Stephen Noffsinger, MD

Mr. F, age 35, is homeless and has a history of cocaine and alcohol use disorders. He is admitted voluntarily to the psychiatric unit because he has homicidal thoughts toward Ms. S, who works in the shelter where he has been staying. Mr. F reports that he is thinking of killing Ms. S if he is discharged because she has been rude to him. He states that he has access to several firearms, but he will not disclose the location. He has been diagnosed with unspecified depressive disorder and exhibited antisocial personality disorder traits. He is being treated with sertraline. However, his mood appears to be relatively stable, except for occasional angry verbal outbursts. The outbursts have been related to intrusive peers or staff turning the television off for group meetings. Mr. F has been joking with peers, eating well, and sleeping appropriately. He reports no suicidal thoughts and has not been physically violent on the unit. However, Mr. F has had a history of violence since his teenage years. He has been incarcerated twice for assault and once for drug possession.

How would you approach assessing and managing Mr. F’s risk for violence?

We all have encountered a patient similar to Mr. F on the psychiatric unit or in the emergency department—a patient who makes violent threats and appears angry, intimidating, manipulative, and/or demanding, despite exhibiting no evidence of mania or psychosis. This patient often has a history of substance abuse and a lifelong pattern of viewing violence as an acceptable way of addressing life’s problems. Many psychiatrists suspect that more time on the inpatient unit is unlikely to reduce this patient’s risk of violence. Why? Because the violence risk does not stem from a treatable mental illness. Further, psychiatrists may be apprehensive about this patient’s potential for violence after discharge and their liability in the event of a bad outcome. No one wants their name associated with a headline that reads “Psychiatrist discharged man less than 24 hours before he killed 3 people.”

The purported relationship between mental illness and violence often is sensationalized in the media. However, research reveals that the vast majority of violence is in fact not due to symptoms of mental illness.^1,2 A common clinical challenge in psychiatry involves evaluating individuals at elevated risk of violence and determining how to address their risk factors for violence. When the risk is primarily due to psychosis and can be reduced with antipsychotic medication, the job is easy. But how should we proceed when the risk stems from factors other than mental illness?

This article reviews risk factors for violence, discusses targeted violence against a specific victim, and offers practical tips for assessing and managing risk, particularly when the risk for violence is not due to mental illness.

Violence and mental illness: A tenuous link

Violence is a major public health concern in the United States. Although in recent years the rates of homicide and aggravated assault have decreased dramatically, there are approximately 16,000 homicides annually in the United States, and more than 1.6 million injuries from assaults treated in emergency departments each year.³ Homicide continues to be one of the leading causes of death among teenagers and young adults.⁴

The most effective methods of preventing widespread violence are public health approaches, such as parent- and family-focused programs, early childhood education, programs in school, and public policy changes.³ However, as psychiatrists, we are routinely asked to assess the risk of violence for an individual patient and devise strategies to mitigate violence risk.

Continue to: Although certain mental illnesses...

Although certain mental illnesses increase the relative risk of violence (compared with people without mental illness),^5,6 recent studies suggest that mental illness plays only a “minor role in explaining violence in populations.”⁷ It is estimated that as little as 4% of the violence in the United States can be attributed to mental illness.¹ According to a 1998 meta-analysis of 48 studies of criminal recidivism, the risk factors for violent recidivism were “almost identical” among offenders who had a mental disorder and those who did not.⁸

Approaches to assessing violence risk

Psychiatrists can assess the risk of future violence via 3 broad approaches.^9,10

Unaided clinical judgment is when a mental health professional estimates violence risk based on his or her own experience and intuition, with knowledge of violence risk factors, but without the use of structured tools.

Actuarial tools are statistical models that use formulae to show relationships between data (risk factors) and outcomes (violence).^10,11

Continue to: Structured professional judgment

Structured professional judgment is a hybrid of unaided clinical judgment and actuarial methods. Structured professional judgment tools help the evaluator identify empirically established risk factors. Once the information is collected, it is combined with clinical judgment in decision making.^9,10 There are now more than 200 structured tools available for assessing violence risk in criminal justice and forensic mental health populations.¹²

Clinical judgment, although commonly used in practice, is less accurate than actuarial tools or structured professional judgment.^10,11 In general, risk assessment tools offer moderate levels of accuracy in categorizing people at low risk vs high risk.^5,13 The tools have better ability to accurately categorize individuals at low risk, compared with high risk, where false positives are common.^12,14

Two types of risk factors

Risk factors for violence are commonly categorized as static or dynamic factors. Static factors are historical factors that cannot be changed with intervention (eg, age, sex, history of abuse). Dynamic factors can be changed with intervention (eg, substance abuse).¹⁵

Static risk factors. The best predictor of future violence is past violent behavior.^5,16,17 Violence risk increases with each prior episode of violence.⁵ Prior arrests for any crime, especially if the individual was a juvenile at the time of arrest for his or her first violent offense, increase future violence risk.⁵ Other important static violence risk factors include demographic factors such as age, sex, and socioeconomic status. Swanson et al⁶ reviewed a large pool of data (approximately 10,000 respondents) from the Epidemiologic Catchment Area survey. Being young, male, and of low socioeconomic status were all associated with violence in the community.⁶ The highest-risk age group for violence is age 15 to 24.⁵ Males perpetrate violence in the community at a rate 10 times that of females.¹⁸ However, among individuals with severe mental illness, men and women have similar rates of violence.^19,20 Unstable employment,²¹ less education,²² low intelligence,¹⁶ and a history of a significant head injury⁵ also are risk factors for violence.⁵

Continue to: Being abused as a child...

Being abused as a child, witnessing violence in the home,^5,16 and growing up with an unstable parental situation (eg, parental loss or separation) has been linked to violence.^16,23,24 Early disruptive behavior in childhood (eg, fighting, lying and stealing, truancy, and school problems) increases violence risk.^21,23

Personality factors are important static risk factors for violence. Antisocial personality disorder is the most common personality disorder linked with violence.¹⁷ Several studies consistently show psychopathy to be a strong predictor of both violence and criminal behavior.^5,25 A psychopath is a person who lacks empathy and close relationships, behaves impulsively, has superficially charming qualities, and is primarily interested in self-gratification.²⁶ Harris et al²⁷ studied 169 released forensic patients and found that 77% of the psychopaths (according to Psychopathy Checklist-Revised [PCL-R] scores) violently recidivated. In contrast, only 21% of the non-psychopaths violently recidivated.²⁷

Other personality factors associated with violence include a predisposition toward feelings of anger and hatred (as opposed to empathy, anxiety, or guilt, which may reduce risk), hostile attributional biases (a tendency to interpret benign behavior of others as intentionally antagonistic), violent fantasies, poor anger control, and impulsivity.⁵ Although personality factors tend to be longstanding and more difficult to modify, in the outpatient setting, therapeutic efforts can be made to modify hostile attribution biases, poor anger control, and impulsive behavior.

Dynamic risk factors. Substance abuse is strongly associated with violence.^6,17 The prevalence of violence is 12 times greater among individuals with alcohol use disorder and 16 times greater among individuals with other substance use disorders, compared with those with no such diagnoses.^5,6

Continue to: Steadman et al...

