1. Introduction

We produce child and infant mortality statistics that are published under the National Statistics logo, the designation guaranteeing that those outputs have been produced to high professional standards set out in the Code of Practice for Statistics, and have been produced free from any political interference.

This page provides information on the collection, production and quality of our Child mortality (death cohort) tables and Infant mortality (birth cohort) tables.

The main differences between these two sets of tables are that:

  • the death cohort tables are based on the year the death occurred while the birth cohort are based on the year infant was born, whether they died in the same year or the following year

  • the death cohort tables cover both infant deaths (under 1 year of age) and child deaths (between 1 and 15 years of age), while the birth cohort tables are for infants only

In both cases, the death record has been linked to the relevant birth registration record so that tables broken down by information available on the birth registration (but not the death registration) can be produced. We also link the data to a third source, the birth notification records provided by the NHS. These records contain information about gestation length and ethnicity of the child which is not available on the birth registration.

Figures from the death cohort tables are described and explained in a single Child and infant mortality publication, because they provide the most timely statistics.

The Child and infant mortality statistics Quality and Methodology Information report contains important information on:

  • the strengths and limitations of the data and how it compares with related data
  • the quality of the output, including the accuracy of the data
  • uses and users of the data
  • how the output was created

The User guide to birth statistics provides more detailed information on the collection, production and quality of birth statistics based on birth registration data.

Following the results of an infant mortality user consultation carried out in spring and summer of 2017, we combined the Birth cohort tables for infant deaths and the Pregnancy and ethnic factors influencing births and infant mortality into one publication called Infant mortality (birth cohort) tables. In addition, there have been revisions to both these tables and the death cohort tables to improve presentation and to meet our user needs. More detail is available in the response to the consultation.

There are also some changes that have been made to the 2019 release, published in February 2021:

  • the death cohort has now been additionally linked to birth notification data, to allow for timely analysis of infant mortality by gestational age
  • the death cohort tables now include all infant deaths rather than just those infant deaths that linked to their corresponding birth registration
  • there have been several updates made to the death cohort tables to improve clarity for users and increase consistency across tables and across other ONS publications; no existing data have been removed, unless they can be found in other ONS publications

These changes are explained further in Section 7: Linkage of births and deaths.

The Office for National Statistics policy on protecting confidentiality in birth and death statistics is available.

Nôl i'r tabl cynnwys

2. Occurrences and registrations

Deaths

The majority of the Office for National Statistics death statistics are based on when a death is registered rather than when it occurred. In most cases, this makes little difference in monitoring trends because the difference between when the death occurred and when it is registered is small.

However, in the case of infant deaths, this delay can be much longer and many deaths occurring in a year will be registered after that year has finished. Table 1 shows that between 10% and 12% of infant deaths that occurred in a reference year were registered the following year. This proportion is larger than for all deaths, for which approximately 5% that occurred in a reference year were registered in a different year. Figures for all deaths that occurred in a reference year by the year of registration are available in Table 6 of the impact of impact of registration delays to mortality statistics tables.

We recommend using occurrence-based statistics to monitor infant mortality patterns over time. The downside to this is that we take the data extract used to produce the statistics some months after the end of the reference year to ensure we capture as many of the deaths that actually occurred in that year as possible. We will still miss some later registrations but this is a balance we must strike between accuracy and timeliness.

Births

The annual totals of live births and stillbirths included in these tables are derived from the standard annual extract of live births and stillbirths. This extract includes all births that occurred and were registered in England and Wales in a calendar year, but also include a small number of late registrations from the previous year.

Nôl i'r tabl cynnwys

3. Child and infant deaths

In Child mortality (death cohort) tables, child deaths are defined as between 1 and 15 years of age.

Infant deaths (under 1 year) can be broken down as follows:

Embed code

Stillbirths

The Stillbirth (Definition) Act 1992 defines a stillbirth as:

“A child which has issued forth from its mother after the twenty-fourth week of pregnancy, and which did not at any time after becoming completely expelled from its mother breathe or show other signs of life”.

