Economic Strengthening for Female Sex Workers: a Review of the Literature

Economic Strengthening

for Female Sex Workers: A

Review of the Literature

Economic Strengthening for Female Sex Workers: A Review of the Literature

May 2014

Whitney Moret ASPIRES, FHI 360

This report was produced under United States Agency for International Development (USAID) Cooperative Agreement No. AID-OAA-LA-13- 00001. The contents are the responsibility of FHI 360 and do not necessarily reflect the views of USAID or the United States Government.

Contents

Introduction ...... 3 The Literature on the Lives and Needs of Sex Workers ...... 4 The State of the Research ...... 4 Themes in the Literature ...... 7 Characteristics of FSWs ...... 7 Health Needs ...... 8 Economic Lives ...... 8 Gender Inequality and Stigma ...... 9 Approaches to HIV-Prevention Interventions for Female Sex Workers ...... 10 Combination Intervention ...... 10 Structural Interventions with FSWs...... 10 Economic Empowerment ...... 11 Best Practices ...... 11 Intervention Examples ...... 11 Community Mobilization ...... 12 The Sonagachi Project ...... 12 Economic Strengthening for FSWs...... 14 Evidence on Economic Strengthening for HIV Prevention ...... 14 Microfinance ...... 15 Vocational Training: SiRCHESI’s Hotel Apprenticeship Program (HAP) ...... 16 Combined Microfinance and Vocational and/or Business Training ...... 17 Undarga ...... 17 Strengthening STD/HIV Control Project in Kenya (SHCP) ...... 18 Combined Economic Strengthening and Community Mobilization ...... 18 Karnataka Health Promotion Trust (KHPT) ...... 19 Summary Table of Interventions ...... 19 Analysis ...... 22 Recommendations for Future Research and Piloting ...... 23 Preliminary Research for Pilot Interventions ...... 23 Research Gaps and Recommendations for Piloting ...... 24 Conclusion ...... 25 Sources...... 25

INTRODUCTION

Female sex workers (FSWs) have been identified as a key population in the global fight against AIDS. In concentrated, mixed, and even generalized epidemics, the contribution of sex work toward the onward transmission of HIV is substantial. High rates of partner change in “upstream” sex work networks have long been recognized to drive “downstream” transmission of sexually transmitted infections (STIs), including HIV. Although modes of transmission analysis once predicted fairly low onward transmission by sex work, in the last two years, more complex analysis with more data has found the contribution by sex work to the spread of HIV/AIDS to be much greater than expected in countries as diverse as Nigeria, Kenya, and India. In light of this evidence, interventions with sex workers to prevent HIV/AIDS could contribute to avert the bulk of infections in many countries.

Globally, it is recognized that prevention and treatment efforts among sex workers require not only a clinical approach, but interventions to address the vulnerabilities of FSWs on a structural level, including legal, physical, and economic factors. The complex structural factors associated with HIV among sex workers have been recognized in the WHO Guidelines for Sex Work and are increasingly addressed through integrated structural approaches used in HIV prevention and treatment programs on the ground.

Because sex work is illegal is many countries, the sex industry has been driven underground, leaving sex workers legally vulnerable. Given the clandestine and socially stigmatized nature of their work, sex workers are left without recourse to legal protection from threats to their safety, vulnerable to arrest and abuse, and without access to health and social services. Stigma has been shown to impact sex workers on a deeper level and is associated with lower levels of self-esteem (Scorgie et al., 2012). Physically, FSWs are frequently subjected to violence perpetuated by pimps, clients, police and madams. Sex work is associated with high levels of drug and alcohol abuse. Sex workers have limited bargaining power with clients and are vulnerable to violent sex, rape, and sex without condoms. Due to frequent engagement in risky sex, sex workers have a global HIV rate of 11.8%, which is 13.5 times greater than the general population (Kerrigan et al., 2013).

These problems are directly related to the economic vulnerability of FSWs, the vast majority of whom report entering sex work for financial reasons due to lack of alternative employment opportunities. Stigma reduces FSWs’ access to formal financial services, and many resort to incurring large debts to exploitative money-lenders in order to meet their financial needs, including paying off debts to pimps and brothel-owners.

The literature on HIV interventions targeting FSWs underscores the inextricable connections between gender, political-legal, and economic structures on the vulnerability of FSWs and their susceptibility to HIV. The good news is that interventions that target sex workers are cost-effective and they work. Analysis by the World Bank demonstrated that FSW-targeted interventions in Ukraine and Kenya have resulted in cost savings of $39M and $8.9M respectively, contributing to aversion of thousands of potential HIV infections (D. Kerrigan et al., 2012, p. xxix).

Despite evidence supporting the efficacy of structural interventions and the key role that economic vulnerability plays in contributing to structural vulnerability, there is relatively little research focused on the financial lives and needs of FSWs. Furthermore, though control over resources has been identified as a central component of economic vulnerability, there is limited research on this topic as pertains to FSWs.

This report reviews the literature on economic general guidelines must be tailored to a given strengthening (ES) interventions for risk reduction context to be truly effective. Even in a specific and HIV prevention among FSWs to identify best geographic location, FSWs are not a practices as well as opportunities for further homogeneous group. Overall, key themes emerge research. It begins with an overview of from the literature on the characteristics of FSWs, intervention approaches, including evidence their health needs, economic lives, and the barriers supporting structural interventions. It then that they face due to gender inequality and stigma. discusses existing research on the needs and financial practices of sex workers, followed by an The State of the Research overview and analysis of interventions designed to In this review, ten studies were selected from the meet these needs. The report concludes with literature for in-depth analysis according to their recommendations for future research to lay the focus on the economic needs and practices of groundwork for a future ES pilot with FSWs. FSWs. Most of the literature on the lives of FSWs uses interview and survey data, with common questions around reasons for entering sex work THE LITERATURE ON THE and engagement in risky behaviors. Most studies LIVES AND NEEDS OF SEX seek to answer a specific question; few undertake WORKERS an exploratory analysis of structures of vulnerability to HIV. Given the importance of Best practices for sex worker interventions have economic factors in structural interventions, been promoted by UNAIDS and other surprisingly few studies focus on the financial international organizations based on broad needs and habits of FSWs, though most discuss dimensions of interest that can apply to sex the experiences of FSWs in ways that shed light workers in multiple contexts. However, these on these practices.

Below is a summary table of the studies reviewed:

Study Goals Methods Findings Notes (K, 2004) Explore link between Interviews with sex High levels of Demonstrates that poverty and sex workers in Mumbai poverty, low levels of FSWs would prefer work (N = 501) vocational skills, low alternate levels of income, employment among FSWs

Does not demonstrate causality

Sheds light on socioeconomic status and livelihood choices of FSWs

Study Goals Methods Findings Notes (Evans, Jana, & Explore structure of Participant Elaborated structural Examines structures Lambert, 2010) vulnerability among observation, determinants of of vulnerability on FSWs in Kolkata, interviews with sex vulnerability among micro and macro India workers (N = 61) and FSWs based on levels senior project concepts of “identity managers, focus and agency” and Incorporated multiple groups “economic security” ethnographic methods

Done during Sonagachi intervention, not prior

(Nagot et al., 2002) Explore sex work Quantitative survey Elaborated typology Accounts for network, including with FSWs (N = 447) of six kinds of sex heterogeneity among those who do not workers: seaters, FSWs self-identify as sex roamers, sellers, workers cabarets, students, bar waitresses

(L.C. Tsai et al., Explore financial Computer-based, Discussed income One of few papers 2013) lives of FSWs interviewer- strategies of FSWs that focus on administered financial lives of baseline assessment FSWs in randomized control trial for Demonstrated how feasibility of savings high levels of intervention for financial FSWs in Mongolia responsibility and (N = 204) limited bargaining power increase risk for HIV and other STIs

(Fitzgerald-Husek, Explores Qualitative Patriarchal power Points out Martiniuk, determinants of interviews (N = 10) relations are a major heterogeneity of sex Hinchcliff, condom use among determinant in workers’ experiences Aochamus, & Lee, FSWs in Oshakati condom use and motivations for 2011) area of Namibia using condoms

Condom use is Very small sample determined by size factors beyond what is usually Language difficulties considered: in translation of education, interviews empowerment, and economic

Study Goals Methods Findings Notes independence

Highlights role of gender inequality in HIV epidemic

(Onyeneho, 2009) Describes the Structured interviews Perception scores Gained qualitative current knowledge, using quantitative revealed themes of info before attitude and knowledge, attitude economic and social quantitative survey practices on and practice (KAP) vulnerability and high interview schedule levels of ignorance HIV/AIDS issues (N = 135) about HIV among FSWs in Surveyed all FSWs Enugu and in selected preferences for communities income generating Focus group Most FSWs in this skills to inform HIV- discussions area were school- prevention aged or students intervention