Steadman et al²⁸ compared 1,136 adult patients with mental disorders discharged from psychiatric hospitals with 519 individuals living in the same neighborhoods as the hospitalized patients. They found that the prevalence of violence among discharged patients without substance abuse was “statistically indistinguishable” from the prevalence of violence among community members, in the same neighborhood, who did not have symptoms of substance abuse.²⁸ Swanson et al⁶ found that the combination of a mental disorder plus an alcohol or substance use disorder substantially increased the risk of violence.

Other dynamic risk factors for violence include mental illness symptoms such as psychosis, especially threat/control-override delusions, where the individual believes that they are being threatened or controlled by an external force.¹⁷

Contextual factors to consider in violence risk assessments include current stressors, lack of social support, availability of weapons, access to drugs and alcohol, and the presence of similar circumstances that led to violent behavior in the past.⁵

How to assess the risk of targeted violence

Targeted violence is a predatory act of violence intentionally committed against a preselected person, group of people, or place.²⁹ Due to the low base rates of these incidents, targeted violence is difficult to study.^7,30 These risk assessments require a more specialized approach.

Continue to: In their 1999 article...

In their 1999 article, Borum et al³⁰ discussed threat assessment strategies utilized by the U.S. Secret Service and recommended investigating “pathways of ideas and behaviors that may lead to violent action.” Borum et al³⁰ summarized 3 fundamental principles of threat assessment (Table 1³⁰).

What to do when violence risk is not due to mental illness

Based on the information in Mr. F’s case scenario, it is likely that his homicidal ideation is not due to mental illness. Despite this, several risk factors for violence are present. Where do we go from here?

Scott and Resnick¹⁷ recommend considering the concept of dangerousness as 5 components (Table 2¹⁷). When this model of dangerousness is applied to Mr. F’s case, one can see that the magnitude of the harm is great because of threatened homicide. With regard to the imminence of the harm, it would help to clarify whether Mr. F plans to kill Ms. S immediately after discharge, or sometime in the next few months. Is his threat contingent on further provocations by Ms. S? Alternatively, does he intend to kill her for past grievances, regardless of further perceived insults?

Next, the frequency of a behavior relates to how often Mr. F has been aggressive in the past. The severity of his past aggression is also important. What is the most violent act he has ever done? Situational factors in this case include Mr. F’s access to weapons, financial problems, housing problems, and access to drugs and alcohol.¹⁷ Mr. F should be asked about what situations previously provoked his violent behavior. Consider how similar the present conditions are to past conditions to which Mr. F responded violently.⁵ The likelihood that a homicide will occur should take into account Mr. F’s risk factors for violence, as well as the seriousness of his intent to cause harm.

Continue to: Consider using a structured tool...

Consider using a structured tool, such as the Classification of Violence Risk, to help identify Mr. F’s risk factors for violence, or some other formal method to ensure that the proper data are collected. Violence risk assessments are more accurate when structured risk assessment tools are used, compared with clinical judgment alone.

It is important to review collateral sources of information. In Mr. F’s case, useful collateral sources may include his criminal docket (usually available online), past medical records, information from the shelter where he lives, and, potentially, friends or family.

Because Mr. F is making threats of targeted violence, be sure to ask about attack-related behaviors (Table 1³⁰).

Regarding the seriousness of Mr. F’s intent to cause harm, it may be helpful to ask him the following questions:

1. How likely are you to carry out this act of violence?
2. Do you have a plan? Have you taken any steps toward this plan?
3. Do you see other, nonviolent solutions to this problem?
4. What do you hope that we can do for you to help with this problem?

Continue to: Mr. F's answers...

Mr. F’s answers may suggest the possibility of a hidden agenda. Some patients express homicidal thoughts in order to stay in the hospital. If Mr. F expresses threats that are contingent on discharge and declines to engage in problem-solving discussions, this would cast doubt on the genuineness of his threat. However, doubt about the genuineness of the threat alone is not sufficient to simply discharge Mr. F. Assessment of his intent needs to be considered with other relevant risk factors, risk reduction strategies, and any Tarasoff duties that may apply.

In addition to risk factors, consider mitigating factors. For example, does Mr. F express concern over prison time as a reason to not engage in violence? It would be more ominous if Mr. F says that he does not care if he goes to prison because life is lousy being homeless and unemployed. At this point, an estimation can be made regarding whether Mr. F is a low-, moderate-, or high-risk of violence.

The next step is to organize Mr. F’s risk factors into static (historical) and dynamic (subject to intervention) factors. This will be helpful in formulating a strategy to manage risk because continued hospitalization can only address dynamic risk factors. Often in these cases, the static risk factors are far more numerous than the dynamic risk factors.

Once the data are collected and organized, the final step is to devise a risk management strategy. Some interventions, such as substance use treatment, will be straightforward. A mood-stabilizing medication could be considered, if clinically appropriate, to help reduce aggression and irritability.³¹ Efforts should be made to eliminate Mr. F’s access to firearms; however, in this case, it sounds unlikely that he will cooperate with those efforts. Ultimately, you may find yourself with a list of risk factors that are unlikely to be altered with further hospitalization, particularly if Mr. F’s homicidal thoughts and intent are due to antisocial personality traits.

Continue to: In that case...

In that case, the most important step will be to carry out your duty to warn/protect others prior to Mr. F’s discharge. Most states either require or permit mental health professionals to take reasonable steps to protect victims from violence when certain conditions are present, such as an explicit threat or identifiable victim (see Related Resources).

Once dynamic risk factors have been addressed, and duty to warn/protect is carried out, if there is no further clinical indication for hospitalization, it would be appropriate to discharge Mr. F. Continued homicidal threats stemming from antisocial personality traits, in the absence of a treatable mental illness (or other modifiable risk factors for violence that can be actively addressed), is not a reason for continued hospitalization. It may be useful to obtain a second opinion from a colleague in such scenarios. A second opinion may offer additional risk management ideas. In the event of a bad outcome, this will also help to show that the decision to discharge the patient was not taken lightly.

The psychiatrist should document a thoughtful risk assessment, the strategies that were implemented to reduce risk, the details of the warning, and the reasoning why continued hospitalization was not indicated (Table 3).

CASE CONTINUED

Decision to discharge

In Mr. F’s case, the treating psychiatrist determined that Mr. F’s risk of violence toward Ms. S was moderate. The psychiatrist identified several static risk factors for violence that raised Mr. F’s risk, but also noted that Mr. F’s threats were likely a manipulative effort to prolong his hospital stay. The psychiatrist carried out his duty to protect by notifying police and Ms. S of the nature of the threat prior to Mr. F’s discharge. The unit social worker helped Mr. F schedule an intake appointment for a substance use disorder treatment facility. Mr. F ultimately stated that he no longer experienced homicidal ideas once a bed was secured for him in a substance use treatment program. The psychiatrist carefully documented Mr. F’s risk assessment and the reasons why Mr. F’s risk would not be significantly altered by further inpatient hospitalization. Mr. F was discharged, and Ms. S remained unharmed.

Continue to: Bottom Line

Use a structured approach to identify risk factors for violence. Address dynamic risk factors, including access to weapons. Carry out the duty to warn/protect if applicable. Document your decisions and actions carefully, and then discharge the patient if clinically indicated. Do not be “held hostage” by a patient’s homicidal ideation.