This definition has been in use since 1 October 1992. Prior to this, the Births and Deaths Registration Act 1953 defined a stillbirth as previously stated, but at 28 or more weeks completed gestation. Figures for stillbirths from 1993 are not comparable with those for previous years. The effect of this change on figures for 1992 is analysed in the annual volume of birth statistics for that year (Office of Population, Censuses and Surveys 1994).

Registration and certification of stillbirths, neonatal and infant deaths

General information about the registration and certification of stillbirths, neonatal and infant deaths in England and Wales can be found in the User guide to mortality statistics. It also provides information about the specific details collected when a death is certified and registered.

Death rates

We calculate the following rates for England and Wales in the Child and infant mortality publication:

  • stillbirth rate
  • perinatal mortality rate
  • early neonatal mortality rate
  • neonatal mortality rate
  • postneonatal mortality rate
  • infant mortality rate
  • age-specific child mortality rate

More information on how we calculate these rates is available in Section 8 of this report.

Nôl i'r tabl cynnwys

4. Referral to the coroner

While the majority of infant deaths are certified by a doctor, some may be reported to the coroner by the certifying doctor or the registrar. The circumstances under which a death has to be referred are covered in the User guide to mortality statistics.

Table 2 provides the numbers of deaths by method of certification for those infants aged under one year, in 2019.

The conditions for certifying neonatal deaths are as for other deaths – that the doctor should have been in attendance during the deceased’s last illness, should have seen the patient prior to death or seen the body, and that the cause of death is known.

Only a small proportion of neonatal deaths require inquests. In 2019, 85% of neonatal deaths were certified by a doctor, 13% by a coroner and only 6% were subject to a coroner’s inquest. This reflects the fact that nearly all neonatal deaths occur in hospitals, and that infant deaths can be certified as to the result of sudden infant death syndrome (SIDS) without being subject to inquest.

Nôl i'r tabl cynnwys

5. Area coverage and base populations

Area coverage

Births and deaths to residents of England and Wales that occur and are registered outside of England and Wales are excluded.

Births and deaths registered in England and Wales to persons whose usual residence is outside England and Wales are included in any total figures for England and Wales, but are excluded from any subdivision of England and Wales.

Figures for live births and stillbirths to women whose usual residence is outside of England and Wales can be found in Tables 5 and 6 in the publication Birth characteristics.

Table 3 provides the number of infant deaths that occurred in England and Wales by calendar year, for those infants who were not usually resident in England and Wales.

Base populations

The population figures used to calculate child mortality rates are mid-year estimates of the resident population of England and Wales based on the census of population. The Office for National Statistics mid-year population estimates are based on updates from the most recent census allowing for births, deaths, net migration and ageing of the population.

The population estimates used for the calculation of mortality rates are the latest consistent estimates available at the time of production. Further information on population estimates and their methodology is also available.

UK comparisons

Considerations need to be made when drawing comparisons between infant mortality statistics for England and Wales and statistics for Scotland and Northern Ireland.

It is a legal requirement across the UK that all births and deaths are registered, which means that infant mortality can be expressed as the number of infant deaths per 1,000 live births. This is an internationally recognised measure of infant mortality and means that fair comparisons can be made over time and between countries.

In England and Wales, there can be long delays between when an infant dies and when the death is registered. Deaths should be registered within five days, unless they are referred to a coroner for investigation, where the delay between the date of occurrence and date of registration can be longer. The Office for National Statistics (ONS) produces figures based on when the death occurred and also when it was registered.

Statistics based on death registrations are timelier, however, deaths registered in any given year will include deaths that happened in that year as well as previous years. Statistics based on death occurrences cannot be produced until later as there must be time to account for late registrations. However, these statistics are the most accurate representation of deaths in a given year. Infant deaths in England and Wales based on death occurrences are the preferred figures and the ones used to monitor trends in infant mortality over time.

National Records of Scotland (NRS) publish infant death figures in Scotland based on date of registration. This is because registration delays are shorter in Scotland than in England and a Scottish series based on date of occurrence would be almost identical to the one based on date of registration.

Similarly, Northern Ireland Statistics and Research Agency (NISRA) present all of their data on infant deaths using the date of registration rather than the date of occurrence, acknowledging that infant deaths are likely to be referred to the coroner, which means that the death may be registered later.