(Ngo et al., 2007) Examined the sex In-depth interviews Discusses effects of Demonstrates workers' social and (N = 30) stigma on self- exceptions to every economic lives, their esteem and the generalization working problems with environment, social government sanction Compares different relationships and Focus groups (N = of FSWs kinds of sex workers presentation of self 69) in everyday social Only qualitative data contacts and interactions in two cities in Vietnam

(Reed, Gupta, Explores the Surveys and semi- Demonstrated high Strong sample size Biradavolu, relationship between structured interviews levels of debt among Devireddy, & debt, violence, and using respondent- FSWs Strong statistical Blankenship, 2010) HIV risk among driven sampling (N = associations FSWs 673) Explored experience between debt and and causes of debt violence despite self- among FSWs reporting

Linked debt to Participants not violence and randomly selected, increased self-selected vulnerability to HIV

(Scorgie et al., Systematic review of Review of qualitative FSW vulnerability is Systematic review of 2012) socio-demographic and quantitative linked to poverty, a large number of characteristics and studies using violence, relevant studies (N = behavioral risk criminalization, drug 128) factors Medline, Web of use, high mobility Science and

Study Goals Methods Findings Notes of female sex Anthropological workers in Sub- Index databases Saharan Africa Discusses regional for articles published variation between

January 2000 and April 2010

(Oyefara, 2007) Examines Quantitative Food insecurity and Strong sample size relationships questionnaires and poverty linked to between food in-depth interviews entry into sex work Qualitative and security, sexual of FSWs in Lagos, quantitative data behavior, STI Nigeria (N = 320) Food insecurity transmission and linked to risky 20% of abortion rates behavior questionnaires not returned or not used High levels of pregnancy and abortion

Studies on the lives of FSWs suffer similar job these women could find” (Vuylsteke & Jana, limitations. Nearly all use cross-sectional data and 2001, p. 202). In Sub-Saharan Africa, women can are therefore limited in their account for change be motivated to enter sex work in response to over time. Since most studies are limited to a conflict and displacement. In fact, in West Africa specific geographical location, they are of limited and other regions affected by conflict, more than generalizability. Finally, nearly all of the studies half of FSWs report that they are foreign nationals rely on self-report data for behavioral practices, (Scorgie et al., 2012). such as condom use, and face potential social desirability bias. It is important to note, however, that not all FSWs choose their line of work at all: many are trafficked or forced into sex work as minors. A Themes in the Literature study in Andhra Pradesh, India, found that one in Characteristics of FSWs five sex workers met the UN definition of Throughout the literature, economic need and experiencing sex trafficking (Gupta, Reed, coercion are mentioned as the primary motivating Kershawa, & Blankenship, 2011). Many report factors for entering sex work (Odek et al., 2009; being “tricked” into sex work with promises of Pillai, Bhattacharjee, Ramesh, & Isac, 2012). When work or simply to meet survival needs (Scorgie et interviewed, sex workers across settings express al., 2012). Trafficked persons are especially desire for alternative forms of employment, but vulnerable to both violence and HIV risk lack the skills to find another job (Fitzgerald- behaviors. Husek et al., 2011; K, 2004; Ngo et al., 2007). As explained by one study, “Compared to Not all sex workers self-identify as such. In Sub- commercial sex, few other jobs offer the same Saharan Africa, the definition of a sex worker is advantages for women, including ease of entry, a complicated by stigma and a high prevalence of ready market and higher earnings than any other transactional sex, blurring the lines between

girlfriend and boyfriend or sex worker and client health services, with barriers to utilization (Scorgie et al., 2012). Indirect sex workers do sex including “stigma, inconvenient hours of work to supplement income from another job operation and economic, language or other (Lee et al., 2010). In their study in Burkina Faso, cultural barriers” (Vuylsteke & Jana, 2001, p. 200). Nagot and colleagues (2002) identified six types of Contraception use is limited among FSWs, who sex workers, only two of which self-identified as frequently undergo repeated abortions (Oyefara, sex workers: seaters, who wait for clients in a 2007; Vuylsteke & Jana, 2001). FSWs also face given location, and roamers, who travel around high risks of rape and violence by clients and bars and nightclubs to seek clients. Mobile street- pimps. Finally, many studies report high levels of side vendors, bar waitresses, students, and drug and alcohol use among FSWs, often cited as cabarets (those who make and sell beer) were all a coping mechanism (Lee et al., 2010; Mbonye et identified as nonprofessional, or indirect sex al., 2012; L.C. Tsai et al., 2013; Vuylsteke & Jana, workers. Other studies distinguish between street- 2001), or a habit perpetuated by brothel-owners based and venue- or brothel-based sex workers that hope to get FSWs hooked (Ngo et al., 2007). (Ngo et al., 2007). Here, venue- and brothel-based FSWs are described as earning higher income and Food security is an important dimension for relying on more diverse sources of income than inquiry because it has implications for both the their street-based counterparts, who are described economic vulnerability of FSWs as well as their as older, poorer, and more dependent on sex biological susceptibility to HIV. Food security work. Overall, FSWs tend to be a mobile and affects HIV transmission according to nutritional, clandestine population that is difficult to locate mental health, and behavioral pathways (S. D. without the assistance of key informants Weiser et al., 2011). In one study of sex workers in (Johnston, Sabin, Hien, & Huong, 2006; Vuylsteke the Lagos metropolitan area in Nigeria, 35% of & Jana, 2001). respondents entered sex work as a response to food insecurity (Oyefara, 2007). The same study The relationships sex workers form with each found a significant statistical relationship between other and with others are just as varied as sex poverty, food insecurity and consistent condom workers themselves. Ngo and colleagues (2007) use by FSWs, with authors summarizing: “hunger describe escorts and call girls as working and malnutrition were the factors that pushed independently from other FSWs, therefore young women into prostitution in Nigeria and forming fewer relationships. Venue-based sex these same factors hindered them from practicing workers, however, do form relationships. Whereas safe sex within the sex industry” (p. 626). in some areas FSWs work independently from pimps (Sherman et al., 2010), in others, pimps are Economic Lives used to provide physical protection for FSWs in The literature is abundant with evidence designated areas in exchange for money and sex connecting economic desperation to risky (Ngo et al., 2007). Pimps benefit from an unequal behavior among FSWs. Poverty and limited power relationship, and often use violence to options are the most-cited reasons for entering sex control sex workers. work, affecting risk in direct and indirect ways. Food insufficiency, for example, has been linked Health Needs to transactional sex, inconsistent condom use, and In addition to their susceptibility to HIV, FSWs other risk behaviors in Swaziland and Botswana face a number of barriers to good health. A major (S. Weiser et al., 2007), while lower rates of problem for FSWs is lack of access to quality condom use have been shown in young women

from poorer households in South Africa Sweat, 2008). It also makes sense to consider how (Hargreaves et al., 2007). For FSWs, economic economic factors affect food security, a major risk need limits bargaining power with clients, who factor for vulnerability to HIV that can be frequently offer higher prices for sex without a ameliorated by economic strengthening condom (L.C. Tsai et al., 2013). intervention (Weiser et al., 2007; Weiser et al., 2011). FSWs are often caught in a web of exploitative relationships that place them at an economic Gender Inequality and Stigma disadvantage, often in persistent debt. Some Gender inequality affects FSWs on three levels: brothel-based sex workers are charged by owners economically, physically through violence and risk for food and rent, others are debt-bonded to for disease, and socially through stigma and pimps, madams and traffickers (Evans et al., limited legal rights. As discussed above, FSWs face 2010). Due to stigma and discrimination, it can be financial pressures from pimps, gangs, madams, difficult for FSWs to open formal bank accounts, and moneylenders. They can also face pressure leading them to resort to borrowing from high- from significant others that limit their control over interest moneylenders, resulting in even more their income. debt. Debt itself has sinister connections to violence and HIV risk. A study in Andhra Pradesh The role of gender inequality is particularly stark revealed very strong statistical associations in its relation to the vulnerability of FSWs to between debt and the experience of violence, violence at the hands of pimps, partners, and which itself is associated with HIV risk, among clients (Ngo et al., 2007). In a study based in FSWs (Reed et al., 2010). Enugu Town, Nigeria, FSWs described the continuous threat of rape and coercion into rough Although economics play a major role in sex, resulting in abrasion and increased risk for vulnerability, there is no linear relationship HIV (Onyeneho, 2009). Where efforts to limit between poverty and HIV risk, and the economic violence against FSWs have been made, the results status of FSWs varies greatly. As put by Kim and have been critical to stemming the spread of HIV. colleagues, “AIDS is a disease of inequality, often In Kenya, it is estimated that reductions in associated with social or economic transition, violence against FSWs “could avert over 5,300 rather than a disease of poverty itself “ (2008, p. new infections among sex workers and 10,000 S57). At a national level, there is no correlation new infections among adults” (D. L. Kerrigan, between GDP and HIV rates in Sub-Saharan Fonner, Stromdahl, & Kennedy, 2013, p. xxxi). Africa, but inequality is associated with higher rates of HIV in Sub-Saharan Africa, Asia, and Meanwhile, stigma remains the undercurrent of Latin America. Rather than focusing on poverty, it the compound risks that FSWs face, creating makes sense to address socioeconomic status and barriers to accessing health and banking services inequality, including gender inequality. These as well as compromising their legal rights. Given issues are directly related to control over income, that sex work is illegal in many countries, FSWs something that many FSWs do not have. A study must live and work in secrecy and have limited to of women participating in a microfinance program no legal recourse when they face rape or on-the- in the Dominican Republic demonstrates that job exploitation. They are often harassed by increased income, in the form of loans, does not police, which interrupts their income and enhance HIV-related bargaining power, but increases their dependence on each transaction, control over income does (Ashburn, Kerrigan, & increasing their likelihood to agree to sex without

