Docstoc

Violence

Document Sample
Violence Powered By Docstoc
					Schizophrenia and Violence:
  from correlations to preventive
            strategies

              Paul E Mullen
           Monash University and
Victorian Institute of Forensic Mental Health
     Improved Management of High Risk
      Groups with Schizophrenia Could

1.    Reduce seriously violent crime by 2-4%.
2.    Reduce incarceration rates (prison and
      forensic hospitals) of those with
      schizophrenia by 30% - 50%.
3.    Improve quality of life for the most
      disturbed and disadvantaged among
      those with schizophrenia.
                  SCHIZOPHRENIA




Clinically Significant        Socially Significant




                         VIOLENCE
Schizophrenia Among Violent Offenders
         Homicides and Schizophrenia
  Authors        Country      Number of    Period       Homicides
                              Homicides                    with
                                                      Schizophrenia
Petursson,     Iceland           47        1900-79       14.9%
Gudjonsson
(1981)
Hafner &     Western            3367      1954-1964       8.0%
Böker (1982) Germany

Wilcox         Contra Costa      71       1978-1980       9.9%
(1985)         County,
               California
Eronen et al   Finland          1037       1984-91        6.1%
(1996)
       Homicides and Schizophrenia
  Authors         Country   Number of    Period       homicides
                            Homicides                    with
                                                    schizophrenia

Wallace et al   Victoria,      168       1993-95        7.2%
(1998)          Australia


Simpson et al New Zealand      846      1988-2000       7.7%
(2004)


Schanda et al Austria         1087      1975-1999       5.4%
(2004)


Fazel &      Sweden           2005      1988-2001        9%
Grann (2004)
  OR for Schizophrenia and Homicide

         STUDY                     OR (95% CI)

Hafner & Böker (1982)       12.7 (11.2 – 14.3)*

Eronen et al (1996)         9.7 (7.4 -12.6)

Wallace et al (1998)        10.1 (5.5 – 18.6)*

Erb (2001)                  16.6 (11.2 – 24.5)

Schanda (2004)              8.8 (6.7 – 11.5)

* Highest probable ascertainment
Violent Offending in those with Schizophrenia
Violent Offending in Schizophrenia(2861)
Lifetime          Schiz.   Control   O.R. 95% C.I.
Individuals with  235          51     4.8   3.6-6.26
violent offences (8.2%)    (1.8%)
Total violent       855      76
offences          (3.6)    (1.5)
Males             13%        2.9%     5     3.6-6.9
Females           1.4%       0.3%     5.4   1.6-18.6

5 yrs after 1st   4.6%       0.7%     7.3   3.9-13.8
admission
   Violence and Homicide in 1,705 patients
    with Schizophrenia. Soyka et al 2004


       7-12 years post discharge:

            Any conviction:              224 (13.1%)

            Violent convictions:         45 (2.6%)

            Homicide/attempted homicide: 5 cases (0.3%)
                  SCHIZOPHRENIA




Clinically Significant        Socially Significant




                         VIOLENCE
 WHY HAS IT NOT BEEN
OBVIOUS TO CLINICIANS &
    RESEARCHERS ?
CLINICAL RISKS & COMMUNITY
RISKS AN APARENT PARADOX


   5 – 10% of violent crime including
    homicide is attributable to the 0.5
    to 0.6% of the population with
    schizophrenia
    CLINICAL RISKS & COMMUNITY
    RISKS AN APARENT PARADOX
BUT in schizophrenia the risks for individuals are:-

   Homicide                        1 in 10000 per
    year           For males                            1
    in 2000 per year
   Convictions serious violence    1 in 500 per year
   Any violent convictions         1 in 180 per year
     for violence
    For males                       1 in 100 per year
   Violent incidents              5-10 in 100 per year
           Violence in Schizophrenia
                 l
DELUSIONALLY DRIVEN          MULTIFACTORIAL



Older                     Younger

Organised Delusions       Disorganised

Domestic                  Domestic and non domestic

                          Psychopathic Traits

Not Antisocial            Conduct Disorder

                          Substance Abusing

(look like patients)      (look like criminals)
                         Schizophrenia

Developmental Difficulties                                 Active Symptoms

                       Personality Vulnerabilities

Education Failure


Unemployment          Social Dislocation                Substance Abuse


Criminal Peer Group                                  Rejection by Services




                       Violent Behaviours
Substance Abuse ?
      Between Schizophrenia,
       Substance Abuse and
            Offending
1.   Substance abuse causes the offending

