This ad-hoc release contains statistics on children in need in England and assessments completed by children's social care services. Each reporting year covers the year ending 31 March.
Children in need are a legally defined group of children (under the Children Act 1989 (opens in new tab)), assessed as needing help and protection as a result of risks to their development or health.
This release focuses on the ‘sexual abuse’ and ‘sexual exploitation’ assessment factors recorded against individual children who were referred to children’s social care services and assessed.
The latest statistics and charts relate to the year ending 31 March 2025. The standard period for data in this release is for the year ending 31 March 2016 to the year ending 31 March 2025.
Throughout this release, ‘CSA’ will be used when referring to child sexual abuse and ‘CSE’ will be used when referring to child sexual exploitation.
29,560
Broadly stable (up 5% since 2016)
Number of children in the year ending 31 March 2025 with sexual abuse recorded during their assessment.
12,120
Down 29% since 2016
Number of children in the year ending 31 March 2025 with child sexual exploitation recorded during their assessment.
2,190
Down 26% since 2016
Number of children in the year to 31 March 2025 assessed as being at risk of harm and starting a child protection plan with initial category of abuse sexual abuse.
66%
The percentage of children assessed as being affected by child sexual exploitation who were persistently absent from school in the 2023/24 academic year.
View tables that we have built for you, or create your own tables from open data using our table tool
Browse and download open data files from this release in our data catalogue
Learn more about the data files used in this release using our online guidance
Download all data available in this release as a compressed ZIP file
All supporting files from this release are listed for individual download below:
Scatter chart showing the percentage of children assessed as being affected by sexual abuse or exploitation by local authority in 2025.
The majority of statistics in this release are based on assessment factors data from the Children in Need census. Factors identified at the end of assessment are additional factors that social workers record as being relevant in a case. More information on assessment factors including a list of all factors can be found in Appendix B of the Children in need census 2025 to 2026 guide (opens in new tab).
Users should be aware of the following when interpreting these statistics:
This release focuses on factors that relate to child sexual abuse or exploitation, but data relating to all assessment factors can be found by exploring the data by navigating to ‘View or create your own tables’.
Factors identified at the end of assessment are additional factors that social workers record as being relevant in a case.
These figures are based on assessment factors recorded against individual children who were referred to children’s social care services and assessed as being in need of children’s social care services during the year (ending 31 March 2025).
The majority of children have more than one factor recorded. It should be noted that not all episodes have factors recorded, but this has improved over time. Nonetheless, there can be differences in the recording and practice between local authorities therefore this data should be treated with caution and shouldn’t be taken to represent the national, regional or local authority level prevalence of particular issues.
The number of children assessed as having been affected by any form of child sexual abuse (CSA) has remained largely stable since 2016, whereas the number of children assessed as having been affected by child sexual exploitation (CSE) has been falling since it peaked at 19,410 in 2018.
6% of all children assessed in the year to 31 March 2025 were assessed as having been affected by CSA, equating to 29,560 children. 4% of children assessed in the year were affected by adult on child sexual abuse and 3% were assessed as being affected by child on child sexual abuse. 3% of all children assessed in the year were affected by CSE, equating to 12,120 children.
The most common factors reported in 2025 were concerns about mental health of the parent/carer where 35% children were assessed as having been affected and domestic abuse where the parent is a victim where 33% children were assessed as having been affected. Substance misuse, which relates to alcohol or drug misuse for the child, their parent/carer or someone else in their household, was also common, with 29% of all children assessed having been affected.
At regional level, there is significant variation in the proportion of children for whom child sexual abuse (CSA) and/or child sexual exploitation (CSE) was recorded as a factor during local authority assessments. In 2025, 10% of children who were assessed in Yorkshire and the Humber had CSA and/or CSE recorded, whereas the proportion was lower in Inner London at 6%.
Analysis found no clear correlation between this variation and commonly cited socio-economic indicators—such as population size, levels of deprivation, or local authority type—aside from a broad regional pattern, as region rates remaining largely stable over time.
Further examination of intra-regional differences revealed multiple instances where neighbouring local authorities, or those with comparable socio-economic profiles, recorded CSA/CSE as an assessment factor in markedly different ways.
Data for all local authorities is available in the featured table.
For comparative purposes, the analysis was replicated for all children assessed as having been affected by mental health concerns about the parent. While similar socio-economic patterns were observed, a key distinction emerged: reporting of CSE/CSA remained broadly stable across regions, whereas parental mental health showed a consistent increase across all regions, with the most pronounced rise in the North East.
