Levels and causes of mortality in mothers and babies are intrinsically linked, occurring at the same time and often to the same mother–baby dyad, although mortality rates are substantially higher in babies. Measuring levels, trends and causes of maternal, neonatal and foetal mortality are important for understanding priority areas for interventions and tracking the success of interventions at the global, national, regional and local level. However, there are many measurement challenges.
This paper provides an overview of the definitions and indicators for measuring mortality in pregnant and post-partum women (maternal and pregnancy-related mortality) and their babies (foetal and neonatal mortality). We then discuss current issues in the measurement of the levels and causes of maternal, foetal and neonatal mortality, and present options for improving measurement of these outcomes. Finally, we illustrate some important uses of mortality data, including for the development of models to estimate mortality rates at the global and national level and for audits.
Highlights
- •
Levels and causes of mortality in mothers and babies are intrinsically linked.
- •
Measuring levels, trends and causes of maternal, neonatal and foetal mortality are interventions and track their success.
- •
There are standard definitions and indicators of measuring mortality in pregnant and post-partum women and their babies.
- •
Measurement challenges exist, awareness and attention to these would improve the measurement for these outcomes.
Introduction
Monitoring levels of maternal mortality has been a priority on the global health agenda. Millennium development goal (MDG) 5 aimed to reduce the maternal mortality ratio (MMR) by 75% between 1990 and 2015. However, measuring progress over this time period was challenging, primarily because of the scarcity of empirical data. Global tracking relied instead on modelled estimates to monitor the success . These estimates suggested that maternal mortality decreased by 44% worldwide in the MDG era . Similar challenges were faced in tracking foetal and neonatal mortality. Neonatal deaths were not explicitly mentioned in MDG 4, which sought to reduce under-5 child mortality by two-thirds, but they were increasingly recognised as comprising almost half of child mortality globally and progressing more slowly. Neonatal mortality was estimated to have decreased by 47% worldwide during this period . Stillbirths (late foetal deaths) were excluded from the MDG targets, and consequently received less attention, although the major associated burden has been quantified more recently . At the end of the MDG era, the number of deaths, albeit based on modelled estimates, remains unacceptably high: 303,000 maternal deaths , 2.6 million stillbirths (late foetal deaths) and 2.7 million neonatal deaths .
Measuring the levels and trends of maternal, neonatal and foetal mortality is important for quantifying disease burden, understanding risk factors and determinants, identifying priority areas for interventions, programmes and policies, and evaluating the success of interventions at the global, national, regional and local level . Knowing the biomedical causes of mortality in pregnant or recently delivered women, or in their babies, is essential to direct interventions to prevent such deaths. Unfortunately, there are many challenges to measurement, but there are also numerous potential options and solutions.
This paper provides an overview of current issues and options in measuring the levels and causes of maternal, foetal and neonatal mortality. We define these deaths and associated indicators, and then focus on the measurement methods, challenges and solutions, and where possible, present potential opportunities to improve measurement of maternal, neonatal and foetal deaths.
Definitions
To compare maternal, foetal and neonatal mortality across populations or over time requires standardised definitions for each outcome. These definitions were included in the 10th revision of the International Classification of Diseases (ICD-10) , as summarised in Table 1 and described below. Various dimensions of these definitions require an ability to assess pregnancy status of women, the timing of death in relation to delivery, gestational age (or alternatively birth weight or birth length) at delivery, vital status at the start of labour and at birth and, cause of death. The dimensions and critical time periods are shown schematically in Fig. 1 .
