WPA2_09_Stuart 3-06-2003 11:42 Pagina 121 (Black/Process Black pellicola)

MENTAL HEALTH POLICY PAPER

Violence and mental illness: an overview

HEATHER STUART
Department of Community Health and Epidemiology, Abramsky Hall, Queens University, Kingston, Ontario, K7L 3N6, Canada

This paper evaluates the relationship of mental illness and violence by asking three questions: Are the mentally ill violent? Are the men-
tally ill at increased risk of violence? Are the public at risk? Mental disorders are neither necessary nor sufficient causes of violence.
Major determinants of violence continue to be socio-demographic and economic factors. Substance abuse is a major determinant of vio-
lence and this is true whether it occurs in the context of a concurrent mental illness or not. Therefore, early identification and treatment
of substance abuse problems, and greater attention to the diagnosis and management of concurrent substance abuse disorders among
seriously mentally ill, may be potential violence prevention strategies. Members of the public exaggerate both the strength of the associ-
ation between mental illness and violence and their own personal risk. Finally, too little is known about the social contextual deter-
minants of violence, but research supports the view the mentally ill are more often victims than perpetrators of violence.

Key words: Mental illness and violence, stigma, violence prevention, victimization

Are the mentally ill violent? Are they more violent than with violence are not representative of the behaviours of
people without a mental illness? Are they a risk to public the majority of mentally ill. Social changes in the practice
safety? These questions have framed both the scientific of psychiatry, particularly the widespread adoption of the
and the public debate surrounding the relationship of vio- dangerousness standard for civil commitment legislation,
lence to mental illness. means that only those with the highest risk of violence
Unless otherwise stated, violence will refer to acts of receive treatment in acute care settings.
physical violence against others, since these are the most In fact, a serious limitation of clinical explanations of
fear-inducing for the public and the greatest determinants of violent and disruptive behaviour is their focus on the
social stigma and discrimination. The term mental illness attributes of the mental illness and the mentally ill to the
will be reserved for non-substance related disorders, usual- exclusion of social and contextual factors that interact to
ly major mental illnesses such as schizophrenia or depres- produce violence in clinical settings. Even in treatment
sion. Substance related disorders and concurrent substance units with a similar clinical mix and acuity, rates of aggres-
abuse will be identified and discussed as separate risk fac- sive behaviours are known to differ dramatically, indicat-
tors. ing that mental illness is not a sufficient cause for the
occurrence of violence (4). Studies that have examined the
antecedents of aggressive incidents in inpatient treatment
ARE THE MENTALLY ILL VIOLENT?
units reveal that the majority of incidents have important
Over time, there seems to have been a progressive con- social/structural antecedents such as ward atmosphere,
vergence of mental illness and violence in day-to-day clin- lack of clinical leadership, overcrowding, ward restric-
ical practice. From early declarations disavowing the com- tions, lack of activities, or poorly structured activity tran-
petence of mental health professionals to predict violence, sitions (4-6).
there has been a growing willingness on the part of many The public are no less accustomed to experiencing vio-
mental health professionals to predict and manage violent lence among the mentally ill, although these experiences
behaviour. With the advent of actuarial risk assessment are mostly vicarious, through movie depictions of crazed
tools, violence risk assessments are increasingly promoted killers or real life dramas played out with disturbing fre-
as core mental health skills: expected of mental health quency on the nightly news. Indeed, the global reach of
practitioners, prized in courts of law and correctional set- news ensures that the viewing public will have a steady diet
tings, and key aspects of socially responsible clinical man- of real-life violence linked to mental illness. The public
agement (1,2). most fear violence that is random, senseless, and unpre-
Many psychiatrists, particularly those working in emer- dictable and they associate this with mental illness. Indeed,
gency or acute care settings, report direct experiences with they are more reassured to know that someone was stabbed
violent behaviour among the mentally ill. In Canada, for to death in a robbery, than stabbed to death by a psychotic
example, where violence in the population is low relative man (7). In a series of surveys spanning several real-life
to most other countries, the majority of psychiatrists are events in Germany, Angermeyer and Matschinger (8)
involved in the management and treatment of violent showed that the publics desire to maintain social distance
behaviour, and 50% report having been assaulted by a from the mentally ill increased markedly after each publi-
patient at least once (3). However, clinical experiences cized attack, never returning to initial values. Further, these