Based on these findings, we propose a working hypothesis that local variation in the recording of CSA and CSE assessment factors is more likely attributable to differences in local authority recording and practice, rather than underlying socio-economic conditions.
This analysis focuses on factors that relate to child sexual abuse or exploitation. Child characteristics data relating to all assessment factors can be found in the underlying data file.
57% of children assessed as having been affected by child sexual abuse (CSA) and 67% children assessed as having been affected by child sexual exploitation (CSE) were female. This is an over-representation when compared with all children in need (43%).
The proportions differed slightly according to the type of CSA where 60% of children assessed with adult on child sexual abuse were female, whereas the proportion was lower for children assessed with child on child sexual abuse at 53%.
Children assessed as having been affected by CSA were older with 58% aged 10 or over which was comparable with the wider children in need population where 60% were aged 10 or over, but higher when compared to 47% of the overall child population (as shown in the latest ONS mid-year population estimates (opens in new tab)). However, children assessed as having been affected by CSE were older still, with 81% aged 10 or over.
Where ethnicity was known, 79% and 74% of children assessed as having been affected by CSA and CSE respectively were White, which was higher when compared with all children in need (67%). Asian and Black children were under-represented. 1 in 10 (10%) of children assessed as having been affected by CSE were from a Mixed ethnicity which was an over-representation when compared to 7% of the overall child population as reported in the 2021 census (opens in new tab).
5% of children assessed with as having been affected by CSA and 8% of children assessed as having been affected by CSE had a recorded disability. These levels were lower than all children in need where 14% of children had a recorded disability.
Over two fifths of children (44.1%) assessed as having been affected by CSE had a special educational need in 2023/24. This proportion was higher than children assessed as having been affected by CSA (37.2%) and twice that of all pupils (18.1%), but lower than all children in need (51.7%).
Children assessed as having been affected by CSA and CSE were also more likely to have an education, health and care plan (11.2% and 14.7% respectively) when compared with all pupils (4.7%), but less likely than all children in need (27.1%).
Pupils in state-funded schools in England are entitled to receive free school meals if a parent or carer was in receipt of certain benefits. More information on free school meal eligibility is provided in the department's Schools, pupils and their characteristics statistical release.
The proportion of children eligible for free school meals in 2023/24 was similar for those assessed as having been affected by CSE (61.9%) and those assessed as having been affected by any CSA (60.7%). These proportions were similar to those recorded for all children in need (61.6%), but much higher than all pupils where 24.6% were eligible for free school meals.
Pupils with one or more suspensions refers to pupils that had at least one suspension across the full academic year. A permanent exclusion refers to a pupil who is excluded and will not come back to that school (unless the exclusion is overturned). This data only includes permanent exclusions which have been upheld by the governing body or Independent Review Panel (IRP), and not those which are still ongoing. The permanent exclusion percentage is calculated as the number of permanent exclusions divided by the number of pupils (x100).
Nearly a fifth of children (19.42%) assessed as having been affected by CSE had one or more suspension from school in 2022/23. This was much higher than children assessed as having been affected by any CSA (9.91%). This was also higher than the proportion of all children in need (14.13%) and all pupils with a suspension (3.74%).
Similarly the proportions of children with permanent exclusions was highest for children assessed as having been affected by CSE (1.51%) compared with 0.81% for all children in need and 0.11% for all pupils.
A pupil is identified as persistently absent if they miss 10% or more of possible sessions.
Two thirds of children (65.7%) who were assessed as having been affected by CSE were persistently absent from school in 2023/24. The proportion was lower for children who were assessed as having been affected by any CSA at 43.4%, which was similar to the rate for all children in need (44.0%), but still twice as high compared to all pupils where 20.3% of pupils were persistently absent.
This analysis focuses on factors that relate to child sexual abuse or exploitation, but data relating to co-occurring factors for all assessment factors can be found by exploring the underlying data file.
The majority of children have more than one factor recorded for each episode of need, and ‘co-occurring factors’ in this section relate to the factors which were recorded alongside sexual abuse and sexual exploitation.
A higher average number of recorded assessment factors may suggest increased case complexity, or a wider range of needs experienced by the child. It may also reflect improvements in recording practices. A lower average may not necessarily indicate reduced complexity, as individual cases may still involve substantial severity despite fewer assessment factors.
