1. Methodology background


 National Statistic   
 Frequency  Annual
 How compiled  Based on third party data
 Geographic coverage  England and Wales
 Last revised  20 April 2017

Back to table of contents

2. Executive summary

Child mortality statistics presents figures and rates on stillbirths, infant deaths (under 1 year of age) and childhood deaths (between 1 and 15 years of age) occurring annually in England and Wales. It also contains historical data back to 1981. The statistics are derived from information recorded when births and deaths are registered in England and Wales as part of civil registration, a legal requirement.

Some tables are based on infant deaths that have been successfully linked to their corresponding birth records. This linkage enables analysis of risk factors and other demographic characteristics collected at birth registration including: birthweight, mother’s age at birth of child, mother’s country of birth, marital status, number of previous children and parents’ socio-economic status.

The Birth cohort tables for infant deaths present data on infant deaths for those babies that were born in a reference year and died before their first birthday. The tables also provide data by some of the risk factors affecting infant deaths derived from linking the death to the corresponding birth registration record.

Child mortality statistics and Birth cohort tables for infant deaths replaced the annual reference volume Mortality statistics: childhood, infant and perinatal (DH3) in 2010. The current releases contain selected tables from DH3, although some tables have been amended to improve presentation.

More information about the certification and registration of childhood and infant deaths can be found in the User guide to child mortality statistics. More general information on the collection, production and quality of mortality data is available in the User guide to mortality statistics. Additional information on the collection and quality of data for births can be found in the User guide to birth statistics.

This document contains the following sections:

  • Output quality
  • About the output
  • How the output is created
  • Validation and quality assurance
  • Coherence and comparability
  • Concepts and definitions
  • Other information, relating to quality trade-offs and user needs
  • Sources for further information or advice
Back to table of contents

3. Output quality

This document contains a range of information that describes the quality of the data and details any points that should be noted when using the output. The Office for National Statistics (ONS) has developed Guidelines for measuring statistical quality; these are based upon the five European Statistical System (ESS) Quality Dimensions. This document addresses these quality dimensions and other important quality characteristics, which are:

  • relevance
  • timeliness and punctuality
  • coherence and comparability
  • accuracy
  • output quality trade-offs
  • assessment of user needs and perceptions
  • accessibility and clarity

More information is provided about these quality dimensions in the following sections.

Back to table of contents

4. About the output

Relevance

(The degree to which statistical outputs meet users’ needs.)

These statistics are based on details collected when deaths are certified and registered. The live births and stillbirths information is based largely on the details collected when births are registered.

The statistics present the numbers of births and deaths that occurred in the reference year and the corresponding rates. Figures are published for live births, stillbirths, infant mortality and childhood mortality.

Births and deaths of all 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 of persons whose usual residence is outside England and Wales are included for any total figures for England and Wales, but are excluded from any sub-division of England and Wales.

The published tables of data are accompanied by a statistical bulletin, which provides analysis and context. The User guide to child mortality statistics and the User guide to mortality statistics provide further information on the collection, production and quality of mortality data.

Age at death is broken down into the broad categories:

  • stillbirth – born after 24 or more weeks completed gestation and which did not, at any time, breathe or show signs of life
  • early neonatal (deaths under 7 days)
  • neonatal (deaths under 28 days)
  • late neonatal (deaths between 7 and 27 days)
  • postneonatal (deaths between 28 days and 1 year)
  • infant (deaths under 1 year)
  • childhood (deaths between 1 and 15 years of age)

Stillbirths and infant deaths are published by:

  • age of mother at the time the child was born, providing a useful comparison between younger and older age groups of mothers
  • birthweight, presented as a grouped variable to draw distinction between death rates for low and normal birthweight babies
  • marital status of the parents as recorded at the birth registration (inside or outside marriage) and registration type (joint or sole registration)
  • National Statistics Socio-economic Classification (NS-SEC) as defined by occupation; for further information on socio-economic classification as defined by occupation, refer to section 7.1 of the User guide to child mortality statistics – additional information is given in the Coherence and comparability section in this report
  • number of previous children born to the mother
  • place of delivery
  • ONS cause of death group

A hierarchical classification of cause groups using the International Classification of Diseases (ICD)-10 has been developed to derive a single cause group for neonatal deaths and stillbirths. The classification enables direct comparison of stillbirths, neonatal and postneonatal deaths. This would not otherwise be possible due to differences in the way the cause of death for stillbirths and neonates is recorded at registration. This is the result of the introduction of a special death certificate for stillbirths and neonates in 1986. The ONS cause groups allow the death to be assigned to a category based on the likely timing of the damage leading to the death. For further information on the ONS cause of death groups hierarchical classification, refer to section 6.2 of the User guide to child mortality statistics or see Causes of neonatal deaths and stillbirths: a new hierarchical classification in ICD-10.

