University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Home > Global Health Literature Digest > Reduced Mortality
Reduced mortality associated with breast-feeding-acquired HIV infection and breast-feeding among HIV-infected children in Zambia
Global Health Sciences Literature Digest
Published September 15, 2008
Journal Article

Fox MP, Brooks D, Kuhn L, Aldrovandi G, Sinkala M, Kankasa C, et al. Reduced mortality associated with breast-feeding-acquired HIV infection and breast-feeding among HIV-infected children in Zambia. Acquir Immune Defic Syndr 2008;48:90-6.


To determine if children infected with HIV postpartum have longer survival from time of infection compared with children infected during gestation or delivery

Study Design

A prospective cohort study. This analysis uses a subset of participants from a larger cohort study called the Zambia Exclusive Breastfeeding Study (ZEBS). The ZEBS was designed to determine whether exclusive breastfeeding, along with abrupt weaning at 4 months, could reduce the rate of transmission of HIV from mother to child.


Two clinics in Lusaka, Zambia


Between May 2001 and September 2004, HIV-infected women were enrolled in ZEBS at antenatal visits. This analysis includes all 213 live-born, singleton children enrolled in ZEBS who tested HIV-positive during follow-up. Time of HIV transmission was estimated from the midpoint between dried blood heel-stick samples obtained at each visit (1 week, monthly until 6 months, then every 3 months). Sixty-one (29%) children were estimated to be infected between 0 and 3 days after birth (intrauterine group: IU), 71 (33%) at between 4 and 40 days (intrapartum/early postpartum: IP/EPP), and 81 (38%) after 40 days (postpartum: PP).


The intervention was designed specifically for the larger cohort of women, who were counseled intensively to breastfeed their children exclusively through 4 months of age. At 1 month after delivery, all women were randomized to abrupt cessation of breastfeeding at 4 months or exclusive breastfeeding through 6 months with weaning as they normally would and the duration of all breastfeeding based on mother's informed choice. Children were given cotrimoxazole prophylaxis from 6 weeks to 12 months of age; highly active antiretroviral therapy was unavailable.

Primary Outcomes

The primary outcome was time to child death.


The median follow-up time was 183 days for the IU group, 276 for the IP/EPP group, and 365 days for the PP group. A little more than 50% of the children were male, and 16% of the children had low birth weight. Among the mothers of the HIV-infected children, 60% had a viral load >100,000 copies/mL, 43% had a CD4 count of <200 cells/µL, and 67% had significant anemia (Hgb <11 g/dL).

There were 81 child deaths over the year since infection. Fifty-six percent of children infected in the IU group, 51% in the IP/EPP, and 27% in the PP group had died by 1 year of follow-up from infection. The IU and IP/EPP groups had an initial high mortality rate during the first 3 months of follow-up compared with infected children in the PP group (P=0.001 and 0.005, respectively); no differences were found between the IU and IE/PP groups (P=0.5). Children in the PP group had less than half the crude mortality rate over 1 year after infection compared with those in the IU group (hazard ratio [HR] = 0.39, 95% confidence interval [CI]: 0.22-0.79). Adjusting for low birth weight, low maternal CD4 cell count, cessation of breastfeeding, and maternal death results in an even greater protective association with child mortality when comparing the PP group with the IU group (HR=0.27, 95% CI: 0.15-0.50). A second analysis conditional on survival to 40 days to account for non-HIV-related mortality in the first 40 days gave a similar result (adjusted HR=0.32, 95% CI: 0.17-0.61).

There also were harmful associations observed between child mortality and breastfeeding cessation (HR=3.1, 95% CI: 1.8-5.3), low birth weight (HR=3.7, 95% CI: 2.1-6.5), and maternal death (HR=3.1, 95% CI: 1.4-6.9).


The authors concluded that this study demonstrated better survival among children infected postpartum compared with children infected during pregnancy or delivery and demonstrated a benefit to an increased duration of breastfeeding among infected children. The authors suggested that testing children early for HIV may provide a means for earlier intervention.

Quality Rating

Based on the Newcastle-Ottawa quality rating for cohort studies, this study is of adequate quality. The study has some limitations. The loss to follow-up was fairly high among the groups: 24% in the IU group, 13% in the IP/EPP group, and 12% in the PP group. A greater probability of death among those lost to follow-up in one of the groups could have biased the results. In addition, children excluded from the study because of no HIV test or no return visit may have been both more likely to be in the IU group and more likely to have died, introducing a bias in comparisons. This possible bias may explain the lack of difference between the IU group and the IP/EPP group seen in earlier studies. One also might argue that the analysis conditional on surviving 40 days was more appropriate as the primary rather than the secondary analysis, although the results differed only slightly.

In Context

Several studies have demonstrated increased survival among children infected postpartum compared with children infected in utero. A pooled analysis has shown a 25% reduction in mortality for children infected after 4 weeks of age compared with those infected before 4 weeks but did not show a difference between children infected at birth and those infected soon after.(1) The finding of reduced mortality among children infected postpartum is consistent with previous studies that found relative associations of 0.10 (2) and 0.74 (3) comparing mortality among late versus early infection. The results of this study are similar to a recent finding from a cohort of children in Zimbabwe, which found similar mortality rates in children infected in utero compared with those infected intrapartum, but a strongly reduced risk of mortality in children infected postpartum compared with those infected in utero.(3)

In contrast to the results of this study, the pooled analysis of Newell et al., which adjusted for breastfeeding behavior, found no difference in mortality between breastfed and non-breastfed children.1 This effect may have been due to residual confounding, because their analysis only compares children ever breastfed with children never breastfed; however, 80% of the women had done some breastfeeding, which would bias the association of breastfeeding and child death towards the null. Other studies, however, using time-dependent measures of breastfeeding as did this study, have found a protective effect of longer breastfeeding among children born to HIV-infected mothers, even when the child is also infected.(4)

Programmatic Implications

The results of this study show that the window for intervention in children infected earlier is shorter than for those infected postpartum (after 40 days). It is therefore critical to improve prevention measures for in utero HIV transmission as well as to test for HIV early post-partum. In addition, children who are infected early in life should continue to be breastfed and given medical care and treatment to increase their survival.


  1. Newell ML, Coovadia H, Corina-Borja M, et al. Mortality of infected and uninfected infants born to HIV-infected mothers in Africa: a pooled analysis. Lancet 2004;364:1236-43.
  2. Spira R, Lepage P, Msellati P, et al. Natural history of human immunodeficiency virus type 1 RNA levels, timing of infection, and disease progression in African HIV-1-infected children. Pedriatrics 1999;104:e54.
  3. Marinda E, Humphrey JE, Iliff PJ, et al. Child mortality according to maternal and infant HIV status in Zimbabwe. Pediatr Infect Dis J 2007;26:519-26.
  4. Taha TE, Kumwenda NI, Hoover DR, et al. The impact of breastfeeding on the health of HIV positive mothers and their children in sub-Saharan Africa. Bull World Health Organ 2006:84:546-54.