Alongside calls made globally for violence against women to be recognized as an important public health issue, an increasing body of academic and policy research has identified violence as an important structural determinant of HIV and sexually transmitted infections (STIs).1,2 Violence perpetrated by an intimate partner has been associated with higher HIV/STI risk through several pathways. Indirectly, much of women’s risk derives from sex-based power differentials favoring the male partner and exist both on an interpersonal level within their relationships and on a social-structural level through sex-based inequities, lower social status of women and masculine ideologies.3,4 Intimate partner violence has been linked to increased HIV/STI risk through lower condom use, higher numbers of partners or more frequent sex, substance use, and involvement in transactional sex or sex work.5–7 Sexual violence has also been directly linked to an increased risk for HIV/STIs because of potential damage to vaginal tracts that facilitate transmission of infection and the reduced likelihood of condoms being used. Evidence increasingly suggests that men who commit violent acts against women are themselves more likely to have higher-risk behavior and be infected with HIV/STIs.8,9
Despite high rates of violence experienced by women in sex work globally and their high risk for HIV and STIs, violence against female sex workers (FSWs) is frequently overlooked in international agendas to prevent violence against women.10 In addition to sex-based social inequities and attitudes that increase women’s risk for interpersonal violence, FSWs are also subject to increased risk of violence through male perceptions of sex work as a violation of the traditional ideals of female sexual purity and higher social tolerance for violence against sex workers as being “deserved,” as well as notions that perpetrators will likely go unpunished. Female sex workers experience high rates of occupational violence (i.e., within the context of sex work) from a variety of perpetrators including clients, police and managers, and exploitative business owners, as well as violence by intimate or other nonpaying partners (NPPs).10 Increasing calls have been made to address interpersonal violence as an important feature of HIV/STI prevention programs for FSWs,10 including in India,11 where the HIV epidemic is concentrated to a large degree among FSW and client populations. Violence has been identified as a concern for FSWs in Karnataka state, India, as part of the Avahan India AIDS Initiative,11 the largest community-structural HIV prevention initiative focused on most-at-risk-populations in southern India.12 Although the overall adult population in Karnataka state has an HIV prevalence estimated to be less than 1%,13 HIV prevalence among FSWs in several districts in Karnataka is much higher (10%–34% in 2004–2005, 8%–27% in 2007–2008).14
While quantitative data linking interpersonal violence against FSWs and HIV/STIs are scarce relative to among general populations of women, a recent systematic review of violence and HIV/STI risk among FSWs suggests emerging evidence has begun to link occupational violence, including physical and/or sexual violence, against FSWs with higher risk for HIV/STIs through increased odds of HIV infection, HIV seropositivity, and STI infection.15 However, substantial research gaps remain in describing the mechanisms through which violence may relate to increased HIV/STI risk among FSWs. Only 4 studies in this review found an association between violence and condom use outcomes (e.g., condom use with clients at last sex; inconsistent condom use [ICU] with clients in the last month). Measurement of condom use outcomes, as well as violence variables, is inconsistent across studies. Just one study distinguished between condom use with occasional/one-time clients and that with regular/repeat clients, and this study was only able to assess sexual (and not physical) violence.11 No studies have examined a relationship between violence by intimate partners and condom use by intimate partners. A lack of evidence has consistently been identified as a barrier to developing community-structural interventions that specifically target violence as an upstream factor that increases risk for HIV/STIs and bringing attention to violence against FSWs to an international scale.
Guided by the lack of available evidence to suggest how and if addressing violence should be incorporated into HIV/STI programming, the objectives of this study were therefore to characterize violence experienced by FSWs in 3 districts in Karnataka state, southern India, according to perpetrator (i.e., occasional or repeat clients; NPPs) and type of assault (physical and/or sexual violence) and to examine the relationships between (1) violence perpetrated by clients and ICU with occasional and repeat clients and (2) violence perpetrated by the main NPP of FSWs and ICU with the NPP. As a secondary analysis, we also examined the factors that were associated with experiencing client violence.