121
WPA2_09_Stuart 3-06-2003 11:42 Pagina 122 (Black/Process Black pellicola)

incidents corresponded with increases in public percep- non-mentally ill counterparts. Therefore, the question of
tions of the mentally ill as unpredictable and dangerous. whether the mentally ill are at a higher-than-average risk
In some countries, such as the United States, public of violence is central to the scientific debate.
opinion has become quite sophisticated. The public judge Definitive statements are difficult to make and it is
the risk of violence differently, depending on the diagnos- equally possible to find recent literature supporting the
tic group, with rankings that broadly correspond to exist- conclusions that the mentally ill are no more violent, they
ing research findings. For example, Pescosolido et al (9) are as violent, or they are more violent than their non-
surveyed the American public (N=1,444) using standard- mentally ill counterparts (14). Prior to 1980, the dominant
ized vignettes to assess their views of mental illness and view was that the mentally ill were no more, and often less
treatment approaches. Respondents rated the following likely to be violent. Crime and violence in the mentally ill
groups as very or somewhat likely of doing something vio- were associated with the same criminogenic factors
lent to others: drug dependence (87.3%), alcohol depend- thought to determine crime and violence in anyone else:
ence (70.9%), schizophrenia (60.9%), major depression factors such as gender, age, poverty, or substance abuse.
(33.3%), and troubled (16.8%). While the probability of Any elevation in rates of crime or violence among mental-
violence was universally overestimated, respondents cor- ly ill samples was attributed to the excess of these factors.
rectly ranked substance abusers among the highest risk When they were statistically controlled, the rates often
groups. Similarly, they significantly overestimated the risk equalized. However, although the main risk factors for
of violence among schizophrenia and depression, but cor- violence still remain being young, male, single, or of lower
rectly identified these among the lower ranked groups. socio-economic status, several more recent studies have
Public perceptions of the link between mental illness reported a modest association between mental illness and
and violence are central to stigma and discrimination as violence, even when these elements have been controlled
people are more likely to condone forced legal action and (1-2,7,13-16).
coerced treatment when violence is at issue (9). Further, Because of the significant methodological challenges
the presumption of violence may also provide a justifica- faced by researchers in this field, the nature of this associa-
tion for bullying and otherwise victimizing the mentally ill tion remains unclear. For example, violence has been diffi-
(10). High rates of victimization among the mentally ill cult to measure directly, so that researchers have often
have been noted, although this often goes unnoticed by relied on official documentation or uncorroborated self-
clinicians and undocumented in the clinical record. In a reports. The prevalence of violence has been demonstrated
study of current victimization among inpatients, for exam- to differ dramatically depending on the source (17). Most
ple, 63% of those with a dating partner reported physical samples have not been representative of all mentally ill indi-
victimization in the previous year. For a quarter, the vio- viduals, but only of those with the highest risk of becoming
lence was serious, involving hitting, punching, choking, dangerous, such as those who are hospitalized or arrested.
being beaten up, or being threatened with a knife or gun. Study designs have not always eliminated individuals with
Forty-six percent of those who lived with family members a prior history of violence (a major predictor of future vio-
reported being physically victimized in the previous year lence), controlled for co-morbid substance abuse, or clear-
and 39% seriously so. Three quarters of those reporting ly determined the sequencing of events, thereby weakening
violence from a dating partner retaliated, as did 59% of any causal arguments that might be made (14).
those reporting violence from a family member (11). In The MacArthur Violence Risk Assessment Study recent-
addition, many people with serious mental illnesses are ly completed in the United States (1,18,19) has made a
poor and live in dangerous and impoverished neighbour- concerted effort to address these problems, so it stands out
hoods where they are at higher risk of being victimized. A as the most sophisticated attempt to date to disentangle
recent study of criminal victimization of persons with these complex interrelationships. Because they collected
severe mental illness showed that 8.2% were criminally extensive follow-up data on a large cohort of subjects
victimized over a four month period, much higher than (N=1,136), the temporal sequencing of important events is
the annual rate of violent victimization of 3.1 for the gen- clear. Because they used multiple measures of violence,
eral population (12). A history of victimization and bully- including patient self-report, they have minimized the
ing may predispose the mentally ill to react violently when information bias characterizing past work. The innovative
provoked (13). use of same-neighbour comparison subjects eliminates
confounding from broad environmental influences such as
socio-demographic or economic factors that may have
ARE THE MENTALLY ILL AT INCREASED
exaggerated differences in past research.
RISK OF VIOLENCE?
In this study, the prevalence of violence among those
Scientists are less interested in the occurrence of isolat- with a major mental disorder who did not abuse sub-
ed acts of violence among those with a mental illness, and stances was indistinguishable from their non-substance
more interested in whether the mentally ill commit acts of abusing neighbourhood controls. A concurrent substance
violence with greater frequency or severity than do their abuse disorder doubled the risk of violence. Those with