On average, children have two assessment factors recorded at assessment. The most common co-occurring factors reported for children who were assessed as having been affected by CSA were mental health concerns about parent, and mental health concerns about child which were reported for 27% and 25% of children respectively. Domestic abuse (concerns parent is a victim) and emotional abuse were also common, the latter was reported for a fifth (20%) of children who were assessed as having been affected by CSA.
There were some differences depending on the type of abuse. A higher proportion of children assessed as having been affected by child on child sexual abuse were also recorded with child on child physical abuse (8%) when compared with children assessed as having been affected by adult on child sexual abuse (4%). Additionally, a higher proportion of children assessed as having been affected by child on child sexual abuse also had socially unacceptable behaviour recorded (13%) when compared with children assessed as having been affected by adult on child sexual abuse (6%).
Nearly 1 in 10 children (9%, 2,740 children) who were assessed as being affected by CSA, were also assessed as being affected by sexual exploitation (CSE).
Almost a third of children (32%) assessed as being affected by sexual abuse: adult on child or child on child did not have any other factors recorded.
Similarly to CSA, the most common factors reported alongside CSE were also mental health, although higher proportions of children were affected. Over half (51%) of children assessed as having been affected by CSE were also assessed with mental health concerns about the child, and 33% were also assessed with mental health concerns about the parent.
Just over a third (34%) of children assessed as having been affected by CSE were also assessed as at risk of going missing. This was much higher than children assessed as having been affected by any CSA (5%). Additionally, children assessed as having been affected by CSE were also commonly identified with socially unacceptable behaviour (28%), self-harm (28%) and drug misuse concerns about the child (27%).
15% of children assessed as having been affected by CSE were also assessed as having been affected by adult-on-child sexual abuse and 11% were also assessed as having been affected by child-on-child sexual abuse.
Only 11% of children assessed as having been affected by CSE had no other factors recorded at assessment.
The median average number of assessment factors recorded for all children who were assessed has remained constant at two during the period from 2016 to 2025.
The mean average has gradually increased from 2.52 in 2016 to 2.75 in 2025, indicating that over this ten-year period more factors are being considered at assessment. This may suggest that case complexity has increased slightly, but could also be a result of improved recording practices. The mean average has remained stable over the last two years.
In 2025, the median number of assessment factors for children ranges from 2 to 6, with the exception of ‘Other’ which has a median of 1.
The highest median of 6 assessment factors was found among children assessed with alcohol misuse (concerns about the child or other person) and drug misuse (concerns about other person), indicating a broader range of factors typically considered in these assessments.
Children assessed as being having been affected by CSE had a relatively high median of 5 assessment factors, compared to children assessed as having been affected by CSA —whether adult on child or child on child—which have a median of 2.
This analysis focuses on factors that relate to child sexual abuse or exploitation. Data relating to referral sources for all assessment factors can be found by exploring the underlying data file.
A referral is defined as a request for services to be provided by children’s social care and is regarding a child who is not currently in need. Referrals to children’s social care services happen before an assessment and can be made from various sources.
Police referrals are the most common source of all referrals, including for children assessed as having been affected by CSA and CSE. However, they are slightly more common for children assessed as having been affected by CSA (38%) or CSE (35%) compared to all children assessed (28%).
Referrals from schools are more common for children assessed as having been affected by child on child sexual abuse (26%) compared to adult on child sexual abuse (15%) or CSE (19%).
Referrals from LA services are more common for children assessed as having been affected by CSE (17%) compared with children assessed as having been affected by CSA (13%) and all children assessed (12%).
Referrals from health services are less common for children assessed as having been affected by CSA (9%) or CSE (9%) compared with all children assessed (15%).
A re-referral occurs when a child is referred within 12 months of a previous referral.
Re-referrals within 12 months can be used as a metric for measuring the success of interventions by social workers. High levels of re-referrals may indicate that cases are being closed prematurely or that interventions were not successful in bringing about lasting change in circumstances for children.
However, there will always be some level of re-referrals that occur due to changes in a child’s circumstances that require new interventions.
This analysis focuses on factors that relate to child sexual abuse or exploitation, but data relating referral rates for all assessment factors can be found by exploring the underlying data file.
The re-referral rate in this section refers to the proportion of referrals for each assessment factor that are re-referrals. For example, there were 9,740 referrals where gangs were recorded at assessment, and 3,250 of these were re-referrals. Therefore, the re-referral rate for gangs is 33%.
Children assessed as having been affected by sexual exploitation (CSE) had one of the highest re-referral rates with 29% of referrals being re-referrals in 2023. This is above the rate for all assessments (23%). The only assessment factors with higher re-referral rates were socially unacceptable behaviour (30%), going missing (32%), child criminal exploitation (33%) and gangs (33%).
