TASK 1: ALL THOSE VIOLENT OPINIONS
MENTAL DISORDERS, VIOLENCE AND CRIMINALITY
THE INTRICATE LINK BETWEEN VIOLENCE AND MENTAL DISORDER (ELBOGEN, JOHNSON)
Research on relationship between mental disorders & violence has mixed results
Some studies support clear link
Other studies support alcohol & drug use contribute to violence risk among adults
with mental disorders
Reasons for inconclusive literature
Necessary to demonstrate that mental illness precedes later violence
o Most studies are correlational
Longitudinal research mainly focused on risk of violence for people already in
clinical settings
o We don’t know to what extent severe mental illness is risk factor in general
population
Empirical studies often combine all violent acts into one composite variable
Current study: longitudinal data set, examines (1) risk factors predicting violence, (2) do
severe mental disorders predict future violent behaviour, (3) how do different risk
factors predict different types of violence
METHODS
Sample 2-wave face-to-face survey
Target population: civilian population in the US
Collected through National Epidemiologic Survey on Alcohol & related
Conditions (NESARC)
Sample Sample adjusted to be representative of US population for region, age,
weighting sex, race, ethnicity
Aims to correct for bias due to undercoverage
Measures Severe mental Wave 1: determined lifetime & recent diagnoses of major
illness & SA depression, bipolar disorder, substance use disorder
(SUD), schizophrenia
Subjects put into 8 groups: (1) no major mental illness or
SUD, (2-5) each mental disorder on their own, (6-8) each
mental illness + SUD
Dispositional, Historical factors – history of any violent behaviour
historical, clinical, Clinical factors – diagnosis + perceived threats
contextual factors Contextual factors – any factors in their life that could
(RISK factors) influence criminal behaviour
, Violent behaviour Wave 2: subjects were asked about criminal / violent
betw. waves 1 & 2 behaviour between the two waves
RESULTS
Approx. 42% had a lifetime diagnosis of severe mental disorder and / or SUD
46% of those with severe mental illness had a lifetime history of comorbid SUD
3 multivariate models: violence not predicted by schizophrenia, major depression, bipolar
disorder alone!
Predictors of any violence: (1) historical factors, (2) contextual factors, (3)
dispositional factors, (4) comorbid mental health & SUD
o Accounts for ¼ of the variance in violent behaviour
Predictors of serious / severe violence: (1) historical factors, (2) contextual factors,
(3) dispositional factors, (4) perception of hidden threats
o Accounts for ¼ of the variance in serious / severe violent behaviour
Predictors of substance-related violence: (1) historical factors, (2) contextual
factors, (3) dispositional factors, (4) comorbid mental health & SUD
o Accounts for 30% of variance in substance-related violent behaviour
Multivariate Predicted probability of violence for people with severe mental illness
model alone = predicted probability of violence for people with NO severe
mental illness
Highest risk: dual-disordered subjects with history of violence
Simple People with severe mental illness had sig. increased probability of
regression having history of violence
Severe mental illness sig. associated with number of factors associated
with elevated risk of violence
VIOLENT BEHAVIOUR, SLIGHTLY HIGHER AMONG PEOPLE WITH SEVERE MENTAL ILLNESS, WAS ONLY SIG. HIGHER
FOR THOSE WITH COMORBID SUD
HISTORICAL, DISPOSTIONAL, CONTEXTUAL FACTORS WERE ASSOCIATED WITH FUTURE VIOLENCE
DISCUSSION
People with severe mental illness: (1) more vulnerable to past histories, (2) prone to
experience environmental stressors that elevate violence risk
Post hoc mediation analysis: link between severe mental illness & violence is reduced but
remains sig. after controlling for contextual factors
Results point to dynamic factors promising targets for develop approaches to
reduce violence risk
Interventions
, Vocational training, supported employment, other means of assisting people to
find stable jobs
Family therapy / legal mediation in context of spousal conflict
Integrated dual-disorder treatment
CBT & psychotropic medications against anxiety related problems due to e.g.,
physical abuse
Limitations Advantages
Self-reported violence likely Variables included are conceptually
underestimates actual violence grounded in scientific literature of
Severe/serious violence was not violence assessment
further specified (e.g., murder, Findings provide data to support a simple
attempted murder?) no generalisation decision rule physicians could use to detect
possible to severe mental illness people at higher risk for violence
associated with homicidal behaviour Findings may help identify individuals
Not all potential risk factors were who should undergo more formal
analysed violence risk assessment
MENTAL ILLNESS IS CLEARLY RELEVANT TO VIOLENCE RISK BUT ITS CAUSAL ROLES ARE COMPLEX
THE RELATIONSHIP BETWEEN MENTAL ILLNESS & VIOLENCE IN A MENTALLY DISORDERED OFFENDER
SAMPLE: EVALUATING CRIMINOGENIC & PSYCHOPATHOLOGICAL PREDICTORS (KINGSTON, ET. AL.)
