Violent behavior: One of the consequences of failing to treat individuals with severe mental illness

E-mail Print PDF

Briefing Paper
Updated April 2011

Click here to download a printable PDF file  


1.    Most individuals with serious mental illnesses are not dangerous.
2.    Most acts of violence are committed by individuals who are not mentally ill.
3.    Being a young male or being a substance abuser (alcohol or drugs) is a greater risk factor for violent behavior than being mentally ill.
4.    Individuals with serious mental illness are victimized by violent acts more often than they commit violent acts (see Backgrounder: “Victimization: One of the consequences of failing to treat individuals with serious mental illnesses”).
5.    If people with serious mental illnesses are being appropriately treated, there is no evidence that they are any more dangerous than individuals in the general population.
6.    All of the above statements are true, but it is also true that a small number of individuals with serious mental illnesses commit acts of violence, including 5 to 10 percent of all homicides. Almost all these acts of violence are committed by individuals who are not being treated, and many such individuals are also abusing alcohol or drugs.



•    A meta-analysis of 204 studies of psychosis as a risk factor for violence reported that “compared with individuals with no mental disorders, people with psychosis seem to be at a substantially elevated risk for violence.” Psychosis “was significantly associated with a 49%–68% increase in the odds of violence.”
Douglas KS, Guy LS, Hart  SD. Psychosis as a risk factor for violence to others: a meta-analysis. Psychological Bulletin 2009;135:679–706.

•    A review of 22 studies published between 1990 and 2004 “concluded that major mental disorders, per se, especially schizophrenia, even without alcohol or drug abuse, are indeed associated with higher risks for interpersonal violence.” Major mental disorders were said to account for between 5 and 15 percent of community violence.
Joyal CC, Dubreucq J-L, Gendron C et al. Major mental disorders and violence: a critical update. Current Psychiatry Reviews 2007;3:33–50.

•    Among 3,743 individuals with bipolar disorder, 8.4 percent committed violent crimes compared to 3.5 percent of the general population in Sweden.
Fazel S, Lichtenstein P, Grann M et al. Bipolar disorder and violent crime. Archives of General Psychiatry 2010;67:931–938.

•    Of 8,003 individuals with schizophrenia, 13.2 percent committed at least one violent crime compared with 5.3 percent of the general population in a study in Sweden. Concurrent abuse of alcohol or drugs accounted for much of the increased rate.
Fazel S, Langstrom N, Hjern A et al. Schizophrenia, substance abuse, and violent crime. Journal of the American Medical Association 2009;301:2016–2023.

•    Data on mental disorders and violence was collected on 34,653 individuals as part of the US National Epidemiologic Survey on Alcohol and Related Conditions. According to one analysis (Elbogen and Johnson), “the incidence of violence was higher for people with severe mental illness, but only significantly so for those with co-occurring substance abuse and/or dependence.” According to another analysis (Van Dorn et al.), which criticized the first analysis, “those with SMI [serious mental illness], irrespective of substance abuse status, were significantly more likely to be violent than those with no mental or substance use disorders.”
Elbogen EB, Johnson SC. The intricate link between violence and mental disorder. Archives of General Psychiatry 2008;66:152–161.
Van Dorn R, Volavka J, Johnson N. Mental disorder and violence: is there a relationship beyond substance use? Social Psychiatry and Psychiatric Epidemiology 2011 Feb 26. [Epub ahead of print].

•    After reviewing the psychiatric literature from 1970 to 2007, the author of the study concluded that “sound epidemiologic research has left no doubt about a significant relation between psychosis and violence, although one accounting for little of society’s violence.” She also noted that “when violence is a problem, individuals in the immediate social circle of an individual suffering from psychosis are most at risk.”
Taylor PJ, Psychosis and violence: stories, fears, and reality. Canadian Journal of Psychiatry 2008;53:647–659.

•    A study in Ohio compared 122 patients with schizophrenia who had committed violent acts with 111 patients with schizophrenia who had not committed such acts. The violent patients had significantly more prominent symptoms and significantly less awareness of their illness.
Friedman L, Hrouda D, Noffsinger S et. al. Psychometric relationships of insight in patients with schizophrenia who commit violent acts. Schizophrenia Research 2003;60:81.

•    A study of 961 young adults in New Zealand reported that individuals with schizophrenia and associated disorders were two-and-one-half times more likely than controls to have been violent in the past year. If the person was also a substance abuser, the incidence of violent behavior was even higher.
Arseneault L, Moffitt TE, Caspi A et. al. Mental disorders and violence in a total birth cohort. Archives of General Psychiatry 2000;57:979–986.