2.   Those with schizophrenia with a propensity for
     offending behaviours also having a propensity to
     abuse substances when they are available


3.   A mixture of the above
       OFFENDING AMONG THE
       MENTALLY DISORDERED
                 1,136 public
                 inpatients            VIOLENCE

MMD                                     17.9%

MMD & S.A.                              31.1%

MMD without S.A.                          4%

Other M.D.                               43%

Controls without S.A. 43% R.R.           4.6%

 Schizophrenia Protective Factor   Steadman et al (1998, 2000)
        Substance Abuse and
           Schizophrenia

Far higher rates of alcohol and drug abuse
are found in those with schizophrenia
particularly younger males.
                              (Soyka 2000)
                Reminder

If you control for a variable which is itself
significantly associated with schizophrenia
then you are controlling in part for the
disorder itself and risk obscuring causal as
well as statistical associations.
 Lifetime Convictions in the Schizophrenia Cohorts
                 Offences                  95%
                                  O.R.   Confidence     P
              Cases    Controls           Interval    value


1975 (8%)


1980 (10%)


1985 (16%)


1990(18%)


1995 (27%)
 Lifetime Convictions in the Schizophrenia Cohorts
                 Offences                  95%
                                  O.R.   Confidence     P
              Cases    Controls           Interval    value


1975 (8%)    14.8%      5.1%      3.3    (2.0-5.3)    0.000


1980 (10%)   21.3%      4.7%      5.5    (3.3-6.0)    0.000


1985 (16%)   19.4%     10.0%      2.2    (1.6-3.1)    0.000


1990(18%)    24.9%      6.6%      4.7    (2.9-7.5)    0.000


1995 (27%)   25.0%      9.6%      3.1    (2.4-4.0)    0.000
         Schizophrenia & S.A.
   Comorbid S.A. is associated with increased
    offending
   This is in part because S.A. in almost
    anyone increases the risks of offending
   This is in part because those with a
    predisposition to offending have a
    particular avidity for substance abuse
   S.A. may explain part but by no means all
    of the correlation.
Manage Substance Abuse
   Do active symptoms mediate offending in
    schizophrenia?
   Yes: Link & Stuve 1994-1998; Taylor
    1985-1998; Arsenault 2000
   Probably not: MacArthur Studies 1998-
    “Clear emergence of schizophrenia before the
      onset of significant violence does suggest that
      in some way the illness may have a direct role
      in the violence”
                              Taylor & Estroff (2002)


   Do the criminal careers differ between
    those with and those without
    schizophrenia?
   Yes: Hafner & Boker 1973; Taylor 1993;
    Wessley et al 1994; Taylor & Hodgins
    1994
            Temporal patterns of convictions
8,791 convictions in cases - 1,119 convictions in controls




               72.7% convicted for first time prior to first admission
    The Role of Symptoms in Violent
               Behaviour


   Apparently undeniable in individual cases.
   Clear increased rates predate and
    continue independent of obvious
    symptoms in many.
   An important but not the major mediator
    in populations.
       ACTIVE SYMPTOMS
Improved Symptom Control.

Stabilisation in I.P. context using
compulsory powers and extended
admissions if indicated.

 (Forensic services as primary preventative
services not just containing services)
  Current social conditions and dislocation
does mediate the correlation to some extent



                         Silver et al (2000)
1.   Social Conditions

        Avoid discharging to disorganised
         accommodation in high crime neighbourhoods.
        Provide appropriate level of support and
         supervision.
        Ensure opportunity for meaningful activity and
         recreation within structured programs or work
         environment.
        Address peer groups which support substance
         abuse and offending.
       Developmental Histories
    Those with schizophrenia who show violent and
    criminal proclivities more frequently:-

   come from deprived and disadvantaged
    backgrounds;
   have family histories of criminality;
   have had poor peer relationships through
    childhood and adolescents;
   had conduct disorder;
   failed educationally.

(Schanda et al 1992; Tihonen et al 1997; Fresán et al
  2004; Cannon 2002)
           Early Intervention
1.   Target children from disadvantaged
     backgrounds for school enhancement
     programmes
2.   Intervene early in educational failure
3.   Develop active management of conduct
     disorder
  Personality Vulnerabilities in
Schizophrenia explain part of the
  Association with Offending
                Genetic Vulnerability




Schizophrenia                           CD & ASPD
                  Genetic Vulnerability




Schizophrenia                                  CD & ASPD




            Socially Disadvantaged Childhood
Mean Number of Registered Total Criminal Offences per Year at Risk
 from Age 15 to Index Offence for Six Offender Groups Subjected to
                  Forensic Psychiatric Assessment

10
 9
 8
 7
 6
 5
 4
 3
 2
 1
 0




                                                                      SA
                SA




                                       +P
         z.