a condom (Ngo et al., 2007; Onyeneho, 2009). in recognition of structural factors that affect the Stigma and gender inequality combine to make course of the HIV epidemic. discussions around HIV with significant others taboo and even dangerous. The AIDS Stigma Behavioral approaches attempt to reduce “risky Kenya (ASK) study showed that for the women behaviors and reinforce protective ones, typically involved, economic empowerment was not by addressing knowledge, attitudes, skills, and protective of participants in terms of fear of beliefs” (AIDSTAR-One, 2014). These include domestic violence from husbands as a response to interventions that promote condom use, discussions about AIDS (Gnauck et al., 2013). encourage delayed sexual debut, encourage partner reduction, and foment peer education, among other activities. Biomedical approaches leverage medical technologies against infection, such as APPROACHES TO HIV- antiretroviral medications, pre- and post-exposure PREVENTION prophylaxis, and condom promotion (Rotheram- INTERVENTIONS FOR Borus, Swendeman, & Chovnick, 2009). FEMALE SEX WORKERS Interventions include setting up clinics to provide medical services, infant feeding for mothers with Given the multiple vulnerabilities experienced by HIV, and provision of contraception, among FSWs, HIV-prevention interventions targeted to others. There is consensus in the literature that this population are well-documented, supported biomedical and behavioral approaches are by a growing body of evidence, and guided by mutually-reinforcing, and indeed must be standards of practice endorsed by international connected to ensure adherence and efficacy. health and development organizations. Best practices have shifted from behavioral and A more recent focus on structural interventions biomedical approaches to combination has brought attention to factors beyond the approaches featuring structural intervention. The individual level that mediate susceptibility to HIV, Joint United Nations Programme on HIV/AIDS including social, legal, cultural, and economic (UNAIDS) published a comprehensive guidance factors (Ashburn et al., 2008). Structural note on HIV-prevention interventions for FSWs interventions introduce an extra layer of in 2012, synthesizing these best practices into complexity to HIV prevention and can be difficult three guiding pillars (UNAIDS, 2012). to evaluate due to the complexity of causality and challenges of randomization. However, a growing Combination Intervention body of evidence supports the use of combination interventions that address behavioral, biomedical, Until recently, HIV prevention interventions were and structural issues (UNAIDS, 2010). not well-supported by evidence, with many trials failing to demonstrate decreases in HIV infection (Padian et al., 2011). However, recent studies Structural Interventions with FSWs demonstrate increasing success in prevention, with Structural interventions address factors beyond particularly strong results when biomedical, the level of the individual, including overarching behavioral, and structural interventions are social and economic structures, that contribute to combined. This marks a distinct shift away from HIV transmission. Structural issues are particularly interventions focused exclusively on either salient in addressing the HIV-vulnerability of behavioral or biomedical factors, as well as the rise high-risk groups such as sex workers. Poverty,

gender inequality, stigma, political and legal and TRY have demonstrated reduced intimate structures all contribute to both engagement in partner and gender-based violence and increased sex work and increased likelihood of contracting likelihood to refuse sex without a condom, HIV. Structural interventions can be top-down, respectively (Dworkin & Blankenship, 2009). state-led national programs, or bottom-up mobilization activities to create social and political One of the factors that affects the physical and change (Evans et al., 2010). Among structural economic vulnerability of FSWs is gender factors considered in FSW interventions, poverty inequality (Dworkin & Blankenship, 2009; and gender inequality are typically most Fitzgerald-Husek et al., 2011), which not only emphasized (Dworkin & Blankenship, 2009). limits their control over financial resources and increases their risk for poverty, but places them at Economic Empowerment a disadvantage when bargaining with clients about Economic strengthening, or empowerment, safe sex practices and increases their risks for interventions for FSWs usually provide violence. Gender inequality plays into the microfinance services or vocational training. stigmatization of sex work, reducing sex workers’ Though poverty is a contributing factor to access to legal protection and social and health vulnerability, its relationship with HIV is not services. linear. Inequality, gender norms, and education levels all affect how and at what levels poverty Best Practices impacts vulnerability (Kim et al., 2008). Though Throughout this review, the three pillars of HIV economic empowerment interventions show prevention interventions for FSWs outlined by promise, there is limited evidence demonstrating UNAIDS will be referenced as the standard for whether and how they work. A systematic review such interventions. UNAIDS’ guidance note of income-generating interventions for HIV emphasizes human rights-based approaches, prevention found inconclusive evidence on the noting that the most successful interventions efficacy of both microfinance and vocational occur when “sex workers are able to assert control training interventions (Kennedy, Fonner, O’Reilly, over their working environments and insist on & Sweat, 2013). However, there is evidence that safer sex” (UNAIDS, 2012, p. 4). The three economic strengthening interventions can pillars, with illustrative intervention activities, are improve food security (Xiong, 2012), which is presented on the following page. both a strong motivating factor for entering sex work as well as engaging in risky sexual behavior (Oyefara, 2007; S. Weiser et al., 2007). INTERVENTION EXAMPLES

Though microfinance is often discussed as a This review discusses eight interventions that structural, economic intervention, some authors consider the structural factors required to address question whether it truly affects larger economic the needs of FSWs, with a focus on economic and social structures (Kim et al., 2008). Evans and empowerment. These interventions feature either colleagues distinguish between structural and microfinance and/or vocational training as an individual empowerment (2010), arguing that for approach to meeting the economic needs of an intervention to be truly structural, it must affect FSWs. The literature on economic strengthening factors beyond the individual. Despite limited interventions for FSWs is limited, and evidence on the effect of microfinance on HIV, intervention studies were selected based on the well-documented interventions such as IMAGE quality of information available. In addition to

on multiple levels, resulting in both individual and collective empowerment (Blanchard et al., 2013). Pillar 1: Assure universal access to comprehensive HIV prevention, A common feature of community mobilization treatment, care and support. interventions is the use of peer-education, typically • Removing structural barriers to access through training FSWs to become peer-educators • Information and education in their communities to teach other FSWs about • Preventive services HIV-related topics. Peer education is widely • Linking and integrating services supported as a best practice in HIV intervention • Eliminating violence against (Gay et al., 2012). A systematic review of 30 SWs studies on peer education interventions between

Pillar 2: Build supportive 1990 and 2006 showed that interventions environments, strengthen increased knowledge of HIV, decreased partnerships and expand choices. equipment sharing for drug users, and increased • Addressing stigma condom use (Medley, Kennedy, O'Reilly, & Sweat, • Expanding choices 2009). Other tactics include forming FSW Alternative employment o cooperatives to provide vital services to the o Housing o Education, including community, such as microfinance or running vocational training clinics, and lobbying for political change. o Control of family assets Microfinance mentioned o The Sonagachi Project here The Sonagachi Project is one of the best-known

