Home visits by female community health workers have produced a one-third reduction in neonatal mortality in a trial in Bangladesh (Lancet 2008;371:1936–44; see also Comment ibid: 1893–5). In the northeast of the country (Sylhet district) 24 population clusters (each with about 20 000 people) were randomised to three options: home care, community care or control arm. In the home care arm female community health workers received 6 weeks of training. They then identified pregnant women, and made two antenatal visits and three postnatal visits (days 1, 3 and 7). In the community care group maternal and newborn care were promoted through group sessions provided by female and male community mobilisers. Usual care was provided in the control arm. The interventions lasted for 30 months and in the last 6 months neonatal mortality rates were 29.2, 45.2 and 43.5 per 1000 livebirths in the home care, community care and control arms respectively, a 34% reduction in the home care arm compared with the control arm. Community care did not affect neonatal mortality. The cost per neonatal death averted in the home care arm (including the cost of strengthening of health systems) was US $2995, indicating high cost-effectiveness.

A genomewide association study (New England Journal …