In contrast, children assessed as having been affected by sexual abuse (CSA) had one of the lowest re-referral rates at 19%. This was the same rate as children assessed as having been affected by abuse linked to faith or belief and the only assessment factor with a lower re-referral rate was female genital mutilation (15%).
The analysis below disaggregates referrals into three groups:
- Initial referrals that will lead to re-referrals;
- Initial referrals that will not lead to re-referrals; and
- Re-referrals
These categories are mutually exclusive. In cases of repeat re-referrals, all are classified as re-referrals.
The analysis in this section measures the proportion of all referrals that had each assessment factor recorded. For example, child sexual exploitation is recorded at assessment in 3% of all initial referrals where the child is not re-referred, 2% of all initial referrals where the child is re-referred and, 4% of all re-referrals.
For more information see the methodology.
In 2023, sexual abuse was recorded in 7% of all referrals and sexual exploitation was recorded in 3% of all referrals.
The proportion of all referrals which did not go on to become re-referrals was higher when sexual abuse was recorded (7%) compared with sexual exploitation (3%). For referrals where sexual exploitation was recorded, there was little difference between the child later being re-referred and the child not being re-referred , whereas where sexual abuse was recorded, children were less likely to go on to be re-referred.
The proportion of any type of referral where CSE was recorded has declined since 2017, however, initial referrals where the child was later re-referred has seen the most change. In 2017, 4% of initial referrals where the child was later re-referred involved children assessed as having been affected by CSE, which has fallen to 2% in 2023.
In contrast, the proportion of any type of referral for children assessed as having been affected by CSA has remained largely stable between 2017 and 2023. Initial referrals where the child was re-referred and re-referrals remained around 5% and initial referrals where the child was not re-referred remained at around 7% for the entire series.
A child becomes the subject of a child protection plan if they are assessed as being at risk of harm at an initial child protection conference (ICPC).
The initial category of abuse under which a child is made the subject of a child protection plan will have been decided upon at the child protection conference. When there is more than one main category of abuse, or where no category is recommended, ‘Multiple’ is chosen. Children included in this category are not included in any other category of abuse, therefore a child is counted only once in this measure.
Over the past 31 years, the number of child protection plans (CPPs) starting in the year has more than doubled from 28,500 in 1994 to 61,840 in 2025. However, between 1994 and 2001, the number of CPPs starting with ‘sexual abuse’ as the initial category of abuse halved. Since 2001, the number of CPPs with ‘sexual abuse’ as the initial category of abuse has slowly declined and is now at it's second lowest level at 2,190 in 2025, accounting for just 3% of all CPPs.
In this time, the number of CPPs with ‘neglect’ as the initial category of abuse has increased and is now the most common, accounting for around half of all CPPs followed by emotional abuse which accounts for around a third.
In 2025, 4,130 children who were assessed as having been affected by sexual abuse (CSA) started a CPP, accounting for 7% of all children who started a CPP during the year. This number has been falling since a peak of 5,360 in 2018, however the proportion has remained stable at 7%.
The number of children assessed as having been affected by sexual exploitation (CSE) who started a CPP has also declined from a peak of 3,400 in 2018 to a series low of 1,820 in 2025. The proportion of all children who started a CPP was lower at just 3% in 2025 and has fallen from 5% in 2018.
This reflects a wider trend of the total number of children starting on a CPP falling gradually from 2018.
In 2025, 14% of children assessed as having been affected by CSA started a child protection plan during the year. This proportion has fallen slightly from 17% in 2017.
The proportion of children assessed as having been affected by CSE who started a child protection plan in 2025 was 15% which has remained fairly stable since 2016 (16%).
The initial category of abuse under which a child is made the subject of a child protection plan will have been decided upon at the initial child protection conference. The initial categories of abuse are emotional abuse, neglect, multiple, physical abuse and sexual abuse. When there is more than one main category of abuse, ‘Multiple’ is chosen. Children included in this category are not included in any other category of abuse, therefore a child is counted only once in this measure.
In 2025, 1 in 6 (16%) of children assessed as having been affected by CSE who started CPPs had sexual abuse recorded as the initial category of abuse. The proportion has fallen over time from 1 in 4 (25%) in 2016.
Neglect is the most common initial category of abuse in 2025 (47%) which has become more common since 2016 (44%). Emotional abuse is the second most common initial category of abuse in 2025 (31%) which has also become more common since 2016 (21%).
