Effect of community-based peer counsellors on exclusive breastfeeding practices in Dhaka, Bangladesh: a randomised controlled trial

Background Most mothers breastfeed in Bangladesh, but they rarely practise exclusive breastfeeding. Hospital-based strategies for breastfeeding promot ion cannot reach them because about 95% have home deliveries. We postulated that with the intervention of trained peer counsellors, mothers could be e nabled to breastfeed exclusively for the recommended duration of 5 months. Methods 40 adjacent zones in Dhaka were randomised to intervention or contr ol groups. Women were enrolled during the last trimester of pregnancy betwe en February and December, 1996. In the intervention group, 15 home-based co unselling visits were scheduled, with two visits in the last trimester, thr ee early postpartum (within 48 h, on day 5, between days 10 and 14), and fo rtnightly thereafter until the infant was 5 months old. Peer counsellors we re local mothers who received 10 days' training. Findings 363 women were enrolled in each group. Peer counselling significan tly improved breastfeeding practices. For the primary outcome, the prevalen ce of exclusive breastfeeding at 5 months was 202/228 (70%) for the interve ntion group and 17/285 (6%) for the control group (difference=64%; 95% CI 5 7%-71%, p>0.0001). For the secondary outcomes, mothers in the intervention group initiated breastfeeding earlier than control mothers and were less li kely to give prelacteal and postlacteal foods. At day 4, significantly more mothers in the intervention group breastfed exclusively than controls. Interpretation Peer counsellors can effectively increase the initiation and duration of exclusive breastfeeding. We recommend incorporation of peer co unsellors in mother and child health programmes in developing countries.