May I very respectfully suggest again that UNICEF's announcment that it will begin to distribute ABM for babies of HIV+ mothers in Uganda, Rwanda, Zambia etc. is a watershed event, and call for discussion other than the "slander the messenger" approach (on Dana Raphael, or Glenda Gray, who is NOT funded by industry). UNICEF's decision, I believe, calls for thoughtful reflection and an examination of the primary research. Someone asked me by email questions about the 1997-1998 research, and I thought I would reply to the group, as I have received many similar requests for information about primary research. As Kathy Auerbach reminded Lactnet readers on this topic, there is a big difference between primary research reports and secondary articles. It is far easier to find secondary sources on HIV & breastfeeding, in which the author(s) opine that breastfeeding isn't really a bad risk for transmission. I don't think that UNICEF would plan to distribute formula unless there were significant cause to do so. As far as primary research reports are concerned, Laura Guay's data from Uganda show the lowest rate of transmission (only 26% of babies breatfed by HIV+ mothers became HIV+). I always list Guay, but I also always list the other primary authors, who have found different rates of transmission in their populations of breastfeeding HIV+ mothers. These include 25% in Rwanda, 26% in Zaire, 30% in Sao Paulo Brazil, 39% in Zaire, 35% in Malawi, 38% in South Africa, 40% in Congo, 43% in Kenya, 46% in Soweto, 48% in India, 49% in Rio de Janeiro Brazil,52% in Central African Republic, 57% in Spain, 63% in Italy, 90% in Zambia, 100% in Saudi Arabia. From the 1998 and 1997 medical journals come more reports. I will refer to four,which are typical. There are *no* primary reports showing that breastfeeding is realtive risk-free. 1) Lewis et al. 1998. Cell-free Human Immunodeficiency virus type 1 in breast milk. Journal of Infectious Diseases 177: 34-39. In previous research, HIV-1 DNA had frequently been detected in breastmilk cells. In this study Lewis at el described the quantity of *cell-free* HIV-1. It was higher in mature milk than in colostrum. For infants consuming 700 mL HM/day, the oral exposure is more than 630,000 copies of HIV-1 RNA/day. This is a substantial exposure. Lewis et al also refer readers to research on simians that showed that oral exposure can readily lead to infeciton. Adult rhesus monkeys could be infected by the oral route with 6000 times less virus than was needed for infection by the non-traumatic rectal route. 2) Becquart et al. 1998. Early postnatal Mother-to-childl transmisison of HIV-1 inBangui, Central Africa Republic. Presented at the Fifth Conference on Retroviruses & Opportunistic Infections, Chigago February 105. All babies were breastfed by HIV+ mothers; 52% of babies beame HIV-infected. 3) Bobat et al. 1997. Breastfeeding by HIV-1-infected women and outcome in their infants: a cohort study from Durban, South Africa. The HIV transmission rate was 39% in those exclusively breastfed, 32% in those mixed fed and 24% in those formula-fed. The longer babies were exclusively breastfed, the higher the risk of transmission. Among babies exclusively breastfed for one month, 45% became HIV-infected. Among those exclusively breatfed for two months, 64% became HIV-infected. Among those exclusively breatfed for three months, 75% became HIV-infected. Mortality was 19% in those exclusively breastfed, 13% in those mixed fed and 0% in those formula-fed. 4) Maguire etal. 1997. Potential risk factors for vertical HIV-1 transmission in Catalonia, Spain: the protective role of cesarean section. AIDS 11:1851-1857. In Spain, 57% of babies breastfed by HIV+ mothers became infected, compared to 17% of bottle fed babies. I know that older studies such as the work in Haiti & Malawi showed lower rates. In older studies,babies who died beore a year of age were sometimes excluded from analysis (truly). ALso, older studies stopped testing babies once they had one HIV-negative test result. All babies of HIV+ mothers would be HIV-antibody-positive for awhile (maternal antibodies in babies). Then many babies revert to negative - i.e. HIV-antibody-negative by test. But when babies of HIV+ mothers continue to breastfeed, they continue to risk HIV infection. In the Nairboi cohort, out of 90 children who were eventually HIV-positive - 40 of them went through a period of months of being HIV-negative. These children then were HIV+, HIV-, then HIV+. Children who convert to positive in the later months or years are very likely to have been infected by breast milk. Other possible means of HIV infecition were excluded (blood transfusions, scarification etc). Ref = Datta et al 1995 Journal of Infectious Diseases 172:1419. Also Datta et al 1994 Journal of Infectious Diseases 11:974-976. Edith White, IBCLC since 1985