                                                          y
                            y




                                                       th
                         th
       hi




                                     SA
              S+




                                                                    +
                                                    pa
                        pa
     Sc




                                                                  P
                                   S+



                                                ho
                      ho




                                              yc
                    yc




                                            Ps
                  Ps
                S+




                             Source: Tengstrom, Grann, Langstrom, Hodgins & Kullgren, 2000
        CBT & Psychotherapies for
        Personality Vulnerabilities

ASPD (Psychopathic) traits
      Callousness and insensitivity
      Suspiciousness                }
       external locus of control    }
      Novelty seeking                   }
      Impulsiveness (fecklessness)      }
      Antagonism/negativity
      Poor insight

Plus cognitive deficits
       WHAT IS TO BE DONE
   Give high risk patients high priority
                 BUT
How do you recognise high risk groups?
   Keep it simple.

   Keep it focused

   Keep it clinical

   Make it Systematic

   Make it multidisciplinary

   Keep it management focused
        WHAT IS TO BE DONE
   Give high risk patients high priority
   Improve the social conditions under
    which those with schizophrenia live
   Ensure employment
   Address the criminogenic „personality‟
    factors
   Manage substance abuse
   Improve symptom control
   Improve risk management
                        Breaking the Links
                             Schizophrenia

Early Intervention                               Vigorous Management of
                                                 Active Illness

                        CBT for Personality      Manage
                        Vulnerabilities          Substance Abuse


Education Enhancement                    Compulsory I.P.
                                         Management if indicated

                        Social Skills Training


                        Placement in Supported
Works Skills Training   Accommodation in low
                        crime neighbourhoods
     Improved Management of High Risk
      Groups with Schizophrenia Could

1.    Reduce seriously violent crime by 2-4%.
2.    Reduce incarceration rates (prison and
      forensic hospitals) of those with
      schizophrenia by 30% - 50%.
3.    Improve quality of life for the most
      disturbed and disadvantaged among
      those with schizophrenia.
Are the associations between schizophrenia
   and offending an artifact of differential
       detection and conviction rates?
   Rates in crimes with a very high clear up
    rates (e.g. homicides) are greater than
    those with low clear up rates (e.g. theft)

   Probability that police are more reluctant
    to proceed to charge obviously mentally
    disordered individuals
 Offending in a Population of People
      with Schizophrenia(2861)
            (Wallace, Mullen, & Burgess, 2003)


  Lifetime violent convictions
                All       With SA No SA          control
1975 [8%]
1980 [10%]
1985 [16%]
1990 [18%]
1995 [27%]
 Offending in a Population of People
         with Schizophrenia
            (Wallace, Mullen, & Burgess, 2003)


  Lifetime violent convictions
                All       With SA No SA          control
1975 [8%]       6%
1980 [10%]      7%
1985 [16%] 7%
1990 [18%]      8%
1995 [27%]      10%
 Offending in a Population of People
         with Schizophrenia
            (Wallace, Mullen, & Burgess, 2003)


  Lifetime violent convictions
                All       With SA No SA          control
1975 [8%]       6%        21%          5%        1%
1980 [10%]      7%        35%          4%        0.5%
1985 [16%] 7%             22%          4%        2%
1990 [18%]      8%        25%          4%        0.8%
1995 [27%]      10%       27%          4%        3%
Address known mediators (risk factors)



Some evidence for two broad types among
  schizophrenics who are violent

1.   With psychopathic traits

2.   Without such traits ( ? Symptom driven )
What is to be Done?
   Keep it simple.
       eg. Male, young, substance abusing, histories
        of conduct disorder and/or offending
        behaviours, antagonistic, poorly compliant.
   Keep it clinical
       e.g. Angry and Threatening, Delusional
        Jealousy, Feckless, Poor insight, Frightened
        and Suspicious.
   Make it Systematic
       e.g. use HCR 20, or other structured clinical
        judgement approaches.
1.   Is there a significant correlation between
     having schizophrenia and violent offending?

2.   Is the correlation socially & clinically important?

3.   If so what might explain that correlation?

4.   What is to be done?
               Conclusions

1.   There is an association between
     schizophrenia and violent offending.
    Is the correlation explained by
         deinstitutionalization ?

