VIOLENCE RISK ASSESSMENT
From the Forensic Psychiatry Review Course, presented by the American Academy of Psychiatry and the Law
Phillip J. Resnick, MD, Professor of Psychiatry, Case Western Reserve University School of Medicine, Cleveland, OH
| Introduction: violence peaks in late teens and early 20s; in general population, 10 times more common in men than
in women, but in severely mentally ill population, incidence equal (common to underestimate risk potential in
women); mild mental retardation increases risk for violence 5-fold for men and 22-fold for women; in Epidemiologic
Catchment Area survey, 2% admitted to having committed violent act in past year, but those with major mental
disorder (eg, schizophrenia, bipolar disorder) were 5 times more likely to be violent than those with no mental
disorder; alcohol abusers twice as likely as people with schizophrenia to be violent, and abusers of drugs other than
alcohol or marijuana three times as likely; among those with schizophrenia, those who abuse substances 17 times
more likely to be violent than those in general population, but those with schizophrenia who do not abuse substances
only 3 times more likely
|
| Components of violence: law more concerned with harm done to people (as opposed to property) and with physical
(as opposed to psychologic) harm; other factors include magnitude, likelihood, imminence, and frequency; in civil
commitment statutes, all states have criterion of “a substantial risk of physical harm to others,” but few have criterion
for psychologic harm; “substantial risk” defined as “strong possibility” of harm (in eyes of law, possibility
equals <50%; probability equals ≥50%); magnitude and likelihood combined to determine whether to commit involuntarily;
ie, if magnitude high (eg, murder) but likelihood low (eg, 2%), individual more likely to be committed
than if magnitude low (eg, slapping someone’s face) and likelihood high (eg, 10%); intervention more justified if
risk imminent (eg, in next 6 hr as opposed to next 6 mo); frequency also consideration (some harms are high frequency
but low magnitude, eg, exhibitionism)
|
| Long-term prediction of risk: “we are not very skillful in the long-term prediction of violence we do better in
short-term prediction”; prediction difficult because many variables involved
|
| Standardized instruments for assessing violence: Psychopathy Checklist (PCL) has high correlation with future violence
in prisoners and in people in forensic hospitals; Violence Risk Appraisal Guide (VRAG) accurate in Canada,
but norms not established for United States; Historical and Clinical Risk Management 20-item scale (HCR-20)
“has proven useful”; Classification of Violence Risk (COVR) so complex it requires special software to use and interpret
|
| Actuarial vs clinical risk assessment: actuarial risk assessment more accurate than clinical assessment; however,
actuarial assessment depends on unmodifiable historical risk factors, such as age and sex, so clinical judgment important
in assessing patient who is acutely psychotic; actuarial risk assessment plays important role in determining
whether to release patient or prisoner
|
| Violence in psychosis: the more positive symptoms patient has (eg, hallucinations, delusions), the more likely he or
she is to be violent, and the more negative symptoms patient has (eg, blunted affect, concreteness, lack of motivation),
the less likely he or she is to be violent; most dangerous would be high positive symptoms and low negative
symptoms
|
 | Paranoia: in outpatient setting, paranoid psychoses more likely to result in violence than any other type of psychosis
(person with paranoid psychosis has access to weapons, is organized, and can plan violence against perceived
persecutor); of all people found not guilty of murder by reason of insanity, 80% diagnosed with comorbid paranoid
psychosis; once admitted to hospital, person with paranoid psychosis no longer has access to weapons or
target, and is not as dangerous as person with disorganized psychosis
|
| Hallucinations: overall less ominous than delusions; hallucinations most likely to lead to violence are those that
evoke negative emotions, such as anger and anxiety; command hallucinations—instruction to do something,
which may be benign (“open a window”) or malevolent (“harm someone”); most common command hallucination
is to harm oneself; studies confirm that people with command hallucinations more likely to act violently;
person more likely to act on hallucination-related delusion; of those with auditory hallucinations, 60% recognize
voice and are more likely to carry out command than those who do not recognize voice; dangerous commands
less likely to be executed than nondangerous commands, unless person has hallucination-related delusion and
recognizes voice
|
| Delusions: much more ominous than hallucinations, especially threat and control-override delusions; study showed
