Aims: To evaluate the effectiveness of home‐based peer counselling to increase breastfeeding rates for unfavourably low birthweight babies. Methods: Randomized clinical trial carried out in maternity hospitals and households in Fortaleza, one of the regions in Brazil with very low income; 1003 mothers and their newborns were selected in eight maternity hospitals. Newborns needed were healthy and weighed less than 3000 g. Intervention: Breastfeeding counselling, conducted by lay counsellors from the community, during home visits carried out on days 5, 15, 30, 60, 90 and 120 after birth. Main outcome measure: Feeding methods in the fourth month of life. Results: The intervention increased exclusive breastfeeding (24.7% vs 19.4%; p=0.044), delayed the introduction of formula and increased the time infants substituted breastfeeding to bottle milk (bottle milk 33.4% in the control group and 20.1% in the intervention group; p=0.00002). When comparing the frequency of artificial breastfeeding versus all other forms of breastfeeding (exclusive+predominant+partial), the intervention increased breastfeeding rates in 39% (RR=0.61; CI 95%: 0.50–0.75); 15% of children were free from artificial feeding (absolute risk reduction). The number of families to be visited to avoid one child receiving artificial feeding (NNT) was 7 (CI 95%: 5–13).

Conclusions: Breastfeeding counselling, promoted by lay counsellors, can impact favourably on exclusive breastfeeding rates and contribute to delaying the utilization of milk formula and weaning. The intervention has great application potential because most cities in the northeast of Brazil count on community health workers that could do the counselling.