| Indicator | Primary threshold | Alternative threshold/definition |
|---|---|---|
| Maternal death | A death while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and the site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management but not from accidental or incidental causes | 90 days or 40 days |
| Late maternal death | A maternal death from direct or indirect obstetric causes >42 days, but <1 year, after termination of pregnancy | |
| Pregnancy-related death | A death while pregnant or within 42 days of termination of pregnancy, irrespective of the cause of death | |
| Early foetal death* | A baby born with no signs of life with birth weight ≥500 to <1000 g | Gestational age ≥22 weeks or length ≥25 cm (if birth weight is not available) |
| Late foetal death | A baby born with no signs of life with birth weight ≥1000 g | Gestational age ≥28 weeks or length ≥35 cm (if birth weight is not available) |
| Intrapartum foetal death | A foetal death occurring after the onset of labour, but before birth | A baby born with no signs of life and no evidence of skin maceration (fresh stillbirth) is commonly used as a surrogate marker |
| Antepartum foetal death | A foetal death occurring before the onset of labour | A baby born with no signs of life, with evidence of skin maceration (macerated stillbirth) is commonly used as a surrogate marker |
| Perinatal death | Composite indicator including all late foetal deaths and early neonatal deaths | Other composite indicators for perinatal deaths are described in the text |
| Early neonatal death | A death of a live-born baby at 0–6 days of age regardless of gestational age or birth weight | |
| Late neonatal death | A death of a live-born baby at 7–27 days of age regardless of gestational age or birth weight | |
| Neonatal death | A death of a live-born baby at 0–27 days of age regardless of gestational age or birth weight | Deaths in the first month of life |
Maternal and pregnancy-related mortality
‘Maternal death’, is defined in the ICD-10 as ‘the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management, but not from accidental or incidental causes’ ( Table 1 ). This definition encompasses direct obstetric deaths, when death occurs because of an obstetric complication such as haemorrhage or eclampsia, and indirect obstetric deaths, when an underlying, previously existing medical condition or non-obstetric medical condition developed during pregnancy, is aggravated by pregnancy. Since deaths that are accidental or incidental to the pregnancy need to be excluded, information on cause-of-death is required to apply this definition.
However, the definition of maternal death is conceptually problematic from a measurement perspective . Distinguishing indirect maternal death from incidental or accidental deaths during pregnancy or post partum is epidemiologically challenging, and consequently coding can be difficult. The decision whether a condition is aggravated by pregnancy or its management can either be made on a case-by-case basis, be ascribed to conditions based on epidemiologic data showing elevated incidence or case fatality in pregnant women with the condition compared with non-pregnant women, or be decided for entire classes of conditions (e.g., deaths from external causes). Guidance is provided but is not particularly helpful; for example, ICD maternal mortality (ICD-MM) instructs that HIV-related deaths should be classified as maternal when ‘there is an aggravating effect of pregnancy on HIV and the interaction between pregnancy and HIV is the underlying cause-of-death’ . It further states that if ‘the woman’s pregnancy status is incidental to the course of her HIV infection’ then the death should not be classified as maternal. Unfortunately, ICD-MM provides no guidance on how to identify when HIV disease progression has been accelerated by pregnancy, making the coding of these deaths very difficult, particularly in the absence of detailed data. Furthermore, epidemiological studies suggest that certain causes of death that are often excluded from maternal mortality estimates, such as suicide or homicide, are more likely to occur in certain subsets of pregnant women compared with non-pregnant women (notably amongst younger age groups) .
Although maternal death is the most widely used mortality definition in pregnant and post-partum women, the ICD-10 gives two further definitions that expand the deaths captured in two different ways. First, ‘late maternal death’ lengthens the time period to capture maternal deaths occurring from 42 days up to 1 year post partum. The 42-day post-partum cut-off has a weak evidence base, and a few studies show women remain at elevated risk for several months after delivery . Historically, a 90-day cut-off has been used , and some even argue that the increased mortality risk may extend beyond 1 year post partum . Second, ‘pregnancy-related death’ includes any ‘death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the cause-of-death’, without excluding accidental or incidental deaths in this specified time period. This latter definition only requires information on the timing of death in relation to pregnancy (or the end of pregnancy), and not on the cause of death ( Fig. 1 ). As such, pregnancy-related death is comparable to neonatal and foetal deaths that are also defined primarily by time periods, as described below.
Foetal and neonatal mortality
Live birth is defined in ICD-10 as ‘the expulsion or extraction from its mother of a product of human conception, irrespective of the duration of the pregnancy, which, after such expulsion or extraction, breathes or shows any other evidence of life such as beating of the heart, pulsation of the umbilical cord, or definite movement of voluntary muscles, whether or not the umbilical cord has been cut or the placenta is attached. Heartbeats are to be distinguished from transient cardiac contractions; respirations are to be distinguished from fleeting respiratory efforts or gasps’. The ICD-10 definition for neonatal death is the death of a live-born infant in the first 28 days of life; this definition is applied nearly universally ( Table 1 and Fig. 1 ).
