Stephen J.LurieMD, PhD, Contributing EditorIndividualAuthorPhil B.FontanarosaMD, Executive Deputy EditorIndividualAuthor
Copyright 2000 American Medical Association. All Rights Reserved.
Applicable FARS/DFARS Restrictions Apply to Government Use.2000
In Reply: Dr Bulterys raises concern about
comparability of the 2 arms of our trial because of the differential HIV-1
infection rates at birth. The most likely explanation was low specificity
of the HIV-1 poymerase chain reaction test on cord blood combined with a high
risk of breast milk transmission during the first 6 weeks of life (the next
scheduled testing time). The likelihood that a positive test at birth would
be followed by a second positive test result was higher in the breastfeeding
arm than the formula arm (50% vs 18%) because of ongoing exposure to HIV-1
through breastfeeding. The timing for such infections was defined by the first
positive result (even though the cord blood result may have been a false positive),
hence a higher apparent infection rate at birth. We developed statistical
methods to adjust for this problem.
Nduati R, Mbori-Ngacha D, John G, Richardson B, Kreiss J. Breastfeeding in Women With HIV—Reply. JAMA. 2000;284(8):956-957. doi:10.1001/jama.284.8.956