Figure 1 shows infant mortality rates per 1,000 live births for England, Wales, Scotland and Northern Ireland. The England and Wales rates are based on death occurrences, the figures presented in our death cohort tables. The figures for both Scotland and Northern Ireland are based on death registrations.

Nôl i'r tabl cynnwys

6. Cause of death

In England and Wales, stillbirths and neonatal deaths are registered using a special death certificate introduced in 1986 (Annex D), which enables reporting of relevant diseases or conditions in both the infant and the mother (see section 11.3 of the User Guide to Mortality Statistics for more detail). Equal weighting is given to main conditions recorded in the infant and in the mother, so it is no longer possible to identify a single underlying cause of death for neonatal deaths and stillbirths. For postneonatal deaths (between 28 days and one year), a single underlying cause of death can be reported using the standard death certificate.

The Office for National Statistics (ONS) developed a hierarchical classification system (PDF, 72KB) producing broad cause groups to enable direct comparison between neonatal and postneonatal deaths. This classification is referred to as the ONS cause groups, and allows the death to be assigned to a specific category, based on the likely timing of the damage leading to the death.

A computer algorithm assigns any mention, in the case of neonatal deaths, and underlying cause in the case of postneonatal deaths, to the first appropriate class of the following mutually exclusive categories:

Before the onset of labour:

1. congenital anomalies
2. antepartum infections
3. immaturity related conditions

In, or shortly after labour:

4. asphyxia, anoxia, or trauma

Postnatal:

5. external conditions
6. infections
7. other specific conditions
9. sudden infant deaths

Unclassified:

0. other conditions

A similar algorithm is used for stillbirths.

The grouping of ICD-10 codes into these nine categories for neonatal and postneonatal deaths is shown in Annexes B and C respectively. Corresponding groupings for stillbirths are shown in Annex A. (Annexes A.1, B.1 and C.1 refer to 2001 to 2010; Annexes A.2, B.2 and C.2 refer to 2011 to 2013 and Annexes A.3, B.3 and C.3 refer to 2014 onwards).

However, for the data years 2001 to 2013, postneonatal deaths were assigned to the ONS cause groups based on mentions rather than underlying cause.

The ONS cause groups were revised in 2014 and only figures since then are comparable with the latest data.

Nôl i'r tabl cynnwys

7. Linkage of births and deaths

Since 1975, infant death records have been linked to their corresponding birth registration record. This is because the birth registration includes a considerable amount of information about the parent(s) that is not available on the death record. This enabled analysis of certain risk factors and demographic characteristics that would otherwise be unavailable, including:

  • age of each parent
  • number of previous children born to the mother
  • country of birth of parents
  • place of birth
  • whether the baby was a singleton or multiple birth

Up until the 2019 data year, the death cohort only included infant deaths that linked to their corresponding birth registration. Any death cohort tables that presented analysis of birth registration variables presented the total number of linked infant deaths, that is infant deaths that were successfully linked to their birth registration.

However, the basis of the death cohort has been changed for the 2019 data year, published in February 2021. The death cohort now includes all infant deaths, not just infant deaths that linked to a corresponding birth registration. Each year there are approximately 3% of infant deaths that cannot be linked to a birth registration record. These records consist of deaths of babies born outside England and Wales, foundlings, adopted children and deaths that should have been linked but for which no birth registration record could be found.

In 2019, for death cohort tables that analyse birth registration variables such as age of mother, any infant deaths that did not successfully link to their birth registration are either included in a “Not stated” category, or where rates are being produced on linked infant deaths they are presented separately.

Linking death cohort data to birth notification

The birth notification is a document completed by the doctor or midwife present at the birth. This data source is supplied to the ONS by the NHS and includes information on gestational length and ethnicity of the baby (as defined by the mother) that is not available on the birth registration.

Our birth cohort tables use the birth notification as the basis for a dataset of all babies born in a reference year, linked to their birth registration and then linked to infant death records to determine whether the infant died before their first birthday.