        “It is hard not to see de-
institutionalisation as a major factor in
   the increasing number of crimes
 committed by schizophrenic patients”


                           Kramp (2005 )
Is the Correlation explained by
      Active Symptoms?
    Deinstitutionalisation in Victoria

Mental Hospital Beds in 1965        200 per 100,000

Deinstitutionalization begins in 1980

1st asylum closed 1985

Last asylum closed 1993

Beds in general and forensic hospital in 2000
                                  40 per 100,000
 Lifetime Convictions in the Schizophrenia Cohorts
                  Offences                 95%
              Cases       Controls O.R. Confidenc     P
                                        e Interval   valu
              N (%)        N (%)
                                                      e
1975 cohort 70 (14.8%)
lifetime
1980 cohort 90 (21.3%)
lifetime
1985 cohort 113(19.4%)
lifetime
1990 cohort 90 (24.9%)
lifetime
1995 cohort 255 (25.0%)
lifetime
 Lifetime Convictions in the Schizophrenia Cohorts
                  Offences                95%
              Cases      Controls O.R. Confidenc        P
                                       e Interval      valu
              N (%)       N (%)
                                                        e
1975 cohort 70 (14.8%)   24 (5.1%) 3.3     (2.0-5.3)   0.000
lifetime
1980 cohort 90 (21.3%)   20 (4.7%)   5.5   (3.3-9.0)   0.000
lifetime
1985 cohort 113(19.4%)   58          2.2   (1.6-3.1)   0.000
lifetime                 (10.0%)
1990 cohort 90 (24.9%)   24 (6.6%) 4.7     (2.9-7.5)   0.000
lifetime
1995 cohort 255 (25.0%) 98 (9.6%) 3.1      (2.4-4.0)   0.000
lifetime
         Schizophrenia Cohorts
              1975-1995

   Do not differ significantly from controls in
    age of onset of offending or in subsequent
    temporal patterns of offending
   Majority of offending careers start before
    diagnosis of schizophrenia
Victimisation for genders combined
  Offence Type      N=2861          Rates of
                     N(%)         Victimisation


Total offences     497 (17.37%)       1 in 6


Property-Related   436 (15.24%)       1 in 7
offences
Violent offences   108 (3.77%)       1 in 27


Homicide            5 (0.17%)        1 in 572


Sexual offences     21 (0.73%)       1 in 136
Victimisation for genders combined


 Cohort       Offence          N=2861          Rates of
               Type             N(%)         Victimisation
Combined   Total offences     497 (17.37%)       1 in 6

           Property-Related   436 (15.24%)       1 in 7
           offences
           Violent offences   108 (3.77%)       1 in 27

           Homicide            5 (0.17%)        1 in 572

           Sexual offences     21 (0.73%)       1 in 136
Reported Victimisation for Males

 Cohort       Offence          N=2861          Rates of
               Type             N(%)         Victimisatio
                                                  n
Combined   Total offences     316 (18.71%)      1 in 5

           Property-Related   286 (16.93%)      1 in 6
           offences
           Violent offences    73 (4.32%)       1 in 23

           Homicide            5 (0.30%)       1 in 338

           Sexual offences     1 (0.06%)       1 in 1689
Reported Victimisation for Females

 Cohort    Offence Type        N=2861          Rates of
                                N(%)         Victimisation
Combined   Total offences     181 (15.44%)       1 in 6

           Property-Related   150 (12.80%)       1 in 8
           offences
           Violent offences    35 (2.99%)       1 in 33

           Homicide            0 (0.00%)

           Sexual offences     20 (1.71%)       1 in 59
                     Conclusions

2.       Why there is an association could include:

          The direct influence of active symptoms
          The influence of comorbid substance abuse
          The social and economic conditions imposed by
           chronic disability
          Personality traits
         MENTAL DISORDERS IN
             OFFENDERS

Wallace Mullen Burgess et al (1998)



Sample             3,838 men and 315 women convicted
                   in higher courts 1993-1995

Ascertainment      Linkage with state wide register
                   of psychiatric contacts established
                   1961
        Offending in a Population of
        People with Schizophrenia
         (Wallace, Mullen, & Burgess, 2000 & 2004)


   5 cohorts of patients with Schizophrenia in
    Victorian public mental health registry
    (1975, 1980, 1985, 1990, 1995)
       Total = 2861 (1689 males, 1172 females)
   Matched control group (#, age, gender)
   Linked to criminal convictions
Scale Structure of the PCL-R: 2nd Ed. (Hare, 2003)


               Psychopathy
       Factor 1                        Factor 2
Interpersonal/Affective             Social Deviance