•    A study of 63 inpatients with schizophrenia in Spain reported that the best predictors of violent behavior were being sicker (i.e., higher scores on symptom measures) and less insight into their illness. "The single variable that best predicted violence was insight into psychotic symptoms."
Arango C, Barba AC, Gonzalez-Salvador T et. al. Violence in schizophrenic inpatients: a prospective study. Schizophrenia Bulletin 1999;25:493–503.

•    A 10-year follow-up of 1,056 severely mentally ill patients discharged from mental hospitals in Sweden in 1986 reported that "of those who were 40 years old or younger at the time of discharge, nearly 40 percent had a criminal record as compared to less than 10 percent of the general public." Furthermore, "the most frequently occurring crimes are violent crimes."
Belfrage H. A ten-year follow-up of criminality in Stockholm mental patients. British Journal of Criminology 1998;38:145–155.

•    A study of 331 individuals with severe mental illness in the United States reported that 17.8 percent "had engaged in serious violent acts that involved weapons or caused injury." It also found that "substance abuse problems, medication noncompliance, and low insight into illness operate together to increase violence risk."
Swartz MS, Swanson JW, Hiday VA et. al. Violence and severe mental illness: the effects of substance abuse and nonadherence to medication. American Journal of Psychiatry 1998;155:226–231.

•    In a study in Finland, an unselected birth cohort of 11,017 individuals was followed for 26 years. Men with schizophrenia without alcoholism were 3.6 times more likely to commit a violent crime than men without a psychiatric diagnosis. Men with both schizophrenia and alcoholism were 25.2 times more likely to commit a violent crime.
Rasanen P, Tiihonen J, Isohanni M Schizophrenia, alcohol abuse, and violent behavior: a 26-year follow-up study of an unselected birth cohort. 1998;Schizophrenia Bulletin 24:437–441.

•    There is very little data that can be used to estimate the percentage of severely mentally ill individuals who become violent. The best study used the Danish psychiatric case register, covering the whole country, and convictions for criminal offenses. Between1978 and 1990, 6.7 percent of males and 0.9 percent of females with "major mental disorders" (psychoses) were convicted of a violent crime ("all offenses involving interpersonal aggression or a threat thereof") compared with 1.5 percent males and 0.1 percent females among individuals with no psychiatric diagnosis. Since these are only convictions, it can be assumed that another unknown percentage committed a violent act for which they were not charged or convicted.
Hodgins S, Mednick SA, Brennan PA, Mental disorder and crime. Archives of General Psychiatry 1996;53:489–496.

•    A study in Switzerland compared 282 men with schizophrenia with a matched control group in the general population. The patients were five times more likely to have been convicted of violent crimes, mostly "assaults resulting in bodily harm." The more acutely ill the patient was, the more likely he was to have been violent.
Modestin J, Ammann R. Mental disorder and criminality: male schizophrenia. Schizophrenia Bulletin 1996;22: 69–82.

•    A study of 538 individuals with schizophrenia living in London reported that the men had a 3.9 times and women a 5.3 times greater risk for conviction for assault and serious violence compared to a control group with other psychiatric diagnoses.
Wesseley SC, Castle D, Douglas AJ et. al. The criminal careers of incident cases of schizophrenia. Psychological Medicine 1994;24:483-502.

•    A Swedish study examined the criminal records of all individuals born in Stockholm in 1953 and still living there 30 years later. Men and women with a severe mental illness were 4.2 times (men) and 27.5 times (women) more likely to have been convicted of a violent crime compared to individuals with no psychiatric diagnosis.
Hodgins, S. Mental disorder, intellectual deficiency, and crime. Archives of General Psychiatry 1992;49:476–483.

•    In a follow-up of patients released from a psychiatric hospital, Dr. Henry Steadman et al reported that "27 percent of released male and female patients report at least one violent act within a means of four months after discharge."
Monahan J. Mental disorder and violent behavior. American Psychologist 1992;47:511–-521.

•    In a carefully controlled study comparing individuals with severe mental illness living in the community in New York with other community residents, the former group was found to be three times more likely to commit violent acts such as weapons use or "hurting someone badly." The sicker the individual, the more likely they were to have been violent.
Link BG, Andrews H, Cullen FT. The violent and illegal behavior of mental patients reconsidered. American Sociological Review 1992;57:275–292.