Foetal death is ‘death prior to the complete expulsion or extraction from its mother of a product of human conception, irrespective of the duration of pregnancy and which is not an induced termination of pregnancy’. Death is indicated by the foetus not showing signs of being a live birth, as described above. ICD-10 defines foetal deaths as occurring from ≥500 g, or ≥22 weeks, or ≥25 cm only. Deaths before this period are spontaneous abortions or miscarriages in lay terminology. Definitions and terminology for foetal deaths are applied more inconsistently – especially amongst high-income countries with thresholds ranging from 20-week gestational age upwards ( Fig. 1 ) . ICD-10 distinguishes early from late foetal deaths using birth weight, gestational age or length criteria. ICD-10 recommends reporting both early and late foetal mortality rates, while WHO recommends using the stillbirth rate or late foetal death rate for international comparisons. The term ‘stillbirth’ is often used in clinical practice and common parlance to refer to any foetal death; however, it is used epidemiologically and in global estimates to refer to late foetal deaths only.
Since ICD-10 was developed several decades ago, the foetal death threshold was set to be based first on birthweight criterion then gestational age and then length. However, birthweight and gestational age thresholds do not give equivalent results. For example, in the USA the Stillbirth rate (SBR) would be 40% lower than with a 500-g threshold compared with a 22-week gestational threshold. Hence, the threshold should be based on one parameter as it is not accurate to assume equivalence. In practice, most health facilities could measure birth weight at the time of delivery, yet in reality less than half of the world’s births are weighed and fewer stillbirths are weighed. Gestational age can be difficult to assess without records from early ultrasound as the gold standard or dating based on last menstrual period . Nevertheless, we would argue that assessment of gestational age is essential to enable correct classification of a foetal death to the early or late category to allow for international comparisons. This is used in practice in middle- and high-income countries, and increasingly in low-income settings. It is proposed that the 11th ICD revision change to a gestational-age-based foetal death threshold, in line with most high-income country reporting.
Assessing the intrapartum versus antepartum timing of foetal death is another area where definitions may be applied differently in different settings with lower-level care. If evidence of a foetal heartbeat at the start of labour is not available, classification as intrapartum or antepartum often relies on an assessment of the skin of the baby (fresh vs. macerated), which is not a very reliable indicator of antepartum or intrapartum timing of foetal death .
Definitions
To compare maternal, foetal and neonatal mortality across populations or over time requires standardised definitions for each outcome. These definitions were included in the 10th revision of the International Classification of Diseases (ICD-10) , as summarised in Table 1 and described below. Various dimensions of these definitions require an ability to assess pregnancy status of women, the timing of death in relation to delivery, gestational age (or alternatively birth weight or birth length) at delivery, vital status at the start of labour and at birth and, cause of death. The dimensions and critical time periods are shown schematically in Fig. 1 .
| Indicator | Primary threshold | Alternative threshold/definition |
|---|---|---|
| Maternal death | A death while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and the site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management but not from accidental or incidental causes | 90 days or 40 days |
| Late maternal death | A maternal death from direct or indirect obstetric causes >42 days, but <1 year, after termination of pregnancy | |
| Pregnancy-related death | A death while pregnant or within 42 days of termination of pregnancy, irrespective of the cause of death | |
| Early foetal death* | A baby born with no signs of life with birth weight ≥500 to <1000 g | Gestational age ≥22 weeks or length ≥25 cm (if birth weight is not available) |
| Late foetal death | A baby born with no signs of life with birth weight ≥1000 g | Gestational age ≥28 weeks or length ≥35 cm (if birth weight is not available) |
| Intrapartum foetal death | A foetal death occurring after the onset of labour, but before birth | A baby born with no signs of life and no evidence of skin maceration (fresh stillbirth) is commonly used as a surrogate marker |
| Antepartum foetal death | A foetal death occurring before the onset of labour | A baby born with no signs of life, with evidence of skin maceration (macerated stillbirth) is commonly used as a surrogate marker |
| Perinatal death | Composite indicator including all late foetal deaths and early neonatal deaths | Other composite indicators for perinatal deaths are described in the text |
| Early neonatal death | A death of a live-born baby at 0–6 days of age regardless of gestational age or birth weight | |
| Late neonatal death | A death of a live-born baby at 7–27 days of age regardless of gestational age or birth weight | |
| Neonatal death | A death of a live-born baby at 0–27 days of age regardless of gestational age or birth weight | Deaths in the first month of life |
Maternal and pregnancy-related mortality
‘Maternal death’, is defined in the ICD-10 as ‘the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management, but not from accidental or incidental causes’ ( Table 1 ). This definition encompasses direct obstetric deaths, when death occurs because of an obstetric complication such as haemorrhage or eclampsia, and indirect obstetric deaths, when an underlying, previously existing medical condition or non-obstetric medical condition developed during pregnancy, is aggravated by pregnancy. Since deaths that are accidental or incidental to the pregnancy need to be excluded, information on cause-of-death is required to apply this definition.
