In rural southern Nepal, where the vast majority of births occur at home, the risk of neonatal mortality is significantly reduced when visiting birth attendants wash their hands before the delivery, mothers wash their hands before handling their infant, or both, according to an observational prospective cohort study.1 When birth attendants and mothers both cleaned their hands, the risk of infant death was decreased by nearly half (relative risk, 0.6).

Hand washing has long been a recommended method of reducing the risk of infection, but there is little research assessing its effect on neonatal mortality in developing countries, let alone in high-risk rural environments like Nepal’s Sarlahi district. The neonatal mortality rate in Nepal is about 33 deaths per 1,000 births, and infectious diseases like meningitis and pneumonia account for almost 40% of these deaths. To examine the impact of hand washing in this setting, the authors performed a secondary analysis of data from a pair of community-based trials conducted in 2002–2006 that were designed to evaluate fieldworkers’ use of the antiseptic chlorhexidine for cleansing the skin and umbilical cord.

All pregnant women in 30 villages in Sarlahi were invited to participate in the studies. Upon enrollment, women were counseled on safe birthing practices, including the importance of having birth attendants wash their hands with soap and of washing their own hands before handling their baby. They received, among other items, a clean birthing kit that contained soap. More than three quarters of the women lived in households that lacked electricity or a latrine, and a similar proportion had never attended school. Outcomes for 23,662 infants born alive from September 2002 to February 2006 were included in the analysis. More than 90% were born at home or outdoors (sometimes en route to a facility). About three in 10 infants weighed less than 2,500 grams, and two in 10 were preterm.

Visiting fieldworkers checked the newborns for signs of illness up to 11 times during their first 28 days of life. About half of these home visits occurred during the first week. The majority (63%) of the infants were first visited within 24 hours of birth; during the initial visit, all mothers completed a questionnaire that included an item about whether the birth attendant had washed his or her hands before the delivery. In 59% of births, the attendant had done so. On a follow-up questionnaire given two weeks later, about 15%of mothers reported always or sometimes washing their hands before caring for their baby.

The neonatal mortality rate in the study was 32 per 1,000 live births. After the exclusion of very early deaths, which were generally related to conditions like asphyxia or congenital abnormalities rather than to infection, infants delivered by birth attendants who had washed their hands prior to delivery had a 19% lower risk of death than newborns delivered by attendants who had not washed their hands (relative risk, 0.8). The risk of death was 60% lower for infants whose mothers washed their hands before caring for them compared with those whose mothers did not (0.4). Although the combination of attendant and maternal hand washing also reduced neonatal mortality (0.4), it was no more effective than maternal handwashing alone.

When the data were adjusted for possible confounding factors, such as birth weight and gestational age, the risk reductions were slightly smaller. Neonatal mortality declined by 19% when birth attendants washed their hands (relative risk, 0.8), by 44% when mothers washed their hands (0.6) and by 41% when both mothers and birth attendants washed (0.6).

Some high-risk groups particularly benefited from the interventions. The risk of mortality was reduced for low-birth-weight infants (relative risk, 0.6), but not for those of regular weight, when birth attendants washed their hands before delivery. Maternal hand washing reduced the risk of neonatal death by 64% among infants born in homes without a latrine (0.4).

The authors qualify these findings by noting that all reports on hand washing were subjective, and that there may have been differences between women who reported washing their hands and those who did not that were not captured by this study. Moreover, because early neonatal deaths were excluded from most analyses, the findings should be interpreted as showing benefits "among infants who survived the first few days of life," rather than among all newborns. Nonetheless, the findings indicate that "a substantial proportion" of neonatal deaths in Nepal "may be preventable with routine hand washing practices." Although considered normal practice in developed countries, "the concept of washing with soap…to protect against infection is not well understood" in developing countries, and the authors suggest that the establishment of hand washing as a routine practice in these settings would have a "tremendous impact" on reducing the four million annual neonatal deaths worldwide.—S. Ramashwar


1. Rhee V et al., Maternal and birth attendant hand washing and neonatal mortality in Southern Nepal, Archives of Pediatric and Adolescent Medicine, 2008, 162(7):603–608.