The WHO and UNICEF recommend home visits to improve health outcomes for mothers and newborns. We evaluated the effect of home visits by community volunteers during pregnancy and postpartum on breast-feeding practices, women’s knowledge about benefits, beliefs and myths of breast-feeding, obstetric and neonatal warning signs, preparation for childbirth and initial care for newborns, and diarrhoea and respiratory diseases in children.

Community quasi-experimental design. We estimated difference-in-difference models with fixed effects at the community level weighted by propensity score and investigated implementation barriers through focus groups and semi-structured interviews.

Poor rural communities in Mexico; 48 intervention and 29 control.

Baseline and follow-up information were reported from two independent cross-sectional samples of women with babies aged between 6 and 18 months (baseline: 292 control, 320 intervention; follow-up: 292 control, 294 intervention).

The intervention increased reports of exclusive breast-feeding in the first 6 months by 24·4 percentage points (pp) (95 % CI: 13·4, 35·4), mothers’ knowledge of obstetric warning signs by 23·4 pp (95 % CI: 9·2, 37·5) and neonatal warning signs by 26·2 pp (95 % CI: 15·2, 37·2) compared to the control group. A non-linear dose–response relation with the number of home visits was found. Diarrhoea and respiratory diseases among children decreased in the intervention v. control group but were not statistically significant.

Home visits should be implemented as a complementary strategy to the provision of prenatal and postnatal care in rural communities due to their potential positive effects on the health of mothers and their children.