Related Resources

• Dolan M, Doyle M. Violence risk prediction. Clinical and actuarial measures and the role of the psychopathy checklist. Br J Psychiatry. 2000;177:303-311.
• Douglas KS, Hart SD, Webster CD, et al. HCR-20V3: Assessing risk of violence–user guide. Burnaby, Canada: Mental Health, Law, and Policy Institute, Simon Fraser University; 2013.
• National Conference of State Legislatures. Mental health professionals’ duty to warn. http://www.ncsl.org/research/health/mental-health-professionals-duty-to-warn.aspx. Published September 28, 2015.

Drug Brand Names

Sertraline • Zoloft

Mr. F, age 35, is homeless and has a history of cocaine and alcohol use disorders. He is admitted voluntarily to the psychiatric unit because he has homicidal thoughts toward Ms. S, who works in the shelter where he has been staying. Mr. F reports that he is thinking of killing Ms. S if he is discharged because she has been rude to him. He states that he has access to several firearms, but he will not disclose the location. He has been diagnosed with unspecified depressive disorder and exhibited antisocial personality disorder traits. He is being treated with sertraline. However, his mood appears to be relatively stable, except for occasional angry verbal outbursts. The outbursts have been related to intrusive peers or staff turning the television off for group meetings. Mr. F has been joking with peers, eating well, and sleeping appropriately. He reports no suicidal thoughts and has not been physically violent on the unit. However, Mr. F has had a history of violence since his teenage years. He has been incarcerated twice for assault and once for drug possession.

How would you approach assessing and managing Mr. F’s risk for violence?

We all have encountered a patient similar to Mr. F on the psychiatric unit or in the emergency department—a patient who makes violent threats and appears angry, intimidating, manipulative, and/or demanding, despite exhibiting no evidence of mania or psychosis. This patient often has a history of substance abuse and a lifelong pattern of viewing violence as an acceptable way of addressing life’s problems. Many psychiatrists suspect that more time on the inpatient unit is unlikely to reduce this patient’s risk of violence. Why? Because the violence risk does not stem from a treatable mental illness. Further, psychiatrists may be apprehensive about this patient’s potential for violence after discharge and their liability in the event of a bad outcome. No one wants their name associated with a headline that reads “Psychiatrist discharged man less than 24 hours before he killed 3 people.”

The purported relationship between mental illness and violence often is sensationalized in the media. However, research reveals that the vast majority of violence is in fact not due to symptoms of mental illness.^1,2 A common clinical challenge in psychiatry involves evaluating individuals at elevated risk of violence and determining how to address their risk factors for violence. When the risk is primarily due to psychosis and can be reduced with antipsychotic medication, the job is easy. But how should we proceed when the risk stems from factors other than mental illness?

This article reviews risk factors for violence, discusses targeted violence against a specific victim, and offers practical tips for assessing and managing risk, particularly when the risk for violence is not due to mental illness.

Violence and mental illness: A tenuous link

Violence is a major public health concern in the United States. Although in recent years the rates of homicide and aggravated assault have decreased dramatically, there are approximately 16,000 homicides annually in the United States, and more than 1.6 million injuries from assaults treated in emergency departments each year.³ Homicide continues to be one of the leading causes of death among teenagers and young adults.⁴

The most effective methods of preventing widespread violence are public health approaches, such as parent- and family-focused programs, early childhood education, programs in school, and public policy changes.³ However, as psychiatrists, we are routinely asked to assess the risk of violence for an individual patient and devise strategies to mitigate violence risk.

Continue to: Although certain mental illnesses...

Although certain mental illnesses increase the relative risk of violence (compared with people without mental illness),^5,6 recent studies suggest that mental illness plays only a “minor role in explaining violence in populations.”⁷ It is estimated that as little as 4% of the violence in the United States can be attributed to mental illness.¹ According to a 1998 meta-analysis of 48 studies of criminal recidivism, the risk factors for violent recidivism were “almost identical” among offenders who had a mental disorder and those who did not.⁸

Approaches to assessing violence risk

Psychiatrists can assess the risk of future violence via 3 broad approaches.^9,10

Unaided clinical judgment is when a mental health professional estimates violence risk based on his or her own experience and intuition, with knowledge of violence risk factors, but without the use of structured tools.

Actuarial tools are statistical models that use formulae to show relationships between data (risk factors) and outcomes (violence).^10,11

Continue to: Structured professional judgment

Structured professional judgment is a hybrid of unaided clinical judgment and actuarial methods. Structured professional judgment tools help the evaluator identify empirically established risk factors. Once the information is collected, it is combined with clinical judgment in decision making.^9,10 There are now more than 200 structured tools available for assessing violence risk in criminal justice and forensic mental health populations.¹²

Clinical judgment, although commonly used in practice, is less accurate than actuarial tools or structured professional judgment.^10,11 In general, risk assessment tools offer moderate levels of accuracy in categorizing people at low risk vs high risk.^5,13 The tools have better ability to accurately categorize individuals at low risk, compared with high risk, where false positives are common.^12,14

Two types of risk factors

Risk factors for violence are commonly categorized as static or dynamic factors. Static factors are historical factors that cannot be changed with intervention (eg, age, sex, history of abuse). Dynamic factors can be changed with intervention (eg, substance abuse).¹⁵

Static risk factors. The best predictor of future violence is past violent behavior.^5,16,17 Violence risk increases with each prior episode of violence.⁵ Prior arrests for any crime, especially if the individual was a juvenile at the time of arrest for his or her first violent offense, increase future violence risk.⁵ Other important static violence risk factors include demographic factors such as age, sex, and socioeconomic status. Swanson et al⁶ reviewed a large pool of data (approximately 10,000 respondents) from the Epidemiologic Catchment Area survey. Being young, male, and of low socioeconomic status were all associated with violence in the community.⁶ The highest-risk age group for violence is age 15 to 24.⁵ Males perpetrate violence in the community at a rate 10 times that of females.¹⁸ However, among individuals with severe mental illness, men and women have similar rates of violence.^19,20 Unstable employment,²¹ less education,²² low intelligence,¹⁶ and a history of a significant head injury⁵ also are risk factors for violence.⁵

Continue to: Being abused as a child...

Being abused as a child, witnessing violence in the home,^5,16 and growing up with an unstable parental situation (eg, parental loss or separation) has been linked to violence.^16,23,24 Early disruptive behavior in childhood (eg, fighting, lying and stealing, truancy, and school problems) increases violence risk.^21,23

Personality factors are important static risk factors for violence. Antisocial personality disorder is the most common personality disorder linked with violence.¹⁷ Several studies consistently show psychopathy to be a strong predictor of both violence and criminal behavior.^5,25 A psychopath is a person who lacks empathy and close relationships, behaves impulsively, has superficially charming qualities, and is primarily interested in self-gratification.²⁶ Harris et al²⁷ studied 169 released forensic patients and found that 77% of the psychopaths (according to Psychopathy Checklist-Revised [PCL-R] scores) violently recidivated. In contrast, only 21% of the non-psychopaths violently recidivated.²⁷

Other personality factors associated with violence include a predisposition toward feelings of anger and hatred (as opposed to empathy, anxiety, or guilt, which may reduce risk), hostile attributional biases (a tendency to interpret benign behavior of others as intentionally antagonistic), violent fantasies, poor anger control, and impulsivity.⁵ Although personality factors tend to be longstanding and more difficult to modify, in the outpatient setting, therapeutic efforts can be made to modify hostile attribution biases, poor anger control, and impulsive behavior.