Mental illness is sig. overrepresented in the criminal justice system
Correctional settings: 10-40%
When definition includes SUD: 80-90%
RELATIONSHIP BETWEEN MENTAL ILLNESS & VIOLENCE
Psychopathologica Untreated mental illness considered a direct cause of criminal
l theory of criminal behaviour
behaviour Support: quantitative reviews & studies
Individuals with psychiatric diagnosis more likely to be
arrested for a violent offense as compared to those who have
never been hospitalised
Limitations: number of studies can’t replicate this result
Possible confounder: sample composition – mental illness as
risk factor for violence among general population BUT effect
does not generalise to offenders
Social Most studies using offender samples showed psychiatric
psychological diagnoses are unrelated to recidivism
explanations of General Personality & Cognitive Social Learning (GPCSL) model
MENTAL DISORDERS, VIOLENCE AND CRIMINALITY
THE INTRICATE LINK BETWEEN VIOLENCE AND MENTAL DISORDER (ELBOGEN, JOHNSON)
Research on relationship between mental disorders & violence has mixed results
Some studies support clear link
Other studies support alcohol & drug use contribute to violence risk among adults
with mental disorders
Reasons for inconclusive literature
Necessary to demonstrate that mental illness precedes later violence
o Most studies are correlational
Longitudinal research mainly focused on risk of violence for people already in
clinical settings
o We don’t know to what extent severe mental illness is risk factor in general
population
Empirical studies often combine all violent acts into one composite variable
Current study: longitudinal data set, examines (1) risk factors predicting violence, (2) do
severe mental disorders predict future violent behaviour, (3) how do different risk
factors predict different types of violence
METHODS
Sample 2-wave face-to-face survey
Target population: civilian population in the US
Collected through National Epidemiologic Survey on Alcohol & related
Conditions (NESARC)
Sample Sample adjusted to be representative of US population for region, age,
weighting sex, race, ethnicity
Aims to correct for bias due to undercoverage
Measures Severe mental Wave 1: determined lifetime & recent diagnoses of major
illness & SA depression, bipolar disorder, substance use disorder
(SUD), schizophrenia
Subjects put into 8 groups: (1) no major mental illness or
SUD, (2-5) each mental disorder on their own, (6-8) each
mental illness + SUD
Dispositional, Historical factors – history of any violent behaviour
historical, clinical, Clinical factors – diagnosis + perceived threats
contextual factors Contextual factors – any factors in their life that could
(RISK factors) influence criminal behaviour
, Violent behaviour Wave 2: subjects were asked about criminal / violent
betw. waves 1 & 2 behaviour between the two waves
RESULTS
Approx. 42% had a lifetime diagnosis of severe mental disorder and / or SUD
46% of those with severe mental illness had a lifetime history of comorbid SUD
3 multivariate models: violence not predicted by schizophrenia, major depression, bipolar
disorder alone!
Predictors of any violence: (1) historical factors, (2) contextual factors, (3)
dispositional factors, (4) comorbid mental health & SUD
o Accounts for ¼ of the variance in violent behaviour
Predictors of serious / severe violence: (1) historical factors, (2) contextual factors,
(3) dispositional factors, (4) perception of hidden threats
o Accounts for ¼ of the variance in serious / severe violent behaviour
Predictors of substance-related violence: (1) historical factors, (2) contextual
factors, (3) dispositional factors, (4) comorbid mental health & SUD
o Accounts for 30% of variance in substance-related violent behaviour
Multivariate Predicted probability of violence for people with severe mental illness
model alone = predicted probability of violence for people with NO severe
mental illness
Highest risk: dual-disordered subjects with history of violence
Simple People with severe mental illness had sig. increased probability of
regression having history of violence
Severe mental illness sig. associated with number of factors associated
with elevated risk of violence
VIOLENT BEHAVIOUR, SLIGHTLY HIGHER AMONG PEOPLE WITH SEVERE MENTAL ILLNESS, WAS ONLY SIG. HIGHER
FOR THOSE WITH COMORBID SUD
HISTORICAL, DISPOSTIONAL, CONTEXTUAL FACTORS WERE ASSOCIATED WITH FUTURE VIOLENCE
DISCUSSION
People with severe mental illness: (1) more vulnerable to past histories, (2) prone to
experience environmental stressors that elevate violence risk
Post hoc mediation analysis: link between severe mental illness & violence is reduced but
remains sig. after controlling for contextual factors
Results point to dynamic factors promising targets for develop approaches to
reduce violence risk
Interventions
, Vocational training, supported employment, other means of assisting people to
find stable jobs
Family therapy / legal mediation in context of spousal conflict
Integrated dual-disorder treatment
CBT & psychotropic medications against anxiety related problems due to e.g.,
physical abuse
Limitations Advantages
Self-reported violence likely Variables included are conceptually
underestimates actual violence grounded in scientific literature of
Severe/serious violence was not violence assessment
further specified (e.g., murder, Findings provide data to support a simple
attempted murder?) no generalisation decision rule physicians could use to detect
possible to severe mental illness people at higher risk for violence
associated with homicidal behaviour Findings may help identify individuals
Not all potential risk factors were who should undergo more formal
analysed violence risk assessment
MENTAL ILLNESS IS CLEARLY RELEVANT TO VIOLENCE RISK BUT ITS CAUSAL ROLES ARE COMPLEX
THE RELATIONSHIP BETWEEN MENTAL ILLNESS & VIOLENCE IN A MENTALLY DISORDERED OFFENDER
SAMPLE: EVALUATING CRIMINOGENIC & PSYCHOPATHOLOGICAL PREDICTORS (KINGSTON, ET. AL.)
Mental illness is sig. overrepresented in the criminal justice system
Correctional settings: 10-40%
When definition includes SUD: 80-90%
RELATIONSHIP BETWEEN MENTAL ILLNESS & VIOLENCE
Psychopathologica Untreated mental illness considered a direct cause of criminal
l theory of criminal behaviour
behaviour Support: quantitative reviews & studies
Individuals with psychiatric diagnosis more likely to be
arrested for a violent offense as compared to those who have
never been hospitalised
Limitations: number of studies can’t replicate this result
Possible confounder: sample composition – mental illness as
risk factor for violence among general population BUT effect
does not generalise to offenders
Social Most studies using offender samples showed psychiatric
psychological diagnoses are unrelated to recidivism
explanations of General Personality & Cognitive Social Learning (GPCSL) model