However, the definition of maternal death is conceptually problematic from a measurement perspective . Distinguishing indirect maternal death from incidental or accidental deaths during pregnancy or post partum is epidemiologically challenging, and consequently coding can be difficult. The decision whether a condition is aggravated by pregnancy or its management can either be made on a case-by-case basis, be ascribed to conditions based on epidemiologic data showing elevated incidence or case fatality in pregnant women with the condition compared with non-pregnant women, or be decided for entire classes of conditions (e.g., deaths from external causes). Guidance is provided but is not particularly helpful; for example, ICD maternal mortality (ICD-MM) instructs that HIV-related deaths should be classified as maternal when ‘there is an aggravating effect of pregnancy on HIV and the interaction between pregnancy and HIV is the underlying cause-of-death’ . It further states that if ‘the woman’s pregnancy status is incidental to the course of her HIV infection’ then the death should not be classified as maternal. Unfortunately, ICD-MM provides no guidance on how to identify when HIV disease progression has been accelerated by pregnancy, making the coding of these deaths very difficult, particularly in the absence of detailed data. Furthermore, epidemiological studies suggest that certain causes of death that are often excluded from maternal mortality estimates, such as suicide or homicide, are more likely to occur in certain subsets of pregnant women compared with non-pregnant women (notably amongst younger age groups) .
Although maternal death is the most widely used mortality definition in pregnant and post-partum women, the ICD-10 gives two further definitions that expand the deaths captured in two different ways. First, ‘late maternal death’ lengthens the time period to capture maternal deaths occurring from 42 days up to 1 year post partum. The 42-day post-partum cut-off has a weak evidence base, and a few studies show women remain at elevated risk for several months after delivery . Historically, a 90-day cut-off has been used , and some even argue that the increased mortality risk may extend beyond 1 year post partum . Second, ‘pregnancy-related death’ includes any ‘death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the cause-of-death’, without excluding accidental or incidental deaths in this specified time period. This latter definition only requires information on the timing of death in relation to pregnancy (or the end of pregnancy), and not on the cause of death ( Fig. 1 ). As such, pregnancy-related death is comparable to neonatal and foetal deaths that are also defined primarily by time periods, as described below.
Foetal and neonatal mortality
Live birth is defined in ICD-10 as ‘the expulsion or extraction from its mother of a product of human conception, irrespective of the duration of the pregnancy, which, after such expulsion or extraction, breathes or shows any other evidence of life such as beating of the heart, pulsation of the umbilical cord, or definite movement of voluntary muscles, whether or not the umbilical cord has been cut or the placenta is attached. Heartbeats are to be distinguished from transient cardiac contractions; respirations are to be distinguished from fleeting respiratory efforts or gasps’. The ICD-10 definition for neonatal death is the death of a live-born infant in the first 28 days of life; this definition is applied nearly universally ( Table 1 and Fig. 1 ).
Foetal death is ‘death prior to the complete expulsion or extraction from its mother of a product of human conception, irrespective of the duration of pregnancy and which is not an induced termination of pregnancy’. Death is indicated by the foetus not showing signs of being a live birth, as described above. ICD-10 defines foetal deaths as occurring from ≥500 g, or ≥22 weeks, or ≥25 cm only. Deaths before this period are spontaneous abortions or miscarriages in lay terminology. Definitions and terminology for foetal deaths are applied more inconsistently – especially amongst high-income countries with thresholds ranging from 20-week gestational age upwards ( Fig. 1 ) . ICD-10 distinguishes early from late foetal deaths using birth weight, gestational age or length criteria. ICD-10 recommends reporting both early and late foetal mortality rates, while WHO recommends using the stillbirth rate or late foetal death rate for international comparisons. The term ‘stillbirth’ is often used in clinical practice and common parlance to refer to any foetal death; however, it is used epidemiologically and in global estimates to refer to late foetal deaths only.
Since ICD-10 was developed several decades ago, the foetal death threshold was set to be based first on birthweight criterion then gestational age and then length. However, birthweight and gestational age thresholds do not give equivalent results. For example, in the USA the Stillbirth rate (SBR) would be 40% lower than with a 500-g threshold compared with a 22-week gestational threshold. Hence, the threshold should be based on one parameter as it is not accurate to assume equivalence. In practice, most health facilities could measure birth weight at the time of delivery, yet in reality less than half of the world’s births are weighed and fewer stillbirths are weighed. Gestational age can be difficult to assess without records from early ultrasound as the gold standard or dating based on last menstrual period . Nevertheless, we would argue that assessment of gestational age is essential to enable correct classification of a foetal death to the early or late category to allow for international comparisons. This is used in practice in middle- and high-income countries, and increasingly in low-income settings. It is proposed that the 11th ICD revision change to a gestational-age-based foetal death threshold, in line with most high-income country reporting.