Pillar 3: Reduce vulnerability and FSW interventions. It has received the honor of address structural issues. being designated a model HIV-prevention project • Gender norms by the World Health Organization and has • Poverty received recognition in the New York Times • Education for all (Basu et al., 2004). Although it wasn’t designed • Address needs of refugees, IDPs, migrants, asylum seekers with an “empowerment” focus, it evolved from a behavioral HIV-prevention intervention to a full- Adapted from UNAIDS, 2012 p. 7 scale social movement. The Sonagachi HIV STD Project (SHIP) began in 1992 in Songachi, a red- light area in Kolkata, India, where baselines interventions focused primarily on economic showed that 80% of sex workers had STIs, only strengthening, the Sonagachi Project, a 1% used condoms, and 1% had HIV (Evans et al., community mobilization project with a secondary 2010). impact on economic strengthening, was also included as an example of best practice. The goal of SHIP was to address vulnerability to HIV among FSWs, among whom India’s HIV Community Mobilization epidemic is concentrated, through clinics in red- Community mobilization is an intervention tactic light areas and peer education on HIV-related that encourages collectivization to bring about topics. The focus of the project shifted to structural change. Interventions using this strategy promoting collective action and empowerment. seek to change social and political structures by As peer educators built relationships with other organizing FSWs to confront structural barriers FSWs, ideas began to spread and FSWs began to

gain the organizing capacity to develop their own proportion of consistent condom users increased collectives. This resulted in a change to the by 25% in the intervention community, where the internalized stigma in the discourse among sex control community experienced a 16% decrease. workers and in the community, encouraging political mobilization and changes in larger Because only two study communities were political norms. Eventually, one of the collectives, assessed, the generalizability of these results is Durbar, took charge of the project. The mobilized limited. Furthermore, information on condom use FSWs lobbied the government to register their was generated using self-report data, which could collective as a state-recognized cooperative, be biased. Finally, some contamination between overcoming opposition from the state as well as control and intervention communities was social stigmatization. The FSWs received “formal possible, further limiting the validity of study recognition as an occupational group by the State results. Nonetheless, the study adds significantly for the first time” (p. 459). In response to to the literature on HIV interventions with FSWs demand, the USHA Multi-purpose Cooperative by testing the Sonagachi model using a control was formed to provide microfinance and banking group. services. Although no control study was A follow-up on this study evaluated the model’s conducted, the impact of the intervention has impact against five common factors of effective, been clear. HIV levels among sex workers in evidence-based HIV/STD prevention programs, Kolkata have been steady at 11%, with consistent including: 1) providing a frame to motivate condom use at 80%. This is a marked difference change; 2) increasing knowledge of risk and from surrounding major cities such as Mumbai, protective factors; 3) building cognitive, affective, where HIV levels among FSWs were at 23.62% in and behavioral skills; 4) reducing environmental 2010 after a peak at 54.29% in 2003 (Evans et al., barriers to change; and 5) building ongoing social 2010, p. 459). support to sustain change over time” (Swendeman There are a number of descriptive and qualitative et al., 2009, p. 1158). Given the difficulty of studies on the Sonagachi Project, but few measuring empowerment as an abstract construct, quantitative studies (Swendeman, Basu, Das, Jana, this study sought to contribute to the evidence & Rotheram-Borus, 2009). One such study base for empowerment-based interventions by replicated the intervention to test the efficacy of measuring empowerment according to the above the Sonagachi model in increasing condom use in proxies. The study used assessment activities and a controlled environment, using community interviews with FSWs to construct a summary organizing and advocacy, peer education, condom outcome index. social marketing, and establishing a health clinic One limitation of the study was that there was (Basu et al., 2004). The study randomly selected greater attrition, with only 80% retention, in the 100 sex workers each from two communities in control community than the intervention India, receiving assessment every 5-6 months over community, where the retention rate was 93%. 15 months. These communities were matched Another issue had to do with the construction of with control communities based on size, the index, which was not balanced across socioeconomic status, and number of sex workers. domains. However, the authors conclude that The results of the study showed that condom use though both the control and intervention increased in the intervention community to 39%, communities received “STD/ HIV prevention compared to 11% in the control community. The and treatment services, the Sonagachi

empowerment intervention improved the majority Evidence on Economic Strengthening for of the HIV/STD linked measured outcome HIV Prevention variables, demonstrating that empowerment Microfinance is a well-documented intervention intervention components can enhance health- for the purposes of women’s empowerment and related prevention outcomes above and beyond poverty-reduction, though it is still not a primary the impacts of clinical services and health response to HIV, and evidence for its effects in education” (p. 1165). the context of HIV prevention for FSWs is more limited (AIDSTAR-One, 2010). Microfinance, The success of the Sonagachi Project underscores particularly microcredit, has received some the role of community mobilization and criticism in recent years based on evidence of empowerment in HIV prevention interventions potential negative consequences for participants. for FSWs. Though Sonagachi was not originally In some cases, women take out microloans merely intended to be an economic strengthening project, as proxies for their husbands, who retain control it later incorporated microfinance through the over the household finances. In other cases, FSW-run USHA cooperative. There is little microcredit, microenterprise, and economic evidence of the specific impact of economic empowerment activities for women can trigger strengthening on the success of Sonagachi, but its domestic violence, divorce, or greater workloads structural approach can serve as a model for other for women tasked with domestic responsibilities HIV-prevention interventions with FSWs. in addition to microenterprise responsibilities. Overall evidence on the efficacy of microfinance Economic Strengthening for FSWs remains mixed (Duvendack et al., 2011). Economic strengthening interventions directed On the other hand, microfinance has yielded some toward FSWs typically incorporate microfinance stunning, positive results. A World Bank study, for and/or vocational training, with the goal of example, attributed 40% of the rural poverty decreasing FSWs’ vulnerability to HIV by reduction experienced in Bangladesh from 1991 to decreasing their financial reliance on sex work. As 1999 to microfinance (Khandker, 2005). One of discussed, economic need is associated with the most well-known examples of the impact of decreased bargaining power and decreases in the microfinance on HIV is the Intervention with likelihood that FSWs will refuse clients who insist Microfinance for AIDS and Gender Equity on sex without a condom. It is also associated (IMAGE) Project, which began in the Limpopo with higher levels of debt, and, in turn, violence. province of South Africa in 2001. IMAGE began Economic strengthening interventions seek to as a pilot with randomized control trial evaluation diversify FSWs; income sources or to enable them and has since been scaled up. Today, it continues to leave sex work altogether for alternative forms to provide HIV education and microfinance to of employment. They can also serve as platforms women. According to a 2007 evaluation, IMAGE for HIV education and conduits for social capital resulted in a 55% decrease in domestic violence as development (Kennedy et al., 2013). Overall, well as enhanced well-being, confidence, and however, the evidence remains mixed on the influence in household decision-making efficacy of income-generating interventions in (AIDSTAR-One, 2010). preventing HIV. Additionally, the IMAGE study compared a

microfinance intervention to a control as well as a

microfinance and health education intervention to

a control (Kennedy et al., 2013). Compared to the skills necessary to leave sex work (McCormick & control, the microfinance intervention did not Munguti, 2003). In addition to business loans, K- yield significant differences in condom use. The VOWRC provided women with counseling and microfinance intervention that incorporated peer support. In a study conducted between 1999 gender and HIV education reported greater and 2000, over 75% of participants reported that likelihood among participants to seek counseling the program had effected a change in their work. and less likelihood to report unprotected sex, but Positive changes were reported in “health status, no difference in frequency of multiple drug use, social interaction and status, and partnerships. economic status” (p. 59). Although a number of women in the program found alternative work, In their systematic review of income-generating about 60% of participant loans were in arrears. interventions and their impact on HIV, Kennedy Overall, though the program delivered on its and colleagues (2013) list two studies that tested intended social outcomes, its model was proven to the combination of microfinance, health be financially unsustainable. education, and business development education. The first, an FSW intervention that introduced K-VOWRC has since been incorporated into condoms, group-based loans, and business state-run HIV-prevention programming and is training and mentorship, resulted in fewer sex now known as HerStory Centre, including the partners among participants and increased following programs: Health Education, condom use with regular partners, but no change Counseling and Paralegal Training; Peer Lead in casual partners (Odek et al., 2009). A second Support System; Girls Vocational Training, which study of the Shaping the Health of Adolescents in was recently funded by the Bill and Melinda Gates Zimbabwe (SHAZ) intervention did not Foundation; and Family and Kinship Support demonstrate changes in condom use (Kennedy et (HerStory Centre, 2012). Although data is not al., 2013). available on the effectiveness of the current programs, it is clear that K-VOWRC has altered Finally, studies of interventions combining its strategy and increased its sustainability as vocational training and health education also show HerStory. mixed results. In Uganda, a youth intervention yielded significant decreases in number of sex Pragati partners, more abstinence and condom use, but Pragati is a large-scale HIV prevention program no change in sexual activity (Rotheram-Borus, targeting FSWs in Bangalore, India. It was Lightfoot, Kasirye, & Desmond, 2012). Similar launched in 2005 with support from the Gates interventions with FSWs, discussed below, did not Foundation, UNDP, Karnataka State AIDS significantly affect condom use (Lee et al., 2010; Prevention Society, and Vrutti, and is Sherman et al., 2010). implemented by Swathi Mahila Sangha (SMS), a sex worker collective, and Swasti, a health Microfinance resource center (Souverein et al., 2013). The program provides a holistic set of services Kenya Voluntary Women Rehabilitation Centre including microfinance services, de-addiction (K-VOWRC) services, condom promotion, STI prevention and The Kenya Voluntary Women Rehabilitation detection, and the provision of safe spaces for Centre (K-VOWRC) was established in 1992 to self-care. The program is based on the theory that provide sex workers with the social and economic the empowerment of FSWs will serve as