Dynamic risk factors. Substance abuse is strongly associated with violence.^6,17 The prevalence of violence is 12 times greater among individuals with alcohol use disorder and 16 times greater among individuals with other substance use disorders, compared with those with no such diagnoses.^5,6

Continue to: Steadman et al...

Steadman et al²⁸ compared 1,136 adult patients with mental disorders discharged from psychiatric hospitals with 519 individuals living in the same neighborhoods as the hospitalized patients. They found that the prevalence of violence among discharged patients without substance abuse was “statistically indistinguishable” from the prevalence of violence among community members, in the same neighborhood, who did not have symptoms of substance abuse.²⁸ Swanson et al⁶ found that the combination of a mental disorder plus an alcohol or substance use disorder substantially increased the risk of violence.

Other dynamic risk factors for violence include mental illness symptoms such as psychosis, especially threat/control-override delusions, where the individual believes that they are being threatened or controlled by an external force.¹⁷

Contextual factors to consider in violence risk assessments include current stressors, lack of social support, availability of weapons, access to drugs and alcohol, and the presence of similar circumstances that led to violent behavior in the past.⁵

How to assess the risk of targeted violence

Targeted violence is a predatory act of violence intentionally committed against a preselected person, group of people, or place.²⁹ Due to the low base rates of these incidents, targeted violence is difficult to study.^7,30 These risk assessments require a more specialized approach.

Continue to: In their 1999 article...

In their 1999 article, Borum et al³⁰ discussed threat assessment strategies utilized by the U.S. Secret Service and recommended investigating “pathways of ideas and behaviors that may lead to violent action.” Borum et al³⁰ summarized 3 fundamental principles of threat assessment (Table 1³⁰).

What to do when violence risk is not due to mental illness

Based on the information in Mr. F’s case scenario, it is likely that his homicidal ideation is not due to mental illness. Despite this, several risk factors for violence are present. Where do we go from here?

Scott and Resnick¹⁷ recommend considering the concept of dangerousness as 5 components (Table 2¹⁷). When this model of dangerousness is applied to Mr. F’s case, one can see that the magnitude of the harm is great because of threatened homicide. With regard to the imminence of the harm, it would help to clarify whether Mr. F plans to kill Ms. S immediately after discharge, or sometime in the next few months. Is his threat contingent on further provocations by Ms. S? Alternatively, does he intend to kill her for past grievances, regardless of further perceived insults?

Next, the frequency of a behavior relates to how often Mr. F has been aggressive in the past. The severity of his past aggression is also important. What is the most violent act he has ever done? Situational factors in this case include Mr. F’s access to weapons, financial problems, housing problems, and access to drugs and alcohol.¹⁷ Mr. F should be asked about what situations previously provoked his violent behavior. Consider how similar the present conditions are to past conditions to which Mr. F responded violently.⁵ The likelihood that a homicide will occur should take into account Mr. F’s risk factors for violence, as well as the seriousness of his intent to cause harm.

Continue to: Consider using a structured tool...

Consider using a structured tool, such as the Classification of Violence Risk, to help identify Mr. F’s risk factors for violence, or some other formal method to ensure that the proper data are collected. Violence risk assessments are more accurate when structured risk assessment tools are used, compared with clinical judgment alone.

It is important to review collateral sources of information. In Mr. F’s case, useful collateral sources may include his criminal docket (usually available online), past medical records, information from the shelter where he lives, and, potentially, friends or family.

Because Mr. F is making threats of targeted violence, be sure to ask about attack-related behaviors (Table 1³⁰).

Regarding the seriousness of Mr. F’s intent to cause harm, it may be helpful to ask him the following questions:

1. How likely are you to carry out this act of violence?
2. Do you have a plan? Have you taken any steps toward this plan?
3. Do you see other, nonviolent solutions to this problem?
4. What do you hope that we can do for you to help with this problem?

Continue to: Mr. F's answers...

Mr. F’s answers may suggest the possibility of a hidden agenda. Some patients express homicidal thoughts in order to stay in the hospital. If Mr. F expresses threats that are contingent on discharge and declines to engage in problem-solving discussions, this would cast doubt on the genuineness of his threat. However, doubt about the genuineness of the threat alone is not sufficient to simply discharge Mr. F. Assessment of his intent needs to be considered with other relevant risk factors, risk reduction strategies, and any Tarasoff duties that may apply.

In addition to risk factors, consider mitigating factors. For example, does Mr. F express concern over prison time as a reason to not engage in violence? It would be more ominous if Mr. F says that he does not care if he goes to prison because life is lousy being homeless and unemployed. At this point, an estimation can be made regarding whether Mr. F is a low-, moderate-, or high-risk of violence.

The next step is to organize Mr. F’s risk factors into static (historical) and dynamic (subject to intervention) factors. This will be helpful in formulating a strategy to manage risk because continued hospitalization can only address dynamic risk factors. Often in these cases, the static risk factors are far more numerous than the dynamic risk factors.

Once the data are collected and organized, the final step is to devise a risk management strategy. Some interventions, such as substance use treatment, will be straightforward. A mood-stabilizing medication could be considered, if clinically appropriate, to help reduce aggression and irritability.³¹ Efforts should be made to eliminate Mr. F’s access to firearms; however, in this case, it sounds unlikely that he will cooperate with those efforts. Ultimately, you may find yourself with a list of risk factors that are unlikely to be altered with further hospitalization, particularly if Mr. F’s homicidal thoughts and intent are due to antisocial personality traits.

Continue to: In that case...

In that case, the most important step will be to carry out your duty to warn/protect others prior to Mr. F’s discharge. Most states either require or permit mental health professionals to take reasonable steps to protect victims from violence when certain conditions are present, such as an explicit threat or identifiable victim (see Related Resources).

Once dynamic risk factors have been addressed, and duty to warn/protect is carried out, if there is no further clinical indication for hospitalization, it would be appropriate to discharge Mr. F. Continued homicidal threats stemming from antisocial personality traits, in the absence of a treatable mental illness (or other modifiable risk factors for violence that can be actively addressed), is not a reason for continued hospitalization. It may be useful to obtain a second opinion from a colleague in such scenarios. A second opinion may offer additional risk management ideas. In the event of a bad outcome, this will also help to show that the decision to discharge the patient was not taken lightly.

The psychiatrist should document a thoughtful risk assessment, the strategies that were implemented to reduce risk, the details of the warning, and the reasoning why continued hospitalization was not indicated (Table 3).

CASE CONTINUED

Decision to discharge

In Mr. F’s case, the treating psychiatrist determined that Mr. F’s risk of violence toward Ms. S was moderate. The psychiatrist identified several static risk factors for violence that raised Mr. F’s risk, but also noted that Mr. F’s threats were likely a manipulative effort to prolong his hospital stay. The psychiatrist carried out his duty to protect by notifying police and Ms. S of the nature of the threat prior to Mr. F’s discharge. The unit social worker helped Mr. F schedule an intake appointment for a substance use disorder treatment facility. Mr. F ultimately stated that he no longer experienced homicidal ideas once a bed was secured for him in a substance use treatment program. The psychiatrist carefully documented Mr. F’s risk assessment and the reasons why Mr. F’s risk would not be significantly altered by further inpatient hospitalization. Mr. F was discharged, and Ms. S remained unharmed.

