What this is
The Home Office annual statistical release on police use of force in England and Wales, covering April 2024 to March 2025. Recording has been mandatory since 1 April 2017, collected from the 43 Home Office funded forces from the year ending March 2018 and from British Transport Police from March 2024.
The release was updated on 4 June 2026 after one force resubmitted its data, so the totals below are the revised ones.
What it records
According to Home Office figures, the 44 forces recorded 812,447 use of force reports in the year ending 31 March 2025, against 747,396 the year before: an increase of 65,051, or 9%.
Restraint tactics, meaning handcuffing, limb or body restraints and ground restraint, appeared in 628,307 reports, 77% of the total and the same proportion as the previous year. Of the 1,146,016 tactics recorded across all reports, restraint accounted for 719,502, or 63%, according to Home Office figures.
Handcuffing made up 80% of restraint tactics and half of all tactics, and 66% of handcuffing was recorded as compliant.
The release also records that the rate of use of force for Black ethnic groups was 3.3 times the rate for White ethnic groups, excluding the Metropolitan Police, according to Home Office figures.
The Home Office attributes the rise principally to continued improvement in recording, supported by a 5% rise in arrests in the same year, according to Home Office figures.
How to read it
A report is one officer’s use of force on one person. Two officers restraining one person produce two reports, so 628,307 is not a count of people. According to Home Office figures, the 315,827 reports of sufficient quality represent an estimated 230,462 individuals, about 73% of the number of reports.
The data cannot link an injury to a tactic: a handcuffing injury and a ground restraint injury are indistinguishable. Within the restraint group there is no breakdown of duration, position, or whether the person was on the ground when the tactic was applied.
And the statistics do not count deaths. There is no outcome field that would let anyone derive a restraint death rate from this release, and dividing these totals by another body’s death count would be arithmetic across two definitions that do not meet.
The most useful figure may be the least quoted: two thirds of recorded handcuffing was compliant, according to Home Office figures, so most recorded restraint in England and Wales is a control technique applied to someone who was not resisting.
Where it stops
One year, England and Wales, and only what was recorded. A restraint that no officer recorded is not in the dataset, and the release cannot say how much of that rise is practice and how much is bookkeeping.
The 4 June 2026 revision changed recorded tactics and officer injury data, and the tactic breakdowns are the part most exposed to it, after Derbyshire Police resubmitted following a system change.
It is a single publisher, and no independent body audits the returns. Nothing in it records whether a restraint was reviewed afterwards, or by whom.
Sources
- Home Office, police use of force statistics, England and Wales: April 2024 to March 2025, updated 4 June 2026 · https://www.gov.uk/government/statistics/police-use-of-force-statistics-april-2024-to-march-2025/police-use-of-force-statistics-england-and-wales-april-2024-to-march-2025
If your service records restraint and never reads the recording back, our restraint reduction training starts with what the data can and cannot tell you.
Reviewed 20 October 2026.