AIDSTAR-One | CASE STUDY SERIES

March 2011

The Avahan-India AIDS Initiative

Promising Approaches to Combination HIV Prevention Programming in Concentrated Epidemics
Prashant Panjiar for the Bill & Melinda Gates Foundation

Drop-in center for groups at highest risk in Mysore.

n an old schoolhouse on the outskirts of Chennai, the scene is hectic at midday. The rehabbed building is now home to a local HIV prevention program, with a busy sexually transmitted infection clinic, drop-in counseling facilities, and a safe house for victims of violence. At one large table sits a group of peer educators, all womenmost of them poor and illiterate reviewing data they have gathered in the neighborhoods where they conduct HIV prevention activities for other sex workers. Dressed in bright saris of every color, their self-confidence and pride in their achievements could not be more evident as they discuss their work. These women, like thousands of other peer educators who work with the Avahan-India AIDS Initiative, play a critical role on the front lines of Indias epidemic. Avahan means a call to action in Sanskrit. It is a fitting name for one of the largest and most promising HIV prevention programs in the world. Launched in 2003 with funding from the Bill & Melinda Gates Foundation, this major HIV prevention program stretches over six of the Indian states most affected by HIV, as well as key trucking routes. Now in a second phase, Avahan1 works in partnership with the Indian government, which will take over most of the programs activities by 2014. Over its short life, Avahan has become a global model for achieving multiple goals that many prevention programs find challenging. First, it has successfully built a comprehensive prevention program that combines the most effective responses to the multiple and complex needs of mostat-risk populations (MARPs). Second, its activist peer educatorswhose responsibilities include ongoing data collection and analysis and some

By Bill Rau

AIDSTAR-One
John Snow, Inc. 1616 North Ft. Myer Drive, 11th Floor Arlington, VA 22209 USA Tel.: +1 703-528-7474 Fax: +1 703-528-7480 www.aidstar-one.com

Throughout this case study, Avahan refers to the program as a whole, including its numerous partners.

This publication was produced by the AIDS Support and Technical Assistance Resources (AIDSTAR-One) Project, Sector 1, Task Order 1. USAID Contract # GHH-I-00-07-00059-00, funded January 31, 2008.
Disclaimer: The authors views expressed in this publication do not necessarily reflect the views of the United States Agency for International Development or the United States Government.

AIDSTAR-One | CASE STUDY SERIES

aspects of program planning and management experience true ownership of and commitment to the program. Third, it successfully addresses difficult structural issues that impede programs worldwide. Finally, Avahan has achieved nearly unprecedented levels of scale-up that have made a real impact on the epidemic in the regions where it works.

Avahan has achieved nearly unprecedented levels of scale-up that have made a real impact on the epidemic in the regions where it works.
monthly prevention services to more than 220,000 female sex workers, 80,000 high-risk MSM and transgendered people, 18,000 IDUs, and 5 million men working in the trucking industry. In addition to Indias National AIDS Control Organization (NACO) and the state government AIDS Control Societies, Avahan partners with nine lead implementing grantees, 134 local nongovernmental organizations (NGOs), and hundreds of formal and informal community-based groups (CBGs), as well as 5,700 peer educators and outreach workers. Avahans current interventions reflect the key elements of combination HIV prevention programming for MARPs: Peer-led outreach to promote behavior change Clinical services to treat STIs Condom social marketing and distribution of free condoms Distribution of clean needles and syringes Support for community mobilization Advocacy to reduce structural barriers to safer sexual practices. These elements combine standard components of a minimum package of services for MARPs with community mobilization and efforts to address structural barriers to risk reduction. To tie these components together, Avahan has strong management and capacity-building structures at many levels throughout its operations. For example, Avahans peer educators not only carry out extensive outreach but

Combination HIV Prevention

At first, Avahan offered a traditional mix of services: HIV behavior change messages delivered primarily through outreach workers and peer educators, condom promotion and distribution, and sexually transmitted infection (STI) treatment. After recruiting local partners and peer educators, it added structural interventions designed to reduce police harassment of and violence against high-risk groups and to expand national HIV prevention policies and community engagement efforts. Over time, Avahan evolved into a combination HIV prevention program targeted to MARPs, a model that includes a mix of risk-reduction strategies based on the latest epidemiological and programmatic evidence. Such programs target different populations with simultaneous behavioral, biomedical, and structural interventions designed to address local realities. Avahan focuses on the regions MARPs: female sex workers, men who have sex with men (MSM), transgender people, injecting drug users (IDUs), and truckers and their helpers. Avahans overall goal is to work with existing government interventions in six states (Andhra Pradesh, Karnataka, Maharashtra, Tamil Nadu, Nagaland, and Manipur), where prevalence of HIV is highest among MARPs, to reach 80 percent of MARPs with prevention services. These states accounted for 83 percent of HIV infections in 2002, and represent a geographic area and population size similar to the United States. Avahan reports providing
2 AIDSTAR-One | March 2011

AIDSTAR-One | CASE STUDY SERIES

also plan their activities. That planning feeds into a multilevel management structure that gives the program adaptability and strong accountability. Structural interventionswhich address social, economic, and political issues that affect health at the individual, community, and societal levelsare a key element of Avahan. To address structural problems, Avahan advocates for policies, regulations, litigation, and enforcement of existing laws to create an enabling legal and social environment for prevention. The organization also supports local empowerment and antiviolence initiatives, as well as access to public health and food security programs. (APAC) project, implemented by the Voluntary Health Services in Chennai, funded by USAID in collaboration with the Government of India. Avahans planners focused the bulk of the programs resources on targeted interventions in the six states where the epidemiological data indicated the prevalence rates were highest, saturating them with far more intensive coverage of MARPs. These interventions emphasized behavior change among high-risk groups and sought to provide adequate services for the sexual health of those groups.