protection against health risks and HIV Through Pragati, a cooperative Women’s Bank transmission. (Swathi Jyothi) was set up by SMS. A study evaluating the reach of the program yielded Pragati is based on three intervention strategies: evidence of promising uptake of microfinance services, with more than half of the over 3,000 (1) protect and respond, which involves women that joined as shareholders between 2007 implementing an outreach strategy in and 2010 opening savings accounts, and 2.2 which peer educators and outreach million rupees ($38,500 USD) in savings workers start a dialogue with the FSWs accumulated during this period (Euser et al., about the services and benefits offered 2012). This study also emphasized the demand for through the Pragati programme (this the crisis response component of the program, strategy also involves sensitising primary highlighting the need to address violence against and secondary stakeholders of the sex FSWs. worker industry, e.g. brothel owners, pimps, and the police, to the problems Finally, Pragati has demonstrated success in and needs of FSWs); reaching scale and maintaining cost efficiency. Between 2005 and 2010, the cost of the program (2) improve their quality of life through per sex worker was about $38, in accordance to identifying and addressing long-term the suggested guidelines by the Indian development needs such as providing government’s National AIDS Control support for alcohol de-addiction, Organisation, Ministry of Health and Family providing savings and credit facilities, and Welfare (NACO) (Euser et al., 2012, p. 11). creating options for alternative and diversified livelihoods; and Vocational Training: SiRCHESI’s Hotel Apprenticeship Program (HAP) (3) build the capacities of the FSWs to SiRCHESI (Siem Reap Citizens for Health, address the issues that threaten their lives Educational and Social Issues) is a Cambodian and livelihoods through strengthening NGO that founded the Hotel Apprenticeship group action and developing a strong Program (HAP) in 2006 in order to reduce HIV collective of FSWs. (Souverein et al., risk for beer sellers by helping them find 2013, p. 3) alternative employment (Lee et al., 2010). In Cambodia, beer sellers are women who sell a A recent study examined program exposure, particular brand of beer at bars. They typically condom distribution, and STI rates of program engage in sex work to supplement their limited participants between 2005 and 2008 (Souverein et earnings, and thus are considered indirect sex al., 2013) using data collected by the clinics run by workers. Beer sellers face abuse, and often report Swasti. Statistical analysis revealed significant overuse of alcohol, as clients frequently encourage increases in program exposure, decreases in STI or coerce them into excessive drinking on the job, incidence, and increases in reported condom use placing them at greater vulnerability to sexual at last paid sex. Although this study suffered from predation. SiRCHESI partners with local hotels to a lack of a control group and a relatively short enroll beer sellers in an 8-month hotel follow-up time (1.48 years), it is one of the few apprenticeship program to transition them into studies on similar programs that is longitudinal, more secure and safer employment. The program rather than cross-sectional. provides training on literacy, English, social skills,

health education, and hotel skills. in income earned from sex work from baseline to follow-up” (p. 655). Both groups reported A matched sample of 14 beer sellers were studied reduction in number of sex partners. No using a quantitative survey and semi-structured significant changes in condom use were reported, interviews, finding that although positive changes although the authors attribute this to the effects of were observed in health-related knowledge, an existing condom promotion program in the behavior, self-image and empowerment, condom area. use did not increase (Lee et al., 2010). It was also found that women do not use condoms However, because program participation included consistently with their husbands, despite the remuneration at a higher level than participating in likelihood for infidelity or paying for sexual sex work, increases in income can be interpreted services elsewhere, placing them at continued risk as a result of the project design rather than the for HIV infection. impact of microenterprise. The financial sustainability of the project is also questionable, Combined Microfinance and Vocational given the necessity of setting up a whole new non- and/or Business Training profit to market and distribute the products made by participants. Finally, though the intervention Pi Bags The Pi project seeks to decrease sexual risk group experienced significantly greater sex partner behaviors among street-based FSWs in Chennai, reduction than the control group, this could also India, by providing HIV education and a be attributable to participation in the microenterprise project involving training in microenterprise activity, which the authors tailoring canvas bags for export sale (Sherman et recognize as “time-intensive” and a potential al., 2010). The pilot for the project was initiated in practical barrier to participation in sex work. 2008 by the Johns Hopkins Bloomberg School of Undarga Public Health and Y. R. Gaitonde Centre for The Undarga intervention developed out of AIDS Research and Education (YRG CARE). feedback given by FSWs who had participated in an HIV reduction trial in Mongolia (Laura This pilot study evaluated the feasibility of the Cordisco Tsai, Witte, Aira, Altantsetseg, & Riedel, intervention, involving 100 participants in an 2011). To meet the emerging demands for HIV intervention group, which received education and support for gaining alternative microenterprise training and support as well as employment, Undarga was launched as a pilot to HIV education, compared to a control group that test the feasibility of a matched-savings program only received HIV education. Assessment was for FSWs, where matched funds could be used for undertaken at baseline, with a 6-month follow-up. business development or vocational education. The project provided training to FSWs to tailor Undarga also provided training in financial literacy export-quality canvas bags, which were sold by the and business development as well as mentorship Pi Foundation, a non-profit set up for the purpose and support. Participants were compensated for of the project. Participants were paid to participation in trainings. Like similar projects, it incentivize their participation in the program. was based on the rationale that diversifying FSW The results on the impact of the project were income increases bargaining power and reduces quite positive. For the intervention group, vulnerability to HIV. It sought to address some of economic well-being increased due to a “42% the sustainability issues with microcredit by increase in monthly income and a 41% reduction introducing a savings-led model. It was also

intended to help FSWs transition to alternate development training, and the promotion of employment. This intervention was meant to have savings practices. The goals of the intervention a larger impact on HIV rates in Mongolia as well, were to encourage alternate employment and where 50% of reported infections among females decrease risky behaviors by decreasing number of are found among FSWs. sex partners and increasing condom use.

The feasibility study for Undarga was conducted The results of the second pilot evaluation, which in 2010 and featured a purposive sample of nine included a series of surveys and interviews, found FSWs, who were assessed using written financial that 2/3 of participants had started their own literacy and business development tests. They also businesses by the end of the study, half reported participated in focus group discussions conducted exiting sex work, and the overall weekly mean after trainings and a month after the end of the number of sex partners was reduced. The number pilot. Results from the study showed increases in of casual partners, however, did not decrease. The confidence in financial management, improved evaluation was based on a pre- and post- analysis outlook on finding alternative employment, some without any control, limiting its ability to isolate improvements in financial literacy, and “initial secular effects on outcomes. Like other studies, it transition from sex work to alternative income also measured condom use and sexual behaviors generation for five out of nine participants” (p. according to self-report data, which could suffer 26). Results from the matched-savings portion of from social desirability bias. the pilot, however, were inconclusive, as this portion of the intervention was only implemented Combined Economic Strengthening for five weeks. Though the pilot was very small and Community Mobilization and the matched-savings component was not long Most of the literature reviewed featured enough to adequately assess its impact, it is one of interventions that either prioritized emphasis on the few studies to assess the combination of economic strengthening or on community vocational training and savings-led microfinance mobilization. A good example of this is Project for FSWs. Avahan in India, a large-scale HIV prevention Strengthening STD/HIV Control Project in program for FSWs that has received acclaim and Kenya (SHCP) has been cited as a model program (Rau, 2011). The Strengthening STD/HIV Control Project in Influenced by the Sonagachi Project, Avahan is Kenya (SHCP) pilot, implemented by the known for its combination approach to HIV University of Nairobi and University of Manitoba, prevention, including facilitating structural change tested the effects of adding microfinance to HIV through community mobilization. However, interventions for FSWs in Kenya (Odek et al., though microfinance appears as an intervention 2009). Originally an HIV-prevention peer strategy in some Avahan programs (Blanchard et education program, SHCP responded to al., 2013), the economic aspect of the program has participant demands for microfinance services by received little emphasis and little study. The partnering with a local microfinance institution. Its Karnataka Health Promotion Trust (KHPT), also initial pilot of this approach included 102 in India, provides an example of a program that participants from 1998-2002, while a scaled up emphasizes both mobilization and economic pilot with 227 FSWs lasted from 2003 to 2005. strengthening. Program components included microcredit for small business development, business