Postneonatal and childhood deaths are also presented by broad underlying cause of death groups. The cause of death information is based on the details collected when the death is certified and registered. The selection of the underlying cause of death for deaths at ages 28 days and over is based on ICD rules and is made from the condition or conditions reported by the certifier, as recorded on the Medical Certificate of Cause of Death. For further information on coding the underlying cause of death at age 28 days and over, refer to section 6 of the User guide to child mortality statistics.

The Department of Health (DH), Welsh government (WG) and Public Health England (PHE) are important users of child mortality statistics. Child mortality figures feed into data available from the National Child and Maternal Health Intelligence Network. Reducing deaths in babies and young children is included within the NHS Outcomes Framework 2015 to 2016. These important users utilise the data to understand trends in main risks for infant mortality, such as age of mother, gestational age, birthweight, marital status, number of previous children and parents’ socio-economic status. Understanding why children are dying is vital in order to make policy changes that have an impact on the infant mortality rate.

Other users of this output include academics, independent researchers of infant mortality, charities and the media.

Timeliness and punctuality

(Timeliness refers to the lapse of time between publication and the period to which the data refer. Punctuality refers to the gap between planned and actual publication dates.)

Child mortality statistics for a given year are published annually, around 13 to 16 months after the end of the reference year. As these figures use deaths occurring in a reference year, the annual data extract must be taken some months after the end of the data year to allow for late registrations. Deaths that are referred to the coroner may take several weeks or months before they are registered. Therefore, sufficient time is allowed to receive as many death registrations as possible, before the annual dataset is taken. Any late registrations received following the extraction of the death occurrence subset for a reference year are not included in published figures.

Birth cohort tables for infant deaths are published annually, around 25 to 28 months after the end of the data year. These figures are based on births occurring in a reference year, linked to subsequent infant deaths within 1 year of birth.

The annual release of the Child mortality figures is announced on the GOV.UK release calendar at least 4 weeks in advance. From 2013, the Birth cohort tables for infant deaths have been published at the same time as Child mortality statistics, although figures are for the preceding births data year.

For more details on related releases, the GOV.UK release calendar is available online and provides 12 months’ advance notice of release dates. In the unlikely event of a change to the pre-announced release schedule, public attention will be drawn to the change and the reasons for the change will be explained fully at the same time, as set out in the Code of Practice for Official Statistics.

Back to table of contents

5. How the output is created

Child mortality statistics are based on information collected at the registration of a birth or an infant death. Where possible, infant deaths have been linked to their corresponding birth record to obtain information on social and biological factors of the baby and parents.

The denominators used to calculate rates for stillbirths and infant deaths are births that occurred in the same year, that is, the true population at risk. For childhood mortality (deaths between 1 and 15 years of age), the denominators are the midyear population estimates of the resident population in England and Wales for the reference year; the most up-to-date estimates at the time of production are used.

More detailed information on the main processes used in the compilation of mortality and birth statistics and on the accuracy and quality of the data used are available in the User guide to mortality statistics and the User guide to birth statistics.

Rates are not calculated where there are fewer than three deaths in a cell, denoted by (u). It is our practice not to calculate rates where there are fewer than three deaths in a cell, as rates based on such low numbers are susceptible to inaccurate interpretation.

Rates that are based on between 3 and 19 deaths are displayed in tables but are denoted (u) as a warning to the user that their reliability as a measure may be affected by the small number of events.

Some data items collected under the Population Statistics Acts (PSA) 1938, 1960 and 2012 have been aggregated to protect confidentiality. Occasionally, it has been necessary to apply secondary suppression to avoid the possibility of disclosure by differencing. Some of the figures in some tables may not add precisely due to rounding or suppression.

The ONS policy on protecting confidentiality in birth and death statistics (currently under revision) is available on our website.

Back to table of contents

6. Validation and quality assurance

Accuracy

(The degree of closeness between an estimate and the true value.)

Information recorded at birth and death registration in England and Wales passes through a number of processes before becoming usable for analysis. Our publication User guide to mortality statistics provides additional information on the collection, processing and quality of mortality data for England and Wales. More specific information relating to stillbirths and infant deaths is available in the User guide to child mortality statistics. The User guide to birth statistics provides detailed information on the registration, collection and quality of births data in England and Wales.