MATERIALS AND METHODS
Study Design and Sampling
The districts in the state of Karnataka in which the Integrated Biological and Behavioural Assessments were conducted were chosen with the aim of representing the different sociocultural regions of the state and on the sizes of the high-risk populations.16 The target sample size in the 3 districts considered in this analysis was fixed at 400 for each survey round. Owing to the availability of relevant survey questions on violence and policing among FSWs, only data from the second round of surveys (2007–2008) were used. Details on sampling are found in Ramesh et al.16 In brief, the target sample size per district was fixed at 400 completed interviews plus blood samples in Belgaum, Bellary, and Shimoga, the districts included in this study. Several types of probability-based cluster sampling designs were used to construct samples of FSWs in each district (conventional cluster sampling for FSWs practicing sex work at homes, brothels, lodges, and dabhas [road-side eating establishments], where the populations of FSWs were relatively stable; conventional time-location cluster sampling for street-based FSWs). Normalized sampling weights were calculated and attached to individuals in each cluster.
Survey Organization and Methods
Statutory approval for the conduct of the Integrated Biological and Behavioural Assessments and their protocols was obtained by the Government of India’s Health Ministry Screening Committee. All studies were approved by the Institutional Ethical Review Board of St John’s Medical College, Bangalore, India, and the Health Research Ethics Board of the University of Manitoba, Winnipeg, Canada. Current FSWs (>14 years) were eligible for inclusion in the study. Surveys were administered through face-to-face interviews, with no names or personal identifiers recorded. A detailed and standardized consent process was implemented for each respondent.
ICU Outcome Measures
The primary analysis included outcomes measuring ICU by occasional clients, repeat clients, or the main NPP, as reported by FSWs. Inconsistent condom use was defined as less than 100% (responses of “often,” “sometimes,” or “never” as compared with “every time”). The outcome variables were dichotomized for several reasons. Conceptually, because women with less than 100% condom could potentially be exposed to HIV and other STIs, this category was of primary interest in this study. Dichotomizing the variable in this way also allows the results from this study to be comparable with other studies, providing a richer understanding of the association between violence and HIV risk among FSWs.15
Physical violence by different perpetrators (i.e., clients and NPP) was collected in ordinal categories of the numbers of times experiencing violence in the last 6 months (defined as “hurt, hit, slapped, pushed, kicked, punched, choked, burned,” with responses of zero, once, 2–5 times, and 6+ times). Physical violence by different perpetrators was defined as “once or greater.” Sexual violence by different perpetrators in the last year was defined as being “beaten or otherwise physically forced to have sexual intercourse with someone even though you didn’t want to?” Eight violence variables were defined (Table 1). Client violence (physical and/or sexual violence within the last year) was also used as an outcome in secondary analysis.
Variables adjusted for in multivariable models examining the association between violence and ICU (primary analysis) and as factors associated with experiencing client violence (secondary analysis) were social and behavioral factors including age, marital status, age at first sex, duration of sex work, and numbers of clients they had sex with in the most recent week worked. Environmental factors included district of recruitment (Belgaum, Bellary, or Shimoga districts); sex work environment, defined as the place of solicitation or servicing of clients of FSWs and grouped into 3 categories: home-based (home, rented room), brothel-based (lodge, dabha [road-side lodge-type establishment], brothel), and public-places-based (vehicle, public places, bar/night club); whether sex work was FSWs’ sole income; and literacy. We also examined if women were part of a sex work collective and if they were recently arrested (in the last year). Finally, we examined exposure to the intervention by whether or not women had seen a condom demonstration by intervention staff. There are several key variables that have been used in previous studies to measure exposure to the intervention.17 Because all of these variables are highly correlated, only 1 variable was used, “whether or not women had seen a condom demonstration,” because this was perceived to be the most conceptually strong potential association with condom use (the primary outcome in the current study).
Statistical analysis was conducted using SAS version 9.1. Continuous variables were categorized based on previous literature if they did not have a linear relationship with the logit of the binary outcomes. Multivariable logistic regression models were developed to assess the relationship between violence and ICU, adjusting for covariates. Social and environmental factors that were associated with experiencing client violence were assessed in bivariate analyses using χ 2 tests and using multivariable logistic regression. On a bivariate level, covariates were considered statistically significantly associated with outcomes at a P < 0.10 level and, if they met these criteria, were then included in all multivariable models (although district and place where servicing clients were included a priori). On a multivariable level, covariates were considered statistically significantly associated with the outcome on a P < 0.05 level. We used 3 models to assess the relationship between the number of times experiencing physical violence by any perpetrator and ICU (see supplementary material, including Table S1). Sampling weights were used in multiple regression models to account for the complex sampling design. Multicollinearity in multivariable models was assessed using the variance inflation factor and tolerance statistics, corrected for the survey methods used. Adjusted odds ratios (AORs) and 95% confidence intervals (95% CIs) were reported for multivariable logistic regression. All P values reported are 2 sided.