Continue to: Bottom Line

Use a structured approach to identify risk factors for violence. Address dynamic risk factors, including access to weapons. Carry out the duty to warn/protect if applicable. Document your decisions and actions carefully, and then discharge the patient if clinically indicated. Do not be “held hostage” by a patient’s homicidal ideation.

Related Resources

• Dolan M, Doyle M. Violence risk prediction. Clinical and actuarial measures and the role of the psychopathy checklist. Br J Psychiatry. 2000;177:303-311.
• Douglas KS, Hart SD, Webster CD, et al. HCR-20V3: Assessing risk of violence–user guide. Burnaby, Canada: Mental Health, Law, and Policy Institute, Simon Fraser University; 2013.
• National Conference of State Legislatures. Mental health professionals’ duty to warn. http://www.ncsl.org/research/health/mental-health-professionals-duty-to-warn.aspx. Published September 28, 2015.

Drug Brand Names

Sertraline • Zoloft

Mr. F, age 35, is homeless and has a history of cocaine and alcohol use disorders. He is admitted voluntarily to the psychiatric unit because he has homicidal thoughts toward Ms. S, who works in the shelter where he has been staying. Mr. F reports that he is thinking of killing Ms. S if he is discharged because she has been rude to him. He states that he has access to several firearms, but he will not disclose the location. He has been diagnosed with unspecified depressive disorder and exhibited antisocial personality disorder traits. He is being treated with sertraline. However, his mood appears to be relatively stable, except for occasional angry verbal outbursts. The outbursts have been related to intrusive peers or staff turning the television off for group meetings. Mr. F has been joking with peers, eating well, and sleeping appropriately. He reports no suicidal thoughts and has not been physically violent on the unit. However, Mr. F has had a history of violence since his teenage years. He has been incarcerated twice for assault and once for drug possession.

How would you approach assessing and managing Mr. F’s risk for violence?

We all have encountered a patient similar to Mr. F on the psychiatric unit or in the emergency department—a patient who makes violent threats and appears angry, intimidating, manipulative, and/or demanding, despite exhibiting no evidence of mania or psychosis. This patient often has a history of substance abuse and a lifelong pattern of viewing violence as an acceptable way of addressing life’s problems. Many psychiatrists suspect that more time on the inpatient unit is unlikely to reduce this patient’s risk of violence. Why? Because the violence risk does not stem from a treatable mental illness. Further, psychiatrists may be apprehensive about this patient’s potential for violence after discharge and their liability in the event of a bad outcome. No one wants their name associated with a headline that reads “Psychiatrist discharged man less than 24 hours before he killed 3 people.”

The purported relationship between mental illness and violence often is sensationalized in the media. However, research reveals that the vast majority of violence is in fact not due to symptoms of mental illness.^1,2 A common clinical challenge in psychiatry involves evaluating individuals at elevated risk of violence and determining how to address their risk factors for violence. When the risk is primarily due to psychosis and can be reduced with antipsychotic medication, the job is easy. But how should we proceed when the risk stems from factors other than mental illness?

This article reviews risk factors for violence, discusses targeted violence against a specific victim, and offers practical tips for assessing and managing risk, particularly when the risk for violence is not due to mental illness.

Violence and mental illness: A tenuous link

Violence is a major public health concern in the United States. Although in recent years the rates of homicide and aggravated assault have decreased dramatically, there are approximately 16,000 homicides annually in the United States, and more than 1.6 million injuries from assaults treated in emergency departments each year.³ Homicide continues to be one of the leading causes of death among teenagers and young adults.⁴

The most effective methods of preventing widespread violence are public health approaches, such as parent- and family-focused programs, early childhood education, programs in school, and public policy changes.³ However, as psychiatrists, we are routinely asked to assess the risk of violence for an individual patient and devise strategies to mitigate violence risk.

Continue to: Although certain mental illnesses...

Although certain mental illnesses increase the relative risk of violence (compared with people without mental illness),^5,6 recent studies suggest that mental illness plays only a “minor role in explaining violence in populations.”⁷ It is estimated that as little as 4% of the violence in the United States can be attributed to mental illness.¹ According to a 1998 meta-analysis of 48 studies of criminal recidivism, the risk factors for violent recidivism were “almost identical” among offenders who had a mental disorder and those who did not.⁸

Approaches to assessing violence risk

Psychiatrists can assess the risk of future violence via 3 broad approaches.^9,10

Unaided clinical judgment is when a mental health professional estimates violence risk based on his or her own experience and intuition, with knowledge of violence risk factors, but without the use of structured tools.

Actuarial tools are statistical models that use formulae to show relationships between data (risk factors) and outcomes (violence).^10,11

Continue to: Structured professional judgment

Structured professional judgment is a hybrid of unaided clinical judgment and actuarial methods. Structured professional judgment tools help the evaluator identify empirically established risk factors. Once the information is collected, it is combined with clinical judgment in decision making.^9,10 There are now more than 200 structured tools available for assessing violence risk in criminal justice and forensic mental health populations.¹²

Clinical judgment, although commonly used in practice, is less accurate than actuarial tools or structured professional judgment.^10,11 In general, risk assessment tools offer moderate levels of accuracy in categorizing people at low risk vs high risk.^5,13 The tools have better ability to accurately categorize individuals at low risk, compared with high risk, where false positives are common.^12,14

Two types of risk factors

Risk factors for violence are commonly categorized as static or dynamic factors. Static factors are historical factors that cannot be changed with intervention (eg, age, sex, history of abuse). Dynamic factors can be changed with intervention (eg, substance abuse).¹⁵

Static risk factors. The best predictor of future violence is past violent behavior.^5,16,17 Violence risk increases with each prior episode of violence.⁵ Prior arrests for any crime, especially if the individual was a juvenile at the time of arrest for his or her first violent offense, increase future violence risk.⁵ Other important static violence risk factors include demographic factors such as age, sex, and socioeconomic status. Swanson et al⁶ reviewed a large pool of data (approximately 10,000 respondents) from the Epidemiologic Catchment Area survey. Being young, male, and of low socioeconomic status were all associated with violence in the community.⁶ The highest-risk age group for violence is age 15 to 24.⁵ Males perpetrate violence in the community at a rate 10 times that of females.¹⁸ However, among individuals with severe mental illness, men and women have similar rates of violence.^19,20 Unstable employment,²¹ less education,²² low intelligence,¹⁶ and a history of a significant head injury⁵ also are risk factors for violence.⁵

Continue to: Being abused as a child...

Being abused as a child, witnessing violence in the home,^5,16 and growing up with an unstable parental situation (eg, parental loss or separation) has been linked to violence.^16,23,24 Early disruptive behavior in childhood (eg, fighting, lying and stealing, truancy, and school problems) increases violence risk.^21,23

Personality factors are important static risk factors for violence. Antisocial personality disorder is the most common personality disorder linked with violence.¹⁷ Several studies consistently show psychopathy to be a strong predictor of both violence and criminal behavior.^5,25 A psychopath is a person who lacks empathy and close relationships, behaves impulsively, has superficially charming qualities, and is primarily interested in self-gratification.²⁶ Harris et al²⁷ studied 169 released forensic patients and found that 77% of the psychopaths (according to Psychopathy Checklist-Revised [PCL-R] scores) violently recidivated. In contrast, only 21% of the non-psychopaths violently recidivated.²⁷

Other personality factors associated with violence include a predisposition toward feelings of anger and hatred (as opposed to empathy, anxiety, or guilt, which may reduce risk), hostile attributional biases (a tendency to interpret benign behavior of others as intentionally antagonistic), violent fantasies, poor anger control, and impulsivity.⁵ Although personality factors tend to be longstanding and more difficult to modify, in the outpatient setting, therapeutic efforts can be made to modify hostile attribution biases, poor anger control, and impulsive behavior.