The Origins of Avahan

When Avahan began, NACO was already operating a prevention program of national scope under its National AIDS Control Program (NACP-I and -II). Initial funding for this effort came from the national government, the World Bank, and several bilateral donors to support both state- and national-level activities. Later, State AIDS Control Societies (SACS) took charge of some of the decisions related to selecting, funding, and monitoring local NGOs, which implemented targeted interventions to prevent HIV. As rates of HIV infection in India increased in the late 1990s, many noted the potential for a much larger epidemic. Although epidemiologic estimates from that time put the number of HIV-infected individuals in India at 4 million, fears grew that prevalence could reach 5 percent by 2005some 37 million people. The national program suffered from uneven implementation, inadequate coverage of MARPs, and insufficient intensity of prevention efforts. However, there were also encouraging developments. By the time Avahan was being designed, initial reports from Tamil Nadu indicated that prevalence in that state was declining in part as a result of the AIDS Prevention and Control

An Effective Management Model

Avahan reflects a data-driven business approach to confronting the epidemic. Its senior managers come from a business background and rely on data for planning and monitoring of program outputs. Data are used to target interventions, make adjustments in the program as needed, and measure progress toward program objectives. The managers recognize that the search for perfect data never ends and thus make decisions based on the best available data rather than wait for the next sample or a more refinedanalysis. The program also followed business models to create a system of decentralized planning and management that gives substantial responsibilities to local organizations. Through its Common Minimum Program (CMP), Avahan established a shared vision for its highly decentralized operations that set basic standards for its NGO partners to ensure highquality services, while providing necessary guidance, training, and monitoring. The centrally defined framework and standards are thus decentralized to the implementing organizations. A third aspect of Avahans business approach is saturating target audiences with adequate staff and services, a key feature found in effective advertising.
3

The Avahan-India AIDS Initiative: Promising Approaches to Combination HIV Prevention Programming in Concentrated Epidemics

AIDSTAR-One | CASE STUDY SERIES

Finally, Avahan is a learning model. Avahan must adapt to respond to changing realities in the field in order to succeed. For example, the CMP was revised three times in the projects first phase, allowing for innovations at the local level to be channeled back to the Avahan program level. Managers listen closely to state implementing agencies and NGOs and adjust their approaches and support when they hear that an aspect of the program is not working optimally. Avahans state partners provide additional staff support and training to NGOs that slip in reaching their targets. with the highest HIV prevalence and the largest concentrations of MARPs within the six states. Avahans partners then undertook detailed mapping and size estimation exercises within districts and subdistricts to determine where to establish services and to use later to gauge progress. As part of that process, they carefully mapped solicitation hot spots and sexual networks at local levels. These data made it possible to focus resources in each state on districts, subdistricts, and sections of up to 10 large cities and 100 towns with the highest HIV prevalence and concentrations of MARPs. The mapping generated estimates of 213,000 female sex workers, 76,000 transgender people and MSM, and 25,000 IDUs in 82 districts across the six states.2 Target groups were further subdivided by demographic characteristics, behaviors, or working locales to allow for different outreach approaches. For example, female sex workers were identified as home-based, as street-based, or as based in brothels and lodges. Building key services: Avahan provides a package of prevention services on a large scale across the six states. Condom promotion and distribution as well as STI diagnosis and treatment are linked to behavior change communicationall essential elements of combination HIV prevention. To maintain high standards of quality in service delivery, the program actively monitors all activities at all levels. Most Avahan clinics are located in areas with high concentrations of MARPs or where two or three such groups tend to overlap. Avahan sponsors STI clinics at sites not covered by government clinics. Many are housed within the nearly 600 community drop-in centers the program has created. Clinician hours are posted, and each STI clinic also includes a counselor, a nurse/dispenser, and an administrative
Estimates in subsequent years increased the number of female sex workers to 221,000 and of most-at-risk MSM to 81,000, while decreasing the number of IDUs to 18,000.
2

Avahans Technical Approach

Avahans technical approach involves the target communities to develop its interventions, using highly participatory approaches to identify hotspots and risk groups, to design a package of prevention services that is relevant and accessible, and to address social factors that increase MARP marginalization and create barriers to prevention services. By investing in monitoring and evaluation activities, Avahan ensures that its interventions continue to effectively address the evolving needs of its target populations. Pinpointing the epidemic: To assess where the need for prevention services for MARPs is most critical, Avahan began its planning process by reviewing Indias extensive epidemiologic and sexual behavioral data. These data showed that the six states where Avahan is based had an estimated 83 percent of all HIV infections in India. These states also had a concentration of female sex workers, MSM, and IDUs. Mapping exercises further confirmed that these states represent the core of the epidemic in India and uncovered even larger numbers of MARPs. First, Avahan state partners and NGOswhich both serve and are often staffed by MARPsmapped areas
4 AIDSTAR-One | March 2011

