Official Statistics

Annex: Devolving child decision-making pilot programme statistics, October 2024 to September 2025

Published 6 November 2025

1. Introduction

As part of wider National Referral Mechanism (NRM) reform activity, the devolved decision-making pilot for child victims of modern slavery aims to test devolving the responsibility to make NRM decisions from the Home Office to local authorities. This approach enables decisions about whether a child is a victim of modern slavery to be made by those involved in their care, and ensures the decisions made are closely aligned with the provision of local, needs-based support and any law enforcement response.

In June 2021 the Home Office launched the pilot in 10 sites. In the pilot, following training and assurance provided by the Single Competent Authority (SCA), decisions for children aged 17 and under (who are more than 100 days away from being aged 18) are made by those relevant local authorities rather than the Home Office. To enable further testing 10 additional pilot sites were launched in early 2023. For the year October 2024 to September 2025, the pilot operated in 20 sites, covering 30 local authorities across England, Scotland and Wales. The third phase of the pilot launched in April 2025 and identified 7 additional sites for expansion of the pilot and one site was extended to incorporate 3 neighbouring local authorities. The successful pilot sites will come on board this autumn (2025) and begin making decisions in November. This analysis provides a comparison of children referred to the NRM, where decision-making is devolved to local authorities, with those where decisions have been made centrally by the Home Office.

Further information regarding the pilot programme can be found in the guidance. This annex is accompanied by data tables.

Figure 1: Pilot site areas by phase of expansion

Notes:

1. Phase 1 of the pilot began in June 2021; phase 2 began between February and April 2023 and phase 3 began in September 2025.

Table 1: Pilot site areas, country and phase of expansion

Location	Site	Phase
England	Hull City Council	1
England	London Borough of Barking and Dagenham	1
England	London Borough of Islington and London Borough of Camden	1
England	North Lincolnshire and North East Lincolnshire	1
England	North Yorkshire Council and the City of York	1
England	Royal Borough of Kensington and Chelsea and the Borough of Westminster	1
England	Solihull Metropolitan Borough Council	1
Scotland	Glasgow City Council	1
Wales	Cardiff Council	1
Wales	Newport City Council and Torfaen, Blaenau Gwent, Monmouth and Caerphilly	1
England	Croydon Council	2
England	East Sussex County Council and Brighton and Hove City Council	2
England	Enfield Council	2
England	Hampshire County Council	2
England	London Borough of Bexley	2
England	London Borough of Lewisham	2
England	London Borough of Redbridge	2
England	Oxfordshire County Council	2
England	Warwickshire County Council	2
England	Wiltshire Council and Swindon Metropolitan Borough Council	2
England	Birmingham Children’s Trust	3
England	Hampshire County Council (expanded to include Isle of Wight Council, Southampton City Council and Portsmouth City Council in 2025)	3
England	Shropshire Council	3
England	Southwark Council	3
England	Telford and Wrekin Council	3
England	West Sussex County Council	3
Northern Ireland	Department of Health Northern Ireland	3
Scotland	City of Edinburgh Council	3

2. Referrals

2.1 Overall

Between October 2024 and September 2025, there were 6,798 children referred to the NRM who were aged 17 and under, which was a 14% increase from the number referred between October 2023 and September 2024 (5,984). Of those, 90% (6,122) were more than 100 days away from being aged 18 upon referral and therefore eligible for the pilot. Of those, 18% (1,113) had their safeguarding responsibility fall to a local authority within the pilot sites (herein called pilot children), and 82% (5,009) did not (herein called non-pilot children) (data table A1; figure 2).

Figure 2: Quarterly NRM referrals for pilot and non-pilot children

Source: SCA

Notes:

1. The pilot began in June 2021 (hence the lower volumes for quarter 2, 2021) with 10 pilot site areas, which was doubled to 20 between February and April 2023.
2. A child is defined as someone more than 100 days away from being aged 18 at the point of their referral.

2.2 Age at referral

Of the pilot children referred to the NRM between October 2024 and September 2025, three-quarters (75%; 840) were aged between 15 and 17 years old at the time of referral. A further 24% (262) were aged between 11 and 14 years old, and the remaining 1% (11) were aged 10 or under. The distribution across age groups is broadly similar to pilot children referred between October 2023 and September 2024 (data table A2).

For non-pilot children, 81% (4,953) were aged between 15 and 17 years old at the time of referral. A further 18% (1,091) were aged between 11 and 14 years old, and the remaining 1% (78) were aged 10 or under (data table A2).

