Child Death Review Data Release: Year ending 31 March 2026
1. Introduction
The National Child Mortality Database (NCMD) was launched on 1 April 2019 and collates data collected by Child Death Overview Panels (CDOPs) in England, from reviews of all children who die at any time after birth and before their 18th birthday. There is a statutory requirement for CDOPs to collect these data and to provide them to NCMD, as outlined in the Child Death Review statutory and operational guidance. The guidance requires all Child Death Review (CDR) Partners to gather information from every agency that has had contact with the child, during their life and after their death, including health and social care services, law enforcement, and education services. This is done using a set of statutory CDR forms and the information is then submitted to NCMD.
Every child who dies is a precious individual and their deaths represent a devastating loss for parents, siblings, grandparents, carers, guardians, extended family and friends. With all child deaths there is a strong need to understand what happened, and why. We must ensure that anything that can be learned to prevent future deaths from happening is identified and acted upon.
The data in the notification section of this report (Tables 1 to 14) summarise quantitative information on all child deaths in England from 1 April 2019 up to 31 March 2026. The review section (Tables 15 to 26) summarises the findings of reviews completed by a CDOP on or before 31 March 2026. Years are presented as 1 April to 31 March, for example 1 April 2019 up to 31 March 2020 is presented as 2019-20 throughout.
It should be read in conjunction with the data tables, where more detail is available.
2. Child death notifications
There were 3,223 child (0 – 17 years) deaths in England in 2025-26, an estimated rate of 27.0 deaths per 100,000 children. The number of deaths decreased by 8% on the previous year, the lowest since 2020-21. Infant (children under 1 year) deaths decreased by 7% on the previous year and deaths of children aged between 1 and 17 years decreased by 8% (Figure 1). The estimated rate of infant deaths per 1,000 live births decreased from 3.8 to 3.5 and deaths of 1 – 17 year olds per 100,000 population (1 – 17 years) decreased from 11.7 to 10.9.
Child deaths by region
The child death rate varied across regions in England, with the rate ranging from 19.4 to 32.1 per 100,000 population of 0-17-year-olds (Figure 2). The child death rate decreased for all regions other than London and the North East, with the biggest decreases in the West Midlands, South West and East Midlands, decreasing by 18%, 11% and 10% respectively from 2024-25.
Figure 2. Estimated child death rate per 100,000 population, by region
Child deaths by ethnicity
The child death rate in 2025-26 remained highest for children of black or black British ethnicity (58.8 per 100,000 population) and Asian or Asian British ethnicity (50.3 per 100,000 population) (Figure 3).
Figure 3. Estimated child death rate per 100,000 population, by ethnicity
Within these ethnicity groupings, over a seven-year period, the child death rate was highest for children of Asian Pakistani ethnicity (58.9 per 100,000 population), followed by Black African (56.3 per 100,000 population), any other Asian background (52.2 per 100,000 population) or Black Caribbean (49.5 per 100,000 population) (Figure 4). This was more than double the rate of children from a White British ethnic background (22.2 per 100,000 population). The child death rate was lowest for those of Chinese ethnicity (16.5 per 100,000 population).
Child deaths by deprivation
The gap in child death rates between the most and least deprived neighbourhoods of England has narrowed for the third consecutive year, with the ratio falling below 2:1 for the first time in our data; however, a substantial gap remains.
The child death rate for children resident in the most deprived neighbourhoods was 36.0 per 100,000 population, a decrease of 14% compared with 2024-25 and the lowest recorded for this group. This is still over double the rate for children resident in the least deprived neighbourhoods (16.5 per 100,000 population), a rate ratio of 2.18 (Figure 5).
Figure 5. Estimated child death rate per 100,000 population, by most/least deprived quintiles
Child deaths by age
For children aged between 1 and 17 years, the highest death rate continued to be for children aged between 15-17 years (17.1 per 100,000 population), followed by 1-4-year-olds (14.0 per 100,000 population) (Figure 6).
Figure 6. Estimated death rates for children aged between 1 and 17 years per 100,000 population, by age group
Child deaths by suspected category of death
All deaths reported to the NCMD are coded independently by three clinicians at the point of notification, before investigation, to identify the most likely category of death (the suspected category). This allows NCMD to estimate changes in suspected category of death in real time. The most common category identified by the three clinicians is used as the suspected category. Where there is no most common category, the category highest in the following order of priority (based on the categorisation used by CDOPs) is used: Suicide, Self-harm or Substance Misuse; Trauma; Malignancy; Underlying Health Condition; Intrapartum or Pre-natal Event; Preterm; Infection; Sudden Unexpected Death in Infancy or Childhood.
The child death rate continues to be highest for children aged 1-17 where the suspected cause of death was due to an Underlying Health Condition (4.1per 100,000 population) or Malignancy (2.2per 100,000 population) (Figure 7).