AIDSTAR-One | CASE STUDY SERIES

In response, Avahan instituted sensitivity training for police. But Avahans peer educators and NGO partners also encourage female sex workers and MSM to proactively confront police who try to intimidate or marginalize them. This approach empowers MARPs to have greater control over their own sexual health and well-being, which is one of Avahans key goals. Largely at the initiative of the MARPs themselves, the program now actively fosters and supports a sense of community ownership to further counter marginalization and neglect. Avahan encourages NGOs to help create CBGs and to support formally registered community-based organizations (CBOs) for MARPs; CBGs and CBOs now operate in many of the districts covered by the program. In turn, the CBGs and CBOs have contributed to the program by focusing on quality-of-life issues beyond HIV prevention that are important to MARPs, such as obtaining food ration cards, sensitizing health providers to their needs, and expanding skills training and income-generating opportunities. Committees within the CBOs monitor the quality and acceptability of services at the Avahan STI clinics. In several cases, committees gained sufficient information to insist on replacing clinicians with irregular attendance or judgmental attitudes.
Prashant Panjiar for the Bill & Melinda Gates Foundation

Peer educators in Thane completing a hotspot map of their outreach area.

assistant. The STI clinics also provide referrals for tuberculosis (TB) diagnosis and HIV testing, care, and treatment. Avahans condom program includes both social marketing and distribution of free condoms by peer educators and at STI clinics. In conjunction with the government, Avahan has intensified the advertising and distribution of socially marketed condoms in more than 100 sexual hot spots along trucking routes. Currently, free and Avahan-supported commercial outlets distribute nearly 20 million condoms monthly. Condoms are also available from many other outlets across India. Altering the structures of risk: One of the key findings from the mapping exercises was that in addition to harassment, female sex workers and MSM experience police and gang violence, which can discourage them from using condoms or changing risky behaviors. For example, police often suspect women found with condoms of being sex workers; the women may be forced to pay a bribe to avoid arrest, or coerced into providing a sexual favor. In health care settings, sex workers and MSM face stigma and discrimination from staff, who often deny them services. Fearful of such treatment, many simply forgo accessing basic health care.

Prashant Panjiar for the Bill & Melinda Gates Foundation

Peers talking to police in Mysore.

The Avahan-India AIDS Initiative: Promising Approaches to Combination HIV Prevention Programming in Concentrated Epidemics

AIDSTAR-One | CASE STUDY SERIES

RESPONSIBILITIES OF PEER EDUCATORS

Mapping peer sexual contact locales and networks Weekly contact with peers, conveying a growing set of messages about safe sex Condom promotion and distribution Weekly data summaries Microplanning and revisions Referrals to STI clinics and follow-up, as needed TB screening and referrals; some accompaniment to clinic HIV testing referrals.

Avahans management relies on the large amount of output data the program generates at all levels, starting with contacts made by peer educators.
Monitoring and evaluation: Avahans management relies on the large amount of output data the program generates at all levels, starting with contacts made by peer educators. Peer educators keep daily records of their contacts and their interactions with their peers, identifying highly vulnerable peers for extra attention. These records are consolidated into weekly summaries of contacts. Implementing NGOs work with the peer educators to analyze these data and plan outreach to ensure maximum coverage and clinic attendance. They then forward output data to the lead agencies in each state to help monitor progress and identify areas for improvement or added attention. Avahan also monitors outcomes. STI rates at clinics are followed closely, as is condom distribution at various sources. Avahan funded two rounds (200607 and 200910) of an Integrated Behavior and Biological Assessment to collect behavior and biological data from a representative sample of the targeted at-risk populations in selected districts. Sample data elements included knowledge, attitudes, and practices related to safer sex; and HIV, STI, and/or hepatitis infection. NACO generates annual estimates of Indias population living withHIV to focus attention on districts with emerging epidemics, and conducts biannual behavioral surveys. Along with its own data, Avahan uses data from NACOs behavior surveys and estimates of people living with HIV to further monitor its own progress and to look for gaps in its districts.

LEARNING FROM PEER EDUCATORS

Peer educators feel they have demonstrated the following to Avahan and its state partners. There are many skills within marginalized communities Communities can manage their problems Communities can mobilize resources Communities can take a leading role in running health and social welfare programs.

Peer Educators Within Avahan

The use of peer educators is a well-established practice in HIV prevention. But many peer educator projects have had limited results. Retention of peer educators, lack of message innovation and change, reliance on uncompensated volunteer peers, the short duration of many projects, and other factors have reduced the impact that such initiatives might have. Within Avahan, peer educators are the central focus of program operations. Peer educators have multiple responsibilities that take advantage of their knowledge and skills. They receive a small stipend that, for many, is the

AIDSTAR-One | March 2011

AIDSTAR-One | CASE STUDY SERIES

difference between living in deep poverty and having the means to work toward a more secure future. While the stipend is important to peer educators, the commitment to personal and community change is also an important motivator. Moreover, the program was structured to provide opportunities that allowed them to develop new skills and positioned them for greater responsibilities. These factors created a strategy to combat the high turnover rates typical in most peer educator programs. Some peer educators help manage the programs activities with peer communities. As they document interactions with peers, they know the data they collect are essential for their own planning, which is key to the programs overall implementation strategy. Avahan has developed non- and low-literacy materials that peer educators use during training and on the job. Many of the materials use pictures and symbols instead of text. Training includes topics on self-esteem development, negotiation and facilitation skills, advocacy and leadership, and peer monitoring. Peer educators and service delivery: Each peer educator has up to 50 clients, meeting with each at least once a month and often more frequently. During meetings, peer educators offer safer sex messages, provide condoms, and note their clients visits to STI clinics. They serve as the primary and initial contacts for access to STI services at Avahan clinics, giving referral notes for diagnosis and treatment services. They encourage and facilitate attendance, sometimes by escorting the peer to the STI clinic, which also serves as a referral point for HIV testing and treatment services.
Peer educators in Pune.