2.3 Unaccompanied asylum-seeking children

An Unaccompanied Asylum-Seeking Child (UASC) is a person aged 17 and under who is applying for asylum, is separated from both parents, and is not being cared for by an adult who in law or by custom has responsibility to do so. Between October 2024 and September 2025, 27% (296) of pilot children referred to the NRM have been identified as UASC.

2.4 Gender

Overall, of pilot children referred between October 2024 and September 2025, around one-quarter were female (25%; 282) and around three-quarters were male (74%; 826). For non-pilot children, 20% were female (984) and 80% were male (4,018) (data table A5).

2.5 Exploitation type

Between October 2024 and September 2025, criminal exploitation was the most commonly reported type of exploitation for both children inside the pilot (54%; 597) and those outside the pilot (48%; 2,387) (Data Table A6).

Among male children, criminal exploitation was the most commonly reported exploitation type for those inside the pilot (65%; 540) and for those outside the pilot (56%; 2,263).

Among female children, sexual exploitation was most commonly reported, both for those inside the pilot (44%; 123) and for those outside the pilot (40%; 395).

Figure 3: NRM referrals for pilot children by gender and exploitation type between October 2024 and September 2025

Source: SCA

Notes:

1. Less commonly reported exploitation types are grouped as ‘Other’ and a full breakdown is provided in data table A6.
2. Does not include referrals where the gender is ‘Other’.

2.6 Nationality

The most common nationality of pilot children referred between October 2024 and September 2025 was UK, which accounted for 65% (723) of referrals. The majority of these were male (73%; 528) with 26% (190) being female (figure 4). The second most commonly referred nationality for pilot children was Sudanese (4%; 50) and the third was Vietnamese (3%; 38).

For non-pilot children, the most common nationality was also UK (53%; 2,639), although as a lower proportion than for pilot children. Most of these were males (79%; 2,087) and 21% (546) were female. The second most commonly referred nationality was Sudanese (6%; 305) and the third was Vietnamese (6%; 289) (data table A7).

Figure 4: NRM referrals for pilot children by gender and nationality between October 2024 and September 2025

Source: SCA

Notes:

1. Less commonly referred nationalities are grouped as ‘Other’ and a full breakdown is provided in data table A7.
2. Does not include referrals where the gender is ‘Other’.

3. Decisions

The pilot programme provides extensive quality assurance and oversight of local authority decision-making. The SCA quality assures and issues decisions made by pilot sites. It is important to note that Local Authorities are the primary service provider for safeguarding and supporting child victims.

3.1 Reasonable grounds decisions

Between October 2024 and September 2025, 1,051 reasonable grounds decisions were issued for pilot children. Of these 88% (925) were positive. This is a decrease from the proportion in the previous year (91%; 902) (data table A8; figure 5).

Over the same period, 5,087 reasonable grounds decisions have been issued for non-pilot children. Of these, 82% (4,179) were positive. This is an increase from the proportion in the previous year (77%; 3,918) (data table A8).

The test for a reasonable grounds decision changed from 30 January 2023, and was further amended from 10 July 2023. From this latter date, the test is objective whereby the decision maker must agree with the statement that there are “reasonable grounds to believe that a person is a victim of modern slavery”.

Figure 5: Quarterly NRM reasonable grounds decisions for pilot children

Source: SCA

Notes:

1. Based on the quarter in which the decision was issued for pilot children.

For pilot children, the average (median) time taken from referral to reasonable grounds decisions made between October 2024 and September 2025 was 61 days, compared to 69 in the previous year. For non-pilot children, the average time taken was 6 days, the same as the previous year (see data table A9 for a quarterly breakdown).

The average time to reasonable grounds decisions is longer for pilot children because the pilot operates on a different timescale, with guidance outlining that a reasonable grounds decision should be made no later than 45 days from the date the pilot area receives the referral. The longer decision-making timescale for reasonable grounds decisions in the pilot allows time for pre-panel information gathering and for the local multi-agency meeting to be organised. Reasonable grounds decision timescales are therefore not directly comparable.

3.2 Conclusive grounds decisions

Overall, between October 2024 and September 2025, 874 conclusive grounds decisions have been issued for pilot children, of which 81% (706) were positive. This is a decrease from the proportion in the previous year (86%; 724) (data table A10; figure 6).

For non-pilot children, there were 6,593 conclusive grounds decisions issued by the SCA over the same period, of which 79% (5,228) were positive. This is a slight increase from the proportion in the previous year (78%; 3,025) (data table 10).

Figure 6: Quarterly NRM conclusive grounds decisions for pilot children

Source: SCA

Notes:

1. Based on the quarter in which the decision was issued for pilot children.

The average (median) time taken from referral to conclusive grounds decisions for pilot children made between October 2024 and September 2025 was 99 days, compared to 102 in the previous year. For non-pilot children, the average time taken was 331 days (see data table A11 for a quarterly breakdown).

