Welcome to the programmatic area on newborn health within MEASURE Evaluation’s Family Planning and Reproductive Health Indicators Database. Newborn health is one of the subareas found in the women’s health part of the sexual and reproductive health (SRH) section of the database. All indicators for this area include a definition, data requirements, data source(s), purpose, issues and—if relevant—gender implications.
- The goal of many programs in developing countries is to improve maternal and newborn health and survival. The World Health Organization (WHO) estimates that globally, 8.8 million children a year die before their fifth birthday; of these, more than 40 percent of deaths occur during the first month of life (WHO, 2010). The majority of these deaths are preventable and virtually all of them occur in developing countries. Although great progress was made in addressing the fourth Millennium Development Goal (reduce child mortality), there is still much work to be done.
- The newborn health indicators selected for this database measure both health outcomes and impacts as well as quality of care. Based primarily on the work of the Saving Newborn Lives Technical Working Group, many of the indicators measure how well the essential components of newborn care are being implemented—both at the facility and community level.
The goal of many programs in developing countries is to improve maternal and newborn health and survival. WHO estimates that globally, 8.8 million children a year die before their fifth birthday; of these, more than 40 per cent occur during the first month of life (WHO, 2010). The majority of these deaths are preventable and virtually all of them occur in developing countries. Although great progress has been made in addressing the fourth Millenium Development Goal (reduce child mortality), there is still much work to be done.
Most newborn deaths in developing countries can be prevented by interventions already widely used. The most common causes of mortality among neonates–prematurity and low-birth-weight, infections, asphyxia and birth trauma– can be prevented by simple cost-effective interventions that also benefit the mother (WHO, 2009). Furthermore, providing all infants with an essential package of newborn care including appropriate resuscitation, warmth, cleanliness and hygiene, clean cord care, and early exclusive breastfeeding also increases survival and reduces the proportion of surviving infants with disability (World Vision International, 2017).
Operations research studies have identified which interventions are likely to effectively reduce newborn mortality, but how these services should be scaled up, by whom, and at what cost is still being determined. Despite the close link between maternal and newborn health, separating maternal and newborn health indicators into two distinct parts may appear a false dichotomy when the antecedents of a poor pregnancy outcome, and the program interventions required to address these, may be the same for both mothers and babies. However, the purpose in doing so is to acknowledge growing awareness of the importance of newborn health and to highlight the fact that despite the many parallels between maternal and newborn health programs, important differences influence the way that programs are monitored and evaluated. These differences arise, not only because program interventions may vary, but because interventions that benefit both mothers and babies may differentially affect mortality. For example, because of their greater impact on newborn survival, interventions such as immunizing pregnant mothers against tetanus and detecting and treating STIs are more likely to be monitored in newborn health programs than in safe motherhood programs.
Methodological Challenges of Evaluating Newborn Health Programs
Some of the challenges of evaluating newborn health programs are similar to those for safe motherhood: the need to consider two outcomes, the large number of proximate determinants, and the difficulties of attributing causality to certain interventions because services are “bundled.” Even though newborn deaths are more frequent than maternal deaths and therefore easier to count, the several measurement challenges include, but are not limited to, the following:
- Countries define births, deaths, and “newborn period” in different ways, making valid international comparisons difficult.
Meaningful use of any indicator is only feasible when standard definitions are used and applied. The first challenge for managers of newborn health programs is the lack of a generally agreed-upon definition of the “newborn period.” In this database, the term newborn refers to the neonatal period (i.e., the first 28 completed days of life). The Neonatal and Perinatal Mortality document defines the perinatal and neonatal periods (WHO, 2006).
A second challenge is the different ways of aggregating newborn deaths according to the timing of the death. Typically, deaths are aggregated in the first month or first week of life or as fetal deaths. Many countries, however, define and record births and deaths in ways that may differ from the standard definitions of fetal, perinatal, and neonatal deaths recommended by WHO. Further difficulties may also arise because national birth and death criteria may be interpreted differently, and thus live births may be misclassified as fetal deaths and vice versa.
- Ideally, all deaths (including fetal, perinatal, and neonatal) should be counted, but in practice, counting neonatal deaths is often the only feasible approach.
Realistically, few programs record information on all neonatal, perinatal, and fetal deaths. In most developing countries, the majority of births and deaths occur at home (WHO, 1996b). Few countries have sufficiently well-developed vital registration systems that can provide valid and reliable information on all births and deaths in the community. Health information systems can only provide information on facility births and deaths and, in most settings, are also poorly developed. Most community-based programs, if they have the capacity to measure mortality at all, will generally only be able to collect valid data on neonatal deaths for reasons explained below.
The quality of newborn mortality data is poorer than the quality of data for other ages.