     • Promiscuous sexual behaviour
     • Many short-term marital relationships
      Scale Structure of the PCL-R: 2nd Ed. (Hare, 2003)

                         Psychopathy
             Factor 1                                   Factor 2
      Interpersonal/Affective                        Social Deviance

    Facet 1              Facet 2                Facet 3               Facet 4
 Interpersonal          Affective               Lifestyle            Antisocial
• Glibness/Super.   • Lack of remorse     • Need for stimul.      • Poor behavioural
   Charm               or guilt             prone. to boredom        controls
• Grandiose self-   • Shallow Affect      • Parasitic lifestyle   • Early behavioural
   worth            • Callous/Lack of     • Lack of realistic,       problems
• Pathological         Empathy               long-term goals      • Juvenile delinq.
   Lying            • Failure to accept   • Impulsivity           • Revocation of
• Conning/             responsibility     • Irresponsibility        condition. release
   Maniupulative       for actions                                • Criminal versatility
                         IN SCHIZOPHRENIA

                       Psychopathic Traits
              Factor 1                                  Factor 2
       Interpersonal/Affective                       Social Deviance

     Facet 1             Facet 2                Facet 3               Facet 4
  Interpersonal         Affective               Lifestyle            Antisocial
• Grandiose self-   • Lack of remorse     • Prone to boredom      • Poor behavioural
  worth                                                              controls
                    • Shallow Affect      • Dependent lifestyle
• Insensitivity                                                   • Early behavioural
                    • Callous/Lack of     • Lack of realistic,       problems
• Suspiciousness       Empathy               long-term goals
                                                                  • Juvenile delinq.
                    • Failure to accept   • Irresponsibility
                       responsibility
 Conceptual Overlap Among DSM-IV
Personality Disorders and Psychopathy


        Histrionic
                     Antisocial
         Histrionic
     Borderline
                Psychopathy

            Narcissistic
         Risk Management

Two basic approaches

1.   Identify whether the individual is in a high
     risk group and either incapacitate them or
     attempt to otherwise manage their risk.

2.   Identify the risk factors operative in the
     population from which the individual comes
     and attempt to reduce those risk factors in
     that population.
      The Appeal of the Individual
              Approach
1.   We are confronted clinically and forensically
     with individuals about whom we have to make
     decisions, not with groups.

2.   Management in practice is about what is to be
     done about this particular individual not about
     populations which may include them.
     Problems with the Individual
             Approach

1.   The established risk factors which inform
     decisions are based on group data. The
     individual approach depends on treating
     someone not according to their individual
     characteristics but as a function of the group
     to which certain, but not all, of those
     characteristics assigns them.
      Problems with the Individual
              Approach
2.   Assigning individuals to a level of risk is usually
     (?always) accompanied with a claim of a
     percentage probability for that person of
     committing a further violent/sexual/whatever
     act.
3.     In reality someone is or is not violent you
     cannot be 70%, 30% or less than 5% violent.
     Thus some individuals (at least with the
     incapacitation approach) will always be
     disadvantaged unfairly.
ROC              80% true positives 20% false
  positives

20% violence     identify 16       false positives 16

10% violence     identify 8        false positives 18

5% violence      identify 4        false positives 19

0.5% violence    identify 1        false positives 40

0.05% violence   identify 1        false positives
   500
   Problems with the Individual
           Approach
3. Low risk groups, or low scoring individuals,
  are usually far more numerous than the high.
  Thus the low risk individuals usually make a
  larger contribution to the behaviour you wish
  to avoid (70% of 100 people is less than 10%
  of 1,000).

  Identifying significant, or better still necessary
  but not sufficient risk factors, and managing
  them in the whole group will therefore usually
  be more effective than searching out the high
  risk individual and managing them.
 How to Resolve Some of the Problems

     Moving from data (group based risk factors) to
     evaluation (of individuals) and action (usually
     directed at individuals) you can either:
1.     Assume individual has the characteristics
       and risk level of group to which they are
       assigned and act on that basis.
2.     Identify presence or absence of mutable
       risk factors in the particular individual and
       manage those variable.
3.     Combine (1) to assign priority and (2) to
       direct action.
What does this mean in practice for
  mental health professionals?
For Assessment

1.   Focus is on the presence or absence of specific
     risk factors not on individual‟s supposed level
     of risk.

1.   Level of overall risk in this individual is about,
     and only about, assigning priority for
     treatment (compulsory versus voluntary
     treatment is not about level of risk but about
     capacity and competence to refuse such
     treatment).
What does this mean in practice for
  mental health professionals