•    In reviewing many of these studies in 1992, Prof. John Monahan concluded: "The data that have recently become available, fairly read, suggest the one conclusion I did not want to reach: Whether the measure is the prevalence of violence among the disordered or the prevalence of disorder among the violent, whether the sample is people who are selected for treatment as inmates or patients in institutions or people randomly chosen from the open community, and no matter how many social and demographic factors are statistically taken into account, there appears to be a relationship between mental disorder and violent behavior."
Monahan J. Mental disorder and violent behavior. American Psychologist 1992;47:511–521.

•    A 1990 study investigated violent behavior among severely mentally ill individuals in 1,401 randomly selected families who were members of the National Alliance for the Mentally Ill (NAMI). In the preceding year, 11 percent of these individuals were reported to have physically harmed another person.
Steinwachs DM, Kasper JD, Skinner EA. Family Perspectives on Meeting the Needs for Care of Severely Mentally Ill Relatives: A nationa Survey (Arlington, Va.: National Allicance for the Mentally Ill, 1992).

•    The Epidemiological Catchment Area (ECA) surveys carried out 1980–1983 reported much higher rates of violent behavior among individuals with severe mental illness living in the community compared to other community residents. For example, individuals with schizophrenia were 21 times more likely to have used a weapon in a fight.
Swanson JW, Hozer CD, Ganju VK et. al. Violence and psychiatric disorder in the community: evidence from the Epidemiologic Catchment Area surveys. Hospital and Community Psychiatry 1990;41:761–770.

•    A Swedish study of 644 individuals with schizophrenia followed for 15 years reported that they committed violent offenses at a rate four times greater than the general population.
Lindqvist P, Allebeck P. Schizophrenia and crime: a longitudinal follow-up of 644 schizophrenics in Stockholm. British Journal of Psychiatry 1990;157:345–350.

•    In reviewing early studies on discharged psychiatric patients, Dr. Judith Rabkin concluded: "Arrest and conviction rates for the subcategory of violent crimes were found to exceed general population rates in every study in which they were measured."
Rabkin J. Criminal behavior of discharged mental patients: a critical appraisal of the research. Psychological Bulletin 1979;86:1–27.
II. Individuals with serious mental illness are responsible for 5 to 10 percent of all homicides
In the United States, there have been few studies bearing directly on this question:

•    Between 1970 and 1975 in Albany County, New York, a study was done on all 48 homicides committed there. Eight homicides (17 percent) were committed by individuals with schizophrenia. Most of them were not being treated at the time of the crime, leading the authors to conclude that “closer follow-ups of psychotic patients, especially schizophrenics, could do a lot to improve the welfare of the patient and community.”
Grunberg F, Klinger BI, Grumet B. Homicide and deinstitutionalization of the mentally ill. American Journal of Psychiatry 1977;134:685–687.
Grunberg F, Klinger BI, Grumet BR. Homicide and community-based psychiatry. Journal of Nervous and Mental Disease 1978;166:868–874.

•    A 1985 study reported that 10 percent (7 out of 71) of all homicides 1978–1980 in Contra Costa County in California were carried out by individuals diagnosed with schizophrenia. All had been evaluated psychiatrically prior to the crime and had refused medication.
Wilcox DE. The relationship of mental illness to homicide. American Journal of Forensic Psychiatry 1985;6:3–15.

•    A 1988 Department of Justice study reported that individuals with a history of mental illness (not including drug or alcohol abuse) were responsible for 4.3 percent of the homicides in the United States, or 897 out of 20,860. In instances in which the homicide occurred among family members, the percentage was much higher, e.g., in 25 percent of cases in which an individual killed his/her parent, that individual was mentally ill.
Dawson JM, Langan PA. Murder in families (Washington, D.C.: Bureau of Justice Statistics, U.S. Department of Justice, 1994).

•    In a more recent study in Indiana, researchers examined the records of 518 individuals in prison who had been convicted of homicide between 1990 and 2002. Among the 518, 53 (or 10.2 percent) had been diagnosed with schizophrenia (n=27), bipolar disorder (n=12), or other psychotic disorders not associated with drug abuse (n=14). An additional 42 individuals had been diagnosed with mania or major depressive disorder. It should be emphasized that the study included only those who had been sentenced to prison and did not include those individuals who had committed homicides and were subsequently found to be incompetent to stand trial or not guilty by reason of insanity and therefore sent to a psychiatric facility instead of prison. Thus, the 10.2 percent is probably an undercount. The authors themselves did not conclude that individuals with serious mental illnesses were responsible for 10 percent of the homicides, but given the data, that seems a reasonable conclusion. The authors also noted that 80 percent of the mentally ill individuals who committed homicides had received past psychiatric treatment but that “many of the offenders were not receiving treatment” at the time of the homicide.
Matejkowski JC, Cullen SW, Solomon PL. Characteristics of persons with severe mental illness who have been incarcerated for murder. Journal of the American Academy of Psychiatry and the Law 2008;36:74–86.
Studies from other countries are consistent with the 5–10 percent figure.