Unlike our birth cohort dataset, our death cohort dataset has not been routinely linked to birth notification information. However, for the Child and infant mortality publication covering infant deaths occurring in 2019, the death cohort has been linked to the birth notification to enable new, more timely statistics on infant mortality by gestational age.

Infant death occurrences for 2019 were linked to their corresponding birth registration and their birth notification. The linkage has also been applied to previous data years from 2007 onwards, which is the earliest year for which linkage to the birth notification is possible. Some death records will not have linked to the birth registration and therefore will not have linked to the birth notification. Table 4 shows the linkage rate of the death cohort to the birth notification across these years.

Socio-economic classification as defined by occupation

For the child and infant mortality publications, the information on occupation and employment is used to derive National Statistics Socio-economic Classification (NS-SEC). To be able to do this, the occupation information of parent(s) from the birth and death certificate must first be coded. For death certificates for infant deaths, and stillborn babies, this coding is completed for all records. However, for live births, only 10% are coded as this is deemed sufficient quality for statistical analysis.

This means that the figures presented in any tables that show NS-SEC based on live births are calculated by multiplying the breakdowns for the 10% of coded records by 10. As a result, the sum of the NS-SEC breakdowns do not match the grand total, which is based on the full birth and death extracts.

From 1991 to 2000, occupation was coded using the Standard Occupational Classification SOC90, and occupation codes were allocated to the Registrar General’s Social Class. Since 2001, the National Statistics Socio-economic Classification (NS-SEC) has categorised the socio-economic classification of people.

The Standard Occupational Classification is revised every 10 years, and in 2011 SOC2010 replaced SOC2000. A report outlining the impact of re-basing the NS-SEC on SOC2010 is available.

The number of classes used will depend on both the analytical purposes and the quality of available data.

Up until the 2011 data year, the ONS published child mortality and birth statistics by NS-SEC using the father’s NS-SEC. Historically, this decision was based on the premise that many mothers either did not have a paid occupation or chose not to state their occupational details at birth registration. Following wider societal changes, this premise is now considered out of date (and has been for years prior to 2011). However, to ensure comparability in our statistics we continued to produce NS-SEC breakdowns on this basis until 2011.

From the 2012 data year, the ONS has used the combined method for reporting NS-SEC for birth and child mortality statistics (using the most advantaged NS-SEC of either parent, and creating a household-level classification rather than just using the father’s classification). More information can be found in a A combined approach to National Statistics socio-economic classification.

Births within or outside marriage or civil partnership, and sole and joint registration

Since 1 September 2009, following the implementation of the Human Fertilisation and Embryology Act (2008), same-sex female couples have been able to register the birth of a child as mother and second parent. The Act also made provision for two men to be officially recognised as the parents of a child through the provision of a parental order, obtainable through the courts.

Because of small numbers, births registered within a civil partnership are included with births registered within marriage. Births registered to a same-sex couple outside a civil partnership are combined with births outside marriage. Given the relatively small numbers of births registered to same-sex couples, the impact on statistics is negligible.

A birth within marriage or civil partnership is that of a child born to parents who were lawfully married or in a civil partnership either:

  • at the date of the child’s birth, or
  • when the child was conceived, even if they later divorced or were granted a civil partnership dissolution or the father or second parent died before the child’s birth

Births occurring outside marriage or civil partnership may be registered either jointly or solely. A joint registration records details of both parents, and requires them both to be present. A sole registration records only the mother’s details. In a few cases, a joint registration is made in the absence of the father or second parent if an affiliation order or statutory declaration is provided.

Information from the birth registration is used to determine whether the mother and father or second parent jointly registering a birth outside marriage or civil partnership, were usually resident at the same address at the time of registration. Births with both parents at the same address are identified by a single entry for the informant’s usual address, while different addresses are identified by two entries.

Some infants born outside marriage are deemed to have been born within marriage when the natural parents subsequently marry between the infant’s birth and death. Birth registrations do not identify children whose parents marry after the birth of the child. All relevant tables in Child mortality (death cohort) tables and Infant mortality (birth cohort) tables relate to marital status at birth. This ensures that the numerators and denominators used to calculate rates are compatible.