Figure 7. Estimated death rates for children aged between 1 and 17 years per 100,000 population, by suspected category of death
Over the 7 year period, the most common suspected cause of death for children aged 1-4, 5-9 and 10-14 was an Underlying Health Condition (5.9, 3.3 and 3.6 per 100,000 population respectively). For children aged 15-17 years old the most common suspected cause of death was Underlying Health Condition and Suicide, Self Harm of Substance Misuse (both 4.5 per 100,000 population) (Figure 8).
Figure 8. Estimated death rates for children aged between 1 and 17 years per 100,000 population, by suspected category of death and age group (1 April 2019 to 31 March 2026)
Infant deaths
Deaths of infants (babies under 1 year of age) accounted for 61% of all child deaths in 2025-26. The infant death rate was 3.5 per 1,000 live births, a decrease from the previous year (3.8) (Figure 9). For infants born at 24 weeks or over, the estimated death rate was 2.4 deaths per 1,000 live births of the same gestational age; a decrease from 2.6 for the previous year and the lowest rate recorded.
Figure 9. Estimated infant death rate per 1,000 live births
Infant deaths by ethnicity
Patterns of infant deaths were similar to those reported for all child deaths. The estimated infant death rate continued to be highest for infants of Black or Black British ethnicity (6.9 per 1,000 live births); more than double the rate of infants of White ethnicity (2.7 per 1,000 live births) (Figure 10).
Figure 10. Estimated infant death rate per 1,000 live births, by ethnicity
Within these ethnicity groupings, over a seven-year period, the infant death rate was highest for infants of Black African ethnicity (9.5 per 1,000 infant population), followed by Black Caribbean (9.0 per 1,000 infant population), and Asian Pakistani (7.5 per 1,000 infant population) (Figure 11). This was higher than the rate for White British ethnic background (2.6 per 1,000 infant population).
Figure 11. Estimated infant death rate per 1,000 infant population, by ethnicity (1 April 2019 to 31 March 2026)
Infant deaths by deprivation
The death rate of infants who were resident in the most deprived neighbourhoods of England (4.7 per 1,000 infant population), remained over twice that of infants resident in the least deprived neighbourhoods (2.0 per 1,000 infant population) (Figure 12). The difference in infant death rates between these areas continues to reduce from its peak in 2022-23 and is now at the lowest ratio recorded since 2021-22.
Figure 12. Estimated infant death rate per 1,000 infant population, by most/least deprived quintiles
Infant deaths by region
Infant mortality varied across regions of England, with rates ranging from 2.4 to 4.6 per 1,000 live births (Figure 13).
All regions other than London and the North East saw decreases in the rate of infant death, with the biggest decreases in the South West, West Midlands and East Midlands, decreasing by 23%, 18% and 14% respectively from 2024-25.
The PERIPrem (Perinatal Excellence to Reduce Injury in Premature Birth) care bundle has been implemented in these three regions (South West from 2020; West and East Midlands from January 2025). The NCMD Thematic report on Newborn Health (2022), which found that prematurity was the leading contributor to deaths in the first decade of life, made recommendations on optimising care before and after preterm birth, and included PERIPrem as a case study.
Figure 13. Estimated infant death rate per 1,000 live births, by region
Infant deaths by suspected category of death
Infant mortality continues to be highest where the suspected cause of death was due to Preterm births (1.8 per 1,000 live births) or an Underlying Health Condition (1.0 per 1,000 live births) (Figure 14).
Figure 14. Estimated infant death rates per 1,000 live births, by suspected category of death
Neonatal deaths
Neonatal deaths (deaths of babies under 28 days of age) accounted for 44% of all child deaths in 2025-2026.
The estimated neonatal death rate was 2.5 per 1,000 live births, a decrease from the previous year (2.6) and remained higher than 2019-20 (Figure 15). The estimated neonatal death rate for babies born at 24 weeks or over was 1.4 deaths per 1,000 live births; a continued decrease from 2023-24 (1.6 per 1,000 live births).
Figure 15. Estimated neonatal death rate per 1,000 live births
Neonatal deaths by region
The estimated neonatal mortality rate for babies born at 24 weeks or over varied by region, ranging from 1.0 deaths per 1,000 live births of babies born at 24 weeks or over in the South West and South East, to 2.0 in the West Midlands (Figure 16).
Figure 16. Estimated neonatal death rate of babies born at or over 24 weeks gestation per 1,000 live births, by region
Neonatal deaths by gestational age
80% of neonatal deaths were of babies born at a premature gestational age (before 37 weeks) (Figure 17), which was an increase from last year (78%). The proportion of neonatal deaths notified to CDOPs of babies born under 24 weeks gestation also remained higher than in 2019-20 (42% vs 34%).