122 World Psychiatry 2:2 - June 2003

WPA2_09_Stuart 3-06-2003 11:42 Pagina 123 (Black/Process Black pellicola)

schizophrenia had the lowest occurrence of violence over the community could be reduced by less than five percent
the course of the year (14.8%), compared to those with a if major mental disorders could be eliminated. The popu-
bipolar disorder (22.0%) or major depression (28.5%). lation attributable risk for those with a substance abuse
Delusions were not associated with violence, even threat- disorder was 34%, and for those with a comorbid mental
control override delusions that cause an individual to illness and substance abuse disorder it was 5%. Therefore,
think that someone is out to harm them or that someone by these estimates, violence in the community might be
can control their thoughts. Previous cross-sectional stud- reduced by only 10% if both major mental disorders and
ies conducted in the United States (20,21) and Israel comorbid disorders were eliminated. However, violence
(22,23) had linked threat-control override delusions to an could be reduced by over a third if substance abuse disor-
increased risk of violence. ders were eliminated.
The importance of substance abuse as a risk factor for Using a similar approach, a Canadian study asked what
violence has been well articulated in other studies. Conse- proportion of violent crimes involving a police arrest and
quently, this may stand out as one of the robust clinical detention could be attributed to people with a mental dis-
findings in the field (24-28). Substance abuse in the con- order. They surveyed 1,151 newly detained criminal
text of medication non-compliance is a particularly offenders representing all individuals incarcerated in a
volatile combination and poor insight also may be a factor geographically defined area. Three percent of the violent
(25). crimes accruing to this sample were attributable to people
with major mental disorders, such as schizophrenia or
depression. An additional seven percent were attributable
ARE THE PUBLIC AT RISK?
to offenders with primary substance abuse disorders.
It is important to keep in mind that both serious violence Therefore, if major mental illness and substance disorder
and serious mental disorder are rare events. Therefore, it is could be eliminated from this population, the proportion
difficult to judge the practical importance of findings that of violent crime would drop by about 10% (32).
may show an elevated risk of violence among samples of
mentally ill as they tell us little about public risk.
CONCLUSIONS
One way of approaching this issue is ask who are the
most likely targets of violence by the mentally ill: members Several general conclusions are supported by this brief
of the general public or members of their close personal overview. First, mental disorders are neither necessary, nor
networks? Most recent studies suggest that violent inci- sufficient causes of violence. The major determinants of
dents among persons with serious mental disorders are violence continue to be socio-demographic and socio-eco-
sparked by the conditions of their social life, and by the nomic factors such as being young, male, and of lower
nature and quality of their closest social interactions (29). socio-economic status.
In the MacArthur Violence Risk Assessment Study (1), for Second, members of the public undoubtedly exaggerate
example, the most likely targets of violence were family both the strength of the relationship between major mental
members or friends (87%), and the violence typically disorders and violence, as well as their own personal risk
occurred in the home. Discharged patients were less like- from the severely mentally ill. It is far more likely that people
ly to target complete strangers (10.7%) compared to their with a serious mental illness will be the victim of violence.
community controls (22.2%). Similarly, in a social net- Third, substance abuse appears to be a major determi-
work study that followed 169 people with serious mental nant of violence and this is true whether it occurs in the
disorder over thirty months (30), violence most frequently context of a concurrent mental illness or not. Those with
erupted in the family when relationships were character- substance disorders are major contributors to community
ized by mutual threat, hostility, and financial dependence; violence, perhaps accounting for as much as a third of
when there was a diagnosis of schizophrenia with concur- self-reported violent acts, and seven out of every 10 crimes
rent substance abuse; and when outpatient mental health of violence among mentally disordered offenders.
services were used infrequently. Of the over 3,000 social Finally, too much past research has focussed on the per-
network members studied, only 1.5% were ever targets of son with the mental illness, rather than the nature of the
violent acts or threats. social interchange that led up to the violence. Conse-
A related question asks to what extent do mentally ill quently, we know much less than we should about the
contribute to the overall prevalence of community vio- nature of these relationships and the contextual determi-
lence. Using data from the Epidemiologic Catchment Area nants of violence, and much less than we should about
studies conducted in the United States, Swanson (31) opportunities for primary prevention (30). Nevertheless,
reported population attributable risks for self-reported current literature supports early identification and treat-
physical violence. Attributable risk refers to the overall ment of substance abuse problems, and greater attention
effect a factor has on the level of violence in the popula- to the diagnosis and management of concurrent substance
tion. For those with a major mental disorder, the popula- abuse disorders among seriously mentally ill as potential
tion attributable risk was 4.3%, indicating that violence in violence prevention strategies (25).