Karnataka Health Promotion Trust (KHPT) safer sex. It also emerged that group members had The Karnataka Health Promotion Trust (KHPT) much greater exposure to program interventions was established by the Karnataka State AIDS than non-members (Pillai et al., 2012). Differences Prevention Society and the University of in outcomes across the three districts highlighted Manitoba in 2003 in Karnataka province, India to differences in organization among FSWs. implement and evaluate HIV prevention programs Increased condom use was also demonstrated (Pillai et al., 2012). Through its research project, among members compared to non-members in STRIVE, KHPT recently evaluated the effects of Bellary and Bangalore Urban, but not Shimoga. programming that combined community Where microfinance services were provided mobilization and savings-led microfinance through groups where FSWs had membership or interventions. the Swathi Jyothi cooperative, such as the FSWs in Bellary and Bangalore Urban, they practiced safer The impact of these programs on safe sex sex. In Shimoga, on the other hand, FSWs practices was tested across the Shimoga, Bellary, accessed credit from other microfinance groups and Bangalore Urban districts of Karnataka, and were comparatively less successful (p. xii). starting with baselines conducted between 7 and These differences underscore the role of 19 months after the start of the program, with microfinance in encouraging safe sex practices. follow-up surveys conducted 28-37 months later. Other lessons learned include the need to include Integrated Biological and Behavioral Assessments financial discipline in economic strengthening (IBBA) and focus groups were conducted with interventions, as well as the role of economic members and non-members of a group or CBO. strengthening initiatives in reducing stigma. This information was used to analyze primary outcomes of collectivization and condom use in The study design may have suffered from social the previous year and secondary outcomes of desirability bias, as sex workers were exposed to HIV/STI prevalence, condom use with clients repeated rounds of behavioral questions. As a and partners, and HIV prevention. longitudinal design across several districts, however, it brings the benefit of comparing across In focus group discussions, FSWs reported that various collectivization strategies. participation in microfinance helped them practice

Summary Table of Interventions

Intervention Intervention Outcomes Strengths Weaknesses Sources Name Type Sonagachi HIV Community Reduction in Considered a No studies with (Basu et al., STD Intervention Mobilization STIs model control available, so 2004; Evans Project (SHIP) empowerment- difficult to isolate et al., 2010; (1992 - ), Kolkata, Clinical Services Political/legal based project causal mechanisms Ghose, India structural Swendeman, Peer Education change Facilitated Organic evolution George, & sustained did not elaborate Chowdhury, Microfinance Increased structural change causal framework 2008; Jana, condom use Basu, Rotheram- Borus, &

Intervention Intervention Outcomes Strengths Weaknesses Sources Name Type Newman, 2004; Swendeman et al., 2009)

Pragati, Swathi Microfinance Reduction in Longitudinal No control group (Euser et al., Mahila Sangha STIs evaluation 2012; (SMS) and Crisis response results Relatively short Souverein et Swasti, (2005 - ), team follow-up time al., 2013) Bangalore, India Combination De-addiction Increase in intervention services condom use approach

Condom Large-scale promotion program with significant reach STI prevention and detection Cost-effective

Rest places/safe spaces

Peer Education

STRIVE, Community Increased Reaches large Social desirability (Pillai et al., Karnataka Health Mobilization condom use number of rural bias in evaluation 2012) Promotion Trust (in Ballary and FSW results (KHPT), (2011- Savings Bangalore 2017), India Urban) Peer Education Decreased Has contributed Clinical Services stigma to Avahan (Bellary) project implementation Decreased and inclusion of violence sex workers in (some sites) national guidelines on Increased HIV prevention knowledge about condoms and STIs Designed in Enhanced consultation with self-perception FSWs

Improved clinical services

Intervention Intervention Outcomes Strengths Weaknesses Sources Name Type Kenya Voluntary Microfinance Alternative Social outcomes 2003 study (McCormick & Women employment reached demonstrated lack Munguti, Rehabilitation Counseling successfully of financial 2003) Centre (K- Increased self- sustainability VOWRC) / Peer support esteem Evolved to HerStory, (1992 - incorporate more No current ), Nairobi, Kenya High loan integrative evaluation data losses programs as available on HerStory HerStory Decreased drug and alcohol use

Hotel Vocational Alternative Reached indirect Did not address (Lee et al., Apprenticeship training employment sex workers, HIV prevention care 2010) Program (HAP), who might not and support (2006 - ), Siem HIV education* Increased identify as sex Reep, Cambodia health workers Did not address knowledge gender norms Addressed No increase in alternative Not condom use employment “transformatory” needs working (Evans) with local partners

Pi Pilot, (2008 - ), Vocational Alternative Randomized Questionable (Sherman et Chennai, India training employment control trial sustainability – al., 2010) evaluation required creation of Microenterprise No change in supplementary condom use Addressed NGO HIV education economic Decreased empowerment Not number of sex issues “transformatory” – partner no community mobilization

Undarga, (2010), Matched savings Alternative Addressed No HIV education (Laura Ulaanbaatar, employment economic component Cordisco Tsai Mongolia Financial literacy empowerment, et al., 2011 Increased including savings No community financial mobilization literacy Business training No integration with Mentorship Greater other services confidence

Strengthening Microcredit Alternative Community No control study (Odek et al., STD/HIV Control

Intervention Intervention Outcomes Strengths Weaknesses Sources Name Type Project (SHCP) Business training employment mobilization Peer education did 2009) pilots, (1990 - not 2006), Kenya Mentorship Fewer sex Transformatory – mention*/address partners demand for MF violence or stigma Savings services from Increased participants condom use Integrated services

Analysis Pragati also illustrates a comprehensive, Impact studies of economic strengthening combination approach to HIV prevention, with demonstrated cost-effectiveness. This project interventions with FSWs provide strong evidence in favor of integrated, structural interventions as addressed health needs beyond STI reduction by best practice. The most successful of the providing de-addiction services. Finally, SHCP interventions discussed here adhere to the took a mobilization approach, with its original peer education program expanding to provide UNAIDS guidelines for interventions with FSWs. Most notably, these interventions tend to be economic strengthening based on participant driven by the FSWs themselves: either directly demand. Like Sonagachi, these two interventions through cooperatives, or indirectly through were strongly led by FSWs themselves. consultation. These interventions better resemble The interventions that yielded less successful Evans and colleagues’ conception of impact results were externally determined, rather “transformatory” interventions, where goals are than internally directed by FSWs. K-VOWRC set by beneficiaries as active agents, rather than demonstrated strong results for its social “instrumental,” where goals are set externally objectives, but suffered from unsustainable micro- (Evans et al., 2010). lending practices, using a strategy that ultimately The most celebrated and successful of the proved untenable (Undarga and KHPT addressed interventions discussed in this review is the this problem by choosing savings-led approaches instead). HAP and Pi also demonstrated Sonagachi Project. Sonagachi addressed all three pillars of the UNAIDS guidelines to create questionable financial sustainability, both sustained, structural change. This was dependent on ongoing donations to provide accomplished through community mobilization, services. HAP did not meet its objective of increasing condom use, and Pi’s reported increases which in turn helped change community perceptions of sex work and reduce stigma, in FSW income and transition away from sex culminating in political-legal change that helped work can be attributed more to the provision of a FSWs organize and claim their rights. It is guaranteed work opportunity than capacity- building. important to note here that organizing was the first step of generating change. Economic Though there is little information on the cost- strengthening activities followed, based on effectiveness of most of the interventions in this demand from the FSW community. review, the evidence for the cost-effectiveness of FSW interventions is strong, particularly for areas

with concentrated HIV epidemics, with an estimated average cost of $102 to $184 per Research Gaps in the Literature participant (D. Kerrigan et al., 2012). Sonagachi is cited as a cost-effective intervention, as is Pragati. • Many studies fail to gather data Interventions with unsustainable microfinance on anal sex, a risky behavior strategies, such a K-VOWRC, or social business strategies, such as Pi, have not demonstrated cost- • Most studies rely on self-report data on condom use and other effectiveness. behavioral patterns

Overall, this review confirms previous analysis • Most studies gather suggesting that there is more evidence in support information on male condom of interventions featuring microfinance services use, neglecting female condom combined with health services than for vocational use training (Kennedy et al., 2013). It also supports From Scorgie et al, 2012 “transformatory” intervention for lasting change. Interventions should always feature HIV education and prevention services. Though their daily lives. For FSW populations, this must evidence on vocational training for HIV include the delicate but critical topic of violence. prevention among FSWs is mixed, the evidence base includes studies of various intervention The literature on the lives of FSWs is not extensive. strategies, many of which do not include a It is especially thin on structural analysis of comprehensive approach. It is possible that vulnerability and economic behaviors and practices. vocational training interventions will be more Analyzing both of these dimensions can yield key successful when combined with other types of insights for intervention design. Pre-pilot research support. should examine the social networks of FSWs, the occupational structures they inhabit, and distinguish between different types of FSWs and their respective circumstances. This includes indirect sex workers, RECOMMENDATIONS FOR who may be hidden or may not identify as sex FUTURE RESEARCH AND workers. Understanding the social capital of FSWs PILOTING can yield insights on the potential for collectivization and peer education.

