NAO report: Stronger Home Office oversight needed as Prevent referrals hit record high

• Referrals to Prevent reached a record 10,293 in the year ending September 2025, while the proportion linked to a clear, dominant extremist ideology has fallen to 30% from 73% in 2017.
• Available evidence suggests Prevent is having a positive impact, although it is inherently difficult to prove that it reduces the incidence of terrorism.
• The NAO recommends stronger financial management, performance oversight, and risk management, so the Home Office can better understand Prevent's impact, manage risks and target resources effectively.
• Download the embargoed report (PDF)
• Download the embargoed statement from the Chair of the Public Accounts Committee (PDF)

Prevent plays an important role in the UK's efforts to counter terrorism, and there is evidence of it having a positive impact on those it supports, but rising referrals and increasingly complex cases have put the system under pressure, a new report by the National Audit Office (NAO) has found.^{1 2}

Demand for the programme has grown significantly, with Prevent referrals increasing year on year since 2010, reaching a record 10,293 in the year ending September 2025, although many of these referrals are assessed as being ineligible for support.³ Increased awareness of Prevent following the 2024 Southport attack may have contributed to this rise.

Despite the increase, the Home Office has significant gaps in its understanding of barriers to referral, and the programme may still be missing some people at risk of radicalisation.

Prevent is delivered locally across Great Britain, with 159 multi-agency Channel panels in England and Wales that assess people at risk of radicalisation and arrange appropriate support. Although it is difficult to demonstrate that Prevent is reducing the incidence of terrorism, available evidence suggests that it is having a positive impact on many whom it supports.⁴

The nature of referrals has changed considerably. Just 30% of referrals in the year ending September 2025 were linked to a clear, dominant extremist ideology, down from 73% in 2017. This change is reflected in new referral categories introduced in April 2024.

One of the new categories identifies individuals who lack a clear ideology but have a fascination with extreme violence or mass causality attacks. These individuals are sometimes referred to as violence-fixated, or violence-fascinated, individuals, or VFIs. In the year ending September 2025, 8% of referrals were in this category.⁵ Currently, the Home Office and Counter Terrorism Policing do not have a unified view of when such individuals should meet the threshold for Channel support.

Despite the change in the type of cases encountered, the NAO found that Channel panels take a strong multi-agency approach to their work, considering referrals carefully and drawing on expertise from policing, health, education and social care. The collaborative approach allows panels to build a fuller picture of individuals' circumstances and identify the support they need.

However, the NAO also saw how panels are often forced to make decisions with incomplete data due to delays in receiving information requests from delivery partners.

Prevent is not an expensive programme, particularly compared with the economic and human costs of a terrorist attack. It  makes up a small proportion of the government's annual £3 billion counter-terrorism budget. However, because of spending locally, the Home Office does not have a complete understanding of the programme's full costs.

In 2025-26, the Home Office underspent its £23.4 million Prevent budget by 28%, continuing a six-year pattern in which it underspent by an average of £5.6 million (17%) a year in real terms. At the same time, funding for Prevent to priority local authorities has fallen from £9.5 million to 45 authorities in 2020-21 to £4.4 million to 27 authorities in 2026-27.⁶

The NAO found that the Home Office has not done enough to evaluate the effectiveness of Channel interventions or to communicate where Prevent is working, contributing to continued scepticism about the programme.

To aid improvement in Prevent's delivery, the NAO recommends that the Home Office should:

1. Agree with Counter Terrorism Policing a unified view of the current thresholds for progressing cases with a fascination with extreme violence or mass-casualty attacks to Channel support.
2. Strengthen financial management to avoid significant underspends in future and to improve its understanding of local delivery costs, including how much local authorities spend on running Channel panels.
3. Improve oversight of Prevent, using stronger performance data and risk management to identify pressures and measure success.

Gareth Davies, head of the NAO, said:

"Prevent is operating in an increasingly complex environment, with record referrals and changing patterns of risk.

"To build on the positive cross-agency work at the heart of the programme, the Home Office needs to strengthen its financial management, oversight and evaluation of the programme so it can build on what is working and ensure the programme is ready for the challenges ahead."

Notes to editors

1. Alongside the report The NAO has created a data visualisation that maps Prevent's referral and support pathway. The data visualisation will be available on the following link from 00:01 on Friday 13^th November: An Interactive View of the Prevent System
2. Many of these referrals (78% in YE September 2025) are assessed by Counter Terrorism Policing as being ineligible for support through local Channel panels. Usually, this is because officers assessed the individual as not being at risk of radicalisation. CTP signposts most of the referrals it closes (61% in YE September 2025) to other safeguarding services. However, some referrals were escalated within CTP because the risk was already assessed as too high for Prevent.
3. 58% of individuals left Channel with no further counter-terrorism concerns in 2024-25 and repeat referrals fell from 34% to 21% over the past decade.
4. Currently, the Home Office and Counter Terrorism Policing do not have a unified view of when such individuals should meet the threshold for Channel support. The NAO recommends that the Home Office and Counter Terrorism Policing resolve this disagreement, which is in an area where Channel panels often debate what is the right thing to do.
5. The Home Office has begun work to improve its understanding of local delivery costs and is reviewing the priority area funding structure for 2028-29 onwards.