123
WPA2_09_Stuart 3-06-2003 11:42 Pagina 124 (Black/Process Black pellicola)

References data from the MacArthur Violence Risk Assessment Study. Am J

Psychiatry 2000;157:566-72.
1. Monahan J, Steadman HJ, Silver E et al. Risk assessment: the 19. Steadman HJ, Mulvy EP, Monahan J et al. Violence by people dis-
MacArthur Study of Mental Disorder and Violence. Oxford: charged from acute psychiatric inpatient facilities and by others in
Oxford University Press, 2001. the same neighbourhoods. Arch Gen Psychiatry 1998;55:393-404.
2. Mullen PE. A reassessment of the link between mental disorder 20. Link BG, Andrews H, Cullen F. The violent and illegal behaviour
and violent behaviour, and its implications for clinical practice. of mental patients reconsidered. Am Sociol Rev 1992;57:275-92.
Aust N Zeal J Psychiatry 1997;31:3-11. 21. Link BG, Stueve A. Psychotic symptoms and the violent/illegal
3. Bourget D, el-Guebaly N, Atkinson MJ. Assessing and managing behavior of mental patients compared to community controls. In:
violent patients. CPA Bull 2002;34:25-7. Monahan J, Steadman HJ (eds). Violence and mental disorder:
4. Katz P, Kirkland FR. Violence and social structure on mental hos- developments in risk assessment. Chicago: Chicago University
pital wards. Psychiatry 1990;53:262-77. Press, 1994:137-59.
5. Shepherd M, Lavender T. Putting aggression into context: an inves- 22. Link BG, Stueve A, Phelan J. Psychotic symptoms and violent
tigation into contextual factors influencing the rate of aggressive behaviours: probing the components of threat/control-override
incidents in a psychiatric hospital. J Ment Hlth 1999;8:159-70. symptoms. Soc Psychiatry Psychiatr Epidemiol 1998;33:S55-S60.
6. Powell G, Caan W, Crowe M. What events precede violent inci- 23. Link BG, Monahan J, Stueve A et al. Real in their consequences: a
dents in psychiatric hospitals? Br J Psychiatry 1994;165:107-12. sociological approach to understanding the association between
7. Marzuk P. Violence, crime, and mental illness. How strong a link? psychotic symptoms and violence. Am Sociol Rev 1999;64:316-22.
Arch Gen Psychiatry 1996;53:481-6. 24. Swanson J, Borum R, Swartz M et al. Psychotic symptoms and dis-
8. Angermeyer MC, Matschinger H. Violent attacks on public figures orders and the risk of violent behavior in the community. Criminal
by persons suffering from psychiatric disorders. Their effect on the Behav Ment Hlth 1996;6:317-38.
social distance towards the mentally ill. Eur Arch Psychiatry Clin 25. Swartz MS, Swanson JW, Hiday VA et al. Violence and severe
Neurosci 1995;245:159-64. mental illness: the effects of substance abuse and nonadherence to