In 2025, the most common initial category of abuse for children starting protection plans who were assessed as having been affected by CSA, was sexual abuse (38%). However, children are less likely to have this category of abuse recorded compared with 2016 (40%).
Neglect was the second most commonly recorded initial category of abuse for children assessed as having been affected by CSA accounting for 35% of children in 2025, an increase from 30% of children in 2016.
Children assessed as having been affected by CSA are also slightly more likely to have emotional abuse recorded as the initial category of abuse in 2025 (20%) compared with 2016 (19%).
A quarter of children who started a CPP and were assessed as having been affected by CSA or CSE at assessment with ‘sexual abuse’ recorded as the initial category had no co-occurring factors. This was much higher than children who started on other initial categories of abuse.
This suggests that children who start CPPs with ‘sexual abuse’ as the initial category of abuse may have less complex cases than children who have other initial categories of abuse such as ‘neglect’ or ‘emotional abuse’. This may explain the trend in more children assessed as having been affected by sexual abuse and exploitation now starting CPPs under ‘neglect’ and ‘emotional abuse’.
Children can be identified with multiple factors at their assessment, and a higher number of assessment factors may indicate that a case is more complex.
In 2025, all children who started on a CPP had 4 factors identified at assessment on average. This was slightly higher for children assessed as having been affected by CSA at 5 factors, and higher again for children assessed as having been affected by CSE at 8 factors.
Children who were assessed as having been affected by CSA and started a CPP with initial category of abuse ‘neglect’ had on average 7 factors recorded at assessment.
Children who were assessed as having been affected by CSE and started a CPP had the highest number of factors recorded on average, and this was true across all initial categories of abuse except ‘multiple’.
Overall, the reason for the decline in CPPs with sexual abuse as the initial category since 2018 is likely to be influenced by multiple factors:
This analysis focuses on factors that relate to child sexual abuse or exploitation, but child journeys data relating to all assessment factors can be found by exploring the underlying data files.
This analysis looked at children assessed in 2017-18 and their journeys through social care over the following 6 years. We looked at their highest level of involvement with social care during the 6 years following their assessment, as well as whether they had ever received each type of intervention during those 6 years.
A lower proportion of children assessed as having been affected by CSE had a no further action referral as their highest level of involvement (12%) compared to children assessed as having been affected by CSA (14%) and all children assessed (15%).
Children assessed as having been affected by CSA or CSE were less likely to have ever had a no further action referral in the 6 years following assessment (35% and 33%) compared to all children assessed (43%).
Children assessed as having been affected by CSE were also more likely to be looked after (14%) compared to children assessed as having been affected by CSA (11%) and all children assessed (12%).
Children assessed as having been affected by CSA were more likely to have CINO (children in need, excluding children on protection plans and children looked after) as their highest level of involvement (55%) compared to children assessed as having been affected by CSE and all children assessed (both 52%).
Similar patterns can be seen when looking at children assessed in 2023-24 and their events occurring after assessment during that year.
If a local authority identifies there is reasonable cause to suspect the child is suffering, or is likely to suffer significant harm, it will carry out an assessment under section 47 of the Children Act 1989 to determine if it needs to take steps to safeguard and promote the welfare of the child. If concerns are substantiated and the child is judged to be at continuing risk of harm then an initial child protection conference (ICPC) should be convened within 15 working days. A child becomes the subject of a child protection plan (CPP) if they are assessed as being at risk of harm at an ICPC.
The analysis looked at the percentage of referrals leading to S47 enquiries, ICPCs and CPPs within the year of assessment, for children assessed in 2017-18 and 2023-24.
In 2023-24, 47% of referrals for children assessed as having been affected by CSA and 46% of referrals for children assessed as having been affected by CSE resulted in a S47 enquiry, compared to 31% of referrals for all children assessed.
S47 enquiries are less likely to become ICPCs and CPPs for CSA (30% of S47s became ICPCs and 26% became CPPs) and CSE referrals (31% of S47s became ICPCs and 25% became CPPs) compared to all children assessed (40% of S47s became ICPCs and 35% became CPPs). Although a smaller percentage of S47 enquiries for CSA or CSE referrals result in ICPCs compared to all children assessed, it is still true that referrals for children assessed as being affected by CSE or CSA are more likely to result in an ICPC compared to referrals for all children assessed. This is because of the higher percentage that escalate from referral to S47.