Characteristics of the Sample
Table 1 describes the prevalence of ICU and different types and frequency of violence experienced by FSWs and characteristics of the sample. The sample sizes for Belgaum, Bellary, and Shimoga were 416, 425, and 414 (total n = 1255), respectively, with 402, 411, and 406 women completing the interview (n = 1219). Overall, 1133 and 1005 FSWs reported having occasional and repeat clients, respectively, and 688 reported having an NPP. Inconsistent condom use was 13.0% (143) with occasional clients, 20.1% (202) with repeat clients, and 72.1% (491) with the NPP. Of the total sample, 13.1% (159) FSWs had experienced physical violence by any perpetrator in the last 6 months, 9.6% (111) had experienced client physical or sexual violence and 3.7% (42) had experienced NPP physical or sexual violence. Across districts, the median age of FSWs was 32 years (interquartile range [IQR], 26–38 years), and the median duration of sex work was 5 years (IQR, 2–11 years) (Table 1).
Violence and ICU
Figure 1 shows the bivariate relationships between binary violence measures and ICU. Compared with those who had not experienced violence, ICU by occasional and repeat clients was statistically significantly higher for women who had experienced physical violence by any perpetrator in the last 6 months (occasional clients: 22.0% vs. 11.6%; repeat clients: 33.4% vs. 17.8%), client physical or sexual violence (occasional clients: 25.2% vs. 11.6%; repeat clients: 32.5% vs. 18.6%), client physical violence only (occasional clients: 24.0% vs. 12.4%; repeat clients: 35.7% vs. 19.1%), and client sexual violence only (occasional clients: 25.2% vs. 11.6%; repeat clients: 37.1% vs. 19.0%) (Fig. 1A, B). Compared with those who had not experienced violence, ICU by the NPP was only statistically significantly higher for women who had experienced sexual violence by the NPP (88.9% vs. 71.9%) (Fig. 1C). Figure 2 shows that ICU by occasional and repeat clients also increased with increasing number of times experiencing physical violence from any perpetrator. There was not a clear relationship between the number of times experiencing physical violence from any perpetrator and ICU by the NPP (results not shown). Women who had experienced physical violence by any perpetrator in the last 6 months were significantly more likely to have experienced sexual violence by any perpetrator in the last year (P < 0.001), suggesting that reports of physical and sexual violence could have resulted from the same experience of violence. However, women who had experienced physical or sexual violence by clients were not significantly more likely to have experienced physical or sexual violence by the NPP (P = 0.321; results not shown).
In multivariable analysis (Table 2), ICU by occasional and repeat clients was significantly associated with the following: physical violence by any perpetrator (AORs, 2.2 [95% CIs, 1.3–3.8] and 2.5 [95% CIs, 1.6–4.0], respectively), client physical or sexual violence (AORs, 2.7 [95% CIs, 1.6–4.4] and 2.2 [95% CIs, 1.4–3.4], respectively), client physical violence (AORs, 2.4 [95% CIs, 1.2–4.9] and 2.8 [95% CIs, 1.5–5.1], respectively), and client sexual violence (AORs, 2.8 [95% CIs, 1.5–5.2] 2.3 [95% CIs, 1.3–4.0], respectively). Owing to small cell size, the relationship between ICU by the NPP and sexual violence by the NPP was not examined in multivariable analysis (Table 2).
Factors Associated With Client Violence
Table 3 displays relationships between social and environmental factors and experiencing client physical or sexual violence. In bivariate analysis, being recently arrested was strongly significantly associated with experiencing client physical or sexual violence, as was having a higher number of clients and having sex work as sole income at a P < 0.1-level. In multivariable analysis, only being recently arrested remained strongly associated with experiencing client violence, after adjusting for social and environmental factors (AOR, 1.8; 95% CIs, 1.0–3.3) (Table 3).
This study investigated the relationship between violence perpetrated by clients toward FSWs and ICU. The odds of ICU with occasional and repeat clients were between 2.2- and 2.8-fold higher for women who experienced different types of client violence (physical and/or sexual) compared with those who did not. In contrast, this analysis did not find a significant association between violence perpetrated by the NPP and ICU.