For Management

1.   Managing risk factors in individuals and/or
     groups not risky individuals.
2.   Identifying what mediates risk factors even
     of the fixed variety (child abuse, prior
     imprisonment, impulsivity, callousness etc)
     and manage those intervening influences.
                          Risk Curves by age

Probability of Violence

                  High




                 Low


                          15   20   25   30    35   40   45   50
                                         Age




                                                         From Tony Florio
Therefore Prevention Must
Primarily Target High Risk
  Groups Not High Risk
        Individuals
                         IN SCHIZOPHRENIA

                       Psychopathic Traits
              Factor 1                                  Factor 2
       Interpersonal/Affective                       Social Deviance

     Facet 1             Facet 2                Facet 3               Facet 4
  Interpersonal         Affective               Lifestyle            Antisocial
• Grandiose self-   • Lack of remorse     • Prone to boredom      • Poor behavioural
  worth                                                              controls
                    • Shallow Affect      • Dependent lifestyle
• Insensitivity                                                   • Early behavioural
                    • Callous/Lack of     • Lack of realistic,       problems
• Suspiciousness       Empathy               long-term goals
                                                                  • Juvenile delinq.
                    • Failure to accept   • Irresponsibility
                       responsibility
   Schizophrenia Among 1998 Males
  Convicted of Serious Violent Offences
      1993-1995 (Wallace et al 1998)

Violent         91 cases      4.2%       OR 4.4 (3.5-5.7)
Offences
Proportion of              0.5% (0.17%
those with                 per year)
Schizophrenia
Convicted
 Victorian Prisoner Health Study
 Mental Health Screening Results
Diagnostic                  Males                   Females
Category
                Aboriginal       Non-       Aborigina       Non-
                  n = 62      Aboriginal        l        Aboriginal
                              n = 262-267    n = 12       n = 88-92
Schizophreni    5 (8.1%)      20 (7.5%)     3 (25%)     4 (4.3%)
a
Manic           5 (8.1%)      26 (9.7%)     1 (8.3%)    3 (3.3%)
Depression
Depression      9 (14.5%)     54 (20.5%)    2 (16.7%)   24 (27.0%)
Total with at 15 (24.2%) 72 (27.5)              4 (33.3%) 25 (28.4%)
Least One
Suspected MI of prisoners met the criteria for at least one of
    116 (26%)
    the diagnostic categories: Schizophrenia - 32 (7%), Major
    depression - 47 (20%), Manic depression - 35 (8%).
Prevalence Rates of Mental Disorders –
Victorian Prisoner Health Study

 Mental Illness                      Victorian PMHS
 Been told by Dr they have an MI        28%
 Schizophrenic Disorders                 7%
 Major Depression                       20%
 Manic Depression (Bipolar Disorder)     8%

 Substance Use
 Alcohol abuse and dependence                   40%
 Used illegal drugs                             66%
 Injected illegal drugs                         49%
 Sought help for drug abuse                     25%
 Sources: Ogloff, Barry-Walsh, & Davis (2003)
      SCHIZOPHRENIA IN 2153
        OFFENDERS (OR)
                All Cases of     Schizophrenia      Schizophrenia
               Schizophrenia   without Substance    with Substance
                                     Abuse              Abuse



Violent              4.4             2.4                 18.8
Offences
Property             2.8             N.S.                13.4
Offences



Odd Ratio p< 0.001
                                           Wallace et al (1998)
  Authors        Country       Homicides     Homicides with    Odds Ratio
                                               schizophreni      [95%
                                                     a          confidence
                                                                 interval]
Hafner &        Western          3367            8.0%         12.7 [11.2-14.3]
   Boker          German
   (1982)         y
Eronen et al    Finland          1037            6.1%          9.7 [7.4-12.6]
(1996)


Wallace et al   Victoria,        168              7.2%        10.1 [5.5-18.6]
(1998)              Australi
                    a


Erb et al       Germany          290             10.0%        16.1 [11.2-12.5]
(2001)                           (incl.
                                attempted)


Schanda et al   Austria          1087            5.4%          8.8 [6.7-11.5]
   (2004)
          Proposition 1
If there is no relationship between mental
             illness and crime…

                THEN


…mental health services have no role in
    managing criminal behaviour.
         Proposition 2
   If the symptoms of mental illness
 (delusions, hallucinations, confusion,
  obsessions) have no relationship to
                crime…

                THEN

…mental health services have no role in
   managing criminal behaviour.
         Proposition 3
  If substance abuse mediates the
  criminal behavior in the mentally
            disordered…

              THEN

…mental health services have no role
 in managing criminal behaviour.
         Proposition 4
     If it is the personality traits
(disorders) which are responsible for
     criminal behaviour, not the
               psychosis…

              THEN

…mental health services have no role
 in managing criminal behaviour.
         Proposition 5
   If it is unemployment, living in a
   disorganised neighbourhood and
associating with lawless and antisocial
     groups which leads to crime…

               THEN

…mental health services have no role in
   managing criminal behaviour.
                 BUT…
    Although these five propositions are
       believed by many mental health
   professionals, and there is considerable
           truth in all but the first…

…mental health services DO have a role and
    a responsibility to manage criminal
                behaviour.


