AEGiS-13IAC: Child mortality associated with failure to breastfeed and voluntary and involuntary weaning; An assessment of the risks and benefits of breastfeeding by HIV-infected mothers.

13th International AIDS Conference

Durban, South Africa - July 9-July 14, 2000

Child mortality associated with failure to breastfeed and voluntary and involuntary weaning; An assessment of the risks and benefits of breastfeeding by HIV-infected mothers.

Int Conf AIDS 2000 Jul 9-14; 13:(abstract no. WeOrC496)

Gray R, Brahmbhatt H
R. Gray, Johns Hopkins School of Public Health, 4028 Hygiene, 615 N.Wolfe Street, Baltimore, MD 21205, United States, Tel.: +1 410 955 7818, Fax: +1 410 955 0792, E-mail: hbrahmbh@jhsph.edu

BACKGROUND: Mother-to-child transmission of HIV via breastmilk presents a policy dilemma. Prior studies of the risks of mortality associated with non-breastfeeding fail to differentiate between voluntary weaning and involuntary non-initiation or involuntary weaning, due to pre-existing maternal/infant morbidity (i.e. confounding due to prior illness). This overestimates the risk associated with formula feeding

METHODS: Data from the Uganda Demographic and Health Survey (1995) which asked questions on breastfeeding and reasons for weaning were used: "voluntary" (i.e work, age of child) and involuntary (i.e preceding maternal/infant morbidity). The child mortality ratio (MR) was determined from the maternal birth history

RESULTS: Among 4007 live born infants, 55 (1.3%) were not breastfed, and of these, 29 (MR = 52.7%) died. Since BF is almost universal and culturally sanctioned in this society, it is likely that the decision not to initiate was involuntary and due to preceding illness. Among the 3952 breastfed infants, 1604 (40.6%) had been weaned by time of interview. 982 (74.0%) were weaned voluntarily and 14 infants died (MR = 1.4%), whereas 345 (26.0%) were weaned involuntarily and 10 died (MR = 2.9%). Relative to those who weaned voluntarily the rate ratio of child mortality was RR = 37.64 for the never breastfed and RR = 2.07 for the involuntarily weaned infants. Thus, either failure to initiate breastfeeding or involuntary weaning was associated with a MR of 7.5% (39/400) which was significantly greater than for voluntary weaning (RR = 5.36). The population attributable fraction of non-initiation or involuntary weaning was 23.2%

CONCLUSIONS: Failure to account for reasons for non-initiation of BF or for involuntary weaning, over estimate the risks of mortality in non-breastfed or weaned infants by 23.2%. Voluntary weaning is associated with a lower mortality risk, and should be used as the benchmark for assessing the risk/benefits of formula feeding by HIV+ mothers.

Keywords: AEGIS, Breast Feeding, Weaning, Mothers, Infant Mortality, HIV Infections, HIV Seropositivity, Infant Nutrition, Infant Food, Health Surveys, Uganda, Infant, Human, Child, Economics, Mortality
000709
WeOrC496