIV risk communication and condom negotiation in the context of non-paying or intimate relationships. The study findings demonstrate that HIV prevention programs have been able to reach the most vulnerable group which has resulted in a lower HIV prevalence among them, these efforts should continue. One the other hand, HIV prevention programs have not been able to reach the FSWs with no non-paying partner to a great extent. Therefore, HIV prevention programs in India need to put more effort in reaching out and providing services to FSWs with no non-paying partners.

Our study findings indicate that the HIV prevention program in India was successful in reaching the most vulnerable section of this key population at risk. However, despite significant uptake of large scale HIV prevention program services, a majority of FSWs did not use condoms with non-paying partners. FSWs having non-paying partners remain a vulnerable key population at risk of HIV infection. Our findings suggest that the next generation of HIV prevention interventions in India should focus on addressing relationship factors such as risk communication and condom negotiation as well as specific vulnerabilities such as indebtedness and street based solicitation among women in sex work.