Recently, peer educators were trained to identify basic TB symptoms and to escort symptomatic peers to clinics for diagnosis, in some cases arranging for TB treatment monitoring. Some of the CBOs that have sprung from the work of the peer educators also help people living with HIV register for government treatment services. Avahans approach to peer education is intensive, both for the peer educators and the supervisory outreach workers of the implementing NGOs. Extensive planning helped establish a viable system of identifying and selecting peer educators, providing initial and follow-up training, and sustaining their interests and skills with close supervision. Drop-in centers offer a supportive place for peer educators to meet, discuss, plan, and offer services to peers. As peer educators began to expand their work and organize themselves, Avahan found it needed further planning to manage and incorporate those community groups. Crisis intervention is a high-profile service of the peer educator system. A peer-led and NGOsupported rapid response network with access to

Peer educators have multiple responsibilities that take advantage of their knowledge and skills.

The Avahan-India AIDS Initiative: Promising Approaches to Combination HIV Prevention Programming in Concentrated Epidemics

Prashant Panjiar for the Bill & Melinda Gates Foundation

AIDSTAR-One | CASE STUDY SERIES

lawyers can respond within 24 hours to reports of violence by police or gangs, domestic violence, or other complaints. Negotiations are conducted at police stations, sometimes with senior officers. Domestic violence has led some CBOs to establish safe houses for women and their children; CBO members may also attempt to intervene with the husband or partner to resolve issues. The dedication of peer educators to their work is one reason Avahan was able to quickly scale-up its program across 82 districts in six states. Most peer educators have come to recognize that even in a program this large, their individual inputs and achievements contribute to the programs objectives. Weekly recordkeeping and biweekly data analysis with outreach workers add to the peer educators sense of responsibility and engagement. Scaling up: Working at a large scale was a primary program objective from the outset. Beginning in late 2003, Avahan moved quickly to expand services sufficiently to cover 80 percent of identified MARPs in the states where it works. This involved identifying lead partner agencies in each of the six states, one for work with truckers, and one for men at sex solicitation locales. Through the lead state partners, Avahan identified local NGOs to implement the work in districts with substantial MARPs. The NGOs received training about Avahans goals, data collection methods and standards, and peer management. It was not always a smooth or linear process, especially in developing an approach to community mobilization. It took time to identify and recruit peer educators, develop low-literacy (and nonliteracy) materials, and provide training. An earlier community mobilization model in India, the well-known Sonagachi Project for sex workers, provided useful lessons in what marginalized communities can achieve. In addition, Avahans state and NGO partners brought their extensive experience in community mobilization to their work. Rather than insisting on one model of community mobilization, Avahan senior managers responded with one of the programs core features: flexibility to adapt based on changing circumstances and information. The willingness of the program to learn from and with its partners fostered a working collaboration that contributed to the rapid scale-up of program activities. Because scale-up began simultaneously in each state, the program had been established in nearly 90percent of the targeted locales, had recruited two-thirds of the projected number of peer educators, and was distributing 6 million condoms per month within two years after the programs launch. Two years after that, Avahan and government-supported interventions covered about 95 percent of female sex workers and 90 percent of MSM in the four southern states.

Program Results
In Avahans first phase, the project was able to quickly go to scale through its massive peer education activities, which resulted in uptake in STI services and condom distribution and a decrease in STI rates, including HIV. In addition to the service delivery outcomes, the program contributed to greater social cohesion and community ownership of HIV prevention programs. Finally, the donor investments in Avahan helped to spur increased

The dedication of peer educators to their work is one reason Avahan was able to quickly scale-up its program across 82 districts in six states.