Dynamic risk factors. Substance abuse is strongly associated with violence.^6,17 The prevalence of violence is 12 times greater among individuals with alcohol use disorder and 16 times greater among individuals with other substance use disorders, compared with those with no such diagnoses.^5,6

Continue to: Steadman et al...

Steadman et al²⁸ compared 1,136 adult patients with mental disorders discharged from psychiatric hospitals with 519 individuals living in the same neighborhoods as the hospitalized patients. They found that the prevalence of violence among discharged patients without substance abuse was “statistically indistinguishable” from the prevalence of violence among community members, in the same neighborhood, who did not have symptoms of substance abuse.²⁸ Swanson et al⁶ found that the combination of a mental disorder plus an alcohol or substance use disorder substantially increased the risk of violence.

Other dynamic risk factors for violence include mental illness symptoms such as psychosis, especially threat/control-override delusions, where the individual believes that they are being threatened or controlled by an external force.¹⁷

Contextual factors to consider in violence risk assessments include current stressors, lack of social support, availability of weapons, access to drugs and alcohol, and the presence of similar circumstances that led to violent behavior in the past.⁵

How to assess the risk of targeted violence

Targeted violence is a predatory act of violence intentionally committed against a preselected person, group of people, or place.²⁹ Due to the low base rates of these incidents, targeted violence is difficult to study.^7,30 These risk assessments require a more specialized approach.

Continue to: In their 1999 article...

In their 1999 article, Borum et al³⁰ discussed threat assessment strategies utilized by the U.S. Secret Service and recommended investigating “pathways of ideas and behaviors that may lead to violent action.” Borum et al³⁰ summarized 3 fundamental principles of threat assessment (Table 1³⁰).

What to do when violence risk is not due to mental illness

Based on the information in Mr. F’s case scenario, it is likely that his homicidal ideation is not due to mental illness. Despite this, several risk factors for violence are present. Where do we go from here?

Scott and Resnick¹⁷ recommend considering the concept of dangerousness as 5 components (Table 2¹⁷). When this model of dangerousness is applied to Mr. F’s case, one can see that the magnitude of the harm is great because of threatened homicide. With regard to the imminence of the harm, it would help to clarify whether Mr. F plans to kill Ms. S immediately after discharge, or sometime in the next few months. Is his threat contingent on further provocations by Ms. S? Alternatively, does he intend to kill her for past grievances, regardless of further perceived insults?

Next, the frequency of a behavior relates to how often Mr. F has been aggressive in the past. The severity of his past aggression is also important. What is the most violent act he has ever done? Situational factors in this case include Mr. F’s access to weapons, financial problems, housing problems, and access to drugs and alcohol.¹⁷ Mr. F should be asked about what situations previously provoked his violent behavior. Consider how similar the present conditions are to past conditions to which Mr. F responded violently.⁵ The likelihood that a homicide will occur should take into account Mr. F’s risk factors for violence, as well as the seriousness of his intent to cause harm.

Continue to: Consider using a structured tool...

Consider using a structured tool, such as the Classification of Violence Risk, to help identify Mr. F’s risk factors for violence, or some other formal method to ensure that the proper data are collected. Violence risk assessments are more accurate when structured risk assessment tools are used, compared with clinical judgment alone.

It is important to review collateral sources of information. In Mr. F’s case, useful collateral sources may include his criminal docket (usually available online), past medical records, information from the shelter where he lives, and, potentially, friends or family.

Because Mr. F is making threats of targeted violence, be sure to ask about attack-related behaviors (Table 1³⁰).

Regarding the seriousness of Mr. F’s intent to cause harm, it may be helpful to ask him the following questions:

1. How likely are you to carry out this act of violence?
2. Do you have a plan? Have you taken any steps toward this plan?
3. Do you see other, nonviolent solutions to this problem?
4. What do you hope that we can do for you to help with this problem?

Continue to: Mr. F's answers...

Mr. F’s answers may suggest the possibility of a hidden agenda. Some patients express homicidal thoughts in order to stay in the hospital. If Mr. F expresses threats that are contingent on discharge and declines to engage in problem-solving discussions, this would cast doubt on the genuineness of his threat. However, doubt about the genuineness of the threat alone is not sufficient to simply discharge Mr. F. Assessment of his intent needs to be considered with other relevant risk factors, risk reduction strategies, and any Tarasoff duties that may apply.

In addition to risk factors, consider mitigating factors. For example, does Mr. F express concern over prison time as a reason to not engage in violence? It would be more ominous if Mr. F says that he does not care if he goes to prison because life is lousy being homeless and unemployed. At this point, an estimation can be made regarding whether Mr. F is a low-, moderate-, or high-risk of violence.

The next step is to organize Mr. F’s risk factors into static (historical) and dynamic (subject to intervention) factors. This will be helpful in formulating a strategy to manage risk because continued hospitalization can only address dynamic risk factors. Often in these cases, the static risk factors are far more numerous than the dynamic risk factors.

Once the data are collected and organized, the final step is to devise a risk management strategy. Some interventions, such as substance use treatment, will be straightforward. A mood-stabilizing medication could be considered, if clinically appropriate, to help reduce aggression and irritability.³¹ Efforts should be made to eliminate Mr. F’s access to firearms; however, in this case, it sounds unlikely that he will cooperate with those efforts. Ultimately, you may find yourself with a list of risk factors that are unlikely to be altered with further hospitalization, particularly if Mr. F’s homicidal thoughts and intent are due to antisocial personality traits.

Continue to: In that case...

In that case, the most important step will be to carry out your duty to warn/protect others prior to Mr. F’s discharge. Most states either require or permit mental health professionals to take reasonable steps to protect victims from violence when certain conditions are present, such as an explicit threat or identifiable victim (see Related Resources).

Once dynamic risk factors have been addressed, and duty to warn/protect is carried out, if there is no further clinical indication for hospitalization, it would be appropriate to discharge Mr. F. Continued homicidal threats stemming from antisocial personality traits, in the absence of a treatable mental illness (or other modifiable risk factors for violence that can be actively addressed), is not a reason for continued hospitalization. It may be useful to obtain a second opinion from a colleague in such scenarios. A second opinion may offer additional risk management ideas. In the event of a bad outcome, this will also help to show that the decision to discharge the patient was not taken lightly.

The psychiatrist should document a thoughtful risk assessment, the strategies that were implemented to reduce risk, the details of the warning, and the reasoning why continued hospitalization was not indicated (Table 3).