The accuracy of information contained in the draft birth entry is the responsibility of the informant(s), usually the mother, or the mother and father where the registration is a joint one outside marriage. Wilfully supplying false information may render the informant(s) liable to prosecution for perjury. It is believed that, in general, the information supplied by the informant(s) is correct.

Occasionally, birth information might be missing from an entry. This can occur for a number of reasons including the informant refusing to give information, or the informant not knowing the information. Under the Population Statistics Act, certain confidential data items are collected at the registration of a birth. If any of these data items are missing, an appropriate value is imputed. More information on the imputation of missing births data can be found in section 1.3 of the User guide to birth statistics. The number of birth records missing age of parents’ can be found in section 2.9. For years prior to 2012, the number of records missing information on previous live-born children can be found in section 2.13.

For deaths, other than the cause of death (including a stillbirth), additional information is supplied to the registrar by the informant when the death is registered. For deaths certified after inquest, the coroner, police officers or other witnesses may supply this information, which cannot later be checked by the registrar.

When a birth or death is registered, the registration system provides the opportunity for the registrar to make validation checks at the point of registration, therefore improving the quality of the data. We conduct internal consistency checks to eliminate any errors made in the supply and recording of birth and death records. Checks are more frequent on those records with extreme values for main variables (such as age of mother and father), as these have a greater impact on published statistics. A small number of registrations are raised with the General Register Office (GRO) on a monthly basis for verification.

Some tables in Child mortality statistics show the latest figures for infant deaths that occurred in a given year, while others are based on infant deaths that occurred in a given year and that have been successfully linked to their corresponding birth record. The Birth cohort tables for infant deaths are births occurring in a reference year that have been successfully linked to their corresponding death registration record for those babies who died before their first birthday.

Around 2% of infant deaths cannot be linked to a birth record. The main reasons for this are either that a birth record cannot be found, or the birth was registered outside England and Wales. Further information on the linkage process is available in section 7 of the User guide to child mortality statistics.

Any late registrations received following the extraction of the death occurrence subset for a reference year are not included in published figures. See the Timeliness and punctuality section for further information.

Coherence and comparability

(Coherence is the degree to which data that are derived from different sources or methods, but refer to the same topic, are similar. Comparability is the degree to which data can be compared over time and domain, for example, geographic level.)

General

Following an evaluation of the suite of mortality releases in summer 2012, further rationalisation of infant mortality products took place to avoid confusion for users and to minimise the variation in infant mortality figures from the use of different datasets. We discontinued Infant and perinatal mortality by social and biological factors, which contained very similar tables. To ensure the tables continue to best meet user needs, a few small changes to existing tables were made. These were outlined in the statistical bulletin: Childhood Mortality Statistics: Childhood, Infant and Perinatal, 2013.

The linkage of birth and infant death records has been conducted since 1975 to obtain information on the social and biological factors of the baby and parents, as registered on the birth record.

Over the years there have been significant changes in the way some of the risk factor variables have been coded. Therefore, comparability over time is limited for some variables.

On 1 October 1992, the legal definition of a stillbirth was changed from a baby born dead after 28 or more weeks completed gestation to one born dead after 24 or more weeks completed gestation. This means that 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 (OPCS) 1994).

In 1986, following recommendation from the World Health Organisation (WHO) of a change of format of the certificate to include maternal conditions, we introduced new certificates for registering neonatal deaths and stillbirths in England and Wales. This allowed certifiers more flexibility in the number and ordering of the causes given, however, this made it impossible to either derive a single underlying cause of death (for neonatal deaths and stillbirths) or to compare with cause of death for postneonatal deaths, which are certified on the standard death certificate.

For this reason, a hierarchical classification, known as ONS cause groups, was developed for stillbirths and neonatal deaths to identify a single underlying “mechanism”, which led to death. A computer algorithm directs any mention, in the case of neonatal deaths, and underlying cause in the case of postneonatal deaths, to the first appropriate cause group. However, for the data years 2001 to 2012, postneonatal deaths were assigned to the ONS cause groups based on mentions rather than underlying cause. We undertook some analysis of the impact of this change and the results showed that it was negligible. Further information on this analysis is available on request from vsob@ons.gsi.gov.uk. Stillbirths and infant mortality data by ONS cause groups are not available prior to 1993.