9. Pescosolido BA, Monahan J, Link BG et al. The publics view of medication. Am J Psychiatry 1998;155:226-31.
the competence, dangerousness, and need for legal coercion of 26. Tiihonen J, Isohanni M, Rsnen P et al. Specific major mental dis-
persons with mental health problems. Am J Public Hlth orders and criminality: a 26-year prospective study of the 1966
1999;89:1339-45. Northern Finland Birth Cohort. Am J Psychiatry 1997;154:840-5.
10. Campbell J. Taking issue: putting violence in context. Hosp Com- 27. Mullen PE, Burgess P, Walace C et al. Community care and crimi-
mun Psychiatry 1994;45:633. nal offending in schizophrenia. Lancet 2000;355:614-7.
11. Cascardi M, Mueser KT, DeGiralomo J et al. Physical aggression 28. Arseneault L, Moffitt T, Caspi A et al. Mental disorders and vio-
against psychiatric inpatients by family members and partners. Psy- lence in a total birth cohort: results from the Dunedin Study. Arch
chiatr Serv 1996;47:531-3. Gen Psychiatry 2000;57:979-86.
12. Hiday VA, Swartz MS, Swanson JW et al. Criminal victimization of 29. Swanson J, Swartz M, Estroff S et al. Psychiatric impairment, social
persons with severe mental illness. Psychiatr Serv 1999;50:62-8. contact, and violent behavior: evidence from a study of outpatient-
13. Hiday VA. The social context of mental illness and violence. J Hlth committed persons with severe mental disorder. Soc Psychiatry
Soc Behav 1995;36:122-37. Psychiatr Epidemiol 1998;33:S86-S94.
14. Wessely S. Violence and psychosis. In: Thompson C, Cowen P 30. Estroff SE, Swanson JW, Lachicotte WS et al. Risk reconsidered:
(eds). Violence. Basic and clinical science. Oxford: Butter- targets of violence in the social networks of people with serious
worth/Heinemann, 1993:119-34. psychiatric disorders. Soc Psychiatry Psychiatr Epidemiol
15. Link B, Stueve A. Evidence bearing on mental illness as a possible 1998;33:S95-S101.
cause of violent behaviour. Epidemiol Rev 1995;17:172-81. 31. Swanson JW. Mental disorder, substance abuse, and community
16. Noffsinger SG, Resnick PJ. Violence and mental illness. Curr Opin violence: an epidemiologic approach. In: Monahan J, Steadman HJ
Psychiatry 1999;12:683-7. (eds). Violence and mental disorder: developments in risk assess-
17. Swanson J, Borum R, Swartz M et al. Violent behaviour preceding ment. Chicago: University of Chicago Press, 1994:101-36.
hospitalization among persons with severe mental illness. Law 32. Stuart H, Arboleda-Flrez J. A public health perspective on violent
Hum Behav 1999;23:185-204. offenses among persons with mental illness. Psychiatr Serv
18. Applebaum PS, Robbins PC, Monahan J. Violence and delusions: 2001;52:654-9.

124 World Psychiatry 2:2 - June 2003