Official Statistics

Serious incident notifications from local authority children’s services 2016 to 2017: main findings

Published 31 October 2017

This release covers the period between 1 April 2016 and 31 March 2017 and contains data on notifiable incidents involving the care of children aged under 18 years. Notifiable incidents are those involving death or serious harm to a child where abuse or neglect is known or suspected, and also deaths of children looked after and children in regulated settings.

The analysis is based on information provided to Ofsted by local authorities at the time of notification and in any further updates received from local authorities and Local Safeguarding Children Boards (LSCBs). The data is experimental, because the statistics do not yet meet the rigorous quality standards of National Statistics.

The number of serious incidents notified to Ofsted has risen since last year.

Between 1 April 2016 and 31 March 2017, Ofsted received 433 serious incident notifications. This was a 14% increase on the number of notifications in 2015 to 16 (379).

The number of child deaths notified to Ofsted has risen since last year.

Between 1 April 2016 and 31 March 2017, 211 cases of child deaths were notified, compared with 171 in the previous year. This was a 23% increase from the previous year but there is no long-term trend.

1. Introduction

Under the Working together to safeguard children statutory guidance, local authorities should notify Ofsted about incidents that meet the following criteria:

• a child has died (including cases of suspected suicide), and abuse or neglect is known or suspected
• a child has been seriously harmed and abuse or neglect is known or suspected
• a looked after child has died (including cases where abuse or neglect is not known or suspected), or
• a child in a regulated setting or service has died (including cases where abuse or neglect is not known or suspected)

There are approximately 12 million children in England. Ofsted received notification of 433 serious incidents relating to 477 children that met the criteria in statutory guidance: this equates to 0.00004% of the total child population in England.

The number of notifications does not equate to the number of incidents that may have occurred within that time period. In each reporting year, a number of notifications are about incidents that occurred before the reporting period. The proportion of notifications received in 2016 to 2017 that related to historical incidents was the lowest in the last 4 years (69, or 16% of all notifications).

Chart 1: Number of notifications received that related to historical incidents by year

Year	Historical incidents	All notifications received
2013 to 2014	69	298
2014 to 2015	101	385
2015 to 2016	68	379
2016 to 2017	69	433

In March 2014, Ofsted introduced an online notification form for local authorities to use to report serious childcare incidents. The new computerised reporting system made it easier for local authorities to report the incidents to Ofsted. This might have been a reason why there was a rise in the number notifications received in 2014 to 2015.

2. Notifications

Overall, the number of incidents notified to Ofsted has risen over the last 4 years, apart from a small dip in notifications received in 2015 to 2016.

Chart 2: The number of serious incident notifications to Ofsted

Of the 433 incidents notified to Ofsted in 2016 to 2017, 211 (49%) related to child deaths compared with 171 (45%) in the previous year, and 222 (51%) related to serious harm compared with 208 (55%) in 2015 to 2016. The 4-year data shows that there has been a steady increase in the number of notifications about incidents of serious harm while the number of notifications about child deaths tend to fluctuate from year to year.

Of all notifications received in 2016 to 2017, 223 (52%) were about boys and 209 (48%) were about girls.^{[footnote 1]} The proportion of notifications about boys and girls over the 4-year period shows no real trend in the data. The figures in 2016 to 2017 are very similar to figures in 2013 to 2014.

Chart 3: Gender split of serious incident notifications by year

Year	Girls	Boys
2013 to 2014	49%	51%
2014 to 2015	52%	47%
2015 to 2016	44%	56%
2016 to 2017	48%	52%

A third (145) of the incidents notified to us in 2016 to 17 were about children under the age of one. Just over half of these (79) were about a non-accidental injury by a parent or carer, or about children who had been killed by a parent or carer. This was similar to the previous year.

Over the last 3 years, the number of notifications about children aged one has increased while the number of notifications about children aged 11 and over has declined following a large increase in 2014 to 2015. Two out of 5 (172) notifications were about children aged 11 years or older.

Chart 4: Split of notifications by age group

As in the previous years, in 2016 to 2017, the majority of notified incidents (307 or 71%) were for children who were White while nearly a quarter (99) related to children from minority ethnic groups.