The Human Fertilisation and Embryology Act 2008 contained provisions enabling two females in a same-sex couple to register a birth from 1 September 2009. Due to the small numbers, births registered to a same-sex couple in a civil partnership are included with marital births, while births registered to a same-sex couple outside a civil partnership are included with births outside marriage. Tables are footnoted to show the number of births to same-sex couples included with marital and non-marital births. Given the relatively small numbers of births registered to same-sex couples, the impact on statistics is negligible.

Cause of death

Information on cause of death in England and Wales has been coded to ICD-10 since January 2001. ICD-10 was implemented on the recommendation of WHO and replaced ICD-9, which had been in use since 1979. Further information on changes in ICD-10 and comparability between ICD-9 and ICD-10 is available on our website.

In January 2011, we switched from using ICD-10 version (v)2001.2 to use v2010. A bulletin presenting the main findings from the Bridge Coding Study of 2009 Stillbirth and Neonatal Death Registrations (in which deaths were independently coded using v2001.2 and v2010), is available to help understand the impact of this change on perinatal mortality statistics for England and Wales. The impact of ICD- 10 v2010 on other deaths has been investigated in a separate study.

On 1 January 2014, we changed the software used to code cause of death from the Medical Mortality Data Software (MMDS) to IRIS. The new IRIS software version 2013 incorporates official updates to ICD-10 that are approved by WHO. The use of the IRIS software has helped improve the international comparability of mortality statistics.

For stillbirths and neonatal deaths, any maternal condition mentioned on the death certificate will be coded to the P chapter (certain conditions originating in the perinatal period) rather than elsewhere in the ICD classification. Previously, these deaths may have been coded to the O chapter (pregnancy, childbirth and the puerperium). The 2014 registrations of stillbirths and neonates that occurred in 2013 were recoded to ensure comparability within the 2013 occurrences dataset. Although this change was implemented on 1 January 2014, there were a very small number of infant deaths that occurred in 2013 and were registered in 2014, that were affected by the change.

Further information on IRIS can be found on our website. There is also a study that looks into the impact of the coding changes on stillbirths and neonatal deaths.

Additional details on specific historical changes to the collection and coding of mortality data are published under section 3.4 of the User guide to mortality statistics.

Country of birth

A new coding system for country of birth was introduced in 2006 and was used to code mother’s and father’s country of birth. The National Statistics Country Classification is based on the International Standards Organisation (ISO) 3166 Codes for the Representation of Names of Countries and their Subdivisions, adapted to meet data needs of UK National Statistics' users and producers.

Socio-economic Classification

Over the years the source of information on father’s social class has undergone numerous changes. Since 2001, the National Statistics Socio-economic Classification (NS-SEC) has been used for all official statistics and surveys. It replaced Social Class (SC) based on Occupation (formerly Registrar General's Social Class) and Socio-economic Groups (SEG). This change was agreed by the National Statistician following a major review of government social classifications commissioned in 1994 and carried out by the Economic and Social Research Council. In 2011, NS-SEC was rebased on the new Standard Occupational Classification (SOC 2010). The new classification is based not on skills but on employment conditions, which are now considered to be central to describing the socio-economic structure of modern societies.

Up until the 2011 data year, child mortality and birth statistics were published based on father’s NS-SEC. Historically, the decision to use father’s NS-SEC was based on the premise that many mothers either did not have a paid occupation or choose not to state their occupational details at birth registration. From the 2012 data year, 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) has been used instead. These changes mean that figures from the 2012 data year onwards are not directly comparable with previous years.

Previous children

Amendments to the Population (Statistics) Act 1938 mean that, from May 2012, 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. A paper describing the changes that have occurred to our birth statistics as a result of improvements to the Population Statistics Act is available on our website. It provides background to the changes and provides high-level findings from the new data collected in 2012 and 2013.

The difference, between the old and new data, in the proportion of married women reporting previous births is larger than was expected purely from the question change. More information can be found in this methodology paper on Quality assurance of new data on birth registrations, as a result of changes to the Population Statistics Act – from May 2012 onwards.

An investigation of childbearing by registration status in England and Wales, using birth registration data for 2012 and 2013 examines the patterns and characteristics in birth registrations following the improvements to the data collected at birth registration. The principal characteristics explored in the paper relate to whether a woman has been previously married and whether the birth is the mother’s first child or subsequent child.

Only minor changes have been made to published tables for 2012 and 2013 since the first full year of new data in 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.

