• HIV infection detected by 4 months among children with no evidence of HIV infection at birth.
  
• Survival to 2 years of age without HIV-infection.
  
• Magnitude of the reduction in mother-to-child HIV transmission and the magnitude of the increase in non-HIV-related under-2-year mortality, attributable to cessation of breastfeeding at 4 months.
  
• HIV infections detected by 2 years among children with no evidence of HIV infection at 4 months
  
• Mortality in children aged 4-24 months among uninfected children
  

It is well established that infants breast fed by their HIV-infected mothers are at risk of acquiring HIV infection through breast milk. However, in low resource settings, where the HIV epidemic now predominates, breast feeding cannot simply be replaced by breast milk substitutes since alternatives to breast milk are unavailable, unaffordable and unsafe. With this application we aim to test the safety and efficacy of short duration exclusive breast feeding to minimize risks of HIV transmission without increasing risks of non-HIV infant mortality. We propose a 5-year study of HIV-positive mothers and their children to be conducted in two urban primary health care clinics in Lusaka, Zambia. All HIV-positive women and their infants will be offered the two-dose nevirapine intervention and will be counseled about the risks and benefits of infant feeding options. Women who indicate their decision to breast feed will be eligible for enrollment into the study. A culturally appropriate, affordable and sustainable breast feeding education and support program to encourage exclusive breast feeding will be developed, and all women who elect to breast feed will be encouraged to exclusively breast feed to 4 months. Half of the women will be randomized to a counseling program which will encourage abrupt weaning to full replacement feeding at 4 months, and half will be randomized to a program to encourage continued breast feeding after 4 months with the usual introduction of weaning foods. Children will be followed for two years with regular medical histories, physical exams and clinical sampling. The primary objective of the study, based on the random assignment, is to compare HIV transmission rates and under-2 year mortality rates in children who abruptly wean at four months of age versus children who are weaned according to local practice. The second primary objective, based on observational comparisons, is to compare HIV transmission among infants whose mothers adhere to recommendations to exclusively breast feed with those who do not. Secondary objectives are to describe acute and chronic effects of abrupt weaning on child morbidity. The study proposes to test an inexpensive and potentially sustainable public health intervention to reduce HIV transmission through breast feeding while preserving benefits of breast feeding for other aspects of child health in a very low resource setting.