Compared to 2017-18, referrals for children assessed as having been affected by CSE are more likely to lead to S47 enquiries (increasing from 39% in 2017-18 to 46% in 2023-24). This has remained stable for CSA referrals and increased slightly for referrals for all children assessed. For all three groups, the percentage of S47 enquiries resulting in CPPs has fallen (by 11 percentage points for CSE referrals, and 5 percentage points for CSA referrals and all children assessed).
Almost a quarter of all children who were assessed in 2023-24, had a previous referral to children's social services in the previous six years which resulted in no further action. The proportion was similar for children assessed as having been affected by sexual abuse (CSA) at 25%. However, the proportion was higher for children assessed as having been affected by sexual exploitation (CSE) at almost a third (32%).
The assessment factors with the highest proportion of children with previous no further action referrals were extra-familial harms including gangs and socially unacceptable behaviour (38%), child criminal exploitation (37%) and going/being missing (37%).
If a local authority identifies there is reasonable cause to suspect the child is suffering, or is likely to suffer significant harm, it will carry out an assessment under section 47 of the Children Act 1989 to determine if it needs to take steps to safeguard and promote the welfare of the child. If concerns are substantiated and the child is judged to be at continuing risk of harm then an initial child protection conference (ICPC) should be convened within 15 working days.
Almost half (49%) of the children assessed as having been affected by CSE in 2023-24 had a section 47 enquiry in the previous six years. The proportion was lower for children assessed as having been affected by CSA at 38%.
The assessment factors with the highest proportion of children with previous section 47 enquiries were gangs (51%) and going/being missing (50%).
In addition to our longitudinal analysis above which looked at child journeys, we conducted analysis on children who were assessed as having been affected by sexual abuse (CSA) and exploitation (CSE) in 2017-18, and went on to enter care in the six years following this assessment. The methodology used is different to our child journey analysis, as we limited the cohort to children who had one assessment in 2017-18 and entered care as part of the same episode.
Overall, 6% of children assessed in 2017-18 entered care in the next six years, and the proportion was similar for children who were assessed as having been affected by CSA at 7%. The proportion was higher for children who were assessed as having been affected by CSE at 10%.
When a child is assessed by children's services their primary need is recorded. This list is hierarchical and where more than one need is identified then the need ‘highest’ up the list is reported.
Just over three quarters (76%) of children assessed as having been affected by CSA in 2017-18 who entered care in the next six years, had the primary need ‘abuse or neglect’. This proportion was lower for children assessed as having been affected by CSE at 63%. Instead, children assessed as having been affected by CSE more commonly had ‘family dysfunction’ (16%) and ‘family being in acute stress’ (11%) recorded as their primary need.
Information is collected on the legal status underlying being 'looked after' , which helps to describe why the child is being looked after.
The most common legal status for children assessed as having been affected by CSA or CSE in 2017-18 who became looked after in the next six years was a care order. However, over two fifths (44%) of children assessed as having been affected by CSE were looked after under voluntary agreement, whereas the proportion was lower for children assessed as having been affected by CSA at just over one fifth (23%).
It became illegal for young people to be placed in former semi-independent and independent living placements from 28 October 2023. Between 28 April 2023 and 27 October 2023 former semi-independent and independent living placement providers were required to submit a completed application to Ofsted to continue to operate as a supported accommodation provider beyond that date. Our analysis looked at the placement of children who entered care following their assessment in 2017-18 up until 31 March 2024, and so children in these placements are grouped into category 'Independent and semi-independent living arrangements/supported accommodation.
The majority of children assessed as having been affected by CSA or CSE in 2017-18 who became looked after in the next six years were in foster placements. The proportion of children assessed as having been affected by CSE in independent and semi-independent living arrangements/supported accommodation (34%) was more than twice as high as the proportion of children assessed as having been affected by CSA (15%). This reflects the age difference between the two groups.
Most children who became looked after in the six years after their 2017-18 assessment changed placement, although the proportion who had 3 or more placement moves during the six years was higher for children assessed as having been affected by CSE at 23% than for children assessed as having been affected by CSA (16%).
Children assessed as having been affected by CSE were also more likely to move placement due to a carer request (20%) than children assessed as having been affected by CSA (14%).
Almost three quarters (74%) of children assessed as having been affected by CSA in 2017-18 who became looked after in the next six years, left care in the same period. The proportion was similar for children assessed as having been affected by CSE at 72%.
The most common reason for the episode of care ceasing was that the child returned home to their parents. Almost a fifth (18%) of all children assessed in 2017-18 who became looked after left in the same period on a special guardianship order but this proportion was lower for children assessed as having been affected by CSA (11%) and much lower for children assessed as having been affected by CSE (5%).