AIDSTAR-One | March 2011

AIDSTAR-One | CASE STUDY SERIES

funding by the state governments to adequately address HIV. STI and HIV prevention: In 2008, midway through its 10-year program, Avahan has achieved its goal of operating at a large scale in six Indian states. The program has recruited and trained thousands of peer educators with a variety of STI/HIV prevention messages, providing them with a range of delivery tools and techniques. Data collection and monitoring materials were developed for use by low-literacy and non-literate peer educators. The program carried out a very extensive mapping exercise of sex hot spots in 82 districts, along with estimations of the number of sex workers, MSM, and IDUs in specific locales in those districts. Based on the monitoring system, peer educators now account for 90 percent of all contacts with the target populations. The rapid increase in the number of condoms dispensed or sold and increases in reported condom use are evidence of the effectiveness of peer educator outreach efforts. By early 2006, more than three-quarters of sex contacts with clients of female sex workers involved condom use. Over the entire project, more than 80 percent of male clients of female sex workers reported condom use at last sex with a sex worker; as of May 2007, that represents a 35 percent increase from a year earlier. By late 2008, each peer educator was distributing an average of 38condoms per month to most-at-risk individuals. Of most-at-risk individuals contacted by the program, 85 percent of female sex workers and 64 percent of most-at-risk MSM used the STI services. STI rates declined after the implementation of Avahan. For example, in the state of Tamil Nadu, where a large, well-established, USAID-funded HIV prevention program implemented by APAC had already shown significant results, the added support of Avahan brought syphilis rates among female sex workers down by one-third between the end of 2003 (preAvahan start-up) and 2006. In addition, chlamydia rates declined by about 80 percent during the same period in Tamil Nadu. Between 2003 and 2007, in the states where Avahan worked, government surveys showed small declines in HIV prevalence, again adding to the impact of earlier interventions. However, in 18 districts where Avahans efforts were most intense, HIV prevalence at antenatal clinics declined from 1.5 percent to 0.6 percent between 2003 and 2007. While other factors may have played a part in the decline of STIs and HIV in these regions, it is clear that Avahan contributed to reduction. In addition to STI management referrals, peer educators added TB screening, HIV testing, and referrals for antiretroviral therapy to their responsibilities, which they recognized as important to the health and well-being of their peers. This reflects how peer educators, as they gained more experience and confidence, came to understand more fully the interconnectedness of HIV with other health conditions. Advocacy and community initiatives: The importance of access to adequate health care and of human and civil rights emerged from the locallevel advocacy practiced by peer educators and the implementing NGOs. Peer educators and other program staff target the police, both street-level patrols and officers, to stop harassment, coercion, and violence. Though during the early mapping exercises fewer than one-quarter of sex workers reported experiencing police harassment, police interference with sex work (and with the ability of sex workers to practice safer sex) has been a concern for many sex workers and MSM. The resistance by sex workers to police interference and the use of rapid crisis response teams to police and gang violence has resulted in changes in working conditions for sex workers. From all anecdotal reports, police interference has fallen off dramatically.

The Avahan-India AIDS Initiative: Promising Approaches to Combination HIV Prevention Programming in Concentrated Epidemics

AIDSTAR-One | CASE STUDY SERIES

Emerging leadership: Avahans peer educators have benefited from shouldering significant responsibilities in addition to traditional peer education tasks. During the design and planning stages, they take charge of microplanning, which includes initial mapping of sexual solicitation and networking locales used by MARPs. Peer educators also play an important role in fostering community mobilization and the development of CBGs and CBOs. These experiences have created a sense of ownership of the program and a commitment to creating environments conducive to reduced risk, leading to increased selfconfidence and self-esteem. One unexpected but welcome outcome has been a blossoming of community solidarity between female sex workers and MSM. Avahan seemed to provide an opportunity and space for these marginalized groups to create communities and organize themselves to articulate their grievances against systems that marginalize them. Several peer educators are now working for NGOs, many of them as outreach workers; others are studying or have acquired new skills, such as hairdressing, that provide income. They recognize themselves as role models for others, which in turn enhances the credibility of the safer sex messages they convey to others. An emphasis on community engagement has been a key focus from the start of Avahan. The Sonagachi model for building sex worker ownership of a safer sex environmentwhich proved to the world that programs can successfully engage MARPswas an important inspiration for Avahan. But Avahan staff and peer educators have gone forward to create their own mechanisms for broader community involvement and ownership, built on the programs experiences and understanding of its target populations. Community ownership: While Avahan has fostered community ownership, the communities

One unexpected but welcome outcome has been a blossoming of community solidarity between female sex workers and MSM.

themselves are assuming ever greater control of the process and outcomes of community mobilization. In many situations where individuals have united around common interests, both informal and formal organizations have appeared. Two aspects of the growth and influence of constituent control and ownership within Avahan are noteworthy. First, the peer educators and their communities are at the forefront of HIV prevention at the local levels, at a scale never before seen in HIV prevention work. Second, the program management has, for the most part, not felt threatened by the growth of community activism and ownership. With the exception of some NGO staff concerned about their jobs, the management at all levels does not worry that community groups are taking over their roles. In fact, the state implementing agencies actively encourage the consolidation of community groups into registered CBOs, with elected officers and committees. In Tamil Nadu, the CBOs have been linked into a registered federation. Some CBOs are also linking with established CBOs and NGOs that deal with microfinance and other areas of advocacy and rights. Cost and affordability: The Bill & Melinda Gates Foundations investment of U.S.$258 million over the first 5 years of the 10-year program is substantial. About $215 million of that amount was actually spent (an average of $48 million per year). As of mid-2009, the Foundation had committed $330 million to Avahan. By comparison, Indias initial budget for NACP-III, to run from 2007 to

