India, the second-most populous country, was experiencing a highly varied HIV epidemic, in the past two decades. HIV in India is concentrated among certain “high-risk” populations like Injecting Drug Users (IDUs), Men having Sex with Men (MSM) and Female Sex Workers (FSWs). Targeted interventions (TIs) with these populations have been one of the major aspects of India’s National AIDS Control Programme (NACP). The recent HIV estimates showed a reduction in Adult HIV prevalence from 0.41% in the year 2000, to 0.36% in 2006 and to 0.31% in 2009. The estimated number of new annual HIV infections declined from 0.27 m in 2000 to 0.12 m in 2009 [1] .

However, Andhra Pradesh (AP) state remains to be critical in the spread and control of HIV epidemic, as it has second highest estimated adult HIV prevalence of 0.9%, with Manipur at 1.40%. While India had an estimated 2.4 m people living with HIV/AIDS in 2009, the state of AP tops the list with about 0.5 m; followed by 0.42 m in Maharashtra, 0.25 m in Karnataka and 0.15 m in Tamil Nadu. Of the estimated people living with HIV/AIDS, about 0.93 m (39%) is women [1] .

HIV prevalence among Men having Sex with Men (MSM) in AP was high at 17.04% in 2007, with an increase from 10.25% in 2006. Sentinel surveillance data of 2007 also showed Andhra Pradesh as the only state with >1% prevalence among ANC attendees [2] .

Targeted Interventions carried out by the Andhra Pradesh AIDS Control Society (APSACS) and Avahan by Bill and Melinda Gates Foundation (BMGF) are the two major HIV/AIDS prevention programs currently in operation in the state. The APSACS in partnership with NACO has been implementing HIV prevention, care and treatment programs since 1999, which have resulted in significant gains. However, the epidemic continues to be a cause for major concern. The HIV sentinel surveillance round in Andhra Pradesh during 2007 revealed that, across the state, the prevalence levels continue to show an upward trend, pointing to a need for intensified prevention efforts. There is an increasing recognition of the substantial role that MSM behaviors have played in HIV epidemics in Indian settings [3] -[13] . However, limited data are available in India on MSM except for some small studies [11] [14] -[19] .

In this backdrop, the present paper utilizes the data from “Integrated Behavioral and Biological Assessment” (IBBA) conducted at two different time points, Round-1 in 2006-2007 and Round-2 in 2009 in select districts of Andhra Pradesh to analyze and present the impact of these Interventions on the sexual behaviours, STI prevalence and HIV prevalence in the state of Andhra Pradesh.

The Integrated Biological and Behavioral Assessment (IBBA), was a cross-sectional study, conducted by adopting same protocol, questionnaires, Methodology, sampling design and sample size during Round-1 and Round-2 in the districts of East Godavari, Guntur, Hyderabad and Visakhapatnam. A sample size of 400 subjects per district per round was arrived to detect an absolute difference of 15% or more from an assumed value of 50% of consistent condom use (use of condom at every sex act) by the target population with 95% confidence, 90% power and design effect of 1.7.

A probability based sampling method with a varied sampling approach was used to select primary sampling units, the sites. Time Location Cluster Sampling approach was used taking into account the different sub-popu- lations that congregate in a single location at different time points of a day or on different days of a week [20] .

This study collated information on socio-demographic characteristics, sexual behaviors, risk perception, prevalence of STI and HIV among self-identified men who have sex with men (MSM) population.

The inclusion criteria were, any male or hijra, self-identified MSM, aged 18 years or above, who had any type of sex (oral, manual, or penetrative), paid or unpaid, with another male in the last one month.

The data collected was scrutinized, fed in to computer by double entry technique to rectify entry mistakes if any and was cleaned using range and consistency checks in order to develop database. Appropriate weights were calculated and applied at district levels for unequal selection probability. These weights were used for statistical analysis of all the parameters [20] . The distributions of data at different time points in the study districts were compared using Pearson c².

Sexual partners of MSM were categorized into 1) Paying male partners (having a partners who paid to have sex), 2) Paid male partners (Paid to have sex with partners), 3) Non-commercial, non-regular male partners 4) Regular male partners, 5) Regular female partners and 6) Paid female Partners. Consistent condom use was defined as using condom at every sex act; and last time condom use was defined as use of condom during the last sex act.

The study was approved by the Institutional ethical committee of National Institute of Nutrition. Written informed consent was obtained from the participants of the study, with a provision to maintain confidentiality.

Community monitoring board and Community advisory boards were established in each of the district with local NGOs, community members, local officials and leaders as members to communicate to the community about the study, to create an ambience of mutual trust, to help and ensure the protection of rights and physical & psychological well-being of the respondents.

Response rate of the participants in the study ranged from a low 56% in Vishakapatnam to a high 70% in E. Godavari. Response rates were relatively less in R2 compared to R1 in all the districts, except E. Godavari where it was similar (data not shown).

