The goal of this research is to provide empirical data for the first time in India on the incidence of HIV and its determinants at the population level through a longitudinal study. This study will be conducted in a representative population-based sample of over 12,600 adult men and women from rural and urban parts of Guntur district in the Indian state of Andhra Pradesh in whom we assessed baseline seroprevalence of HIV, herpes simplex virus-2 (HSV-2) and syphilis, and the sexual and other risk behavior during 2004-2005. We will conduct a 5-year follow-up of this cohort to estimate new HIV, HSV-2 and syphilis infections and ascertain the determinants of HIV incidence and the transmission dynamics at the individual, partnership and broader levels. In addition, we will assess the bias in reporting sexual behavior using a confidential polling box approach for a more realistic understanding of risky sexual behavior at the population level. We will also assess effective coverage of and barriers to key HIV prevention, treatment and care interventions at the population level to understand how their coverage could be enhanced. Assuming a loss to follow-up of 25% of the baseline sample, based on a recent pilot study, we estimate that there will be 85 incident cases of HIV (95% confidence interval 63-107) at 5-year follow-up, which would be a reasonable number to explore the associations of new HIV infections with their determinants. We will utilize field procedures standardized in the baseline study that will be adapted for the follow-up study. Trained field staff will document socio- demographic background, sexual and other risk for acquiring HIV, and utilization of HIV interventions through interview and they will collect dried blood samples on filter paper using the finger prick method. Ethical considerations will be paramount to minimize risk to study participants. The laboratory analysis for HIV, HSV-2 and syphilis will follow a strategy similar to the baseline study, using internationally accepted test kits. Confidentiality of data will be ensured. Multivariate regression techniques will be used to assess the association of new HIV infections with determinants at the individual, partnership, cluster and geographic area levels. Our proposed study will provide original scientific data and analysis of new HIV infections and their determinants at the population level, bias in sexual behavior reporting, and barriers to HIV interventions. These unique and rigorous data will fill several key gaps in the evidence base needed for the understanding of HIV transmission and its control in high burden parts of India.
This is the first study in India that will follow-up a representative sample of the population to provide original data and analysis of new HIV infections and their determinants at the population level, bias in sexual behavior reporting, and barriers to HIV interventions. Rigorous scientific methods will be used for field data collection, laboratory analysis and statistical analysis. This study will fill several crucial gaps in the evidence base needed for the understanding of HIV transmission and its control in high burden parts of India.