10

AIDSTAR-One | March 2011

AIDSTAR-One | CASE STUDY SERIES

2012, was set at over $1 billion, and has since grown to $1.7 billion, using a loan from the World Bank, grants from other sources, and internal resources. (The Gates Foundation funding is included in the countrys total budget for HIV.) About $350 million of the NACP-III budget is for programming for a broad spectrum of high-risk groups across the country. While Avahan covers six states and major trucking routes, Indias national program covers the entire country, a larger and more complex undertaking. Nonetheless, Avahan has demonstrated to the government what can be accomplished with well-targeted and adequate resources. Limited data were available at the time of field visits about the costeffectiveness of the Avahan program or its various components. Avahan does have a subcontractor gathering data for costing analyses. One study puts the NGO-level cost of covering one beneficiary at an average of $48, although there are variations between states and subgrantees. While within the guidelines prepared by NACO, the cost is higher than the $40 of one states SACS. Another study found that there is a wide variation in the costs of reaching high-risk individuals. The variation was explained by different approaches of implementing partners (both state partners and NGOs) and the number of beneficiaries contacted by individual NGOs. However, as scale expanded, the average costs declined. A breakout of financial distributions for program implementation by Avahan and its partners showed that about 41 percent went to prevention activities for female sex workers and MSM. Just over 20 percent went to prevention for high-risk males (e.g., clients of sex workers and potential clients such as truckers found in hot spots). The next largest segment was for capacity building for partners and government (14 percent). A small proportion, four percent, went to prevention activities for IDUs. The remainder was divided between advocacy, communications, monitoring and evaluation, and knowledge building. The population of the six Indian states where Avahan operates totals approximately 300 million people; Avahans annual cost averaged $33 million over its lifespan. Given the costs associated with even small increases in prevalence in India and the subsequent burden of treatment and care, Avahans contributions to reduce STI and HIV prevalence could be interpreted to represent a proportional investment in prevention efforts. For instance, a 0.1 percent increase in the six states would roughly translate to 100,000 people needing treatment at an annual cost of $24.6 million.
The Avahan-India AIDS Initiative: Promising Approaches to Combination HIV Prevention Programming in Concentrated Epidemics 11

THE COST OF PEER EDUCATOR STIPENDS: EVIDENCE FROM TAMIL NADU

Each of about 1,725 peer educators connected to the Avahan program in Tamil Nadu receive a stipend of about U.S.$10.50 per month. This comes to just over $18,100 per month for direct support to all the peer educators in that state. The annual cost comes to $217,200. Management support (training, supervision, materials) is not included in the figures. By contrast, the monthly costs for ART drugs alone total about $22 per patient.

AIDSTAR-One | CASE STUDY SERIES

The Avahan project has been the subject of a recent ongoing debate regarding its perceived high cost structure and sustainability.3,4 As the transition to NACO nears, concerns have been raised that Avahans costs could hinder the Government of Indias maintenance of program operations. Similarly, Avahans cost structure has implications for its replication elsewhere. More specifically, concerns have been raised over whether or not the Avahan model can be sustained if there is a continued expansion in the size of MARPs, for example, with an influx of sex workers from the surrounding region. In February 2010, a special peer-reviewed supplement of the journal, Sexually Transmitted Infections, evaluated results from Avahan from different angles, including cost.5 middle segment of managementbetween target audiences and senior project managementis the weak link, often filled by an implementing NGO. In contrast, almost all of the NGOs originally recruited by Avahan are performing well because of a relatively simple program approach that maintains common standards, sets up an intensive reporting system, and provides ongoing training and interaction. Training and oversight have also been stepped up to address any NGO weaknesses. Intensive use of human resources: The ability and willingness to work at a large scale reflects Avahans commitment to placing human resources where they are most needed to reach their target populations and consistently deliver services to MARPs. Avahans ratio of peer educators to peers ranges from 1:35 to 1:50, much lower than in other peer education programs. Peer educators are also responsible for intensive data collection and analysis, including recording individual contacts and aggregating the data weekly. Using, not just collecting, data: From the beginning, data have been used to identify key drivers of the epidemic and to pinpoint areas with high levels of HIV infection. From the weekly reports of peer educators, data work their way up through the management structure and are used at all levels to monitor progress and to identify and address problems. Peer education with multiple responsibilities: Avahan has demonstrated that peer educators can handle a range of behavioral change topics, activities, and responsibilities. Through the microplanning process, peer educators become part of a whole team, rather than isolated individuals. They build on the trust Avahan management places in them to stimulate solidarity within peer communities. It is useful to note that the turnover rate for peer educators working with Avahan is about 10 percent per year, a rate generally lower than most other peer educator programs.

What Worked Well

The Avahan program demonstrates that a very large HIV prevention program can provide a combination of services to multiple target groups at multiple levels with a flexible management structure. In addition, early planning to foster community ownership and promote structural interventions has added critical value to Avahans overall program. Managing with a clear set of goals and standards: To respond to changing conditions and opportunities, Avahans management structure is both proactive and flexible. Avahan remains focused on its objectives and personnel as it operates at a very large scale, thereby gaining the trust of its target groups. For many projects, the
Flock, Elizabeth. How Bill Gates Blew $258 Million in Indias HIV Corridor. Forbes India, June 5, 2009. Available at http://business.in.com/article/crossborder/how-bill-gates-blew-$258-million-in-indias-hiv-corridor/852/1. 4 Yamada, Tachi. Bill & Melinda Gates Foundation Responds To Forbes Article. Forbes, July 20, 2009. Available at http://www.forbes.com/2009/07/20/gatesfoundation-response-philanthropy-avahan_print.html 5 Sexually Transmitted Infections. Emerging Results from a Scaled HIV Prevention ProgramAvahan, the India AIDS Initiative. February 2010, Volume 86, Suppl 1. http://sti.bmj.com/content/86/Suppl_1
3

12

AIDSTAR-One | March 2011

AIDSTAR-One | CASE STUDY SERIES

strengthens a targeted intervention approach and makes it more accountable. NACO and SACS have learned from and with Avahan and give the program credit for demonstrating what can be done with good data, monitoring, management, advocacy, and intense involvement with affected communities. Combining a mix of interventions: When structural interventions and community mobilization are added to a mix of prevention services, the synergy achieved with combination HIV prevention is evident. The approach adopted by Avahan demonstrates the effectiveness of a large-scale combination HIV prevention approach in concentrated epidemics. The Avahan experience shows that combination HIV prevention is not simply an ideal or theoretical construct, but a viable programmatic approach.