CASE CONTINUED

Decision to discharge

In Mr. F’s case, the treating psychiatrist determined that Mr. F’s risk of violence toward Ms. S was moderate. The psychiatrist identified several static risk factors for violence that raised Mr. F’s risk, but also noted that Mr. F’s threats were likely a manipulative effort to prolong his hospital stay. The psychiatrist carried out his duty to protect by notifying police and Ms. S of the nature of the threat prior to Mr. F’s discharge. The unit social worker helped Mr. F schedule an intake appointment for a substance use disorder treatment facility. Mr. F ultimately stated that he no longer experienced homicidal ideas once a bed was secured for him in a substance use treatment program. The psychiatrist carefully documented Mr. F’s risk assessment and the reasons why Mr. F’s risk would not be significantly altered by further inpatient hospitalization. Mr. F was discharged, and Ms. S remained unharmed.

Continue to: Bottom Line

Use a structured approach to identify risk factors for violence. Address dynamic risk factors, including access to weapons. Carry out the duty to warn/protect if applicable. Document your decisions and actions carefully, and then discharge the patient if clinically indicated. Do not be “held hostage” by a patient’s homicidal ideation.

Related Resources

• Dolan M, Doyle M. Violence risk prediction. Clinical and actuarial measures and the role of the psychopathy checklist. Br J Psychiatry. 2000;177:303-311.
• Douglas KS, Hart SD, Webster CD, et al. HCR-20V3: Assessing risk of violence–user guide. Burnaby, Canada: Mental Health, Law, and Policy Institute, Simon Fraser University; 2013.
• National Conference of State Legislatures. Mental health professionals’ duty to warn. http://www.ncsl.org/research/health/mental-health-professionals-duty-to-warn.aspx. Published September 28, 2015.

Drug Brand Names

Sertraline • Zoloft

The obsessive pursuit of another has long been described in fiction and the scientific literature, but was conceptualized as “stalking” only rela­tively recently—first, under the guise of celebrity stalking and, later, as a public health issue recognized as affecting the general population. A useful working definition of stalking is “… the willful, malicious, and repeated fol­lowing of and harassing of another person that threatens his/her safety.”¹

Stalking victims report numerous, severe, life-changing effects from being stalked, including physical, social, and psychological harm. They typically experience mood, anxi­ety, and posttraumatic stress symptoms that require prompt evaluation and treatment.

Prevalence and other characteristics
Stalking and its subsequent victimization are common. Here are statistics:
   • in the United States, approximately 1 million women and 370,000 men are stalked annually
   • women are 3 times more likely to be stalked than raped²
   • lifetime prevalence of stalking victimization is 20% (women, 23.5%; men, 10.5%)
   • 75% of stalking victims are women
   • 77% of stalking emerges from a prior acquaintance, including 49% that originated in a romantic relationship
   • 33% of stalking encounters eventually lead to physical violence; slightly >10% of encounters lead to sexual violence
   • stalking persists for an extended period; on average, almost 2 years.³

Penalties. Stalking can result in interven­tion by the criminal justice system. Legal sanctions levied on the perpetrator vary, depending on (among other variables) the severity of stalking; type of stalking; motive of the stalker; and the strength of incriminating evidence. Surprisingly, the outcome of the perpetrator’s prosecution (arrest, conviction, length of sentence) is unrelated to whether the victim reported continued stalking at follow-up.^4,5

What are the symptoms and the dam­age? Given the intrusive nature of stalking behaviors and the extended period during which stalking persists, victims typically experience harmful psychological effects that range from subclinical symptoms to overt psychiatric disorders.

Stalking can have a profound impact on the victim and result in numerous psycho­logical symptoms that become the focus of clinical attention. The typically chronic nature of stalking probably plays a signifi­cant role in its contributions to its victims’ psychological distress.⁶ Melton⁷ found that the most common adverse effect of stalk­ing was related to the emotional impact of being stalked—with victims feeling scared, depressed, humiliated, embarrassed, dis­trustful of others, and angry or hateful.

Stalking victims report traumatic stress, hypervigilance, excessive fear, and anxiety coupled with disruptions in employment and social interactions.⁸ Many report hav­ing become highly distrustful or suspicious (44%); fearful (42%); nervous (31%); angry (27%); paranoid (36%); and depressed (21%). In general, victims have elevated scores on the Trauma Symptom Checklist.⁹

Stalking in the setting of intimate part­ner abuse is associated with harmful out­comes for the victim. These include repeat physical violence, psychological distress, and impaired physical or mental health, or both.^3,7,10

Stalking victims who are female; had a prior relationship with the stalker; have experienced a greater variety of stalking behaviors; are divorced or separated; and have received government assistance were found to be more likely to experience mul­tiple negative outcomes from stalking.¹¹

Effects on mental health. Stalking victims have a higher incidence of mental disorders and comorbid illnesses compared with the general population,¹² with the most robust associations identified between stalking victimization, major depressive disorder, and panic disorder. Stalking contributes to symptoms of posttraumatic stress disor­der,¹³ and there is an association between posttraumatic stress and poor general health.¹⁴ Stalking victims report higher cur­rent use of psychotropic medications.¹²

Victims who blame themselves for being stalked report a significantly higher severity of depression, anxiety, and post­traumatic stress symptoms. Those who ruminate more about the stalking expe­rience, or who explicitly emphasize the terror of stalking to a greater extent, also report a significantly higher severity of symptoms.¹⁵

Spitzberg³ reported that stalking victim­ization has several possible effects on vic­tims (Table 1).

The degree of a victim’s symptoms corre­lates partially with the severity of stalking. However, other variables play a crucial role in explaining the level of distress among stalking victims¹⁵; these include the types of coping strategies adopted by victims. Self-blame, catastrophizing, and rumination are significantly associated with malad­justment; on the other hand, positive reap­praisal—thoughts of attaching a positive meaning to the event, in terms of personal growth—is associated with greater psycho­logical adjustment.

The more stalking a victim experiences (and, presumably, experiences greater dis­tress), the greater the variety of coping strat­egies she (he) employs.¹⁶

How should stalking victims be treated?
Stalking victims are an underserved popu­lation. Practitioners often are unsure how to address stalking; furthermore, available treatments can be ineffective.

There is a great deal of variability in what professionals who work with stalk­ing victims believe is appropriate practice. Services provided to victims vary widely,¹⁷ and the field has not yet come to a consen­sus on best practices.¹⁶

Proceed case by case. Practitioners must understand the nuances of each case to con­sider what might work at a particular point in time, and information from victims can help guide decision-making.16 Evidence suggests that stalking victims can feel frus­trated in their attempt to seek help, particu­larly from the criminal justice system; it is possible that such bad experiences may dis­suade them from seeking help later.^5,8,18 It is worth noting that, as the frequency of stalk­ing decreases for any given victim, her (his) perception of safety increases and distress diminishes.¹⁶

Few communities have attempted to address systemically the problem of stalk­ing. Existing anti-stalking programs have focused on the criminal justice aspects of intervention,⁸ with less emphasis on treat­ing victims.

Some stalking victims rely on friends and family for support and assistance, but research shows that most reach out to agen­cies for assistance and, generally, seek help from multiple sources.¹⁸ Typically, stalking victims are served by 2 types of victim ser­vice organizations:
   • specialized, small, private and non­profit agencies (eg, domestic violence shelters, rape crisis centers, victims’ rights advocacy organizations)
   • small units housed in police depart­ments and prosecutors’ offices.¹⁷

Note: When victims seek services at criminal justice agencies, they may be feel­ing particularly unsafe and distressed. This underscores the importance of co-locating victim service providers and criminal jus­tice agencies.¹⁶

Stalking victims might benefit from multi-disciplinary team consultation, including input from psychiatric, psy­chotherapeutic, and law enforcement or security professionals. Key priorities for practitioners to address with stalking vic­tims are given in Table 2.¹⁹

Stalking behavior does not significantly decrease when victims are in contact with victim services.16 Practitioners can integrate this prospect into their understanding of stalking when they work with victims: That is, it is likely that the problem will not go away quickly, even with intervention.