that people who think their minds are being controlled or that thoughts are being put into their heads much more
likely to act violently on delusion; belief that one is dead or that one’s thoughts are being broadcast has no correlation
with likelihood of violence, whereas feelings of paranoia or of being out of control are associated with violence;
persecutory delusions more likely to lead to violence, more likely to be acted upon, and have highest
magnitude of harm; delusion of being poisoned associated with high incidence of violence; anger and fear precede
all violent acts, whether by psychotic or nonpsychotic people
|
| Murder/suicide by parent: when mother who loves her children decides to take her own life, she may believe her
children will be better off dead and in heaven; when woman with young children expresses suicidal ideation, ask
about her plans for her children (study shows 41% of depressed mothers with children <3 yr of age have had
thoughts of seriously harming or killing their children)
|
| Personality traits associated with violence: those associated with antisocial and borderline personality disorders;
eg, impulsivity, low tolerance for frustration and criticism, recklessness, being self-centered, projecting blame onto
others
|
| Childhood antecedents of adult violence: parental brutality (no matter age of child; boys more likely to respond to
being brutalized by becoming aggressive, girls more likely to replicate victimization); juvenile delinquency, which
may begin at any age, risk factor for violence; childhood fire setting and cruelty to animals predictive of adult violence
|
| Taking history of violence: past violence best single predictor of future violence; inquire about patterns of violence
(eg, why did it occur? who said what to set it off? was perpetrator intoxicated or on drugs? how severe was injury?);
interview family as well as patient; look at past records
|
| Patterns of violence: psychotic violence—some people exhibit violent behavior only when psychotic (people with
schizophrenia who are not acutely psychotic no more violent than people without schizophrenia); insult-evoked
violence—some people have code of “death before dishonor”; ego-syntonic violence—comfortable using violence
to resolve disputes; ego-dystonic violence—remorseful after committing violent act; affective violence—patterned
activation of autonomic nervous system that includes certain postural changes such as clenching fists, tightening
jaw muscles, staring at prey, and standing with feet apart for balance; “the person has a grievance; they get the
idea; they attack”; predatory violence—“there’s a grievance; the person gets the idea; they do research, planning,
and preparation before the attack”
|
| Other factors to consider: mood-altering substances—alcohol disinhibiting (of people arrested for violence, 30%
have blood alcohol level that exceeds legal limit, and 30% have some other illicit substance in blood); stimulants
predispose to violence because they are disinhibiting and tend to cause grandiosity and paranoia; access to
weapons—widely available in United States (40% of American households have firearms, and 43% keep them
loaded); owning “a whole arsenal” ominous; ominous if paranoid individual has moved firearm recently, especially
if it has been moved to his or her bed (“that person belongs in a hospital; if they can’t sleep without a loaded gun
within arm’s reach, the possibility of them shooting someone, a misperceived persecutor, is very elevated”); personal
history—age at first criminal arrest highly correlated with lifetime of criminality (younger age at first arrest
more likely to lead to life of crime); Uniform Code of Military Justice, “article XV” indicates punitive action by
commanding officer without court martial; ask about court martial and article XV when determining whether veteran
has history of violence; when asking about history of sexual violence, “it’s particularly important that you
don’t show disgust on your face” if interviewee’s sexual preferences not same as interviewer’s; usual exploratory
sexual behavior between peers not concerning, but very concerning if it involved child and adult (study showed
76% of serial rapists exposed to childhood sexual misconduct by adult)
|
| Assessing dangerousness: anger in absence of empathy dangerous because empathy “helps put on the brakes”; people
who lack empathy more likely to act more immediately and more violently in response to small irritation; in 5
min preceding inpatient assault, most commonly seen behaviors include yelling, swearing, and standing uncomfortably
close; food universal symbol of hospitality, and offering food, even snack from vending machine, may
help deescalate violence; all threats should be taken seriously, whether made in person or on telephone; when patient