Evaluators require information on all births and deaths to derive valid measures of newborn health outcome. As mentioned above, few developing countries have sufficiently developed vital registration systems to provide this data and, in many settings, reporting is very incomplete due to institutional, social, and cultural barriers.
There are major biases in the way deaths are reported. Even in countries with well-developed registration systems, a bias exists towards the reporting of larger, older babies, whereas deaths of very small babies early in the neonatal period are often omitted. Fetal deaths are much less likely to be reported than deaths of live births (WHO, 1996b).
- Survey estimates of newborn mortality may not be suitable for short-term monitoring.
Prospective studies would provide the most reliable mortality rates but are too expensive for regular reporting purposes. In practice, the most reliable estimates of neonatal mortality are derived from large scale surveys that rely on the retrospective report of deaths in early infancy. Surveys focusing on live births provide estimates of neonatal mortality, but perinatal mortality estimates require complete pregnancy histories. Because many population-based surveys focus on obtaining demographic indicators that use live births in the denominator, there has been relatively less experience with the use of pregnancy histories (which collect information about stillbirths). The reliability of any survey estimate depends on the completeness of reporting, and underreporting is generally more pronounced for deaths in early infancy. Because of the relatively small numbers of deaths recorded in this type of survey, national neonatal mortality rates are usually presented for a period of five years before the survey, and sub-national estimates are presented for ten years before the survey. Lack of precision in the estimates may sometimes make it very difficult to assess the significance of small changes between surveys (Rutstein, 1999).
- Measuring perinatal and neonatal morbidity is very difficult.
Estimates of newborn morbidity are important for designing effective program interventions. As with safe motherhood, however, existing estimates of newborn morbidity are usually derived from facility data and are unlikely to reflect the true burden of morbidity in the community unless all births and deaths are institutional. Although community members can learn to diagnose illness in a sick newborn (Bang et al., 1999), illness is often difficult to recognize because babies usually present with relatively non-specific symptoms, such as poor feeding and lethargy. Assigning a cause of death may be difficult because many different diseases may present with the same symptoms, and many babies die at home before ever reaching medical attention. Few facilities have adequate diagnostic facilities when ill babies do eventually present for care.
- New program indicators are required at the individual, community, and facility level.
Process indicators are required for measuring the availability, accessibility, quality, and demand for services at the facility level where the provision of newborn health services has historically been overlooked. A national survey in Kenya, for example, showed that over one third of hospitals lacked even the most basic equipment for resuscitation (MOH, NCPD, and ORC Macro, 2000).
In addition to monitoring at the facility level, indicators are also required for monitoring and evaluating interventions at the individual and community level. Many infants become ill and die before ever reaching medical care. It is particularly important to develop indicators that help programs understand community knowledge, attitudes and behaviors in response to newborn illness and to determine which interventions are the most effective.
The newborn health indicators selected for this database measure both health outcomes and impacts as well as quality of care. Based primarily on the work of the Saving Newborn Lives Technical Working Group, many of the indicators measure how well the essential componenets of newborn care are being implemented — both at the facility and community level.
Bang, A.T., R.A. Bang, S.B. Baitule, M.H. Reddy, and M.D. Deshmukh. 1999. “Effect of Home-based Neonatal Care and Management of Sepsis on Neonatal Mortality: Field Trial in Rural India (comments).” Lancet 354, 9194: 1955-1961.
World Vision International. Intervention 2: Essential Newborn Care. Accessed July 5, 2017.
Narayanan I, Rose M, Cordero D, Faillace S, and Sanghvi T. The Components of Essential Newborn Care. Published by the Basics Support for Institutionalizing Child Survival Project (BASICS II) for USAID. Arlington, Virginia, June 2004.
Rutstein, S.O. 1999. “Guidelines for the MEASURE DHS+ Main Survey Report.” Demographic and Health Surveys. Baltimore, MD: Macro International Inc.
USAID. Essential Maternal and Newborn Health. 2009. Available at: https://www.usaid.gov/what-we-do/global-health/maternal-and-child-health/technical-areas/newborn-health
WHO and UNICEF. Countdown to 2015 decade report (2000-2010): taking stock of maternal, newborn and child survival. Geneva, 2010.
WHO. Newborns: reducing mortality. Fact sheet No333. August, 2009.
WHO. 1996b. Essential Newborn Care: Report of a Technical Working Group, April 1994. Maternal and Newborn Health/Safe Motherhood Unit. Division of Reproductive Health (Technical Support). Geneva: WHO.
WHO. Neonatal and Perinatal Mortality: Country, Regional and Global Estimates. 2006. Available at: http://apps.who.int/iris/bitstream/10665/43444/1/9241563206_eng.pdf