Short courses of drugs can be given to HIV-infected pregnant women to reduce the chance of HIV infection being passed to her child either during pregnancy or during the labour process. However, children can also become infected by drinking the mother's breastmilk which contains the HIV virus. In many poor, developing countries in Africa, breastfeeding is the normal way of infant feeding and is vitally important because of the protection it gives to children from other diseases such as diarrhoea and malnutrition. Ideally there would be a way to make breastfeeding safer from HIV transmission without losing its other advantages.

A medical study recently suggested that HIV-infected women who exclusively breastfed their children i.e. gave breastmilk but without any water, tea, formula milk or any solid foods did not pass on the virus to their children to the same degree as women who MBF with these other fluids and foods. It is important to confirm whether this observation is in fact true or not. We hypothesize that exclusive breastfeeding by HIV-infected mothers carries a lower risk of HIV transmission than mixed breastfeeding.

We propose to follow 2,100 HIV-infected pregnant women and also some HIV-uninfected women from the time that they book at the clinic until 24 months of age. HIV-infected women who say they intend to breastfeed and all the HIV-uninfected women will be visited at their homes by breastfeeding counsellors both before and after delivery to support exclusive breastfeeding. HIV-infected women who choose not to breastfeed will be helped by clinic staff to safely replace all breastmilk with some other milk. An independent team will visit all mothers at their homes and collect information about the way they feed their children. Blood samples will be collected from the children at different times by a simple heel prick and the blood stored on a piece of filter paper. By testing these samples and comparing with the type of feeding at that time, we will be able to see when a child becomes infected and whether exclusive breastfeeding gives any protection.

Short course antiretroviral regimens such as AZT (Thai regimen), or nevirapine (Uganda HIVNET 012 regimen) given to HIV-infected women in pregnancy can reduce in utero and intrapartum transmission to their child by 40-50%. However, these gains are threatened by the continued transmission of HIV through breast milk of poor HIV-infected women in Africa who have no realistic choice but to breastfeed. A recent study from Durban compared for the first time, Mother-to-child transmission of HIV (MTCT) rates in each of the 3 feeding groups viz. exclusive breastfeeding (EBF), mixed breastfeeding (MBF) and formula feeding. The results of this study showed that at 3 months the MTCT in EBF mothers was similar to that in mothers giving only formula feeds, and the MTCT in EBF mothers was significantly lower than in those MBF (14.6% vs. 24.1%). This suggested that EBF may carry no additional risk of MTCT of HIV1 over formula feeding. A prospective study is required to verify these findings.

We propose a cohort study of HIV-infected mothers and their infants, to examine the risk of post-natal transmission of HIV in relation to infant feeding practices after mothers receive intensive breastfeeding counselling and support to practise EBF until about 6 months of age (the optimal duration for EBF).

The hypothesis being tested is: Exclusive breastfeeding by HIV-infected women carries no additional risk of post-natal transmission of HIV than exclusive formula feeding.

The primary objectives of the study are to determine the effect of infant feeding practices on HIV infection rates of infants at 6 and 22 weeks of age, and secondly to to determine the infant survival rate at 24 months of age according to feeding practices and HIV status.

Secondary objectives are:

i) To determine the HIV infection rate of infants as measured of a sample collected within 72 hours of birth; ii) To determine the HIV transmission incidence attributable to the duration of different feeding practices; iii) To determine the cumulative incidence of vertical transmission in EBF, MBF and EFF infants; iv) To describe risk factors, other than feeding practice, for post-natal transmission of HIV, including maternal and infant morbidity and breast health; v) To assess the determinants of transmission in MBF adjusting for exposure factors e.g. type and age of introduction of other food/milk; vi) To describe the morbidity and growth of infants in relation to feeding practices and HIV status; and vii) To describe adherence rates to EBF following a breastfeeding support intervention.

HIV infected women will be counselled according to the WHO/UNICEF/UNAIDS infant feeding guidelines. All HIV-infected women and a random sub-sample of HIV-uninfected women will be recruited antenatally. Those who choose to breastfeed will be visited frequently at home by a team of breastfeeding counsellors before and after delivery to promote and support EBF. HIV-infected mothers who choose not to breastfeed will be supported in their choice by clinic-based staff. An independent team of field monitors will collect data weekly on feeding practices and morbidity of all infants at home visits. Blood samples will be collected by finger/heel prick and stored on filter paper prior to qualitative and quantitative estimation of HIV viral content to determine the timing of transmission.

Inclusion Criteria:

• All pregnant women attending antenatal clinics in the Mpukunyoni and Hlabisa Tribal Areas of Hlabisa District or at Kwadabeka clinic or St Mary's hospital, Mariannhilll Hospital who offer consent to participate in the study

Exclusion Criteria:

• Intention to leave Hlabisa District or Kwadabeka community within 3 months of delivery.
• Failure to give consent.
• Less than 16 years of age.