                 WHY?
SCHIZOPHRENIA




  VIOLENCE
       PREVALENCE OF AUSTRALIAN
    PRISONERS WITH A MENTAL ILLNESS



   Major mental disorder    8% m.         15% f.
    (psychosis)

   Schizophrenias      5% m.      6% f.

   Personality disorders 39% m.   49% f.
    PREVALENCE OF AUSTRALIAN
 PRISONERS WITH A MENTAL ILLNESS

Substance Abuse
  Alcohol                         55% m.    33% f.
  - hazardous drinking levels I community

Regular community use in 12 months prior to imprisonment
  Cannabis                     55% m.         33% f.
  Opiates                      27% m.         50% f.
  Cocaine                      21% m.         26% f.
  Amphetamines                 21% m.         20% f.
BIRTH COHORT STUDIES
      Finland Birth Cohort (12,058
       subjects (Tiihonen et al 1997)

By age 26 years

Schizophrenia    51 cases
      13.7% violent offences      OR 7.0 (3.1-15.9)

SES interacts lowest SES more than doubles risk of a
  conviction compared to highest SES
           COMMUNITY STUDY II
      Birth Cohort 173,668 men 162,322 women

                   RELATIVE RISKS

           ANY OFFENCE        VIOLENT OFFENCE

           Males    Females    Males    Females
MMD         2.3       3.3       2.4        5.9
MR          5.6       10.4      5.8        8.0
APD         4.2       5.4       5.3        7.9
Drugs       5.4       6.8       5.9       10.2
                                Hodgins et al 1996
 Risks for Violent Behaviours in the 12 Months
     Before 21 Years of Age (961 persons)

                Court            Self-            Court Convictions and/or
                Convictions      reported         Self-reported Violence
                for Violence     Violence
DSM-III-R       OR (95% CI)      OR               OR (95% CI) AOR
Axis 1                           (95%CI)                      (95% CI)
Disorder
Alcohol         2.5(1.1-5.6)     5.4 (3.1-9.4)    4.0 (2.4-6.8)    3.4 (2.0-5.9)
dependence
disorder
Marijuana       8.1 (4.1-16.0)   6.1 (3.5-10.5) 6.9 (4.1-11.4)     5.4 (3.2-9.2)
dependence
disorder
Schizophrenia   5.1 (2.0-13.1)   7.1 (3.5-14.6)   5.4 (2.6-10.9)   4.6 (2.2-9.7)
spectrum
disorder
                                                   Arseneault et al 2000
            SCHIZOPHRENIA IN
             OFFENDERS (OR)
                 All Cases of
                Schizophrenia

Any                  3.2
Offence
Violent              4.4
Offences
Property             2.8
Offences
Homicide             10.1

Odd Ratio p< 0.001
                                Wallace et al (1998)
            SCHIZOPHRENIA IN
             OFFENDERS (OR)
                All Cases of     Schizophrenia      Schizophrenia
               Schizophrenia   without Substance    with Substance
                                     Abuse              Abuse
Any                  3.2             1.9                 12.4
Offence
Violent              4.4             2.4                 18.8
Offences
Property             2.8             N.S.                13.4
Offences
Homicide             10.1            7.1                 28.8

Odd Ratio p< 0.001
                                           Wallace et al (1998)
  Odds Ratios for Homicide Schizophrenia
     and Substance Abuse in Males

 Study       All Sz   S.A.   Sz +   Sz - SA
                             S.A.
Eronen et     10.0     ?     17.2    7.25
al 1996

Wallace et    10.1    5.7    28.8     7.1
al 1998

Schanda       6.5      ?     20.7     7.1
et al 2005
 Lifetime Convictions in the Schizophrenia Cohorts
                  Offences                    95%
                                     O.R.   Confidence     P
               Cases      Controls           Interval    value


1975 (8%)   70 (14.8%)   24 (5.1%) 3.3      (2.0-5.3)    0.000

1980 (10%) 90 (21.3%)    20 (4.7%)   5.5    (3.3-9.0)    0.000


1985 (16%) 113(19.4%)    58          2.2    (1.6-3.1)    0.000
                         (10.0%)

1990(18%)   90 (24.9%)   24 (6.6%) 4.7      (2.9-7.5)    0.000


1995 (27%) 255 (25.0%) 98 (9.6%) 3.1        (2.4-4.0)    0.000
           Prior Offending ?
The increased rates of offending among
patients can be explained by their previous
arrest records
                        (Monahan 1981, Cohen 1980).