18% of children assessed as having been affected by CSA in 2017-18 who became looked after were adopted. The proportion was much lower children assessed as having been affected by CSE at 5%. Instead, a higher proportion of children assessed as having been affected by CSE moved into independent living.
The Children Act 2004 (as amended by the Children and Social Work Act 2017 (opens in new tab)) states that where a local authority in England knows or suspects that a child has been abused or neglected, the local authority must notify the Child Safeguarding Practice Review Panel (opens in new tab) (the Panel) and by extension DfE and Ofsted if:
(a) the child dies or is seriously harmed in the local authority's area, or
(b) while normally resident in the local authority's area, the child dies or is seriously harmed outside England.
The local authority must also notify the Secretary of State for Education and Ofsted where a looked after child had died, whether or not abuse or neglect is known or suspected.
The statutory duty to notify the Panel, and by extension DfE and Ofsted, came into effect from 29 June 2018. Prior to this, notifications were only made to Ofsted.
Serious harm includes (but is not limited to) serious and/or long-term impairment of a child's mental health or intellectual, emotional, social or behavioural development. It also covers impairment of physical health.
The statistics show the number of incidents notified in the period, rather than the number of incidents that occurred in the period and are based on one notification per incident, which can relate to more than one child in some instances.
The local authority should notify the Panel within five working days of becoming aware that the incident has occurred. However, in some cases, there can be delays in notifications being submitted. In addition, whilst data quality has improved in recent years, DfE is aware that, in some instances, not all incidents that meet the definition for a serious incident are notified.
The notification system that collects characteristics of the serious incident went live in February 2019. From the year ending 31 March 2020 to 2025, child sexual abuse was defined as intra-familial sexual abuse. From 1 April 2025, the notification system was updated to collect both intra-familial and extra-familial sexual abuse. This new collection of extra-familial sexual abuse is not included in this publication as we do not have a full year of data.
All data in this release is based on single notifiable incidents. In some instances, a notification can relate to more than one child; when this happens, the child characteristic information is based on one child only (either the first child listed on the notification or the child who the harm relates to). Therefore, in some instances, the child characteristics information doesn’t always relate to all children involved in the serious incident and this should be considered when interpreting the figures. The characteristics information is identified as at the point of notification.
Comparisons between different characteristics should be made with caution due to the small sizes of some breakdowns.
In the year to 31 March 2025, the number of notifications related to child sexual abuse fell for the second consecutive year to 26, while notifications related to child sexual exploitation steadily declined from 19 in 2020 to 14 in 2025.
The DfE is aware that, in some instances, not all incidents that meet the definition for a serious incident are notified. It is not possible to ascertain whether the fall in 2025 reflects a decrease in serious incidents or whether fewer notifications were reported compared with earlier years.
Notifications related to both child sexual abuse and child sexual exploitation are most common for children aged 11 to 15 accounting for 42% and 59% respectively. This is an over-representation in comparison with the general population where 11 to 15 year olds accounted for 29% of children aged 0 to 17 according to the 2021 census (opens in new tab).
Between 2020 and 2025, 79% of child sexual abuse notifications and 85% of child sexual exploitation notifications related to females. In comparison, females accounted for a smaller proportion among children (aged 0 to 17) from the 2021 census (opens in new tab) (49%).
Between 2020 and 2025, 69% of child sexual abuse notifications and 67% of child sexual exploitation notifications related to children with White ethnic backgrounds.
Children with Mixed ethnic backgrounds accounted for 10% of child sexual abuse notifications which is slightly higher than the total child population from the 2021 census (opens in new tab) (7%). Children from Black ethnic backgrounds accounted for 13% of child sexual exploitation notifications which is slightly higher than the total child population from the 2021 census (opens in new tab) (5%).
Based on notifications with a known disability status, 10% of child sexual abuse notifications and 15% of child sexual exploitation notifications related to children with a disability which is slightly lower than the proportion of all Serious incident notifications related to children with a disability in 2025 (20%).
Most child sexual abuse notifications related to children living at home, accounting for just over three in every four notifications (76%). Most child sexual exploitation notifications related to children living in residential children's homes (30%) and home (28%).
Between 2020 and 2025, 5% of notifications related to children on a protection plan at the time of the child sexual abuse incident and 12% related to children on a protection plan at the time of the child sexual exploitation incident.
