Mental health

Military combat link to violent crime risk

BBC News reported that, “Younger members of the armed forces returning from duty are more likely to commit violent offences than the rest of the population.”

The news report was on a study following almost 14,000 UK military personnel, most of whom had been deployed in Iraq or Afghanistan. Violent offences were the most common types of offences and they were most common in younger men. The study found that military service in itself was not associated with an increased risk of committing violent offences once other factors were taken into account, but serving in combat was.

Men who had been exposed to more traumatic events during deployment or misused alcohol after deployment were at increased risk, as were men with aggressive behaviour and those with post-traumatic stress disorder.

Compared to the general public, the rate of offending overall among military personnel was lower, but that more of the offences were violent offences.

The authors conclude that more research is needed in this area to identify effective approaches to reduce the risk of offending in military personnel.

Where did the story come from?

The study was carried out by researchers from King’s College London; the Weston Education Centre, and the University of New South Wales. It was funded by the UK Medical Research Council and UK Ministry of Defence. The study was published in the peer-reviewed medical journal, The Lancet.

The study was covered appropriately by the UK media. Most news sources stressed the point that the majority of military personnel returning from combat will not commit criminal offences, and went on to accurately report the results of the study and outline the risk factors for offending among service personnel.

What kind of research was this?

This was a cohort study looking at the risk of violent offending over time in military personnel. The researchers report that there is concern about the proportion of prisoners in the UK and US who have served in the military, including veterans of Iraq and Afghanistan, some of whom have committed violent offences. They say that there is a lack of good quality research into what factors might lead to or contribute to the risk of violent offences by military personnel and their research aimed to address this question.

The current study had the advantage of being able to assess offences committed over a period of time using criminal records, rather than just assessing offences at one point in time.

What did the research involve?

The researchers used a randomly selected group of 13,856 UK military personnel in active service at the start of the study. This included personnel who had been deployed either in Iraq or Afghanistan, and those who had been trained but not deployed. They were recruited in two phases, in 2004-2005 and 2007-2009.

Participants filled in questionnaires about themselves, their experiences and behaviour before and since joining the military (including deployment and combat exposure), and their health and behaviour after deployment. This included an assessment of post-deployment mental health using standard questionnaires to assess symptoms, in particular those of post-traumatic stress disorder (PTSD). The researchers set defined thresholds of symptoms to identify those with PTSD, and those who almost met PTSD criteria but not quite (called ‘subthreshold’ PTSD).

In the second part of the study (2007-2009), the frequency of aggressive behaviour in the past month had been assessed using an accepted measure. This included verbal or physical aggression towards others or taking out aggression on property, such as kicking or smashing things.

To identify violent offences, the researchers used the Police National Computer (PNC) database. This database records all standard offences in the UK and should include any offences dealt with in military courts that are recordable offences (includes those punishable by imprisonment and some non-imprisonable offences).

The researchers used the database to identify the date of the offence, type or offence, and outcome of the offence (conviction, caution, reprimand or warning). The researchers identified all offences committed by individuals from birth up to the end of the study (July 2011).

The researchers then looked at whether there was an association between factors such as pre-military violent offending, socio-demographic characteristics, and characteristics of military service on risk of offending.

Women were not included in analyses of the effect of deployment and combat on offending, as there were few women in the sample, and women are mostly deployed in non-combat roles due to military policy.

What were the basic results?

Most of the participants were full-time military personnel (92.7%) and male (89.7%), and an average age of 37 years (median) at the end of the study. Average time spent in the military was 12.2 years, and 59% were still in service at the end of the study.

Overall, 15.7% of the participants committed one or more offences in their lifetime (17% of men and 3.9% of women). Offences were most common in the post-deployment period (12.2%), than in the pre-deployment service period (8.6%) and the pre-service period (5.4%). The most common types of offences were violent offences (64% of the offenders had committed a violent offence). Among men, any offences (29.8%) and violent offences (20.6%) were both most common in those aged under 30.

Subsequent violent offending was more common in men who had been deployed to Iraq or Afghanistan (7.0%) than in men who had not been deployed (5.4%) the hazard ratio was 1.21, 95% confidence interval (CI) was 1.03 to 1.42. However, after taking into account factors such as age, education, pre-service violent offending and various characteristics of military service (potential confounders) this link was no longer statistically significant.

However, serving in a combat role was associated with an increased risk of offending (6.3%) compared to being deployed in a non-combat role (2.4%), even after taking into account the potential confounders (adjusted hazard ratio 1.53, 95% CI 1.15 to 2.03).

Increased exposure to traumatic events during deployment, post-deployment alcohol misuse, post-traumatic stress disorder, and high levels of self-reported aggressive behaviour were also associated with increasing risk of violent offending.

How did the researchers interpret the results?

The researchers say that their study highlights the role of pre-existing risk factors for violent offences among military personnel. They say that targeting aggressive behaviour and alcohol misuse might be ways to reduce violent offending among service personnel. They add that PTSD is less common but is also a risk factor for violent offending and should be appropriately treated and the risk monitored.


This interesting study gives a valuable picture of offences among military personnel in the UK.

To put the findings into context, the authors of the study note that about 28% of men in England and Wales aged between 18 and 52 years in 2006 had a criminal conviction, compared to 17% of male military personnel in their study. They suggest this difference may relate to the fact that, on average, military personnel spent over a decade in military service, and men tended to enlist at an age when offending is at its peak in the general population (19 years). They also say that other explanations could include that the military may instil more ordered behaviour or be more tolerant of low-grade crime (leading to fewer offences being recorded during service).

Despite this, the authors also note that violent crimes are less common offences among the general public than among military personnel. This suggests that violent offending is of particular concern in this group.

It is worth bearing in mind the limitations of the study, which include:

  • Offences dealt with in military court may not all have been transferred to the police database, particularly those of less severity, and those committed further in the past.
  • As with all observational studies, it is difficult to say whether the associated risk factors themselves directly caused the increase in risk or if other factors are playing a role. The method of identification of participants in the criminal records may not have identified all offenders, as it relied on automatic matching of names, gender, date of birth, which could be misrecorded.

The study has a number of strengths, which include:

  • its relatively large sample size
  • taking into account a range of factors that could influence results, such as pre-service offending
  • being able to identify timing of the offences, so that it was clear which offences occurred before, during, and after service. This is important because if an exposure (in this case military service) is thought to be linked to an outcome (in this case offending), then researchers need to be able to show that the outcome occurs after the exposure rather than the other way around
  • using criminal records to identify offences, which should be more reliable than basing this on self-report

The information in this study could hopefully be used to better identify those at risk of offending in order to take preventative steps. However, as the authors note, exactly the best way to go about this is uncertain, so more research is needed in this area to identify effective approaches to reduce offending.

Analysis by *NHS Choices. Follow Behind the Headlines on twitter*.

NHS Attribution