The average time to conclusive grounds decisions for pilot children is much shorter, because the pilot has been designed to create shorter decision-making timescales from the point of referral to a final decision being made. Guidance outlines that a conclusive grounds decision must take place no later than 45 days after the reasonable grounds decision (a maximum of 90 days from when the referral was received by the pilot site). Moreover, a positive reasonable and conclusive grounds decision can be made at the same time and at the same meeting, if there is sufficient evidence to do so, bringing down the conclusive grounds decision-making timescale.

It is important to note that wait times presented here are for cases that received a conclusive grounds decision in this period and do not reflect the waiting time of all cases within the system. Wait times are taken as the difference between the date of the referral receipt and decision date, and do not exclude any periods during which referrals may be suspended, withdrawn or previously closed. Hence, actual average wait times will be shorter.

3.3 Joint decisions

Pilot children can receive joint positive reasonable and conclusive grounds decisions at the same time and at the same meeting, if there is sufficient evidence to do so. Between October 2024 and September 2025, there were 368 joint positive decisions issued (data table A12). The average (median) time taken from referral to joint decisions made between October 2024 and September 2025 was 64 days (see data table A13 for a quarterly breakdown).

These decisions are included in the statistics for reasonable grounds decisions and conclusive grounds decisions, so should not be combined.

4. Quality information

These statistics are produced in accordance with quality requirements of the Code of Practice for Statistics. It uses guidance published on the Government Analysis Function, which considers 5 aspects of quality:

• relevance
• accuracy and reliability
• timeliness and punctuality
• accessibility and clarity
• coherence and comparability

Relevance

Statistics on the devolving child decision-making pilot programme provide a picture of the volumes of children who have their NRM decisions made by local safeguarding partners, instead of the Home Office. They support efforts to test whether determining if a child is a victim of modern slavery within existing safeguarding structures is a more appropriate model for making modern slavery decisions for children.

Accuracy and reliability

Data is based on an extract from a live spreadsheet and a live case management system taken in October 2025. The recorded data might be updated as new information comes to light. As such, the statistics are subject to change over time and so may differ from data included in previous bulletins.

The number of non-pilot children is calculated by using the age of a potential victim at the time of referral. This is quantified by using the date of birth and date the referral was received by the SCA. Children more than 100 days away from being aged 18 upon referral, who are not part of the pilot, are considered non-pilot children.

All percentages in this bulletin are rounded to the nearest whole percentage point.

Timeliness and punctuality

These statistics were published one month after a data extract was taken. This allowed time for the statistics to be collated and quality assured.

Accessibility and clarity

This bulletin has been published to reflect accessibility criteria outlined in guidance published by the Government Analysis Function.

Coherence and comparability

Statistics on decisions are based on referrals made from 14 June 2021. Not all referrals for pilot and non-pilot children would have had decisions at the date the data cut was taken.

Broadly consistent data recording practises (with exception of changes to pilot site coverage) allow for meaningful comparisons over time. The pilot began in June 2021 with 10 pilot site areas, which was doubled to 20 between February to April 2023. A further 7 sites began to receive referrals in September 2025.

Data source

Data is based on an extract from a live spreadsheet and a live case management system taken in October 2025.

Statistical production

This statistical bulletin has been produced to the highest professional standards and is free from political interference. It has been produced by statisticians working in the Home Office Analysis and Insight Directorate under the Home Office’s Statement of compliance with the Code of Practice for Official Statistics, which covers Home Office policy on revisions and other matters.

Data is cleaned using reproducible and tested code. The code reformats data for statistical production, such as by grouping the data into categories (for example age groups) and standardises spelling. Following this, the statistics are collated and outputted as a bulletin with accompanying graphs.

The Chief Statistician, as Head of Profession, reports to the National Statistician regarding all professional statistical matters and oversees all Home Office National Statistics products regarding the Code, being responsible for their timing, content and methodology.

Quality assurance

Data cleaning and grouping is done using reproducible code, which is split into sections to first provide a cleaned dataset, and then produce the bulletin and graphs. The latter 2 are manually cross-checked with the dataset.

5. Related publications

For further information on the devolved decision-making pilot for children, see the published guidance.

Statistics on all potential victims referred to the NRM can be found in the statistics collection.

Feedback and enquiries

We are always looking to improve the accessibility of our documents. If you find any problems, or have any feedback, relating to accessibility, or general questions regarding this publication, please email us NRMStatistics@homeoffice.gov.uk

For press enquires: pressoffice@homeoffice.gov.uk; 0300 123 3535