This increase in deaths of babies under 24 weeks compared to earlier years is difficult to interpret but may be impacted by multiple factors, such as more consistent recognition of signs of life by clinical teams, babies receiving survival-focussed care, appropriate completion of MCCDs (medical certificate of cause of death), and better reporting to CDOPs.
Figure 17. Proportion of neonatal deaths by gestational age at birth
Further information on child death notifications can be found within Tables 1-14, including a breakdown of infant and child death rates by Integrated Care Boards and County and Unitary Authorities.
3. Child death reviews by CDOPs
3,609 child deaths were reviewed by CDOPs in England between 1 April 2025 and 31 March 2026 (some of these deaths may have occurred in earlier years).
The CDOP is responsible for identifying any modifiable contributory factors in relation to the child’s death. Modifiable factors are defined as factors which, by means of nationally or locally achievable interventions, could be modified to reduce the risk of future child deaths.
The proportion of reviews that identified modifiable factors remains at its highest point with 47% of deaths reviewed in 2025-26 identifying modifiable factors. The proportion of reviews with modifiable factors varied per region from 34% to 59% (Table 15).
Figure 18. Number of child death reviews by CDOPs, by year of review and modifiable factors
The most common primary category (i.e., the likely cause) of death for reviews in 2025-26 was Perinatal/neonatal event, which was recorded for 35% of all child death reviews, followed by Chromosomal, genetic and congenital anomalies (22%) and Acute medical or surgical condition (9%) (Figure 19). These patterns were similar to previous years.
The most common primary category of death for children under 1 was Perinatal/neonatal event (55%); for children aged between 1 and 9 years it was Chromosomal, genetic and congenital anomalies (26%); and for children aged between 10 and 17 years it was Malignancy (19%). Figure 19 shows the number of reviews by primary category of death and age group.
Figure 19. Number of child death reviews by CDOPs by primary category of death and age group, 2025-2026
Among reviews where the CDOP had enough information to judge whether modifiable factors were present, deaths categorised as Deliberately inflicted injury, abuse or neglect had the highest proportion of reviews with modifiable factors (89%), followed by Sudden unexpected, unexplained death (79%), and the Perinatal Asphyxia subcategory of the Perinatal/neonatal event category (72%) (Figure 20 and Table 16).
Figure 20. Number of reviews completed by CDOPs by primary category of death and whether modifiable factors were identified, 2025-26
Contributory and modifiable factors identified in child death reviews
As part of the child death review process, CDOPs must record any contributory factors identified during the review and decide which may be modifiable. Definitions of these terms can be found in the statutory child death analysis form or guidance notes can be found in the contributory factors guidance. More than one factor can be identified in each child death review.
The most common recorded modifiable factors by CDOPs during reviews of infant deaths in the 3 years between 1st April 2023 and 31st March 2026 were High maternal BMI (Body Mass Index) (697, 11% of all infant death reviews), Parent/carer smoked cigarettes/tobacco (647, 10%), and Smoking cigarettes/tobacco in pregnancy (530, 8%) (Figure 21).
Figure 21. Most common modifiable factors identified by CDOPs in reviews for children aged under 1 year, reviewed in the 3 years between 1 April 2023 and 31 March 2026
The most common recorded modifiable factors by CDOPs during reviews of deaths of children aged 1 – 17 years were Poor communication/information sharing between agencies (159, 4% of all child death reviews (1 – 17 years)), Issue with treatment, including delays (e.g., delay in starting treatment, side effects or complications developed as a result of treatment, or medical or surgical error) (126, 3%), Service uncommissioned/unfunded/unavailable (114, 3%) (Figure 22).
Figure 22. Most common modifiable factors identified by CDOPs in reviews of children aged 1 – 17 years, reviewed in the 3 years between 1 April 2023 and 31 March 2026
Known to social care
For reviews completed in 2025-26, 16% of children were known to social care at the time of their death; a decrease from the previous year (17%) (Figure 23). A further 17% were reported as previously known to social care.
Of the 564 reviews where the child was known to social care at the time of their death, 52% identified modifiable factors. The proportion of reviews that identified modifiable factors remains higher for children who were known to social care (52%) than children who were never known to social care (46%).
Figure 23. The proportion of child death reviews by CDOPs, by year of review and whether the child was known to social care
For reviews completed by CDOPs in 2025-26, 4% of the deaths were subject to a local or national Child Safeguarding Practice Review (CSPR) (Figure 24), the same as the previous year. Of the 124 reviews where a CSPR took place, 86% of CDOP reviews recorded modifiable factors; an increase from the previous year (83%).
