Women infected with the human immunodeficiency virus (HIV) can transmit their infection to their baby during pregnancy, delivery, or breastfeeding --a process known as mother-to-child transmission (MTCT). According to the World Health Organization (WHO), an estimated 3.4 million children under the age of 15 were living with HIV at the end of 2011. In 2011 alone, 330,000 children had newly acquired HIV from their mothers. `Prevention of mother-to-child transmission of HIV' (PMTCT) refers to a series of interventions that help protect babies born to HIV-infected mothers against the virus. In order to be effective, these interventions need to be implemented as a cascade - starting with antenatal care and continuing during postpartum care. The maximum efficacy of the interventions to reduce MTCT differs by setting. In a developed country setting, the risk of MTCT has be reduced to less than two percent, whereas in a developing country setting, like India, the risk of MTCT still remains relatively higher. The elimination of mother-to-child transmission of HIV is now considered a realistic public health goal. Considerable efforts to expand PMTCT programs and guaranteeing access to antiretroviral therapy (ART) for pregnant and postpartum HIV-infected women has raised the possibility of achieving the virtual elimination of MTCT of HIV.

In order to achieve the maximum impact of PMTCT and realize the goal of virtually eliminating new HIV infections among children by 2015, high levels of coverage, access, utilization, and adherence to treatment regimens must be attained across India. However, 40 percent of HIV-infected women enrolled in the national PMTCT program in India are estimated to be lost to follow-up (LTF) even before they receive a single dose of Nevirapine (NVP). PRAYAS, a non-government organization (NGO) located in the city of Pune, Maharashtra, runs one of the largest private sector PMTCT programs in India. Between 2002 and 2008, PRAYAS collaborated with 43 hospitals in nine districts across Maharashtra and provided comprehensive antenatal care (ANC) counseling and HIV testing services to 122,005 pregnant women and enrolled 950 HIV-infected women in the PMTCT program. This dissertation uses de-identified data previously collected by PRAYAS for program purposes.

The first goal of this dissertation was to contribute to the knowledge on the factors associated with loss to follow-up during the PMTCT cascade in India. Univariate and multivariate analyses were conducted to estimate the associations between being LTF and socio-demographic factors, using generalized linear models. Results of the multivariate analysis showed that women with less than a college level education, women from poor families, women who were registered after 20 weeks of pregnancy, and women whose partners were HIV-uninfected or of unknown HIV status were more likely to be LTF before delivery. Similarly, the factors associated with being LTF after delivery were less than college level education, being in a poor family and registration after 20 weeks of pregnancy.

PMTCT programs are regarded as an entry point to continued care because they provide an opportunity to link an HIV-infected woman, her partner, and her child (if infected) to long-term treatment and care. However, little is known about the factors associated with utilization of continued care among women who have previously utilized PMTCT services. The second goal of this dissertation was to study the barriers associated with reduced utilization of HIV-related continued care in women who have previously accessed PMTCT services in India. After adjusting for potential confounders, results from the multivariate analysis showed that women with poor HIV-related knowledge, women who were currently married, women whose partners had never utilized HIV-related care and women who could not afford to travel to the HIV-care facility were less likely to utilize HIV-related continued treatment and care.

The number of HIV-exposed uninfected (HIV-EU) infants identified in India is likely to increase due to the scale up of programs aimed at realizing the goal of elimination of MTCT by 2015. While some studies from developed and developing countries have reported stunting in HIV-EU children compared to HIV-unexposed uninfected children, others have found no such association. No studies on the effect of HIV-exposure on postnatal growth patterns in HIV-EU children in India have been published to date. The final goal of this dissertation was also to assess the effect of in utero HIV exposure on birth weight and postnatal growth in HIV-uninfected children in India. Birth weight, height and weight of 297 HIV-EU children and 1611 HIV-unexposed uninfected children, in India, were compared. Linear regression models were used to evaluate the association between in utero HIV exposure and birth weight and in utero HIV exposure and postnatal height and weight, after adjusting for potential confounders. HIV-EU children weighed 123.5 g less (p<0.01) at birth compared to HIV-unexposed children. On an average, HIV-EU children were 2.9 cm shorter (p<0.00) compared to HIV-unexposed children. In children five years of age and younger, with every year of increase in age, HIV-EU children grew 0.8 cm less (p<0.01) than HIV-unexposed uninfected children. At ages three and five, the HIV-EU children were 0.22 cm (p<0.05) and 1.8 cm shorter (p<0.05) than HIV-unexposed uninfected children respectively. After adjusting for potential confounders, no significant difference in weight between HIV-EU and HIV unexposed uninfected children was found at different ages.