5

From an NPR article discussing whether planning a shooting breaks the law:

Advocates often bristle when people lump mental health into conversations about mass shooters. They argue that individuals with severe mental illness are much more likely to be victims of violence than perpetrators of it.

  1. Are individuals with severe mental illness "much more likely" to be victims of violence than perpetrators of it? If the claim is true, is it misleading? What exactly does "victims of violence" (from the claim) mean?

The claim is repeated more specifically on MentalHealth.gov for victims of "violent crime."

In fact, people with severe mental illnesses are over 10 times more likely to be victims of violent crime than the general population.

  1. Can the statistic "10 times more likely" be verified?
Barry Harrison
  • 14,093
  • 4
  • 68
  • 101
  • Question 2 may be poorly phrased. I have tried to be as clear about it as I can be. As usual, I welcome (and appreciate) all suggestions. Thanks in advance! – Barry Harrison Jun 06 '19 at 07:28
  • 2
    "mental illness" may be an overly broad category. The small subset of people suffering severe dissociative psychotic episodes may have a very different rate of violent acts vs the large fraction of the population suffering from depression of anxiety. But you could swamp the small group in the large to hide very different rates. – Murphy Jun 06 '19 at 12:08
  • 3
    Victims of violence outnumber perpetrators on average. – daniel Jun 06 '19 at 13:20
  • 1
    @daniel My thoughts exactly. This is especially true when we are talking specifically about _mass shooters_ as they are here. Mass shootings almost always (possibly statistically always) have more victims than perpetrators. You could also say individuals _without_ severe mental illness are more likely to be victims of mass shootings than perpetrators. What you would need to compare is if perpetrators with severe mental illness are more likely than those without. Their comparison is basically pointless. – JMac Jun 06 '19 at 13:43
  • @daniel Yes good points. I hope an answer will address "true, but misleading." – Barry Harrison Jun 06 '19 at 15:29
  • @JMac Also good points! I have now edited the question because of this. – Barry Harrison Jun 06 '19 at 15:31
  • @Murphy I guess the best thing you can do is interpret "mental illness" in the NPR article the way a reasonable person would (and see if the claim is true, but also misleading because of the points you mentioned). I have been told to not define terms in questions. I know this is a bad answer. – Barry Harrison Jun 06 '19 at 15:35
  • This appears to a likely original source for Claim 2: ["_Crime Victimization in Adults With Severe Mental Illness_"](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1389236/) (2005): "_More than one quarter of persons with [severe mental illnesses] had been victims of a violent crime in the past year, a rate more than 11 times higher than the general population rates [...]_". – Nat Jun 06 '19 at 19:02
  • Looks like [the same source](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1389236/) also makes Claim 1: "_Among persons with [severe mental illness], violent victimization is far more prevalent (more than 25% within 1 year in this study) than perpetration of violence (4%–13%)._". This line is near the end, and it looks like it cites two references. – Nat Jun 06 '19 at 19:10
  • Okay, so looks like both claims do come from research papers. However, I'd stress that people shouldn't blindly believe papers just because they're published in peer-reviewed venues. I mean, some **_really_** stupid papers get through peer-review; debunked one [here](https://skeptics.stackexchange.com/questions/34413/will-global-warming-reduce-available-oxygen/40189#40189). In general, peer-reviewers aren't fact-checkers. I dunno if SE.Skeptics would allow it due to the original-research policy, but this is the sort of claim that someone needs to review the original studies to evaluate. – Nat Jun 06 '19 at 19:15
  • @Nat I feel like you could answer. You would definitely bring (and already have brought) a new perspective. I would upvote. – Barry Harrison Jun 07 '19 at 06:38
  • Say a person of group A will hurt someone in a year with 10% chance, get hurt wirh 25% chance. For group B these values are 0.1% and 0.05% respectively. Who would you like to meet? Now say you meet 500 people of B, but only one of A, per day... which group should have guns? These statistics are worthless out of a detailed context: are the people hurt random ? no. is group a/b randomly assigned over all regions, social contexts, etc.? no. – bukwyrm Jun 08 '19 at 20:20
  • @bukwyrm I see what you are saying. This is why I am asking the question. I want to know what sort of misinterpretation/misleading is happening here. – Barry Harrison Jun 09 '19 at 07:27

1 Answers1

1

Point 1 doesn't appear to be true, at least looking at homicide.

"mental illness" is a category as broad as the ocean but let's try...

First let's look at psychotic patients.

Rates of Homicide During the First Episode of Psychosis and After Treatment: A Systematic Review and Meta-analysis

The observation that almost half of the homicides committed by people with a psychotic illness occur before initial treatment suggests an increased risk of homicide during the first episode of psychosis. The aim of this study was to estimate the rates of homicide during the first episode of psychosis and after treatment. A systematic search located 10 studies that reported details of all the homicide offenders with a psychotic illness within a known population during a specified period and reported the number of people who had received treatment prior to the offense. Meta-analysis of these studies showed that 38.5% (95% confidence interval [CI] = 31.1%–46.5%) of homicides occurred during the first episode of psychosis, prior to initial treatment. Homicides during first-episode psychosis occurred at a rate of 1.59 homicides per 1000 (95% CI = 1.06–2.40), equivalent to 1 in 629 presentations. The annual rate of homicide after treatment for psychosis was 0.11 homicides per 1000 patients (95% CI = 0.07–0.16), equivalent to 1 homicide in 9090 patients with schizophrenia per year.

So there's a 1 in 629 chance that someone having their first psychotic episode will kill somebody.

After that there's a 1 in 9090 chance per year that they will kill somebody.

But this doesn't give us an indicator of how it compares to the chances of being a victim.

This study may give us a better comparison:

Patients with mental illness as victims of homicide: a national consecutive case series

1496 victims of confirmed homicide died between Jan 1, 2003, and Dec 31, 2005, in England and Wales. Patients with mental illness were more likely to die by homicide than were people in the general population (incidence rate ratio 2·6, 95% CI 1·9–3·4). 90 homicide victims (6%) had contact with mental health services in the 12 months before their death. 213 patients with mental illness were convicted of homicide in the same 3 year period. 29 of 90 patient victims were killed by another patient with mental illness. In 23 of these 29 cases, the victim and perpetrator were known to each other, and in 21 of these cases, the victims and perpetrators were undergoing treatment at the same National Health Service Trust. In these 29 cases in which patient victims were killed by another patient with mental illness, alcohol and drug misuse (19 victims [66%], 27 perpetrators [93%]) and previous violence (7 victims [24%], 7 perpetrators [24%]) were common in both victims and, particularly, perpetrators. In seven of the 29 cases in which the victim was killed by another patient with mental illness, both victim and perpetrator were diagnosed with schizophrenia.

So in this sample sampling from Patients with mental illness, 90 were victims of homicide while 213 patients were convicted of homicide.

29 of 90 patient victims were killed by another patient with mental illness.

So while mental health patients are significantly more likely to be murdered ( 2.6X ) it appears that a significant fraction of that risk comes from the other mental health patients in the same trust. The average mental health patient is more likely to commit homicide than be a victim of homicide.

Claim 2 is easily confirmed:

https://www.time-to-change.org.uk/media-centre/responsible-reporting/violence-mental-health-problems

people with severe mental illnesses are more likely to be victims, rather than perpetrators, of violent crime.

Community
  • 1
Murphy
  • 9,486
  • 1
  • 47
  • 45