March 1, 2012

Mentally ill 'often victims of violence'

People with mental health issues were often victims of violence

Mentally ill people are four times more likely to be victims of violence, the BBC has today reported.

This alarming statistic is based on a review of research looking at how often people with a range of disabilities had experienced violence in the previous year, and how this compared with non-disabled people. After combining the results of 26 previous studies, researchers found that more than 24% of those with a mental illness had been physically attacked in the previous year, as had more than 6% of people with intellectual impairments and more than 3% of people with all types of disability. People with disabilities were generally more at risk of violence than non-disabled individuals.

Although it had some limitations, this large well-conducted review supports previous research suggesting that people with disabilities are at increased risk of violence, and those with mental illness are particularly vulnerable. Most of the previous studies it looked at were in high-income countries including the UK, so the findings are particularly relevant for this country.

Further research on this important issue is now required to understand the magnitude of the problem in the UK and to develop further public health strategies to protect vulnerable groups.

Where did the story come from?

The study was carried out by researchers from Liverpool John Moores University and the World Health Organization (WHO). It was funded by the WHO Department of Violence and Injury Prevention and Disability. The study was published in the peer-reviewed medical journal The Lancet.

The BBC’s report was fair and included comments from independent UK experts.

What kind of research was this?

This was a systematic review and meta-analysis combining the results of previous research on violence against people with disabilities. It looked both at studies reporting on the rates of recorded violence against disabled adults, and at those that examined risk of violence to disabled adults compared with non-disabled adults.

The authors point out that about 15% of adults worldwide have a disability, a figure that is predicted to increase because of ageing populations and the rise in chronic disease, including mental illness. People with disabilities seem to be at increased risk of violence because of several factors including exclusion from education and employment, the need for personal assistance with daily living, communication barriers and social stigma and discrimination. The authors also say that there is an increasing number of media reports highlighting cases of physical violence and sexual abuse of disabled individuals living in institutions, but point out that formal research to quantify the problem is scarce.

What did the research involve?

The authors searched 12 online research databases to identify any studies that had reported on the prevalence of violence against adults with disabilities, or their risk of violence compared with non-disabled adults. They searched for all relevant studies published between 1990 and 2010. They also used additional methods to look for further studies, including hand searching reference lists and web-based searches.

To be deemed suitable for inclusion, studies had to meet various criteria. For example, their design had to be either a cross-sectional, case-control or cohort, they had to report on specific disability types, and they had to report violence occurring within the 12 months prior to the study.

All the identified studies were independently assessed by two separate reviewers using accepted criteria for assessing the quality of research. Individuals in the studies were grouped according to the type of disability: non-specific impairments (physical, mental, emotional or other health problems), mental illness, intellectual impairments, physical impairments and sensory impairments. The types of violence examined were physical violence, sexual violence, intimate partner violence and any violence.

The researchers calculated prevalence rates and the risk of violence faced by disabled people compared with non-disabled people, using standard statistical methods.

What were the basic results?

The researchers’ initial search identified 10,663 studies on the subject, but only 26 were eligible for inclusion. Overall, these studies provided data on 21,557 individuals with disabilities.

Of these studies, 21 provided data on the prevalence of violence among disabled people, and 10 provided data on the risk of violence compared with non-disabled people. By combining their results, researchers found that over the previous year:

  • 24.3% of mentally ill adults had been subjected to violence of any type (95% CI: 18.3 to 31.0%)
  • 6.1% of adults with intellectual impairments had been subjected to violence of any type (95% CI: 2.5 to 11.1%)
  • 3.2% of adults with any impairment had been subjected to violence of any type (95% CI: 2.5 to 4.1%)

However, the researchers did note significant differences between individual studies (heterogeneity) in their prevalence estimates. Heterogeneity provides an indicator of how suitable it is to combine the results of different studies, with greater heterogeneity suggesting studies are of lower compatibility with each other.

When they pooled the results of studies comparing disabled with non-disabled individuals they found that, overall, disabled people were 1.5 times more likely to have been attacked than non-disabled people (odds ratio: 1.5; 95% CI: 1.09 to 2.05).

There was also a trend for people with specific types of disability to experience more violence, but not all associations were significant:

  • People with intellectual impairments were 1.6 times more likely to have been physically attacked than people without intellectual impairments (results from three studies; pooled odds ratio: 1.60; CI 95%: 1.05 to 2.45).
  • Mentally ill people were no more likely to have been physically attacked than non-mentally ill people (three studies; pooled odds ratio: 3.86; 95% CI: 0.91 to 16.43).
  • People with non-specific impairments were no more likely to have been physically attacked than those without (six studies; pooled odds ratio: 1.31; 95% CI: 95% 0.93 to 1.84).

How did the researchers interpret the results?

The researchers conclude that adults with disabilities are at a higher risk of violence compared with non-disabled adults, and that those with mental illnesses could be particularly vulnerable. However, they add that the available studies have methodological weaknesses and that gaps exist in the types of disability and violence they address. They also point out that good studies are absent for most regions of the world, particularly low-income a
nd middle-income countries.


Violence and abuse against anyone is not acceptable, but there is an even greater need to ensure that vulnerable groups who may be less able to help themselves receive adequate protection against this type of victimisation. This valuable systematic review helps to establish the proportion of people with disabilities who have experienced violence, as well as how this compares to people without disabilities. The estimates it provides may prove useful for planning services and policies to protect vulnerable individuals, such as people with mental health issues.

However, the review does have several limitations, many of which the authors acknowledge:

  • The studies were limited to looking at violence within the 12 months before each study, which means the review probably underestimates people’s lifetime exposure to violence.
  • It is not clear from some of the studies whether the violence was a cause or a result of people’s health conditions, i.e. whether disability led to violence, or if violence caused people to develop disability such as mental health issues. This factor could particularly affect studies of people with mental illness, which form a large proportion of the studies included.
  • The studies included in the review varied in quality, with only one achieving the assessors’ maximum quality scores. The researchers say that combining the results of individual studies was severely hindered by lack of methodological consistency between studies, including variations in samples used, definitions of disability and violence, and methods of data collection. When they pooled the study results there was significant heterogeneity (differences) between individual studies in the proportion of people who experienced violence, making it difficult to give an accurate estimate of the prevalence. Also, many studies failed to include comparison groups, which are needed to compare risk of violence between those with and without disability.
  • In studies that did compare people with and without disability, overall there were higher odds of experiencing violence in those with any disability compared with those with none, but analyses by individual type of disability did not consistently give significant associations.
  • Regardless of whether or not people have disabilities, they may be unwilling to report violence or abuse and, therefore, the rates reported in the reviewed studies may not reflect what happens in reality.

Despite these limitations, this is a valuable attempt to quantify the prevalence and the risk of violence faced by disabled people. Further high-quality research on this important issue is required to understand the magnitude of this problem if strategies are to be developed that can help prevent it.