3. Child deaths

In the year ending 31 March 2017, Child Death Overview Panels (CDOP) completed 3,575 child death reviews in England. The categories of death reviewed by CDOPs include:

• medical causes, for example a life limiting condition or a neonatal death
• sudden, unexpected death
• trauma and other external factors
• suicides or deliberate self-inflicted harm
• deliberately inflicted injury, abuse or neglect.

Of these child deaths, 211 (6%) met the ‘Working together’ criteria and were also notified to Ofsted. These have been categorised below according to information provided by the relevant local authority at the time of notification about the nature of the incident. Because incidents should be notified to Ofsted within 5 days of occurring, information available to the local authority at this stage about the cause of a child’s death is often limited. There is no requirement in guidance for local authorities to update Ofsted about the cause of a child’s death once they have established it. Consequently, the information held by Ofsted about the cause of death may not be accurate. Where the local authority has given Ofsted further information, for example about post-mortem or criminal investigations, we have included it in the tables below.

All information in this release about child deaths should be treated with caution because our knowledge is based on incomplete data. There may be ongoing post mortems or criminal investigations, where the details are not known to Ofsted. There are also some deaths that remain unexplained or for which the cause of death is not known or for which it has not been possible for the agencies investigating the death to find out all the circumstances.

As stated previously, the figures for child deaths may not all relate to deaths that occurred during the relevant period. This statistical release includes 17 notifications of child deaths that occurred before 1 April 2016.

Table 1: Cause of death for child deaths notified to Ofsted^{[footnote 2]}

Cause of death	2013 to 2014	2014 to 2015	2015 to 2016	2016 to 2017
Accidents (including possible overlay[footnote 3])	12	31	18	16
Dangerous behaviour (incl. substance misuse)	-	5	2	4
Killing/non-accidental injury by parent or carer	26	39	19	34
Killing/non-accidental injury by unrelated person	5	12	8	13
Natural causes (incl. life-limiting disability and illness)	48	31	42	49
Neglect	4	5	8	9
Suicide	-	29	26	35
Unknown or unascertained	38	38	48	51
Total	163	190	171	211

In 2016 to 2017, Ofsted was notified of 211 deaths of children, an increase on the 171 notified deaths in the previous year. As the above table shows, there is no long term trend as the numbers have fluctuated year on year. There has been an increase in both the number of notifications about killing or non-accidental injury by a parent or carer, from 19 (11%) in 2015 to 2016 to 34 (16% of the total deaths) in 2016 to 2017, and in the killing or non-accidental injury by unrelated person, from 8 in 2015 to 2016 to 13 in 2016 to 2017, but there is no long-term trend. The number of notifications about children dying of natural causes has increased steadily, from 31 in 2014 to 2015 to 49 in 2016 to 2017.

As in previous years, the age group that had the largest number of notifications was under ones (74 or 35%). Seventeen (23%) of these children are reported to have died as a result of killing or non-accidental injury by parent or carer. One in 5 died of natural causes, such as life-limiting disabilities or illness, similarly to the previous years. The reason of death for 2 out of 5 child deaths was unknown or unexplained at the time of notification. There continues to be a fall in the number of notifications categorised as accidents (including overlay) for this age group, from 12 (18%) in 2014 to 2015 to 5 (7%) in 2016 to 2017.

Two in 5 (35 of 80) deaths of children aged 11 years or older in 2016 to 2017 were suicides. Twenty of these children were boys and 15 were girls. Three of the children were looked after by the local authority at the time of the incident. These numbers and percentages of the child deaths by suicide are higher than in the previous year but are similar to the level seen in 2014 to 2015.

Of the 211 incident notifications of child death, 123 (58%) related to boys and 87 (41%) related to girls. This is very similar to the picture in 2015 to 2016.^{[footnote 4]}

The majority of child deaths notified to Ofsted during 2016 to 2017 were for White children (141 or 67%) and 24% (50) were for children from minority ethnic groups. This continues the up and down trend over the last 3 years.