makes threat, aim questioning at helping him or her to foresee and acknowledge consequences of threatened
behavior; explore affect thoroughly, and try to help patient find alternative behaviors that are not self-destructive;
specific persecutor—when evaluating paranoid psychotic’s threat, avoid generic questions and focus on patient’s
specific perceived persecutor (eg, patient may deny homicidality, but if asked how he or she would react to specific
perceived persecutor, he or she may respond that there is no alternative but to kill persecutor); perceived
intentionality—how does patient respond to accidental contact (eg, stranger accidentally brushing against him or
her)? research shows that childhood bullies and adult aggressive people more likely than others to respond to ambiguous
stimulus and attribute intentionality to it, leading to physical encounter; victim’s role—victim may respond
to threat or assault in many ways; defiance or counter-provocation on part of victim may escalate violence of attack;
needs to be taken into consideration when deciding whether perpetrator should return to that setting
|
| Consultation: patient’s talking about killing someone can stir strong feelings of countertransference in therapist,
whose first impulse might be to end therapeutic relationship; likewise, therapist awakened in middle of night might
have similar thoughts; in these situations, better to obtain consultation; research shows that when 2 people discuss
risk assessment, conclusion more likely to be correct (also, therapist less likely to be sued successfully if he or she
has consultant)
|
| Dynamic and static risk factors: dynamic risk factors—subject to change by intervention; include such things as
where patient lives, his or her access to weapons, and whether patient psychotic; static risk factors—cannot be
changed; examples include patient’s age, sex, and history of violence; develop systematic violence-prevention plan
by examining each dynamic risk factor and determining what intervention can be done; therapist has duty to protect
foreseeable victims from patient’s violence
|
| Case example and discussion
|
| Case: man (veteran) hospitalized after hearing voice tell him to kill his supervisor; veteran’s brother intercepted
him while leaving home with gun, and took him to hospital; impulse to kill supervisor continued for 4 to 5 days,
then subsided; patient has history of physical fighting
|
| Discussion: most significant risk factors patient’s history of violence and his being undeterred by expectation of dying
at hands of police; other risk factors include aborted homicide attempt, misjudgment of intended victim’s reaction,
expectation of further provocation from intended victim, and patient’s access to guns; history of fight in hospital
setting and impulsivity not optimistic; static risk factors include age, sex, and veteran status (“every veteran has
been trained to kill people every veteran has training with weapons; therefore, his likelihood of succeeding is
better than if he didn’t have that training”); violence ego-syntonic; patient does not have empathy for his supervisor
|
| Interventions to reduce violence: inform supervisor, but not by letter and preferably not by telephone call; best if
supervisor visits therapist’s office, perhaps with patient also present (obtain consent of both before putting them
together); hear story from supervisor’s point of view, and try to determine if he has been “riding” patient, and if
so, why; consider inviting union representative or other person to empower patient (advantage is that if patient
does return to work, he will have advocate who knows background of conflict and suggested resolution; this
might encourage patient to find alternative to violence next time conflict occurs); other interventions include anger
management, treating patient’s alcohol use, removing guns from patient’s home, and reducing patient’s job
isolation
|
| Conclusion: speaker suggests having systematic violence-reduction plan in which risk factors can be determined to
be static or dynamic and interventions can be planned for all dynamic risk factors
|
Suggested Reading
Buckley PF et al: Insight and its relationship to violent behavior in patients with schizophrenia. Am J Psychiatry
161:1712, 2004; Carroll A: Are violence risk assessment tools clinically useful? Aust N Z J Psychiatry 41:301, 2007;
Doyle M, Dolan M: Predicting community violence from patients discharged from mental health services. Br J Psychiatry
189:520, 2006; Evans C et al: Intrusive memories and ruminations related to violent crime among young offenders:
phenomenological characteristics. J Trauma Stress 20:183, 2007; Friedman SH et al: Child murder
committed by severely mentally ill mothers: an examination of mothers found not guilty by reason of insanity. 2005
Honorable Mention/Richard Rosner Award for the best paper by a fellow in forensic psychiatry or forensic psychology.