Preliminary Research for Pilot Understanding economic structures will help Interventions illuminate the economic needs and opportunities FSWs are a heterogeneous group with varied available for FSWs. This is connected to FSW needs in environments that vary politically, capacity for employment, including their skills and socially, and economically. Understanding a education levels. It is important to investigate population’s unique circumstances in a given power dynamics, as many FSWs are caught in setting is critical to designing an intervention to economically exploitative positions due to debt- adequately meet their needs. As such, it is bondage, the illegal status of their work, recommended that research undertaken prior to backgrounds of poverty, and lack of better piloting feature careful attention to the structures opportunities. Other inter-related dimensions for of vulnerability that project participants face in analysis include stigma, violence, and health-

related knowledge. not always generalizable. This is particularly true because, in India, sex work is not illegal, which Though surveys can be used to collect some of greatly expands opportunities for community this information – particularly that pertaining to mobilization and the pursuit of occupational rights health knowledge and consumption – qualitative (Evans et al., 2010). In this review, it is notable research, including interviews and focus groups, that community mobilization was featured only in will be important to obtaining a nuanced picture Indian interventions. In other locations, it is of the circumstances FSWs face. The most unclear how possible it might be to pursue this sustainable structural interventions are FSW-led, strategy. Therefore, research is needed on the so intervention design should also be attuned to potential for mobilization in areas more hostile to the demands of internal conceptions of sex worker rights. vulnerability of the target population. It should be noted that sex work remains illegal in many There is also a trend in many studies to either countries, making FSWs a very sensitive focus on qualitative, “empowerment” variables or population. Any research should be handled with quantitative variables to measure HIV risk extreme care and attention to maintaining the behaviors, such as condom use or number of sex anonymity of those involved. partners. The effects of economic strengthening interventions are often mediated through such Research Gaps and “empowerment” variables, so it is important to measure both. Recommendations for Piloting There is a strong need for further research on Though there is a great deal of research on the economic strengthening interventions for FSWs. role of microfinance on women’s empowerment, Overall, there is still a limited body of research on there is very little research done on savings-led the effect of economic insecurity on HIV approaches for sex workers. Similarly, vulnerability among FSWs (Reed et al., 2010). inconclusive data on the impact of vocational Furthermore, the specific mechanisms by which training demands further investigation, particularly economic strengthening may affect vulnerability with regard to its effects when combined with remain unclear: whether increased income, social microfinance. capital development, or other factors affect outcomes demands further investigation (Dworkin In addition to economic strengthening, any pilot & Blankenship, 2009). Though discussed as a risk initiative must include HIV education, preferably factor for HIV transmission, food insecurity is in coordination with clinical service provision. It is rarely addressed in FSW intervention. Research on also recommended to conduct the study with a how economic strengthening affect HIV risk control arm and a treatment arm to isolate secular among FSWs through the channel of food security variables. Furthermore, most of the research is is an opportunity for further exploration. based on cross-sectional data, with relatively few longitudinal studies. Longitudinal analysis would A great deal of attention has been given to FSW provide important insights on sustainability, interventions in India, which has the largest particularly for a combination savings-led number of HIV cases of any country and where microfinance/vocational training intervention like the HIV epidemic is most concentrated among Undarga, which yielded inconclusive results due to sex workers. Though research on Indian the timeframe of the evaluation (Laura Cordisco interventions has yielded important insights, it is Tsai et al., 2011). Finally, a recurring problem in

the literature is the effect of social desirability bias technical_briefs/microfinance_hiv_and_ on intervention evaluation results. Given that womens_empowerment information on variables like number of sex AIDSTAR-One. (2014). Behavioral Interventions Retrieved January 28, 2014, from partners and consistency of condom use depends http://www.aidstar- on self-report data, there is a risk of misreporting. one.com/focus_areas/prevention/pkb/b This can be ameliorated by separating pilot ehavioral_interventions implementers from evaluators. Evaluations should Ashburn, K., Kerrigan, D., & Sweat, M. (2008). also account for self-selection bias (Dworkin & Micro-credit, Women’s Groups, Control of Own Money: HIV-Related Negotiation Blankenship, 2009). Among Partnered Dominican Women. AIDS and Behavior, 12(3), 396-403. Basu, I., Jana, S., Rotheram-Borus, M. J., Swendeman, D., Lee, S.-J., Newman, P., CONCLUSION & Weiss, R. (2004). HIV Prevention Among Sex Workers in India. Journal of The evidence is clear that targeting HIV Acquired Immune Deficiency Syndromes, 36(3), prevention interventions to sex workers is a cost- 845-852. effective, high-impact approach to combatting the Blanchard, A. K., Mohan, H. L., Shahmanesh, M., global HIV epidemic. However, many Prakash, R., Isac, S., Ramesh, B. M., . . . opportunities for further research and innovation Blanchard, J. F. (2013). Community mobilization, empowerment and HIV remain, particularly with regard to economic prevention among female sex workers in strengthening. More research is needed to south India. BMC Public Health, 13, 234. investigate the causal mechanisms of effective doi: 10.1186/1471-2458-13-234 economic strengthening interventions, including Duvendack, M., Palmer-Jones, R., Copestake, J.G., their effects on food security. Additionally, though Hooper, L., Loke, Y., & Rao, N. (2011). What is the evidence of the impact of the research is clear that participant-led microfinance on the well-being of poor interventions have proven sustainable and people? London: EPPI-Centre, Social effective in confronting various structural factors Science Research Unit, Institute of of vulnerability to HIV, it is not clear how Education, University of London. transferrable community mobilization approaches Dworkin, S. L., & Blankenship, K. (2009). Microfinance and HIV/AIDS Prevention: to structural change are to environments where Assessing its Promise and Limitations. sex work is illegal. Finally, there is more research AIDS and Behavior, 13(3), 462-469. doi: needed on savings-led microfinance and combined 10.1007/s10461-009-9532-3Euser, S. M., microfinance/vocational training interventions. Kumar, S., Den Boer, J. W., Souverein, Addressing gaps through further research and D., Rama Narayana Gowda, P., Shekhar Gowda, C., . . . Jenniskens, F. (2012). piloting will provide a major contribution to the Pragati: an empowerment programme for state of the practice of HIV prevention for FSWs. female sex workers in Bangalore, India. Global health action, 5, 1-11. doi: 10.3402/gha.v5i0.19279 Evans, C., Jana, S., & Lambert, H. (2010). What SOURCES makes a structural intervention? Reducing vulnerability to HIV in community AIDSTAR-One. (2010). Microfinance, HIV, and settings, with particular reference to sex Women's Empowerment Retrieved work. Glob Public Health, 5(5), 449-461. January 7, 2014, from doi: 10.1080/17441690902942472 http://www.aidstar- Fitzgerald-Husek, A., Martiniuk, A. L., Hinchcliff, one.com/focus_areas/gender/resources/ R., Aochamus, C. E., & Lee, R. B. (2011).

"I do what I have to do to survive": an Johnston, L. G., Sabin, K., Hien, M. T., & Huong, investigation into the perceptions, P. T. (2006). Assessment of Respondent experiences and economic considerations Driven Sampling for Recruiting Female of women engaged in sex work in Sex Workers in Two Vietnamese Cities: Northern Namibia. BMC Womens Health, Reaching the Unseen Sex Worker. Journal 11, 35. doi: 10.1186/1472-6874-11-35 of Urban Health, 83(1 Supplement), 16-28. Gay, J., Croce-Galis, M., & Hardee, K. (2012). K, A. K. (2004). Economic aspects of sex work: Female Sex Workers. What Works for can economic empowerment dissuade Women and Girls: Evidence for HIV/AIDS women from entering into and continuing Interventions. 2nd Edition. Retrieved sex work? [Article]. Journal of Family January 15, 2014, from Welfare, 50, 79-82. www.whatworksforwomen.org Kennedy, C. E., Fonner, V. A., O’Reilly, K. R., & Ghose, T., Swendeman, D., George, S., & Sweat, M. D. (2013). A systematic review Chowdhury, D. (2008). Mobilizing of income generation interventions, collective identity to reduce HIV risk including microfinance and vocational among sex workers in Sonagachi, India: skills training, for HIV prevention. AIDS the boundaries, consciousness, Care, 2013. doi: negotiation framework. [Article]. Social 10.1080/09540121.2013.845287 Science & Medicine, 67(2), 311-320. Kerrigan, D., Wirtz, A., Baral, S., Decker, M., Gnauck, K., Ruiz, J., Kellett, N., Sussman, A., Murray, L., Poteat, T., . . . Beyrer, C. Sullivan, M. A., Montoya, M., . . . (2012). The Global HIV Epidemics Mwanthi, M. A. (2013). Economic among Sex Workers. Washington, DC: empowerment and AIDS-related stigma World Bank. in rural Kenya: a double-edged sword? Kerrigan, D. L., Fonner, V. A., Stromdahl, S., & Culture, Health & Sexuality: An International Kennedy, C. E. (2013). Community Journal for Research, Intervention and Care, Empowerment Among Female Sex 15(7), 851-865. Workers is an Effective HIV Prevention Gupta, J., Reed, E., Kershawa, T., & Blankenship, Intervention: A Systematic Review of the K. M. (2011). History of sex trafficking, Peer-Reviewed Evidence from Low- and recent experiences of violence, and HIV Middle-Income Countries. AIDS and vulnerability among female sex workers in Behavior, 17(6), 1926-1940. doi: coastal Andhra Pradesh, India. 10.1007/s10461-013-0458-4 International Journal of Gynaecology and Khandker, S. R. (2005). Microfinance and Poverty: Obstetrics, 114(2), 101-105. doi: Evidence Using Panel Data from 10.1016/j.ijgo.2011.03.005 Bangladesh. The World Bank Economic Hargreaves, J., Bonell, C., Morison, L., Kim, J., Review, 19(2), 263-286. doi: Phetla, G., Porter, J., . . . Pronyk, P. 10.1093/wber/lhi008 (2007). Explaining continued high HIV Kim, J., Pronyk, P., Barnett, T., & Watts, C. prevalence in South Africa: (2008). Exploring the role of economic socioeconomic factors, HIV incidence empowerment in HIV prevention. AIDS, and sexual behaviour change among a 22(Suppl 4), S57-71. rural cohort, 2001-2004. AIDS, 21(Suppl Lee, H., Pollock, G., Lubek, I., Niemi, S., O'Brien, 7), S39-48. K., Green, M., . . . Idema, R. (2010). HerStory Centre. (2012). Programs Retrieved Creating New Career Pathways to Reduce January 22, 2014, from Poverty, Illiteracy and Health Risks, while http://www.herstorycentre.org/programs Transforming and Empowering Jana, S., Basu, I., Rotheram-Borus, M. J., & Cambodian Women’s Lives. Journal of Newman, P. A. (2004). The Sonagachi Health Psychology, 15(7), 982-992. doi: Project: A Sustainable Community 10.1177/1359105310371703 Intervention Program. AIDS Education Mbonye, M., Nalukenge, W., Nakamanya, S., and Prevention, 16(5), 405-414. Nalusiba, B., King, R., Vandepitte, J., & Seeley, J. (2012). Gender inequity in the