Mother’s country of birth

Parents’ country of birth for children born in England and Wales has been recorded at birth registration since 1969, but these data have only been available for an infant mortality analysis of social factors since 1975 when routine linkage was started. A breakdown of the mother’s country of birth groupings can be found alongside the published tables.

Birthweight

Birthweight is measured in grams. For live birth registrations, the birthweight is passed electronically to the ONS from the notification by the midwife or doctor in attendance at the birth. These details are then supplied to the registrar. For stillbirths, details of the weight of the fetus are supplied on a certificate or notification by a doctor or midwife. The certificate or notification is then taken by an informant to the registrar.

In cases where no birthweight is recorded, the birth is included in the total “all weights” but not distributed among the individual categories. Any remaining missing birthweights are included in the “Not stated” total for the relevant tables containing birthweight. Annual figures for records where the birthweight was not recorded for live births and stillbirths can be found in the User guide to birth statistics.

In 2020, the methodology towards birthweights was adjusted. From the 2019 data year onwards, implausible birthweights have been removed from individual categories but are included in the total "all weights". Assessment of what an implausible birthweight was based on the recorded birthweight and gestational age. More detailed information about this change can be found in Section 4 of the User guide to birth statistics.

Number of previous children

In May 2012, the ONS implemented a legislative change to improve the statistical information collected at birth registration in England and Wales. Two amendments were made to the Population (Statistics) Act 1938 – the legislation which requires registrars to collect confidential information for statistical purposes. The changes were made within the Welfare Reform Act 2009.

The two amendments mean that:

  • information is now collected at all birth registrations on the total numbers of previous live births and previous stillbirths that the mother has had (not just those with the current or former husband); this has simplified the question asked by registrars and provides improved coverage
  • information is now collected at all birth registrations on either: whether the mother has been previously married or in a civil partnership (if she is currently married or in a civil partnership); or whether the mother has ever been married or in a civil partnership (if she is not currently married or in a civil partnership)

This brings the birth registration process more in line with equality legislation.

Prior to May 2012, information on the number of previous children with a current or former husband and whether the mother had previously been married was only collected for births occurring within marriage.

When the Population (Statistics) Act came into force in 1938, only 4% of live births in England and Wales occurred outside marriage, so the information required was collected for nearly all mothers. However by 2011 nearly half of births (47%) took place outside marriage or civil partnership and so the legislation no longer reflected modern society.

Only minor changes were made to published tables for 2012 and 2013 as the first full year of new data was 2013, but some childhood deaths in this year will relate to births in 2012 prior to the changes being implemented. The main improvements resulting from the amendments to the Population (Statistics) Act 1938 have been introduced to published tables for child deaths occurring in 2014. Figures for 2014 onwards are not comparable with previous years.

Gestation

Gestational age is measured in completed weeks. For stillbirths, gestation is recorded at birth registration and is therefore available on our annual births datasets.

For live births and infant deaths, gestation is not recorded on the birth registration. For these, gestation comes from the birth notification which is linked to corresponding death records in our annual Infant mortality (birth cohort) tables and Child mortality (death cohort) tables.

Gestational age is grouped as follows:

Embed code

Mother’s age

For 2013 to 2017, if the mother’s age was missing, it was imputed using the most recently processed complete record of similar characteristics to the incomplete record. However, the improvement of these imputations on the quality of the statistics was unclear given the small number of records this affects.

Imputation of mother’s age was discontinued in March 2018 to make processing more efficient and our methods easier for users to understand. The remaining records where the mother’s age is missing are now categorised as “not stated” in our tables. This affects 2018 births data onwards.

Nôl i'r tabl cynnwys

8. Calculating infant and child mortality rates

The rates presented in these publications are described in this section.

Stillbirth rate:

Perinatal mortality rate:

Early neonatal mortality rate:

Neonatal mortality rate:

Postneonatal mortality rate:

Infant mortality rate:

Age-specific child mortality rate:


Significance testing

Within this bulletin, a change which is described as statistically significant has primarily been assessed using confidence intervals. For infant mortality data where we have all the death records, confidence intervals help tell the difference between a change caused by random fluctuations between years, and a real change in the infant mortality rate. If the confidence interval around a figure does not overlap with the interval around another, we can say with more confidence that the difference is likely to be a real change rather than simply down to chance.