J Forensic Sci 50:1466, 2005; Harris GT et al: Applying a forensic actuarial assessment (the Violence Risk Appraisal
Guide) to nonforensic patients. J Interpers Violence 19:1063, 2004; Hart SD et al: Precision of actuarial risk
assessment instruments: Evaluating the margins ‘of error’ of group v. individual predictions of violence. Br J Psychiatry
190:s60, 2007; Hatters Friedman S et al: Filicide-suicide: common factors in parents who kill their children
and themselves. J Am Acad Psychiatry Law 33:496, 2005; Kling R et al: Use of a violence risk assessment tool in an
acute care hospital: effectiveness in identifying violent patients. AAOHN J 54:481, 2006; Monahan J et al: An actuarial
model of violence risk assessment for persons with mental disorders. Psychiatr Serv 56:810, 2005; Monahan J
et al: The classification of violence risk. Behav Sci Law 24:721, 2006; Ogloff JR, Daffern M: The dynamic appraisal
of situational aggression: an instrument to assess risk for imminent aggression in psychiatric inpatients. Behav Sci
Law 24:799, 2006; Paterson B: Developing a perspective on restraint and the least intrusive intervention. Br J Nurs
15:1235, 2006-2007; Robbins PC et al: Mental disorder, violence, and gender. Law Hum Behav 27:561, 2003;
Skeem JL, Mulvey EP: Psychopathy and community violence among civil psychiatric patients: results from the
MacArthur Violence Risk Assessment Study. J Consult Clin Psychol 69:358, 2001; Steadman HJ et al: A classification
tree approach to the development of actuarial violence risk assessment tools. Law Hum Behav 24:83, 2000; Teasdale
B et al: Gender, threat/control-override delusions and violence. Law Hum Behav 30:649, 2006.
Cultural and Linguistic Resources
In compliance with California Assembly Bill 1195, Audio-Digest Foundation offers selected cultural and linguistic
resources on its website. Please visit this site: www.audiodigest.org/CLCresources.
Educational Objectives
| The goal of this program is to improve the listener’s ability to assess the risk of violent behavior by mental health patients.
After hearing and assimilating this program, the clinician will be better able to:
|
| 1. Describe the components of violence that make a patient eligible for involuntary commitment.
|
| 2. State the differences between actuarial and clinical risk assessment, and explain the circumstances under
which each is most applicable.
|
| 3. Compare and contrast the roles paranoia, hallucinations, and delusions play in violence perpetrated by those
with psychosis.
|
| 4. Take a history that will more accurately assess a patient’s risk of behaving violently in the present situation.
|
| 5. Determine the degree of danger associated with any given threat made by a mentally ill person.
|
Faculty Disclosure
In adherence to ACCME Standards for Commercial Support, the Audio-Digest Foundation requires all faculty members
to disclose relevant financial relationships within the past 12 months that might create any personal conflict of
interest. Any identified conflicts were resolved to ensure that this educational activity promotes quality in health care
education and not a proprietary business or commercial interest. For this program, Dr. Resnick indicated nothing to
disclose.
Acknowledgements
Dr. Resnick was recorded at The Forensic Psychiatry Review Course, held October 23-25, 2006, in Chicago, IL, and sponsored
by the American Academy of Psychiatry and the Law. The Audio-Digest Foundation thanks Dr. Resnick and the
Academy for their cooperation in the production of this program.
|
