Some 70% of Brazilian women giving birth in hospitals with staff trained according to the Baby-Friendly Hospital Initiative (BFHI) exclusively breast-feed their infants in the hospital, but the proportion declines rapidly and sharply thereafter unless the practice is reinforced with home visits.1 In a randomized study, the aggregated prevalence of exclusive breast-feeding between the 10th day and sixth month after birth was 45% among mothers who were visited at home and 13% among mothers who were not.

Researchers assessed patterns of breastfeeding among women giving birth to healthy singleton infants at public hospitals in Pernambuco, Brazil, during two periods: a preintervention period in 1998 and a study period in 2001. Between the two periods, maternity staff at the hospitals were trained with BFHI content to administer a hospital intervention to promote exclusive breast-feeding for six months and continued breast-feeding for at least two years, and home visitors were trained with the same content to conduct postnatal visits to promote and support breast-feeding. In the study, women giving birth in the hospitals between March and August 2001 were randomly assigned to receive either 10 home visits or none. The data thus allowed for evaluation of the hospital intervention alone and of the hospital intervention combined with home visits.

Social, demographic and environmental characteristics of mothers and infants were recorded at the time of delivery; breast-feeding practices were assessed at eight points between birth and six months, by use of a questionnaire.

Analyses were based on 364 mother-infant pairs in the preintervention cohort and 350 mother-infant pairs in the study cohort. In each cohort, more than half of mothers had a family income below the poverty level. One-third were younger than 20, and the infant was the first for nearly 40%. Most of the mothers (82-94%) had received some antenatal care. Infants with a low birth weight (less than 2,500 g) were excluded from the preintervention cohort and made up only 5% of the study cohort. A substantial proportion of infants (18-29%) were delivered by cesarean section.

The proportion of mothers exclusively breast-feeding their infants in the hospital was significantly higher in the study period than in the preintervention period (70% vs. 21%). However, the proportion in the study cohort who exclusively breast-fed their infants after hospital discharge fell sharply among mothers who did not receive home visits: Only 30% of mothers in this group were exclusively breast-feeding at 10 days, and just 15% were doing so at 30 days. In contrast, roughly 70% of mothers who were visited at home were following this practice at 10 and at 30 days.

The hospital intervention on its own was associated with a significant increase in the average aggregated prevalence of exclusive breast-feeding between days 10 and 180 after birth, from 7% in the preintervention period to 13% in the study period. Among women in the study cohort, however, the average aggregated prevalence was significantly higher among mothers who received home visits than among those who did not (45% vs. 13%).

The proportion of mothers who at least partially breast-fed their infants in the hospital rose from 70% during the preintervention period to 81% during the study period. In the study cohort, the average aggregated prevalence of breast-feeding at least partially between days 10 and 180 was significantly greater among mothers who received home visits than among those who did not (78% vs. 62%); however, breast-feeding prevalence for those who were not visited did not differ from prevalence in the preintervention period (63%).

Breast-feeding patterns in the group not assigned to home visits varied by poverty status and maternal education. At 30 days after birth, mothers whose family income was above the poverty level were significantly more likely to be exclusively breast-feeding than their poorer counterparts; similarly, mothers with higher levels of schooling were more likely to be exclusively breast-feeding than their less-educated counterparts. In contrast, in the group receiving home visits, levels of exclusive breast-feeding at this time did not differ by poverty status or education.

During the first six months of life, significantly smaller proportions of infants whose mothers received home visits than of those whose mothers were not visited were given pacifiers or bottles, and were fed water, tea or milk other than breast milk. Substantial proportions of infants whose mothers were not visited were fed these other liquids, often soon after birth; for example, at the age of 10 days, 31% were fed other types of milk, 51% water and 56% tea.

Improvements in levels of exclusive breast-feeding achieved with the BFHI are largely limited to the hospital, the researchers observe; moreover, this initiative benefits disadvantaged mothers less than advantaged mothers. "There could be a misplaced sense of security among international agencies and governments that the BFHI will sustain improved breast-feeding practices when mothers return home," they contend, especially in countries where mothers leave the hospital soon after delivery and infants are traditionally given other liquids and pacifiers early in life. Noting that this study and others have found that postnatal home visits improve levels of exclusive breast-feeding, the researchers conclude that the challenge now is to find ways of incorporating such support into routine health care.—S. London


1. Coutinho SB et al., Comparison of the effect of two systems for the promotion of exclusive breastfeeding, Lancet, 2005, 366(9491):1094-1100.