Nôl i'r tabl cynnwys

9. Further information

Our website (www.ons.gov.uk) provides a comprehensive source of freely available vital statistics and Office for National Statistics (ONS) products. More information on our website can be obtained from the contact numbers and addresses found in this section.

Special extracts and tabulations of child mortality data for England and Wales are available to order (subject to legal frameworks, disclosure control, resources and our charging policy, where appropriate). Such enquiries should be made to the Office for National Statistics (ONS) Vital Statistics Outputs Branch to Health.Data@ons.gov.uk or +44 (0)1329 444 110). All user requested data will be published on the website.

Other sources of information on births and deaths

Additional information on background to mortality data that we publish, together with the quality of mortality data can be found in Sections 16 and 12 of the User guide to mortality statistics. Further information and background on birth statistics can be found in the User guide to birth statistics.

Other sources of data on births and deaths

Deaths occurring in a given year

Deaths registered in a given year

Summary data for infant mortality in England and Wales are available in the Deaths registrations summary tables. A geographical breakdown of infant death numbers and rates by local authority and county level is available in Deaths registered in England and Wales by area of usual residence.

The Vital statistics in the UK: births, deaths and marriages provide annual infant mortality data for the United Kingdom and its constituent countries.

Other UK countries

For infant mortality data for other UK countries please see the latest infant death statistics for Northern Ireland and the latest infant death statistics for Scotland.

Births

The Births summary tables, England and Wales provide the main summary statistics for live births in England and Wales.

Other useful information

The ONS response to the review of infant mortality statistics that took place between 20 April and 20 July 2017 is available.

Results from the ICD-10 bridge coding study for stillbirths and neonatal deaths

Disclosure Control Policy for Birth and Death Statistics

We welcome feedback from users on the content, format and relevance of child/infant mortality outputs.

Nôl i'r tabl cynnwys

10. Glossary

Antepartum

Occurring just before birth.

Cause groups

The ONS cause groups is another term used for “Hierarchical classification”; see the relevant subheading in this section.

Child

Children aged between 1 and 15 years.

Congenital anomaly

A structural or functional abnormality of the human body that develops before birth.

Coroner

Public official responsible for the investigation of violent, sudden or suspicious deaths.

Early neonatal

Relating to infants aged under seven days.

Hierarchical classification

The ONS’s method for classifying the causes of neonatal deaths and stillbirths, made up of groups of International Classification of Diseases and Related Health Problems (ICD) codes referred to as “ONS cause groups”.

ICD

International Classification of Diseases and Related Health Problems.

Infant

Child aged under one year.

Inquest

Inquiry into the cause of an unexplained, sudden or violent death, held by a coroner.

Linkage

The matching of infant death records to their corresponding birth registration record or birth notification record.

Neonatal

Relating to infants aged under 28 days.

NS-SEC

National Statistics Socio-economic Classification categorises the socio-economic classification of people, and has replaced the Registrar General’s Social Class and the Socio-economic Group (SEG).

Occurrences

Number of deaths according to the date on which the death occurred.

Perinatal

Stillbirths and early neonatal.

Postneonatal

Relating to infants aged between 28 days and 1 year.

Registrar

Local authority employee responsible for the registration of births, deaths, marriages and civil partnerships.

Registrations

Number of deaths according to the date on which the deaths were registered.

SOC 2010

Standard Occupational Classification 2010 is the current occupational classification. SOC2010 codes, details of employment status and size of organisation are required for the derivation of NS-SEC. See NS-SEC.

Stillbirth

A baby born after 24 or more weeks completed gestation and which did not, at any time, breathe or show signs of life.

Underlying cause of death

“The disease or injury which initiated the train of morbid events leading directly to death, or the circumstances of the accident or violence which produced the fatal injury” in accordance with the rules of the International Classification of Diseases (excludes deaths at age under 28 days).

VSOB

Vital Statistics Outputs Branch (at ONS).

WHO

World Health Organisation.

Nôl i'r tabl cynnwys