The following points on children known to any agency should be considered when interpreting the statistics:
- whether a child was known to any agency is interpreted differently across local authorities. Some local authorities only include children known to Children’s Social Care but in other cases can include agencies such as the police, GP services, health visitors, early help, midwifery etc.
- children known to GP services have been excluded (where this information was recorded) from 2023 onwards but were included prior to 2023.
- they can include children involved with agencies previously, and not just at the time of the incident.
- information is recorded at the time of reporting. Local authorities may not yet know whether the child was known to any agency.
Between 2020 and 2025, 93% of the child sexual abuse incidents and 95% of the child sexual exploitation incidents involved children who were known to an agency which is higher than the proportion of all Serious incident notifications related to children who were known to agency in 2025 (86%).
This release includes statistics for 2024 and 2025 on notifications relating to children looked after by local authorities (CLA). This follows the additions of CLA status in December 2022 and accompanying guidance in April 2023 to the Child Safeguarding Incident Notification System (opens in new tab). Each notification for a CLA death represents the death of one CLA child.
A child is looked after by a local authority if they are:
- Provided with accommodation, for a continuous period of more than 24 hours;
- Subject to a care order; or
- Subject to a placement order.
Between 2024 and 2025, over 1 in 4 of all child sexual abuse notifications (26%) related to CLA and over 3 in 4 of all child sexual exploitation notifications (76%) related to CLA.
This release contains statistics for 2024 and 2025 on notifications relating to children with a child in need plan, following the addition of child in need plan status in December 2022 to the Child Safeguarding Incident Notification System (opens in new tab).
Between 2024 and 2025, 14% of notifications related to children on a child in need plan at the time of the child sexual abuse incident and 10% related to children on a child in need plan at the time of the child sexual exploitation incident.
Each entry in police recorded crime data relates to one victim, with the most serious crime being recorded. A victim of a crime with multiple perpetrators would only be recorded once in the year of the crime being reported.
Contact child sexual abuse excludes offences related to indecent images.
The total police recorded crime rate for contact child sexual abuse at a national level reached a peak of 5.41 per 1,000 children in 2021-22 and increased by 41% between 2015-16 and 2024-25. Some of this increase is likely driven by improvements in reporting and recording practice.
Over the same period the rate of children with either child sexual abuse or child sexual exploitation (CSAE) identified as a factor at assessment has fallen by 9%, and the two rates began diverging from 2016-17. In 2024-25, the police recorded crime rate for contact CSA was 4.75 per 1,000 children in the population, whereas the CSAE rate was 3.20, 33% lower.
Non-recent crime is defined as a crime that has been reported more than a year after it occurred. Breakdowns between non-recent and recent crime data are available from 2016-17 onwards.
A proportion of total contact CSA crimes are non-recent and so the victim may not be a child at the time the crime was reported. The proportion of contact CSA that was non-recent was 32% in 2024-25, and has fallen by 20% since 2016-17. The recent crime rate has largely followed a similar trend to the total crime rate and was lower than the CSAE rate until 2023-24.
Data on the age of the victim of a crime is not available and so the true relevant crime rate affecting only individuals who were children at the time of reporting cannot be determined, but it will lie somewhere between the total and recent crime rates.
The total police recorded crime rate for contact CSA and the CSAE rate both vary considerably at the police force area level, with maximums of 8.26 per 1,000 children in the population (City of London) and 6.06 (Humberside), respectively, and minimums of 2.73 (Metropolitan) and 0 (City of London) in 2024-25. Some police force areas see minimal or no divergence at all between the two rates and others have a gap between the two rates that is greater than at the national level. Derbyshire has a CSAE rate (4.93) above its total police recorded crime rate (4.65).
The breakdown of recent and non-recent crimes for contact CSA is not available at police force area level, however we would expect the true relevant crime rate to lie somewhere between the total and recent crime rates as at the national level.
Find out how and why we collect, process and publish these statistics.
Ad hoc official statistics are one off publications that have been produced as far as possible in line with the Code of Practice for Statistics.
This can be broadly interpreted to mean that these statistics are:
Find out more about the standards we follow to produce these statistics through our Standards for official statistics published by DfE guidance (opens in new tab).
If you have a specific enquiry about Children in need: A focus on sexual abuse and exploitation statistics and data:
If you have a media enquiry:
Telephone: 020 7783 8300
If you have a general enquiry about the Department for Education (DfE) or education:
Telephone: 037 0000 2288
Opening times:
Monday to Friday from 9.30am to 5pm (excluding bank holidays)
Don't include personal or financial information like your National Insurance number or credit card details.