Assessing the intrapartum versus antepartum timing of foetal death is another area where definitions may be applied differently in different settings with lower-level care. If evidence of a foetal heartbeat at the start of labour is not available, classification as intrapartum or antepartum often relies on an assessment of the skin of the baby (fresh vs. macerated), which is not a very reliable indicator of antepartum or intrapartum timing of foetal death .
Indicators
Counting numbers of maternal, foetal and neonatal deaths can identify countries, regions or subgroups with the largest numeric burden, but often we are also interested in knowing where the risk of such deaths is highest. For example, due to its large population, India has a much greater number of maternal deaths than Sierra Leone, yet the risk of a woman in India dying of maternal causes is much lower than in Sierra Leone . Identifying the risk faced by individual women or babies requires the numbers of deaths be considered in relation to a denominator at risk of these deaths. Below we have described commonly used indicators of risk, as well as others used in mortality measurement.
Maternal indicators
Assessing the risk of maternal or pregnancy-related mortality requires relating the number of such deaths in a given time period and a given country or area, to the number of women at risk. The ideal denominator for this – the number of pregnant woman entering into the pregnancy/post-partum period, or time spent pregnant or post-partum – is difficult to obtain without conducting prospective studies of large groups of women. Instead, routine data sources are commonly used to calculate MMR: the number of maternal deaths per 100,000 live births in a given time period: (number of deaths/live births) × 100,000. This live-birth denominator approximates the number of pregnancies, but excludes women who have miscarriages, induced abortions or stillbirths, while women having multiple live births (e.g., twins or triplets) are counted multiple times in the denominator. In some settings, all maternity cases, including those resulting in foetal deaths, and even induced abortions, are included in the denominator .
Three additional, less commonly reported, indicators are defined below:
- 1.
MMR (or pregnancy-related): deaths per 100,000 women aged 15–49 per year (midpoint population)
- 2.
Lifetime risk of maternal (or pregnancy-related) death: the probability that a 15-year-old girl will die eventually from maternal (or pregnancy-related) causes, assuming that current levels of fertility and mortality (including maternal (or pregnancy-related) mortality) do not change in the future, considering competing causes of death .
- 3.
Proportion of deaths: proportion of maternal (or pregnancy-related) deaths among all deaths of women of reproductive age.
The MMR (or pregnancy-related) and the level of fertility influence all three indicators. For any given MMR, the higher the level of fertility, the higher the level of the three indicators. The lifetime risk indicator and the proportion of deaths are also influenced by death rates among non-pregnant/non-post-partum women: all else being equal, the higher the death rates in non-pregnant/non-post-partum women, the lower these two indicators will be.
Foetal and neonatal indicators
Mortality indicators for outcomes in babies are usually measured per 1000 births. Neonatal mortality rates use live births as the denominator: (number of neonatal deaths)/(live births) × 1000. Foetal mortality rates can be calculated as (number of foetal deaths)/(live births + foetal deaths) × 1000. A combined indicator for all ‘perinatal deaths’ is used, which includes all late foetal deaths (≥1000 g or ≥28 weeks) and all early neonatal deaths (days 0–6): (number of perinatal deaths)/(live births + foetal deaths) × 1000.
It is recommended that all deaths in babies <28 days of age, whether in utero above a specified threshold or in the neonatal period, are recorded by gestational age, birth weight and timing (antepartum or intrapartum and day of neonatal death). Such reporting of outcomes is of programmatic relevance. For example, the ‘intrapartum stillbirth and early neonatal death indicator’, may be used to monitor improvements of the quality of obstetric and newborn care provided at birth. It can be calculated at a facility level as (intrapartum stillbirths + neonatal deaths within the first 24 h of life (≥2500 g))/(live births + foetal deaths (≥2500 g)) .
Another, less frequently used, measure is the ‘prospective foetal mortality rate’: (number of foetal deaths at a gestational age per 1000 foetal deaths at that gestational age or greater, plus live births). This is a more accurate denominator for those at risk, and provides an estimate of the risk of foetal death at a given gestational age . In high-income settings, this indicator has been used to compare the risk of foetal death with the neonatal mortality rate to determine the optimal gestational age for delivery .
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