UK comparability

For mortality data for other UK countries, please see stillbirths and infant deaths in Scotland and stillbirths and infant deaths in Northern Ireland. These figures are comparable with those for England and Wales.

Back to table of contents

7. Concepts and definitions

(Concepts and definitions describe the legislation governing the output and a description of the classifications used in the output.)

Since 1 September 1992, a stillbirth has been defined as “a child which has issued forth from its mother after the 24th week of pregnancy, and which did not at any time after becoming completely expelled from its mother breathe or show other signs of life”. Prior to this, a stillbirth was as this definition, but at 28 or more weeks completed gestation. See the Coherence and comparability section for further information.

For information on the categories of infant deaths based on age (for example, neonatal and postneonatal) see the Relevance section.

Amendments to the Population Statistics Act 1938 mean that from May 2012, information is now collected at all birth registrations on the total numbers of previous live births and previous stillbirths that the mother has had (previously information was only collected for married women and only included previous children with the current or former husband). This has simplified the question asked by registrars and provides improved coverage.

Plurality refers to the total number of live births and stillbirths at the maternity.

The rates used in the tables are outlined in section 3.3 of the User guide to child mortality statistics.

The existing provisions for the registration of deaths and the processing, reporting and analysis of mortality can be found in the User guide to mortality statistics.

Back to table of contents

8. Other information

Output quality trade-offs

(Trade-offs are the extent to which different dimensions of quality are balanced against each other.)

Child mortality statistics and the Birth cohort tables for infant deaths are based on the date the death occurred rather than the date on which it was registered. Although this means that there is a delay in the publication of these statistics to allow for late death registrations, there are a number of advantages of using death occurrences over registrations. The long history of infant mortality statistics as an important indicator of the health of the nation is based on occurrence data (as defined by WHO). See section 2 of the User guide to mortality statistics for further information.

Assessment of user needs and perceptions

(The processes for finding out about uses and users, and their views on the statistical products.)

A user consultation to review infant mortality statistics took place between 5 July and 16 August 2011 and our response to the review is available.

A proposal for changes to birth statistics by socio-economic classification was published on our website in February 2013. No feedback was received so the outlined changes were implemented.

Proposed changes to Child mortality statistics were outlined in the Child Mortality Statistics, 2011 bulletin (published February 2013) and user feedback was requested. Very little feedback was received so the proposed changes have been implemented.

User feedback is requested at the bottom of all emails sent by customer service teams within our Vital Statistics Outputs Branch.

We also receive feedback through regular attendance at user group meetings and conferences.

Back to table of contents

9. Sources of further information or advice

Accessibility and clarity

(Accessibility is the ease with which users are able to access the data, also reflecting the format in which the data are available and the availability of supporting information. Clarity refers to the quality and sufficiency of the release details, illustrations and accompanying advice.)

The latest figures on Child mortality statistics and the Birth cohort tables for infant deaths can be accessed free of charge on our website. A statistical bulletin containing context and commentary accompanies the release. The bulletin describes the main trends looking at age-specific mortality rates and reports on the latest statistics on stillbirths and infant deaths.

Our recommended format for accessible content is a combination of HTML webpages for narrative, charts and graphs, with data being provided in usable formats such as CSV and Excel. Our website also offers users the option to download the narrative in PDF format. In some instances other software may be used, or may be available on request. Available formats for content published on our website but not produced by us, or referenced on our website but stored elsewhere, may vary. For further information, please refer to the contact details at the beginning of this document.

For information regarding conditions of access to data, please refer to these links:

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 Vital Statistics Outputs Branch (vsob@ons.gsi.gov.uk or telephone: +44 (0)1329 444 110). All user requested data will be published onto the website.

Access to microdata and disclosive data, that is, data which have the potential to identify an individual record, requires the approval of the ONS Microdata Release Procedure (MRP) before the data can be provided.

Useful links

The Vital statistics: population and health reference tables provide annual infant mortality data for the UK and its constituent countries (based on deaths registered in a year). For data for other UK countries please see the latest infant death statistics for Northern Ireland and the latest infant death statistics for Scotland.

Summary data for infant mortality in England and Wales (based on deaths registered in the year), are available in the Deaths registrations summary tables. A geographical breakdown of infant death numbers and rates by local authority and county (based on deaths registered in a year) is available in Deaths registered in England and Wales by area of usual residence.

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

More general information on the collection, production and quality of mortality data is available in the User guide to mortality statistic

Back to table of contents

Contact details for this Methodology