Interpersonal Violence and Condom Use
Similar to women outside the context of sex work, one of the ways in which the link between violence and HIV among sex workers has been proposed to operate is through the relationship between violence and reduced condom use. Fear or the threat of violence by clients has been linked to being pressured into not using condoms or reduced ability to insist on condom use with clients.18–21 Coercive sex, rough sex, a history of experiencing violence, and sexual assault have been associated with reduced condom use and condom breakages.18,20,22 In the context of these studies, our results provide further evidence suggesting that violence by clients against FSWs is associated with higher rates of ICU with clients. Our study detected a statistically significant relationship between sexual violence and ICU by the main NPP in bivariate analysis, which we were unable to explore in multivariable analysis because of the low prevalence of violence.
We caution that our results cannot conclude that there is no relationship between violence by intimate or other non-paying partners of FSWs and ICU in our setting. Other studies have found that discussions of condom use by FSWs can be a trigger for violence by intimate partners, potentially further inhibiting women from negotiating condom use with these partners.23,24 The prevalence of violence by intimate or other non-paying partners of FSWs is very low, as reported by FSWs in our study compared with reports in a nationally representative study of the general population of women in India (37% compared with 4%).9 It is difficult to assess why there is such a dramatic difference, but this is likely due to a combination of factors that contribute to underreporting and low observed estimates of violence by intimate partners, as reported by FSWs in our study. Given the high rates of “everyday” and institutionalized occupational violence (i.e., by clients, police, pimps/managers, and the public) experienced by women in sex work, these women in particular may not interpret acts of violence by intimate partners as violence.25 Moreover, the survey item used in the nationally representative survey included a much broader definition of violence than used in our study.9 A more detailed question that relates specifically to intimate partner violence, developed along with input from FSWs, could improve reporting among FSWs. Other means of capturing more accurate responses could be used. For example, reports of occupational violence in other Indian settings through polling booth surveys, which are designed to minimized reporting bias,26,27 have been higher than reported in face-to-face interviews.11
The relationship between violence by clients and condom use is not straightforward. This study is cross-sectional, and thus, causal associations between experiencing violence and reduced condom use cannot be made, although evidence suggests that there may be a causal link in some cases (see previous discussion). Bivariate (Fig. 2) and multivariable analyses (supplementary material, Table S1) in this study did suggest a positive relationship between the number of times experiencing physical violence and increased ICU by clients, although a statistically significant monotonic increasing dose-response relationship was not found. Violence toward FSWs is related to broader societal inequities and power imbalances, which are similarly linked to factors that place FSWs at risk for HIV, such as ICU; in other words, it is likely that a complex set of factors leads to increased risk for violence and increased HIV risk, whereas violence and HIV risk are directly related, as well.
Addressing Violence as a Key Component of Structural Interventions for FSWs
There has been increasing support for comprehensive HIV prevention programs that address violence as an upstream factor for HIV risk for FSWs on a structural level, aiming to modify environments that facilitate increased HIV risk through experiences of violence.10 Our results suggest that programs that seek to reduce HIV risk through addressing violence by clients, both occasional and repeat, could be most effective. Antiviolence programming embedded within HIV prevention can be effective at reducing violence by clients; for example, safer-environment interventions that include the development of drop-in centers, where FSWs can visit for protection. The creation of safer sex work spaces can also play a key role in reducing client violence, with strategies including working with managers (e.g., brothel/lodge owners and housing managers) to include protective measures for sex workers23 or support for sex worker–organized indoor sex work spaces. Collectivization and community engagement processes can empower FSWs in their negotiations with clients and help FSWs challenge inequitable sex-based power balances that favor male clients.23 A study evaluating the impact antiviolence programming within the Avahan India AIDS Initiative suggested that there has been a reduction in the proportion of FSWs reporting sexual violence by any perpetrator (primarily clients, police) in follow-up surveys compared with at baseline.11 Such initiatives address occupational violence against FSWs from police and clients using a comprehensive approach, including statewide training to police officers, legal literacy training for FSWs, increased mobilization and collectivization of FSWs, crisis response teams, and legal representation for FSWs.11 It is crucial that the impact of antiviolence programming on HIV risk environments of FSWs be supported and assessed to better understand how antiviolence programming can be included in comprehensive HIV prevention. Programs and structural policies that tackle broader social inequities that contribute to FSWs’ vulnerability to both violence and HIV risk should also be explored.