Even among the mentally ill past criminality
remains a robust predictor of future
criminality

     (Bonta et al 1998; Hodgins & Muller Isberner 2004)
Is the correlation explained by prior
        offending histories?
OFFENDING AMONG THE
MENTALLY DISORDERED

In 63% males and 55% of females
offending preceded first psychiatric
contact. Mean age at first offence did
not differ from controls. Offending
increased in two years prior and
subsequent to first contact.

                       Mullen et al 2000
     SCHIZOPHRENIA AMONG
      VIOLENT OFFENDERS
Taylor & Gunn (1984)

Sample                 1,241 men on remand

Ascertainment          PSE

Schizophrenia   -      expected rate   0.6%

                -      non fatal violence 9%

                -      homicide         11%
Is there an association
        between
schizophrenia & criminal
       violence ?
      SCHIZOPHRENIA IN SERIOUS
   OFFENDERS in 3838 males (1993-1995)

                         % (ODDS RATIOS)
           Schizophrenia in Victorian population
                           0.7%
Any           2.4% (3.2 95%CI 2.6-3.9 )
Offence
Violent       4.2% (4.4 95%CI 3.5-5.7)
Offences
Property      2.1% (2.8 95%CI 1.9-4.2)
Offences
Homicide      7.2% (10.1 95%CI 5.5-18.6)


                                         Wallace et al (1998)
 Potential Mediators (Confounders)
        of the Relationship I


1.   Substance Abuse

2.   Prior Offending

3.   Early Development

4.   Personality Traits
     Potential Mediators (Confounders)
            of the Relationship II

6.    Social Conditions

7.    Assumed differential arrest and conviction
      rates
Mediators
1. Schizophrenia        Mediator           Violence
2. Schizophrenia   Mediator A       Mediator B      Violence

                       Mediator A
                                             Violence
3. Schizophrenia
                       Mediator B
Confounders
4. Schizophrenia       Confounder              Violence

5. Schizophrenia      Confounder               Violence
6. Schizophrenia       Violence              Confounder

                      Mediator

7. Schizophrenia                           Violence
                      Confounder
Homicide Offences in 2861 Patients
       with Schizophrenia
9 cases were charged with murder. 4 convicted, 1
adjudicated insane, 3 convicted of lesser offences and one
acquitted.
       Homicide/attempted homicide: 8 (0.2%)

No controls were charged with, or convicted of, murder.

Homicide Rate:
1975-2000 1 - 1.5 per 100,000 per year
14 year risk 14 - 21 per 100,000
Chances for 3,000 over 14 yr 0.42 - 0.63 (0.01% - 0.02%)
Consequences of Psychopathology
       of Schizophrenia
1.   Pre-existing Vulnerabilities
     Developmental Difficulties, Conduct Disorder, Dissocial,
     Educational Failure, Psychopathic Traits, Substance
     Abuse

2.   Acquired Vulnerabilities
     Active Symptoms, Personality Changes, Social
     Dislocation, Substance Abuse

3.   Imposed Vulnerabilities
     Drug Side Effects, Increased Isolation, Incarceration
     eroding social skills and employment prospects.
There is a correlation between having
 schizophrenia and violent offending.

Is this due to confounding influences ?

                If not

 What might explain this correlation?
What mediates that increase?
   In some, the illness itself
   In some, personality vulnerabilities
   In some, associated substance abuse
   In some, the social dislocation
    associated with the illness
   In some, a disrupted personal and
    social development
   In many, some or all of the above.
                   Mediator B
Confounders
4. Schizophrenia   Confounder      Violence

5. Schizophrenia   Confounder      Violence
6. Schizophrenia    Violence     Confounder

                   Mediator

7. Schizophrenia                Violence
                   Confounder
Mediators
1. Schizophrenia        Mediator        Violence
2. Schizophrenia   Mediator A    Mediator B Violence

                       Mediator A
                                        Violence
3. Schizophrenia
                      Mediator B
Confounders

				
DOCUMENT INFO