Police violence, reports and violence: the statistics behind the review

Special Prosecution Chamber, SU statistics, fatal police interventions and the older DN review on women violence as most common off-duty police offences.

Anyone who applies to police force abuse, report police or report police for misconduct is often not looking for theory. The person wants to know two things: whether what happened was legal, and how to build a dossier that won't disappear when the police's own incident report becomes the first version in the system.

This review is therefore written as a combination of statistics, source criticism, and practical evidence focus. For a clean plan of action, see important when you have been exposed to the police and report the way of the police with documented evidence

Have you been subjected to excessive force by police? In case of acute danger or ongoing threat, call 112. This guide is information and documentation support, not emergency legal advice or

Seek care if you have injuries and ask that the injuries be documented in the medical record. Save original footage, photos, testimonials, and timeline before posting anything

Support is available at the Crime Victim Support Centre, the Crime Victim Authority and in case of violence in close relationships at the Kvinnofridslinie. If you are under 18 years of age, you can also contact Bris.

The official control trail is growing rapidly

Of the 2025 registrations, 12,040 were C cases, where the offence has imprisonment on the penal scale. The measure refers to recorded crimes and criminal suspicions, not the number of notifiers, unique

The Special Chamber of Prosecutors handles criminal suspicions against police officers, prosecutors and judges, among others. The Chamber's own side stresses that most cases involve police employees, and that it can range from suspected use of force to responding to and exercising their duties.

The most common category of crime is misconduct. The prosecutor's office states that intervention cases accounted for 69 percent of the notifications in the period from January 2024 to May 2025. The Police Authority's Special Investigations statistics for 2025 indicate that misconduct made up just over 70 percent of SU's cases, followed by violent crime and computer hacking

same time, the police part of the annual report shows another part of the justice chain. C cases involving police employees increased from 8,384 in 2023 to 10,509 in 2025. The number of criminal suspects where preliminary investigation was initiated decreased over the same period from 2,845 to 2,285, while criminal suspicions prosecuted were 481, 563 and 513 respectively. These ranks should be carefully compared: a person or event may involve multiple criminal suspicions and the prosecution in a given year need not relate to the same registrations entered that year

How many move on and how many fall away?

The police authority reports that 9,536 cases were submitted to SU in 2025 and that 9,187 cases were closed in the same year. 254 cases were reported to prosecutors, which, according to the Police, are usually cases that lead to prosecution. A simple difference means that 8,933 closed cases were not reported to prosecutors. It should be described as “closed without accounting” rather than precisely “closed preliminary investigations”, since the statistics mix cases where preliminary investigation was never initiated with

The critical thing in the figure is not just that few cases go ahead. The critical point is that the lapse occurs in a system where the reported person is often a state perpetrator of violence, where the evidentiary situation is quickly shaped by the police's own documentation, and where the individual often lacks film, witnesses, injuries documented by care

At the same time, one should not pretend that all dropped cases are abuses swept away. The Public Prosecution Service states that prosecution rates have normally been between 4 and 5 per cent, with just over 7 per cent in 2023. It can be due to high evidentiary requirements, statutory violence powers and difficult emergency defence assessments. But for this very reason, the person who reports must work very concretely: time, place, film, injuries, witnesses, what was said and what was written afterwards

Lethal interventions: two metrics must be kept apart

There are at least two different measures of mortality that are often mixed up:

• Death by police officer-involved fire: persons who die in connection with the shooting of a police officer.
• The Forensic Science Service's mode of death “police intervention”: deaths classed in RMV's forensic statistics as police intervention, which is not necessarily the same as shooting.

Police report that there were 17 incidents of accidental fire against a person or vehicle in 2019, 21 events in 2020 and 2021, 21 events in 2022, 19 events in 2023, 14 events in 2024 and 11 events in 2025. Casualties related to industrial fire were 1 in 2019, 5 in 2020, 2 in 2021, 2 in 2022, 0 in 2023, 0 in 2024 and 1 in 2025. The police themselves write that the number of people who died in connection with industrial fire has increased in the last ten years, although recent years

RMV stresses that their statistics are based on case management systems and are not Sweden's official cause of death statistics, but it is nonetheless relevant because it captures

The charts on the statistics page show both tracks separately: actual fire/death and RMV's mode of death police intervention.

Off-duty violence: the uncomfortable DN review

The older data on police officers' crimes outside the service comes from Dagens Nyheter's review of prosecutions against police officers 1998-2008, reproduced by SVT, Aftonbladet, among others. SVT News Characters.

The crux of that data is serious: between 1998 and 2008, 48 police officers were charged with gross violation of women's rights or mistreatment of a wife, partner or ex-partner. SVT reported that just under a third of all prosecutions against off-duty police officers concerned abuse of women. Aftonbladet reiterated that abuse of women was highest among the crimes of police officers outside of work, followed by beatings and drunk driving

It is the relative placement that makes the review particularly sensitive. SVT reiterated that the most common prosecutions against the public in the same period concerned traffic offences/unlawful driving, theft/shoplifting and drug offences for one's own use. Assault of man came in ninth place and prosecutions of female abuse in 13th place. In contrast, in the DN review of off-duty police officers, women abuse ranked highest. The older statistics cannot be attributed to today's police forces without new evidence. But it must also not be used to lower the standards of responsibility. A police officer has the state's violence mandate, education about the risks of violence and a central role in preventing, investigating and moving crimes forward in the justice chain. Whoever has that power and knowledge must therefore be able to be held to a higher level of responsibility in the event of intimate relationship violence. Such violence is not only a common personnel matter but also concerns fitness, trust, the safety of victims and the credibility of the rule of law.

These are older statistics and should not be presented as currently in 2026. But it remains relevant for two reasons. First, it shows that violence is not only a matter of intervention in service, but also of a professional group with a monopoly on violence in which some individuals have also been convicted or charged with violence in intimate relationships. Second, it shows a particular damage to trust: a police officer who himself practices intimate violence is difficult to see as a neutral representative of society when other crime victims encounter the

Critical conclusion: the problem is both legal and documentary

Police have the legal right to use force in certain situations. It is true and must be said. But precisely for this reason, the question of control becomes harder, not softer: was the violence necessary, proportionate and adapted to the situation? Was the person already under control? Was there mental illness, drunkenness, communication difficulties or other vulnerability? Were injuries and witness accounts documented before the police version became the only

For those victimized, the practical lesson is clear:

• Film and save the original file if it can be done safely.
• Photograph damage and location directly.
• Seek medical care or legal certification as soon as possible if you have been injured.
• Gather witnesses before they disappear from the scene.
• Request event report, body camera if available, decision documentation and diary number.
• Report factually to the police office/SU and consider the JO when it comes to the exercise of authority, response or inaction.

A critical tone is not enough. If the control system is narrow and few cases go ahead, the evidence must be as difficult to dismiss as possible

Sources

• Annual Report of the Special Prosecutor's Chamber 2025
• Prosecutor's Office: Special Prosecution Chamber
• Police Authority: Special investigations, statistics
• Police Authority: How often do police use firearms?
• Swedish National Board of Forensic Medicine: Statistics
• SVT: Violence against women is the most common police crime
• Aftonbladet: Mistreatment of women, the police's most common crime