Indicator 15: Intrapartum stillbirth and very early neonatal death rate

In 2022, an estimated 2.3 million newborns died within the first 28 days of life,¹ most of whom died in the first week of life.² In addition, of the 1.9 million late gestation stillbirths that occurred in 2022, nearly half died during the intrapartum period after the onset of labor but before delivery.³⁴ About a third of stillbirths are related to intrapartum asphyxia while one quarter of neonatal deaths are also preceded by intrapartum asphyxia. Most asphyxia-related deaths can be prevented with good quality childbirth care and timely and effective neonatal resuscitation.⁵ Tracking changes in stillbirths and newborn mortality in facilities is therefore central to understanding the impact of EmONC service delivery.

The intrapartum stillbirth and very early neonatal death rate is proposed as a combined indicator of the quality of intrapartum and immediate postnatal care at an individual hospital level, as well as a facility-based impact indicator analyzed at sub-national and national levels. Measuring both intrapartum stillbirth and very early neonatal death mitigates some of the challenges with misclassification frequently seen when either of these components are measured and assessed separately.

Some of the neonatal deaths captured by this indicator are also included in the calculation for Indicator 16: Neonatal inpatient mortality rate. The distinctions between these two indicators are explained in Differences between Indicators 15 and 16.

The numerator for this indicator is the number of in-facility intrapartum stillbirths plus the number of very early neonatal deaths. In-facility intrapartum stillbirths are babies that were known to be alive at admission but suffered fetal death during labor or the birthing process in the facility.⁷ Very early neonatal deaths are deaths during the first 24 hours after a live birth (0-24 hours).⁸ The denominator is the total live births plus in-facility intrapartum stillbirths in the facility. The numerator and denominator should be calculated for the same time period and should capture in-born babies only.

Note that antepartum stillbirths (those occurring before onset of labor or the birthing process⁹) are not included in the numerator or the denominator since this indicator focuses on the quality of care provided during the intrapartum and immediate postnatal period. Data sources must therefore distinguish between antepartum and intrapartum stillbirths. Furthermore, ideally the numerator and denominator consist of in-facility stillbirths, i.e. where the baby was alive on admission. Assessing fetal heart tone on admission (e.g., with a hand-held doptone, an electric fetal heart amplifier) is recommended. However, even with routine use of doptone, false positives and negatives can occur, thus frequent training for health workers is critical.¹⁰

Misclassification can occur between very early neonatal deaths and intrapartum stillbirths. Since both groups are joined in the numerator, this does not affect the accuracy of this indicator. However, disaggregating intrapartum stillbirths and very early neonatal deaths can provide useful additional information (see ‘Analysis and interpretation’ section below), so if possible it is desirable to minimize misclassification between these two groups.

Data for this indicator are likely to be available in the HMIS. The data originate in patient records, maternity registers, pediatric registers, special newborn care unit registers, neonatal intensive care unit registers, and morgue records. Quality and completeness of data in registers varies but generally need improvement, particularly with misclassification between intrapartum stillbirths and early neonatal deaths,¹¹¹² hence why these are combined in this indicator. Initiatives such as Maternal and Perinatal Death Surveillance and Response (MPDSR) and tools such as the EN-MINI¹³ are designed to help countries strengthen and improve HMIS data.

For this guide it is proposed that this indicator be expressed as a rate per 1,000 births, but it is also commonly expressed as a percentage, much like a case fatality rate.

The stability of any rate depends on a large enough number in the numerator and accurate denominators. WHO cautions the use of rates when the numerator is small; this will affect the frequency of calculation. Even if data on stillbirths and very early neonatal deaths are collected routinely in the HMIS, the intrapartum stillbirth and very early neonatal death rate is most usefully calculated and analyzed on an annual basis.

There is no benchmark for this indicator; governments should set their own targets.

High rates indicate that the quality of obstetric or newborn care is suboptimal and may flag the need for additional support to healthcare workers to ensure that they are able to resuscitate a baby with breathing difficulties, diagnose fetal distress and act accordingly, and provide essential newborn care. High rates may also reflect poor monitoring of labor and delays in providing care for women during labor and delivery. Very low rates may indicate a problem with data quality, particularly underreporting and misclassification. Techniques to assess data quality for stillbirths can be found in the guideline Stillbirth Definition and Data Quality Assessment for HMIS, mentioned below in ’Useful links.’

Calculating this indicator at the national level masks variation across facility type, geographical access and characteristics of the catchment area.¹⁴ It is therefore useful to disaggregate data by variables such as: subnational administrative or geographic areas, type and managing authority of facility, EmONC facility classification, and readiness indicators.

Unlike some of the other indicators, this indicator can also be used for individual facility level monitoring of quality of care to help identify hotspots for prioritization and specific newborn related interventions.

Disaggregating this indicator by intrapartum stillbirths and very early neonatal deaths can be helpful. Thus, the full package of indicator and sub-indicators would include:

Evidence suggests that early neonatal deaths and intrapartum stillbirths can decline at different rates and rely on different interventions for their reduction. For example, drawing from data from Pregnancy Outcome Surveys in all facilities in Kigoma, Tanzania, and comparing changes over a 5-year period, the intrapartum stillbirth and early neonatal death rate declined from 2.5% to 1.4%. This decline was driven largely by the 58% decline in the intrapartum stillbirth rate.¹⁵

When a country has reliable access to data at birth and every baby is counted, we recommend further disaggregating data on intrapartum stillbirths and neonatal deaths by weight and/or gestational age to maximize the interpretation of the indicator. More detailed audits could look at subgroups more amenable to survival to optimize the focus on preventable stillbirths and early neonatal deaths. For example, those at highest risk (such as very small or very preterm) could be excluded using a threshold defined by the local context depending on their neonatal care capabilities. For example, data could be looked at by weight thresholds of ≥1500g, ≥2000g, ≥2500g or gestational age thresholds of ≥32 weeks or ≥37 weeks.

• UNICEF, IGME, LSHTM. Stillbirth Definition and Data Quality Assessment for Health Management Information Systems (HMIS).