•    Large et al. identified 18 studies of homicide and schizophrenia and conducted a meta-analysis. They reported that 6.5 percent of the homicide offenders had a diagnosis of schizophrenia.
Large  M, Smith G, Nielssen O. The relationship between the rate of homicide by those with schizophrenia and the overall homicide rate: a systematic review and meta-analysis. Schizophrenia Research  2009;112:123–129.

•    A study of all 2,005 individuals convicted of homicide or attempted homicide in Sweden from 1988 to 2001 reported that 229, or 11 percent, had schizophrenia or bipolar disorder. Substance abuse and medication noncompliance were significant risk factors.
Fazel S, Grann M. Psychiatric morbidity among homicide offenders: a Swedish population study. American Journal of Psychiatry 2004;161:2129–2131.
Fazel S, Buxrud P, Ruchkin V et al. Homicide in discharged patients with schizophrenia and other psychoses: a national case-control study. Schizophrenia Research 2010;123:263–269.

•    In Singapore, 110 individuals were charged with murder between 1997 and 2001. Among these, 7 had schizophrenia, 1 had bipolar disorder, and 2 had delusional disorders. Thus, 10 out of 110 (9 percent) had psychotic disorders.
Koh KGWW, Gwee KP, Chan YH. Psychiatric aspects of homicide in Singpore: a five-year review (1997–2001). Singapore Medical Journal 2006;47:297–304.

•    In England and Wales, a clinical survey of 1,594 people convicted of homicide from 1996 to 1999 reported that 85 (5 percent) had schizophrenia.
Shaw J, Hunt IM, Flynn S et al. Rates of mental disorder in people convicted of homicide: national clinical survey. British Journal of Psychiatry 2005;188:143–147.

•    In Germany, a study of 290 individuals who committed or attempted to commit homicides between 1992 and 1996 reported that 29 (10 percent) had schizophrenia.
Erb M, Hodgins S, Freese R et al. Homicide and schizophrenia: maybe treatment does have a preventive effect. Criminal Behavior and Mental Health 2011;11:6–26.

•    A study of homicides in Finland reported that "the risk of committing a homicide was about 10 times greater for schizophrenia patients of both genders than it was for the general population." For men "schizophrenia without alcoholism increased the odds ratio more than 7 times; schizophrenia with coexisting alcoholism more than 17 times."
Eronen M, Tiihonen J, Hakola P. Schizophrenia and homicidal behavior. Schizophrenia Bulletin 1996;22:83–89.

•    In Denmark, 251 individuals accused of homicide were evaluated psychiatrically. Thirty-seven (15 percent) were diagnosed with psychosis related to schizophrenia or affective disorder.
Gottlieb P, Gabrielson G, Kramp P. Psychotic homicides in Copenhagen from 1959 to 1983. Acta Psychiatrica Scandinvaica 1987;76:285–292.
III. Most acts of violence committed by individuals with serious mental illness are carried out when they are not being treated

•    Meta-analyses of studies of individuals with serious mental illness who commit acts of violence, including homicides, report that a disproportionate number of these acts occur during the person’s first psychotic episode before they have been treated.
Large MM, Nielssen O. Violence in first-episode psychosis: a systematic review and meta-analysis. Schizophrenia Research 2010;125:208–220.
Nielsson O, Large M. Rates of homicide during the first episode of psychosis and after treatment: a systematic review and meta-analysis, Schizophrenia Bulletin 2010;36:702–712.

•    A study in New York assessed 60 severely mentally ill men who had been charged with violent crimes. The author reported that medication noncompliance and lack of awareness of illness both played significant roles in causing the men’s violent behavior.
Alia-Klein N, O’Rourke TM, Goldstein RZ et al. Insight into illness and adherence to psychotropic medications are separately associated with violence severity in a forensic sample. Aggressive Behavior 2007;33:86–96.