Of the 211 reported child deaths, 23 (11%) were of children who were subject to a child protection plan, a decrease from 29 (17%) in 2015 to 2016. One in five (43) reported deaths were for children looked after. The majority of these children (29 out of 43) were reported to have died from natural causes. This reflects the fact that many children become looked after because they have complex life limiting disabilities and a local authority takes on some legal responsibility for the care and wellbeing of the child. The remaining 14 child deaths of children looked after were due to accidents (2), non-accidental injury by an unrelated person (3), neglect (one), suicide (3), and for 5 of them the reason was unknown or unascertained. The proportion of deaths of children looked after is similar to last year’s picture, and an improvement on 2013 to 2014, when 1 in 4 child deaths affected children looked after.

4. Serious harm

Ofsted received 222 notifications of incidents involving serious harm to a child. This related to 261 children, which is 0.00002% of the total child population in England. In some cases, where the incident involved harm to more than one child, the notification form only provided information about the first child identified. All data in this release is based on single notifiable incidents and the information is only about the first child identified on the incident form.

At the time of a notification being made to Ofsted, the local authority may not have had full information about the serious incident. Local authorities are not required to update Ofsted with new information. The data in the table below includes updated information where we have been given it after the notification.

Table 2: Cause of serious harm for incidents notified to Ofsted

Cause of serious harm	2013 to 2014	2014 to 2015	2015 to 2016	2016 to 2017
Accidents	4	3	4	8
Neglect by parent or carer	13	25	26	23
Non-accidental injury by parent or carer	34	55	75	82
Non-accidental injury by unrelated person	11	6	14	9
Other (incl. unknown or unascertained)	19	24	29	30
Self-harm/dangerous behaviour	4	8	12	16
Sexual abuse by family member	15	13	10	10
Sexual abuse/child sexual exploitation by unrelated person	35	61	38	44
Total	135	195	208	222

Like last year, the most frequent overall cause of serious harm was non-accidental injury by a parent or carer. Incidents in this category have continued to increase over the last 4 years, with 82 (37%) notifications in 2016 to 17. All but one incidents of this type affected children aged under 5 (81 or 99%), of whom 41 were boys and 40 were girls. This is slightly above last year’s figure of 91% (68), of whom 48 were boys and 20 were girls. ^{[footnote 5]}

The next highest category of serious harm was incidents of sexual abuse or child sexual exploitation by an unrelated person (44). The majority of these incidents (40 or 91%) affected children aged 11 years or older, of whom 31 were girls and 9 were boys. Half of these children were looked after by the local authority and 3 were subject to a child protection plan at the time of the incident.

The highest number of serious harm notifications was about children aged under one (71 or 32%). This is similar to previous years, with the exception of 2014 to 2015 when children aged between 11 and 15 accounted for the highest number of serious harm incidents. The vast majority of children under one sustained non-accidental injuries by a parent or carer (62 or 87%). Over the 4-year period, this category has continued to be the highest amongst children under one and has steadily been increasing both in number and proportion.

Of all serious harm notifications, 122 (55%) were about girls and 100 (45%) about boys. There is no real trend in the year on year data.

The majority of serious harm notifications were about White children (166 or 75%). In the previous 3 years, the percentage of White children has been around 70%. Children from minority ethnic groups accounted for 22% (49), which is also a slight increase from previous years.

Of all children who suffered some form of serious harm, 26 (12%) were the subject of a child protection plan and 57 (26%) were looked after by a local authority at the time of the serious incident.

5. Serious case reviews

The LSCB in a local authority area is responsible for deciding whether or not to initiate a Serious Case Review (SCR) in response to an incident.

For 113 (26%) of the 433 incidents notified to Ofsted, the relevant LSCB told Ofsted that it had decided to initiate a SCR. LSCBs notified Ofsted of 139 decisions not to initiate a SCR. For the remaining 181 cases, Ofsted received no information from the relevant LSCB about whether or not it had decided to conduct a SCR. Ofsted’s data about SCRs is therefore incomplete.

6. Revisions to previous release

This is the first release of the data for the period 1 April 2016 to 31 March 2017.