Figure 24. The proportion of child death reviews by CDOPs, by year of review and whether there was a Child Safeguarding Practice Review
Timeliness of CDOP reviews
41% of reviews in 2025-26 were completed by the CDOP within 12 months of the death; an increase from the previous year (38%), and a continuing fall from 2019-2020 where 67% of reviews were completed within 12 months (Figure 25). The median time taken to complete reviews in 2025-26 was 414 days (around 14 months).
Figure 25. Proportion of child death reviews completed by CDOPs by the time taken to complete the review
4. Technical information
All reference tables including further breakdown of data can be found here.
All CDOPs continue to submit data to NCMD on an ongoing basis. NCMD is dependent on accurate data entry by the CDOPs, and specifically, category of death is presented within the data release as it was submitted by the CDOP. The data included within this release represent child deaths that were submitted to NCMD that were going to be, or had been, reviewed by a CDOP in England. In a small number of reviews, this may include deaths of children usually resident outside of England.
CDOPs are required to assign a category of death to each death reviewed within the Analysis Form; the final output of the child death review process. The classification of categories is hierarchical where the uppermost selected category is recorded as the primary category, should more than one category be selected. The definitions for each category can also be found in the Analysis Form.
A child for these purposes is defined as a child aged 0 up to their 18th birthday, excluding stillbirths and planned terminations of pregnancy carried out within the law.
An infant is defined as a child under 1 year of age.
The estimated neonatal and infant death rates reported have been calculated using ONS data for live births, and the rate is presented per 1,000 live births.
The estimated child (0-17 years) death rate and death rate of children aged 1-17 years have been calculated using population data of children the same age in England, from the ONS mid-year population estimates, or Census 2021 data (for all years). The rate is presented per 100,000 children of the same age.
ONS publish live births data and population estimates using calendar years. As this CDR data release uses financial years, live births and population estimates that correspond to the largest proportion of the financial year were used; for example, 2019 live births and mid-year population estimates were used to calculate rates for deaths occurring in 2019-20, and so on.
Census 2021 data (population of 0-17 year olds in England) was used to calculate rates of child death by ethnicity (Tables 6 and 13) for all years for the child death rate and death rate of children aged 1-17 years. This was because this population data by ethnicity was not available for previous years.
The data sources are reported under each table and figure.
Table 4 uses the population of infants (0 year-olds) as a proxy measure for live births, as the data for live births by deprivation and region was not available. This may have a small impact on the rates presented. Due to availability of published population data, each table uses different underlying populations. This means that there may be a small difference in rates reported at national and regional level between tables.
The populations for tables 1, 2, 3 and 14 use the most recent data for all years; for tables 4, 11 and 12 populations for 2024-25 are used for the year ending 31 March 2025-26; for tables 6 and 13 Census 2021 data is used for all years.
The births in tables 1, 2, 3, 6, 11, 12 and 14 use the most recent data for all years; for table 4 births for 2024-25 are used for 2025-26; for table 13 population data from Census 2021 is used for all years in place of births.
Where the number of deaths or the underlying population is small, rates are less reliable and can fluctuate considerably between periods. All rates should therefore be interpreted alongside the number of deaths. Rates based on small numbers are also very sensitive to changes in, or assumptions about, the population estimates used as denominators.
Caution should be applied when making comparisons to previous data releases as all data is refreshed within the latest publication, therefore some previous years will include more up to date information.
Changes were made to data collection of ethnicity by NCMD in April 2021, April 2023 and January 2024 so the following categories are likely to be underestimated. In April 2021 ‘Gypsy or Irish Traveller’ was added to ‘White’ and ‘Arab’ was added to ‘Other ethnic group’. In April 2023 ‘Roma’ was added to ‘White’ and added to the paper forms in January 2024.
Please note that the number of reviews reported for Tables 15–26 corresponds to the reviews completed within a given year, not the number of deaths that occurred in that year. The time required to complete a review can exceed one year, and this duration varies by category. Therefore, these numbers cannot be interpreted as trends over time.
All deaths reported to the NCMD are coded independently by three clinicians at the point of notification, before investigation, to identify the most likely category of death (the suspected category). The most common category identified by the three clinicians is used as the suspected category. Where there is no most common category, the category highest in the following order of priority (based on the categorisation used by CDOPs) is used: Suicide, Self-harm or Substance Misuse; Trauma; Malignancy; Underlying Health Condition; Intrapartum or Pre-natal Event; Preterm; Infection; Sudden Unexpected Death in Infancy or Childhood.
For further information on NCMD data processing please see our Privacy Notice.
5. Acknowledgements
The National Child Mortality Database (NCMD) is commissioned by the Healthcare Quality Improvement Partnership (HQIP) and funded by NHS England as part of the National Clinical Audit and Patient Outcomes Programme. www.hqip.org.uk/national-programmes.
© 2026 Healthcare Quality Improvement Partnership (HQIP)