Reducing neonatal mortality in resource poor settings

BMJ 2012; 344 doi: https://doi.org/10.1136/bmj.e2197 (Published 21 March 2012) Cite this as: BMJ 2012;344:e2197
  1. Kim Eva Dickson, senior adviser on maternal and newborn health,
  2. Mickey Chopra, associate director, chief of health
  1. 1Health, Unicef, NY 10017, USA
  1. kdickson{at}unicef.org

What works is now clearer but implementation is a challenge

Since the announcement in 2000 of the millennium development goals (MDGs), progress towards achieving these goals has resulted in considerable reductions in deaths from communicable diseases such as HIV, tuberculosis, and malaria (MDG 6); maternal mortality (MDG 5); and child deaths (MDG 4). Child deaths for instance have declined from more than 12 million in 1990 to 7.6 million in 2010.1 However, progress in reducing neonatal deaths—deaths within the first month of life—has lagged behind. Neonatal deaths now account for a greater proportion of global child deaths than ever before—nearly 41% of all deaths in children under 5 years occur during the neonatal period.2

In this context, the results of the linked trial by Bhandari and colleagues (doi:10.1136/bmj.e1634) are of particular interest and importance.3 It is the first study to evaluate India’s large and complex Integrated Management of Neonatal and Childhood Illness (IMNCI) programme, which is an approach to neonatal and child care that is being implemented across the country. Bhandari and colleagues evaluated the Indian IMNCI programme and found that it significantly reduced infant and child mortality. Worldwide, 99% of neonatal deaths occur in low and middle income countries, and 50% of these deaths occur at home.4 In 1990, more than half of these deaths occurred in just five countries—India, Nigeria, Pakistan, China, and the Democratic Republic of Congo2—and India had the largest number of neonatal deaths in the period 1990-2009.

India’s adapted IMNCI programme is different from the generic programme in that it places an emphasis on the use of community based providers (auxiliary nurse-midwives or Anganwadi workers) and training of these groups for home visits for postnatal care of the newborn baby. This strategy has already proved to be effective.5 The current study reports a significant reduction in the infant mortality rate (hazard ratio 0.85, 95% confidence interval 0.77 to 0.94) and a reduction in neonatal mortality for babies born at home (0.80, 0.68 to 0.93) for the adapted programme. Overall, the findings of the linked study indicate that the adapted IMNCI programme can be implemented on a large scale and can reduce infant mortality and neonatal mortality in settings where a large proportion of babies are born at home. However, they also raise several questions that need to be investigated further.

Firstly, although Bhandari and colleagues showed a significant reduction in neonatal mortality for babies born at home, mortality was not reduced for babies born at the health facility.3 This might be the result of a statistical artefact, such as regression to the mean; because of the greater fidelity to the intervention by the community workers; or because of unmeasured bias between the groups. It would be premature to conclude that this intervention should be restricted to non-facility based births, and further evaluation of the reasons for this difference is needed.

Secondly, it is not clear which part of the comprehensive intervention was most effective. Several different components were implemented, including women’s groups, home visits, and improved clinical care training, and fidelity to the protocol was variable for each. For example, although the study managed to achieve a relatively high number of home visits (73.5%) within the first 10 days of childbirth, only 42.6% of mothers and infants had the recommended three visits and only 56.6% were visited in the first two days. As the intervention is scaled up within the country, it would be useful to know which parts of the complex package of interventions should be most actively promoted.

Lastly, the lack of impact on neonatal mortality rates in the first 24 hours, regardless of the place of birth, is of concern. This finding may be partly explained by the relatively low level of home visits within the first 48 hours. It is crucially important to strengthen interventions that are targeted at reducing early neonatal mortality, because three quarters of the estimated four million neonatal deaths each year occur in the first week, with the highest risk being on the first day of life.4

The MDG for child survival cannot be met without substantial reductions in neonatal mortality. A lack of policies is no longer the main problem. Since the publication of the Lancet series on neonatal survival in 2005, many countries now have adequate policies in place to tackle child mortality.6 Gaps remain in implementation and action. We need to find ways to identify and reach the most vulnerable—and to adopt an equity focused approach to the implementation of policies—with more global and political commitment to invest in community based approaches.

Bhandari and colleagues’ trial adds to the growing body of evidence that cost effective community based interventions can substantially reduce neonatal mortality even when resources are scarce.7 We know what to do, but how to ensure that we reach the most vulnerable when they are at their most vulnerable remains a challenge that requires political commitment, focused programmes, research, and funding to be overcome.


Cite this as: BMJ 2012;344:e2197


  • Research, doi:10.1136/bmj.e1634
  • Competing interests: All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract

Log in

Log in through your institution


* For online subscription