7. Notes

Requirement for local authorities to notify Ofsted

The criteria for notifiable incidents are set out in Working together to safeguard children statutory guidance, page 74, paragraph 13 onwards:

13). A notifiable incident is an incident involving the care of a child which meets any of the following criteria:

• a child has died (including cases of suspected suicide), and abuse or neglect is known or suspected
• a child has been seriously harmed and abuse or neglect is known or suspected
• a looked after child has died (including cases where abuse or neglect is not known or suspected); or
• a child in a regulated setting or service has died (including cases where abuse or neglect is not known or suspected)

14). The local authority should report any incident that meets the above criteria to Ofsted and the relevant LSCB or LSCBs promptly, and within five working days of becoming aware that the incident has occurred.

15). For the avoidance of doubt, if an incident meets the criteria for a Serious Case Review (see below) then it will also meet the criteria for a notifiable incident (above). There will, however, be notifiable incidents that do not proceed through to Serious Case Review.

16). Contact details and notification forms for notifying incidents to Ofsted are available on Ofsted’s website.

Serious case reviews

The Working together to safeguard children statutory guidance says on page 78:

Decisions whether to initiate an SCR

The LSCB for the area in which the child is normally resident should decide whether an incident notified to them meets the criteria for an SCR. This decision should normally be made within one month of notification of the incident. The final decision rests with the Chair of the LSCB. The Chair may seek peer challenge from another LSCB Chair when considering this decision and also at other stages in the SCR process. The LSCB should let Ofsted, DfE and the national panel of independent experts know their decision within five working days of the Chair’s decision.

8. Methodology

Due to concerns about confidentiality, we have decided not to publish the underlying dataset for the Serious Incident Notifications statistical release this year. This allows us to provide users with a rich and detailed analysis of the data while minimising the risk of identification of individuals from the dataset. If you are a researcher with a need to access the underlying data please contact [email protected]

The data in the release is from serious incident notifications received by Ofsted between 1 April 2016 and 31 March 2017.

From 1 April 2016 to 31 March 2017, Ofsted was notified of 442 incidents by local authorities using the online notification form for local authority officers on gov.uk. Out of these notifications, 9 did not meet the criteria in the statutory guidance so have not been counted in the data in this statistical release.

The figures for child deaths caused by killing or non-accidental injury by a parent or carer also includes notifications where the child was killed or fatally injured by another family member. Figures for child deaths by killing or non-accidental injury by an unrelated person include cases where the injury was caused by a person living in the household who was not a family member (for example an adult in a relationship with a parent).

The ethnicity has not been reported for 6% of serious incident notifications received in 2016 to 2017. Also, for 51 child deaths (24%) the main cause was recorded as unknown or unascertained.

The number of SCRs notified to Ofsted might not equal the number of SCRs that have been commissioned in the period. Decisions to initiate SCRs are not always communicated promptly to Ofsted.

The number of incident notifications received during the time period covered in this statistical release does not necessarily equate to the number of incidents that occurred between 1 April 2016 and 31 March 2017. Some of the incidents that were notified in this period occurred before 1 April 2016.

9. Further information

9.1 Contact for comments or feedback

If you have any comments or feedback on this publication, please contact the Social Care Data Team on 03000 130020 or [email protected]

9.2 Acknowledgements

Thanks to Klara Davies, Anna Otczyk and Anne Gair for their contribution to this statistical release.

1. There was one notification in 2016 to 2017 where the gender was unknown as the child died at birth and the gender was not recorded by the local authority. The gender for one child was also not recorded in 2014 to 2015 therefore the percentages in the chart do not add up to 100%. ↩

2. In 2013 to 2014 the category of ‘dangerous behaviour’ included notifications of suspected suicides therefore the number of notifications for these two categories cannot be shown separately. ↩

3. Overlaying describes the accidental death by smothering caused by a larger individual sleeping on top of a child when sharing a bed or sofa. ↩

4. The percentages do not add up to 100 as one child’s gender was unknown/not reported. ↩

5. One incident was recorded with child’s age unknown. ↩