Our study also showed an association between being recently arrested and client violence. Addressing the role of policing policies and practices on violence toward sex workers is particularly important in antiviolence programming. The lack of legal protections for violence against FSWs (both domestic and occupational), as well as fear of police and disengagement of FSWs from legal systems, inhibits women from reporting experiences of violence. Regulation of sex work often results in microlevel interpretations and the enforcement of sex work laws that can increase risk of violence. Sex workers may avoid certain areas or be forced to move frequently because of police presence or crackdowns on sex work.28,29 Locally, this can displace women to more isolated settings where they are at higher risk for violence and further away from resources that may facilitate safer sexual behavior.28 Again, our data do not allow causal inferences to be made, and in some cases, the factors that place women at risk for violence may also increase their visibility to police and risk of arrest (e.g., women who work in streets/public places). Interestingly, the association between sex work environment and experiencing physical or sexual client violence was not statistically significant in multivariable analysis. However, the direction of the associations supports previous research suggesting that street-based and brothel-based FSWs may experience substantial vulnerabilities that could place them at higher risk for violence. Brothel-based FSWs may have limited workplace control (e.g., in terms of the numbers and types of clients they have) and protections from violent clients when working under exploitative management and are often subject to police raids; street-based FSWs are highly visible and may work in isolated settings, placing them at higher risk of police harassment and arrest and violent clients.10 Thus, it is important to consider structural policies and programs that influence vulnerability to both factors (e.g. decriminalizing sex work and safer-environment interventions). Community mobilization efforts of sex work organizations as a means of reducing police violence and harassment30,31 could potentially be useful.
This analysis relied on self-reported answers to questions that may be perceived as sensitive, and the questions are therefore susceptible to social desirability bias.27 Reports of violence in particular may be underestimated in this study. Although the definition of physical violence used in the surveys was relatively broad, the definition of sexual violence was more restrictive and may not have captured all violence.
The findings from this study provide evidence of a relationship between client violence and ICU by occasional and repeat clients, and a relationship between being arrested and experiencing client violence. Comprehensive structural/policy programming for FSWs, including within HIV-focused prevention programs, is urgently needed to help reduce FSWs’ vulnerability to violence.
1. Garcia-Moreno C, Watts C. Violence against women: Its importance for HIV/AIDS. AIDS 2000; 14: S253–S265.
2. Martin SL, Curtis S. Gender-based violence and HIV/AIDS: Recognising links and acting on evidence. Lancet 2004; 363: 1410–1411.
3. Jewkes RK, Dunkle K, Nduna M, et al.. Intimate partner violence, relationship power inequity, and incidence of HIV infection in young women in South Africa: A cohort study. Lancet 2010; 376: 41–48.
4. Santana M, Raj A, Decker M, et al.. Masculine gender roles associated with increased sexual risk and intimate partner violence perpetration among young adult men. J Urban Health 2006; 83: 575–585.
5. Dunkle KL, Jewkes RK, Brown HC, et al.. Gender-based violence, relationship power, and risk of HIV infection in women attending antenatal clinics in South Africa. Lancet 2004; 363: 1415–1421.
6. El-Bassel N, Gilbert L, Krishnan S, et al.. Partner violence and sexual HIV-risk behaviors among women in an inner-city emergency department. Violence Vict 1998; 13: 377–393.
7. Johnson SD, Cunningham-Williams RM, Cottler LB. A tripartite of HIV-risk for African American women: The intersection of drug use, violence, and depression. Drug Alcohol Depend 2003; 70: 169–175.
8. Dunkle KL, Jewkes RK, Nduna M, et al.. Perpetration of partner violence and HIV risk behaviour among young men in the rural Eastern Cape, South Africa. AIDS 2006; 20: 2107–2114. doi:10.1097/01.aids.0000247582.00826.52.
9. Decker MR, Seage GRI, Hemenway D, et al.. Intimate partner violence functions as both a risk marker and risk factor for women’s HIV infection: Findings from Indian husband-wife dyads. JAIDS. 2009; 51: 593–600. doi:10.1097/QAI.0b013e3181a255d6.
10. Shannon K, Csete J. Violence, condom negotiation, and HIV/STI risk among sex workers. JAMA 2010; 304: 573–574.
11. Beattie TSH, Bhattacharjee P, Ramesh BM, et al.. Violence against female sex workers in Karnataka state, south India: Impact on health, and reductions in violence following an intervention program. BMC Public Health 2010; 10: 476.