When structural interventions and community mobilization are added to a mix of prevention services, the synergy achieved with combination HIV prevention is evident.
Working with volunteers: Peer educators join Avahan as volunteers. However, Avahan provides a monthly stipend equivalent of between $10 and $20 each. The stipend is an important supplement for women and men who live very marginally; many are in debt. However, when they describe their motivation for becoming peer educators, they talk not about the extra income but about the sense of accomplishment and self-fulfillment they feel. Removing structural obstacles to HIV prevention: To counter local-level harassment by police of female sex workers and MSM, Avahan launched sensitivity training, peer advocacy, and rapid responses to individual cases of harassment or violence. As relations between the police and MARPs improve, it becomes easier for targeted populations to practice safer sex and to seek care. As confidence has grown within CBGs, other structural impediments have been addressed: peer community members have now acquired ration cards, access to public health facilities, and, for some, health insuranceall services long denied to marginalized groups. Authentic community ownership: The active involvement of target audiences has been critical in building community involvement and ownership throughout Avahan. Not all community groups organized under the Avahan umbrella will survive, but many will, and they will become more powerful in representing their constituents in the process. Stakeholder learning from and with Avahan: Avahan adds value to the governments HIV prevention program as it broadens and

Challenges
There is wide acknowledgment by the Indian National AIDS Control Organization, SACS, Avahans state partners, and most participants that the program has generally worked well. However, there are several areas where program weaknesses have been encountered, and these provide lessons for other large, focused HIV prevention programs. Sustainability: The Avahan program has several more years to run. In 2009, Avahan began to hand over district-level program activities to SACS; such transfers will be completed in 2014. The financial sustainability of the Avahan approach is assured by the governments very substantial budget for HIV and AIDS work and its commitment to communitydriven programming. Yet not all are convinced that the Government of Indias resources can maintain the Avahan program at current operating costs. In addition, Avahans partners are concerned that the support given for CBGs will wane as the government takes over programs. Partners also fear potential
13

The Avahan-India AIDS Initiative: Promising Approaches to Combination HIV Prevention Programming in Concentrated Epidemics

AIDSTAR-One | CASE STUDY SERIES

funding disruptions, especially to local NGOs and CBOs and to those that employ sex workers as peer educators, given the governments more cumbersome bureaucratic structures for transferring funds. These concerns are fueled in part by a long-held suspicion of authority by CBO members, although many have gained confidence about pressing for their rights. Also, CBO members are new at organizational management; it remains to be seen if the capacity-building element of Avahan will be sufficient to build the skills and experience CBO leaders need to manage the organizations. Some of Avahans implementing state partners and NGOs are likely to have their roles taken over by SACS. It is possible that technical assistance and monitoring of local programs may not receive the same intensity of attention as has occurred under Avahan. That said, the acknowledged learning that has occurred within SACSwith support from Avahan, USAID, and other sourcesprovides a solid base for sustaining many of the technical aspects for reaching MARPs, which is a NACO priority. Finally, with four years to manage the transition from Avahan program to government program, opportunities exist to identify and correct problems and strengthen key areas for continued community mobilization. Coordination at scale: Avahan is a largescale coordinated program, with common program elements used in all locations where it works. Its individual components and diverse state realities are held together by those common elements and by standards of service delivery and a strong management structure. While program managers have handled coordination well, the ability to handle differences is an ongoing challenge. CBG involvement is being integrated into the programs processes very slowly. Avahan management is exploring ways to strengthen CBGs by converting some groups into registered CBOs, and then establishing systems to monitor community group activities. The program has done well to support rather than feel threatened bycommunities who are
14 AIDSTAR-One | March 2011

seeking a greater role in running their own affairs. But it also means that the strong coordination features of the Avahan program will begin to loosen to accommodate diverse local interests and methods. Replication: Avahans combination HIV prevention approach appears to be very replicable in countries with concentrated HIV epidemics and likely replicable where there are mixed epidemics. Replication will require a good, but not necessarily perfect, set of epidemiological and behavioral data, good planning, and a sound management structure at all levels. Any replication of the Avahan model should reflect local realities and organizational experiences. The financial and human resources needed to build such a program are, in most cases, within the reach of HIV and AIDS budgets in many countriesif donors, governments, and implementing agencies can agree to work on a common program. Even countries that lack an adequate budget or consensus on focusing resources on the main drivers of the epidemic can learn from the Avahan model and replicate some of its elements. Three of those elements stand out. First, use available data to identify the main drivers, most-at-risk, or underserved groups in the epidemic and put adequate resources toward reaching them. Second, focus on building and sustaining community buy-in and an active role in program development. Third, address the structural barriers that make it difficult for individuals to practice risk reduction. Monitoring qualitative and emotive aspects of the program: Although Avahan has systems to regularly collect and use large amounts of data, it has been slow to develop methods for monitoring the effects of advocacy efforts at all levels, measuring levels of violence against MSM and female sex workers (other than police violence), quantifying expanded access to non-program services (such as food ration cards) for high-risk people, and capturing information about community mobilization. The programs initial emphasis on