Homosexuality, bisexuality and trans-sexuality are highly stigmatized in India. Men who have sex with men (MSM) are frequently abused; face physical violence and harassment from police and the general society [14] [17] [22] . There are limited data available in India on sexual behaviors linked to STI/HIV prevalence among MSM in Andhra Pradesh and India [18] [19] [22] [23] . The current paper presents the results from the studies that are conducted in Andhra Pradesh as part of a large scale intervention programme for the HIV prevention among MSM by the Avahan program. The study is unique in terms of its magnitude and numerous behavioral and biological indicators collected in the process.

Avahan Intervention started in 2006 in the select districts in Andhra Pradesh and continued till the time of the round-2 of this evaluation. The populations who participated in both rounds of this study were self-identified men having sex with men. The first round of the study was conducted in 2006 while the second round was conducted in 2009. Consistent with other studies on MSM conducted in India and elsewhere, this study revealed that most MSM were in the younger age group and therefore, were at a higher risk of acquiring and transmitting HIV infection. Mean age of the respondents who participated mean age at sexual debut during both rounds remained almost similar. Most of the participants were localities in each district, indicating lesser migration in this high-risk group compared to female sex workers.

MSM in E. Godavari district shows a new trend of 9% staying with male partners, which is unusual in India and such behaviours carry a lot of stigma and discrimination. This can be result of the significant increase in community mobilization and empowerment activities like the community based groups seen in the district.

Condom use with all types of male partners (regular, main and commercial) increased significantly from round-1 of the study showing a clear impact of the interventions in changing behaviors. However, there has been a significant increase in the number of commercial partners across the districts, which might be indicating to a new trend of commercial sex among MSM, which according to other studies has been lower till date [21] [22] , since sex among MSM had been mostly for pleasure and not for monetary gains [14] [24] [25] . About 45% of the MSM in Visakhapatnam district reported having a paid male/hijra partner, which is significantly higher than the earlier round and an increasing trend of having paying male partners is seen in Guntur and Hyderabad.

Many studies did indicate that MSM was more difficult to reach; which was always perceived as a bigger challenge for interventions [14] [15] [17] [25] . Results from this study however show a mixed result. While E. Godavari, Guntur and Hyderabad districts showed nearly 80% to 100% outreach by the program for condom distribution and clinical services, while Visakhapatnam districts showed lesser than 60% exposure to the program intervention. Interestingly, Visakhapatnam district had the highest program participation rates during R1, compared to R2.

A significant proportion of MSM reported having regular female partner during both the rounds of the study and the consistent condom use by MSM with these partners remained low during both the rounds, indicating high risk behavior. Impact of behavior change interventions is not seen in this regard, which could be because the program does not focus much on this aspect or because of social stigma. This is a crucial indicator for preventing the spread of HIV in the state. In addition, not all MSM use condom with their male partners. Thus, there exists potential risk of the bridge group spreading HIV to population at large.

Although an overwhelming majority of the respondents had heard of AIDS/HIV, only about 50% to 70% of them had correct knowledge on HIV. Knowledge in STIs among men was observed to be low in Visakhapatnam district in R2, compared to R1. High levels of knowledge about prevention of HIV/AIDS or STI alone is not enough to prevent risky behavior, unless attitudes also change [25] .

Though the HIV prevalence during R2 decreased in Visakhapatnsm, a significant increase was observed in the district of Guntur, while it remained high in the districts of E. Godavari, and Hyderabad, which was a matter of concern and needed to be further investigated. The prevalence of Chlamydial infection, though low increased during R2 in Visakhapatnam, but tended to decrease in other districts. The prevalence of syphilis during R2 increased two-fold in Guntur district, while it tended to decrease in the remaining districts.

HIV prevention programs in Andhra Pradesh need better focus on condom use MSM with female partners, and imparting knowledge on HIV/AIDS prevention. Program needs to take note of the increasing trend in having commercial sex. There is an urgent need to focus on the Visakhapatnam district as the program exposure and STI knowledge among the MSM is low. Though the HIV prevalence is among MSM in Visakhapatnam, in view of the low exposure to interventions, there is need to strengthen IEC activity to control the spread of the infection.

The findings in the study are subject to certain limitations. First, a sample size of 400 was considered for MSM at district level. Though this size is adequate to provide sufficiently precise estimates for high prevalence indicators such as condom use, it may be inadequate to precisely estimate other indicators such as HIV or STI prevalence. Self-identified MSM do not represent the whole MSM population and in the present study, only self-identified MSM contacted in the hot spots during the survey period are included as respondents.

Support for this study was provided by the Bill & Melinda Gates Foundation through Avahan India AIDS Initiative as a grant to FHI.

The authors do not have any conflict of interest. GNVB conceived and supervised the study. KV involved field supervision, development of database, carried out the statistical analysis and prepared the manuscript. HK & SPV monitored supervised the field operations. HR supervised lab estimations and SK supervised the field operations and provided critical comments. All authors read & approved the final version of the manuscript. The authors thank all the MSM respondents who participated in the study. The authors thank Dr. B. Sesikeran, Director, National Institute of Nutrition, for permission to conduct the study.