12. Avahan, the India AIDS Initiative—the Business of HIV Prevention at Scale. New Delhi, India: The Bill & Melinda Gates Foundation, 2008.
13. HIV Sentinel Surveillance and HIV Estimation in India 2007: A Technical Brief. Delhi, India: National AIDS Control Organization, Government of India, 2008.
14. Ramesh BM, Beattie TSH, Shajy I, et al.. Changes in risk behaviours and prevalence of sexually transmitted infections following HIV preventive interventions among female sex workers in five districts in Karnataka state, south India. Sex Transm Infect 2010; 86: i17–i24. [Abstract/Full Text]
15. Shannon K, Nesbitt A, Deering KN, et al.. Violence and Links to HIV Infection Among Sex Workers: A Systematic Review. Geneva, Switzerland: World Health Organization, 2012.
16. Ramesh BM, Moses S, Washington R, et al.. Determinants of HIV prevalence among female sex workers in four south Indian states: Analysis of cross-sectional surveys in twenty-three districts. AIDS 2008; 22: S35–S44.
17. Deering KN, Boily M-C, Lowndes CM, et al.. A dose-response relationship between exposure to a large-scale HIV preventive intervention and consistent condom use with different sexual partners of female sex workers in southern India. BMC Public Health 2011; 11 (suppl 6): S8.
18. Rhodes T, Simic M, Baros S, et al.. Police violence and sexual risk among female and transvestite sex workers in Serbia: Qualitative study. BMJ 2008; 337: a811. [Abstract/Full Text]
19. Shannon K, Kerr T, Strathdee SA, et al.. Structural and environmental barriers to condom use negotiation with clients among female sex workers: Implications for HIV prevention strategies and policy. Am J Pub Health 2009; 99: 659–665.
20. Decker MR, McCauley HL, Phuengsamran D, et al.. Violence victimisation, sexual risk and sexually transmitted infection symptoms among female sex workers in Thailand. Sex Transm Infect 2010; 86: 236–240. [Abstract/Full Text]
21. Panchanadeswaran S, Johnson SC, Sivaram S, et al.. Intimate partner violence is as important as client violence in increasing street-based female sex workers’ vulnerability to HIV in India. Int J Drug Policy 2008; 19: 106–112.
22. Bradley J, Rajaram SP, Moses S, et al.. Why do condoms break? A study of female sex workers in Bangalore, south India. Sex Transm Infect 2012; 88: 163–170.
23. Argento E, Reza-Paul S, Lorway R, et al.. Confronting structural violence in sex work: Lessons from a community-led HIV prevention project in Mysore, India. AIDS Care 2011; 23: 69–74.
24. Panchanadeswaran S, Johnson SC, Sivaram S, et al.. A descriptive profile of abused female sex workers in India. J Health Popul Nutr 2010; 28: 211–220.
25. Deering K, Nesbitt A, Shannon K, editors. A global review of gender-based violence against sex workers: A neglected public health and human rights issue. CIHR Institute for Gender and Health: Innovations in Gender, Sex, and Health; 2010; Toronto, Canada.
26. Gregson S, Zhuwau T, Ndlovu J, et al.. Methods to reduce social desirability bias in sex surveys in low-development settings: Experience in Zimbabwe. Sex Transm Dis 2002; 29: 568–575.
27. Hanck SE, Blankenship KM, Irwin KS, et al.. Assessment of self-reported sexual behavior and condom use among female sex workers in India using a polling box approach: A preliminary report. Sex Transm Dis 2008; 35: 489–494.
28. Shannon K, Rusch M, Shoveller J, et al.. Mapping violence and policing as an environmental-structural barrier to health service and syringe availability among substance-using women in street-level sex work. Int J Drug Policy 2008; 19: 140–147.
29. O’Neil J, Orchard T, Swarankar RC, et al.. Dhandha, dharma and disease: Traditional sex work and HIV/AIDS in rural India. Soc Sci Med 2004; 59: 851–860.
30. Biradavolu MR, Burris S, George A, et al.. Can sex workers regulate police? Learning from an HIV prevention project for sex workers in southern India. Soc Sci Med 2009; 68: 1541–1547.
© Copyright 2013 American Sexually Transmitted Diseases Association
31. Chatterjee P. AIDS in India: Police powers and public health. Lancet 2006; 367: 805–806.