AIDSTAR-One | CASE STUDY SERIES

quantitative indicators was not adequately followed up by developing mechanisms to capture soft data. Thus, while anecdotal reports and accounts exist on impacts of the program on these indicators, more solid data to back up those accounts are only beginning to emerge. Measuring costs: The Avahan program is collecting and analyzing data on the cost of various program components. A challenge will be to determine the cost-effectiveness of the different program components without losing sight of the coherence and mutually reinforcing nature of each. Another challenge will be for NACO to revise its cost estimates for interventions to adequately sustain NGO and CBO activities. implementation are motivated when they are involved deeply and regularly in the program. Identify and work on structural obstacles and opportunities at all levels: Structural factors may be very local, such as police harassment, or national, such as policies that criminalize sex work, or in between, such as clinic rules and processes that discourage MARPs from accessing services.

RESOURCES
Avahan has produced several publications that describe aspects of the program. These can be found on the programs website: www.gatesfoundation .org/avahan/Pages/overview.aspx. Avahans state partners also maintain websites that provide information on their involvement. For Tamil Nadu, see www.taivhs.org/ For Andhra Pradesh, see http://www.hlfppt.org/ andhrapradesh.htm For more information on the APAC project, please visit www.apacvhs.com. In addition, the following publications were useful in preparing this case study: Avahan. 2008. Managing HIV Prevention from the Ground Up: Peer Led Outreach at Scale in India. New Delhi: Avahan. Cornish, Flora, and Catherine Campbell. 2009. The Social Conditions for Successful Peer Education: A Comparison of Two HIV Prevention Programs Run by Sex Workers in India and South Africa. American Journal of Community Psychology 44(1-2):12335. Jana, S., et al. 2004. The Sonagachi Project: a sustainable community intervention program. AIDS Education and Prevention 16:40514. Kenya National AIDS Control Council. 2009. Kenya HIV Prevention Response and Modes of Transmission Analysis: Final Report. Nairobi: NACO.
15

Recommendations for Implementers

The experience of Avahan highlights the need for good evidence for programming; adequate resources to meet identified needs, including strong project management; and the importance of addressing the social and cultural context that may contribute to or hinder a projects success. Identify and understand the main drivers of the epidemic: An effective combination HIV prevention program relies on good (but not necessarily perfect) data and an understanding of what the data tell about the core drivers of the epidemic. Put enough people on the ground to fully cover the target populations: A ratio of 1:50 of peer educators to target groups is sufficient to affect behaviors and foster community ownership. Develop a management structure that uses the skills and resources of all staff and volunteers at all levels: Staff and volunteers who contribute to program planning and

The Avahan-India AIDS Initiative: Promising Approaches to Combination HIV Prevention Programming in Concentrated Epidemics

AIDSTAR-One | CASE STUDY SERIES

Kim, Caron R., and Caroline Free. 2008. Recent Evaluations of the Peer-Led Approach in Adolescent Sexual Health Education: A Systematic Review. International Family Planning Perspectives 34(2):8996. Available at www.guttmacher.org/pubs/journals/3408908. html (accessed September 2010) UNAIDS. 1999. Peer Education and HIV/AIDS: Concepts, Uses and Challenges. Geneva: UNAIDS. Pradesh, P. Shailaja and the staff of the Swagati Project within the Hindustan Latex Family Planning Promotion Trust provided information and access to project sites. Thanks also to V. Sampath Kumar and Kerry Pelzman with USAID/India, Timothy Mah and Shyami DeSilva with USAID/Washington, and the Prevention Technical Working Group for their review and insight.

RECOMMENDED CITATION ACKNOWLEDGMENTS

A special thanks to the staff of the Bill & Melinda Gates Foundation who assisted with information and logistics in the preparation of this case study. At the Foundation, Gina Dallabetta and Matangi Jayaram were especially helpful. In Tamil Nadu, the staff of the Tamilnadu AIDS Initiative of Voluntary Health Services, including Project Director R. Lakshmibai, generously shared the organizations experiences within Avahan. In Andhra Rau, Bill. 2011. The Avahan-India AIDS Initiative: Promising Approaches to Combination HIV Prevention Programming in Concentrated Epidemics. Case Study Series. Arlington, VA: USAIDs AIDS Support and Technical Assistance Resources, AIDSTAR-One, Task Order 1.

Please visit www.AIDSTAR-One.com for additional AIDSTAR-One case studies and other HIV- and AIDS-related resources.

AIDSTAR-Ones Case Studies provide insight into innovative HIV programs and approaches around the world. These engaging case studies are designed for HIV program planners and implementers, documenting the steps from idea to intervention and from research to practice. Please sign up at www.AIDSTAR-One.com to receive notification of HIV-related resources, including additional case studies focused on emerging issues in HIV prevention, treatment, testing and counseling, care and support, gender integration and more.

16

AIDSTAR-One | March 2011