•    A study of 907 individuals with severe mental illness reported that those who were violent were “more likely to deny needing psychiatric treatment.” The authors concluded that “clinical interventions that address a patient’s perceived need for psychiatric treatment, such as compliance therapy and motivational interviewing, appear to hold promise as risk management strategies.”
Elbogen EB, Mustillo S, Van Dorn R et al. The impact of perceived need for treatment on risk of arrest and violence among people with severe mental illness. Criminal Justice and Behavior 2007;34:197–210.

•    A study of 1,011 outpatients with severe psychiatric disorders in five states reported that “community violence was inversely related to treatment adherence,” i.e., the less medication individuals took, the more likely they were to become violent.”
Elbogen EB, Van Dorn RA, Swanson JW et al. Treatment engagement and violence risk in mental disorders. British Journal of Psychiatry 2006;189:354–360.

•    A four-state (New Hampshire, Connecticut, Maryland, and North Carolina) study of 802 adults with severe mental illness (64 percent schizophrenia or schizoaffective disorder, 17 percent bipolar disorder) reported that 13.6 percent had been violent within the previous year. “Violent” was defined as “any physical fighting or assaultive actions causing bodily injury to another person, any use of lethal weapon to harm or threaten someone, or any sexual assault during that period.” Those who had been violent were more likely to have been homeless, to be substance abusers, and to be living in a violent environment. Those who had been violent were also 1.7 times more likely to have been noncompliant with medications.
Swanson JW, Swartz MS, Essock SM et al. The social-environmental context of violent behavior in persons treated for severe mental illness. American Journal of Public Health 2002;92:1523–1531.

•    In the three-site MacArthur Foundation Study of violence and mental illness, 17.4 percent of the patients were violent in the 10-week period prior to hospitalization, during which time they were not being treated, compared to an average of 8.9 percent for the five 10-week periods after hospitalization during which most of them were being treated.
Steadman HJ, Mulvey EP, Monahan J et. al. Violence by people discharged from acute psychiatric inpatient facilities and by others in the same neighborhoods. Archives of General Psychiatry 1998;55:393–401.

•    An English study of 1,015 forensic patients with severe mental illness ("functional psychosis") reported that the diagnosis of "schizophrenia was most strongly associated with personal violence" and that "more than 75 percent of those with a psychosis were recorded as being driven to offend by their delusions." The authors concluded that "treatment appears as important for public safety as for personal health."
Taylor PJ, Leese M, Williams D et. al. Mental disorder and violence. British Journal of Psychiatry 172:218–226, 1998.

•    A study of 348 inpatients in a Virginia state psychiatric hospital found that patients who refused to take medication "were more likely to be assaultive, were more likely to require seclusion and restraint, and had longer hospitalizations."
Kasper JA, Hoge SK, Feucht-Haviar T et. al. Prospective study of patients’ refusal of antipsychotic medication under a physician discretion review procedure. American Journal of Psychiatry 1997;154:483–489.

•    In an analysis of data from the ECA study, the authors noted that "mentally ill individuals with no treatment contact in the past six months had significantly higher odds of violence in the long term” and that “moderate levels of agitation and psychoticism increase the risk of violence." They then conclude: "This would seem to provide a strong argument for providing more interventions targeted specifically to persons with combined mental illness and addictive disorders who are likely not to comply voluntarily with conventional outpatient therapies."
Swanson J, Estroff S, Swartz M et. al. Violence and severe mental disorder in clinical and community populations: the effects of psychotic symptoms, comorbidity, and lack of treatment. Psychiatry 1997;60:1–22.

•    A study of 133 outpatients with schizophrenia showed that "13 percent of the study group were characteristically violent." Having inadequately treated symptoms of delusions and hallucinations was one of the predictions of violent behavior. Specifically, "71 percent of the violent patients . . . had problems with medication compliance, compared with only 17 percent of those without hostile behaviors," a difference that was statistically highly significant (p 0.001).
Bartels J, Drake RE, Wallach MA et. al. Characteristic hostility in schizophrenic outpatients. Schizophrenia Bulletin 1991;17:163–171.

•    A study of severely mentally ill patients in a state forensic hospital found a highly significant correlation (p 0.001) between failure to take medication and a history of violent acts in the community.
Smith LD. Medication refusal and the rehospitalized mentally ill inmate. Hospital and Community Psychiatry 1989 40:491–496.

•    A study of inpatients diagnosed with schizophrenia reported an inverse correlation between their propensity to violence and their blood level of antipsychotic medication.
Yesavage JA. Inpatient violence and the schizophrenic patient: an inverse correlation between danger-related events and neuroleptic levels. Biological Psychiatry 1982;17:1331–1337.


Visit Your State