lives of women involved in sex work in 378(9787), 269-278. doi: 10.1016/S0140- Kampala, Uganda. Journal of the 6736(11)60877-5 International AIDS Society, 15 Suppl Pillai, P., Bhattacharjee, P., Ramesh, B. M., & Isac, 1(3(Suppl+1)), 1-9. doi: S. (2012). Impact of two vulnerability 10.7448/IAS.15.3.17365 reduction strategies - collectivisation and McCormick, D., & Munguti, K. (2003). participation in savings activities - on Microfinance and behaviour change HIV risk reduction among female sex among Nairobi's commercial sex workers. workers: KHTP. Small Enterprise Development, 14(2), 56-65. Rau, B. (2011). The Avahan-India AIDS Initiative Medley, A., Kennedy, C., O'Reilly, K., & Sweat, AIDSTAR-One Case Study Series. M. (2009). Effectiveness of Peer Reed, E., Gupta, J., Biradavolu, M., Devireddy, V., Education Interventions for HIV & Blankenship, K. M. (2010). The Prevention in Developing Countries: A Context of Economic Insecurity and Its Systematic Review and Meta-Analysis. Relation to Violence and Risk Factors for AIDS Education and Prevention, 21(3), 181- HIV Among Female Sex Workers in 206. Andhra Pradesh, India. Public Health Nagot, N., Ouangré, A., Ouedraogo, A., Cartoux, Reports, 125(Suppl. 4), 81-89. M., Huygens, P., Defer, M., . . . Van de Rotheram-Borus, M. J., Lightfoot, M., Kasirye, R., Perre, P. (2002). Spectrum of commercial & Desmond, K. (2012). Vocational sex activity in Burkina Faso: classification Training with HIV Prevention for model and risk of exposure to HIV. Ugandan Youth. AIDS and Behavior, 16(5), Journal of Acquired Immune Deficiency 1133-1137. Syndromes, 15(29), 517-521. Rotheram-Borus, M. J., Swendeman, D., & Ngo, A. D., McCurdy, S. A., Ross, M. W., Chovnick, G. (2009). The Past, Present, Markham, C., Ratliff, E. A., & Pham, H. and Future of HIV Prevention: T. (2007). The lives of female sex workers Integrating Behavioral, Biomedical, and in Vietnam: Findings from a qualitative Structural Intervention Strategies for the study. Cult Health Sex, 9(6), 555-570. doi: Next Generation of HIV Prevention. 10.1080/13691050701380018 Annual Review of Clinical Psychology, 2009(5), Odek, W. O., Busza, J., Morris, C. N., Cleland, J., 143-167. doi: Ngugi, E. N., & Ferguson, A. G. (2009). 10.1146/annurev.clinpsy.032408.153530. Effects of micro-enterprise services on Scorgie, F., Chersich, M. F., Ntaganira, I., HIV risk behaviour among female sex Gerbase, A., Lule, F., & Lo, Y.-R. (2012). workers in Kenya's urban slums. AIDS Socio-Demographic Characteristics and Behav, 13(3), 449-461. doi: Behavioral Risk Factors of Female Sex 10.1007/s10461-008-9485-y Workers in Sub-Saharan Africa: A Onyeneho, N. G. (2009). HIV/AIDS risk factors Systematic Review. AIDS and Behavior, 16, and economic empowerment needs of 920-933. female sex workers in Enugu Urban, Sherman, S. G., Srikrishnan, A. K., Rivett, K. A., Nigeria. Tanzania journal of health research, Liu, S.-H., Solomon, S., & Celentano, D. 11(3), 126. D. (2010). Acceptability of a Oyefara, J. L. (2007). Food insecurity, HIV/AIDS microenterprise intervention among pandemic and sexual behaviour of female female sex workers in Chennai, India. commercial sex workers in Lagos AIDS and behavior, 14(3), 649-657. doi: metropolis, Nigeria. SAHARA-J: Journal 10.1007/s10461-010-9686-z of Social Aspects of HIV/AIDS:, 4(2), 626- Souverein, D., Euser, S. M., Ramaiah, R., Ram, P., 635. Gowda, N., Gowda, C. S., . . . Boer, J. W. Padian, N. S., McCoy, S. I., Karim, S. A., Hasen, D. (2013). Reduction in STIs in an N., Kim, J., Bartos, M., . . . Cohen, M. S. empowerment intervention programme (2011). HIV prevention transformed: the for female sex workers in Bangalore, new prevention research agenda. Lancet, India: the Pragati programme. Global

Health Action, 6(22943). doi: UNAIDS. (2012). UNAIDS Guidance Note on 10.3402/gha.v6i0.22943 HIV and Sex Work. Geneva, Switzerland: Swendeman, D., Basu, I., Das, S., Jana, S., & UNAIDS. Rotheram-Borus, M. J. (2009). Vuylsteke, B., & Jana, S. (2001). Reducing HIV Empowering sex workers in India to Risk in Sex Workers, Their Clients and reduce vulnerability to HIV and sexually Partners. In P. R. Lamptey & H. D. Gayle transmitted diseases. Social Science & (Eds.), HIV/AIDS Prevention Care Medicine, 69(8), 1157-1166. doi: Resource-Constrained Settings. Arlington, VA: 10.1016/j.socscimed.2009.07.035 Family Health International. Tsai, L. C., Witte, S. S., Aira, T., Altantsetseg, B., Weiser, S., Leiter, K., Bangsberg, D., Butler, L., & Riedel, M. (2011). Piloting a Savings- Percy-de Korte, F., & et al. (2007). Food Led Microfinance Intervention with insufficiency is associated with HIV-risk Women Engaging in Sex Work in sexual behaviour among women in Mongolia: Further Innovation for HIV Botswana and Swaziland. PLoS Med, 4(1), Risk Reduction. The Open Women's Health e260. doi: 10.1371/journal.pmed.0040260 Journal, 5, 26-32. Weiser, S. D., Young, S. L., Cohen, C. R., Kushel, Tsai, L. C., Witte, S. S., Aira, T., Riedel, M., M. B., Tsai, A. C., Tien, P. C., . . . Hwang, H. G., & Ssewamala, F. (2013). Bangsberg, D. R. (2011). Conceptual "There is no other option; we have to framework for understanding the feed our families...who else would do it?": bidirectional links between food The financial lives of women engaging in insecurity and HIV/AIDS. American sex work in Ulaanbaatar, Mongolia. Global Journal of Clinical Nutrition, 94(suppl), journal of health science, 5(5), 41-50. doi: 1729S-1739S. 10.5539/gjhs.v5n5p41 Xiong, K. (2012). Review of the Evidence: UNAIDS. (2010). Combination HIV Prevention: Linkages between Livelihood, Food Security, Tailoring and Coordinating Biomedical, Economic Strengthening, and HIV-Related Behavioural and Structural Strategies to Outcomes: MEASURE Evaluation. Reduce New HIV Infections UNAIDS Discussion Paper. Geneva: UNAIDS.