Victims’ needs remain great and broad-based. Spence-Diehl et al¹⁷ conducted a survey of service providers for stalking victims, evaluating the needs of those vic­tims and the response of their communities. Some of their recommendations for better meeting victims’ needs are in Table 3.¹⁶

At the heart of a client-centered model of victim assistance is the provider’s abil­ity to listen to a victim’s story and respond in a nonjudgmental manner. This approach honors the victim’s circumstances and her personal understanding of risk.²¹
 

Bottom Line
Stalking victims are a distinctive population, experiencing numerous emotional, physical, and social effects of their stalking over an extended period. Services to treat this underserved population need to be further developed. A multifaceted approach to treating victims incorporates psychological, somatic, and practical interventions, and a victim-centered approach is associated with better outcomes.
 

Related Resources
• Harmon RB, O’Connor M. Forcier A, et al. The impact of anti-stalking training on front line service providers: using the anti-stalking training evaluation protocol (ASTEP). J Forensic Science. 2004;49(5):1050-1055.
• Spitzberg BH, Cupach WR. The state of the art of stalking: tak­ing stock of the emerging literature. Aggression and Violence Behavior. 2007;12(1):64-86.
 

Disclosure
The author reports no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

The obsessive pursuit of another has long been described in fiction and the scientific literature, but was conceptualized as “stalking” only rela­tively recently—first, under the guise of celebrity stalking and, later, as a public health issue recognized as affecting the general population. A useful working definition of stalking is “… the willful, malicious, and repeated fol­lowing of and harassing of another person that threatens his/her safety.”¹

Stalking victims report numerous, severe, life-changing effects from being stalked, including physical, social, and psychological harm. They typically experience mood, anxi­ety, and posttraumatic stress symptoms that require prompt evaluation and treatment.

Prevalence and other characteristics
Stalking and its subsequent victimization are common. Here are statistics:
   • in the United States, approximately 1 million women and 370,000 men are stalked annually
   • women are 3 times more likely to be stalked than raped²
   • lifetime prevalence of stalking victimization is 20% (women, 23.5%; men, 10.5%)
   • 75% of stalking victims are women
   • 77% of stalking emerges from a prior acquaintance, including 49% that originated in a romantic relationship
   • 33% of stalking encounters eventually lead to physical violence; slightly >10% of encounters lead to sexual violence
   • stalking persists for an extended period; on average, almost 2 years.³

Penalties. Stalking can result in interven­tion by the criminal justice system. Legal sanctions levied on the perpetrator vary, depending on (among other variables) the severity of stalking; type of stalking; motive of the stalker; and the strength of incriminating evidence. Surprisingly, the outcome of the perpetrator’s prosecution (arrest, conviction, length of sentence) is unrelated to whether the victim reported continued stalking at follow-up.^4,5

What are the symptoms and the dam­age? Given the intrusive nature of stalking behaviors and the extended period during which stalking persists, victims typically experience harmful psychological effects that range from subclinical symptoms to overt psychiatric disorders.

Stalking can have a profound impact on the victim and result in numerous psycho­logical symptoms that become the focus of clinical attention. The typically chronic nature of stalking probably plays a signifi­cant role in its contributions to its victims’ psychological distress.⁶ Melton⁷ found that the most common adverse effect of stalk­ing was related to the emotional impact of being stalked—with victims feeling scared, depressed, humiliated, embarrassed, dis­trustful of others, and angry or hateful.

Stalking victims report traumatic stress, hypervigilance, excessive fear, and anxiety coupled with disruptions in employment and social interactions.⁸ Many report hav­ing become highly distrustful or suspicious (44%); fearful (42%); nervous (31%); angry (27%); paranoid (36%); and depressed (21%). In general, victims have elevated scores on the Trauma Symptom Checklist.⁹

Stalking in the setting of intimate part­ner abuse is associated with harmful out­comes for the victim. These include repeat physical violence, psychological distress, and impaired physical or mental health, or both.^3,7,10

Stalking victims who are female; had a prior relationship with the stalker; have experienced a greater variety of stalking behaviors; are divorced or separated; and have received government assistance were found to be more likely to experience mul­tiple negative outcomes from stalking.¹¹

Effects on mental health. Stalking victims have a higher incidence of mental disorders and comorbid illnesses compared with the general population,¹² with the most robust associations identified between stalking victimization, major depressive disorder, and panic disorder. Stalking contributes to symptoms of posttraumatic stress disor­der,¹³ and there is an association between posttraumatic stress and poor general health.¹⁴ Stalking victims report higher cur­rent use of psychotropic medications.¹²

Victims who blame themselves for being stalked report a significantly higher severity of depression, anxiety, and post­traumatic stress symptoms. Those who ruminate more about the stalking expe­rience, or who explicitly emphasize the terror of stalking to a greater extent, also report a significantly higher severity of symptoms.¹⁵

Spitzberg³ reported that stalking victim­ization has several possible effects on vic­tims (Table 1).

The degree of a victim’s symptoms corre­lates partially with the severity of stalking. However, other variables play a crucial role in explaining the level of distress among stalking victims¹⁵; these include the types of coping strategies adopted by victims. Self-blame, catastrophizing, and rumination are significantly associated with malad­justment; on the other hand, positive reap­praisal—thoughts of attaching a positive meaning to the event, in terms of personal growth—is associated with greater psycho­logical adjustment.

The more stalking a victim experiences (and, presumably, experiences greater dis­tress), the greater the variety of coping strat­egies she (he) employs.¹⁶

How should stalking victims be treated?
Stalking victims are an underserved popu­lation. Practitioners often are unsure how to address stalking; furthermore, available treatments can be ineffective.

There is a great deal of variability in what professionals who work with stalk­ing victims believe is appropriate practice. Services provided to victims vary widely,¹⁷ and the field has not yet come to a consen­sus on best practices.¹⁶

Proceed case by case. Practitioners must understand the nuances of each case to con­sider what might work at a particular point in time, and information from victims can help guide decision-making.16 Evidence suggests that stalking victims can feel frus­trated in their attempt to seek help, particu­larly from the criminal justice system; it is possible that such bad experiences may dis­suade them from seeking help later.^5,8,18 It is worth noting that, as the frequency of stalk­ing decreases for any given victim, her (his) perception of safety increases and distress diminishes.¹⁶

Few communities have attempted to address systemically the problem of stalk­ing. Existing anti-stalking programs have focused on the criminal justice aspects of intervention,⁸ with less emphasis on treat­ing victims.

Some stalking victims rely on friends and family for support and assistance, but research shows that most reach out to agen­cies for assistance and, generally, seek help from multiple sources.¹⁸ Typically, stalking victims are served by 2 types of victim ser­vice organizations:
   • specialized, small, private and non­profit agencies (eg, domestic violence shelters, rape crisis centers, victims’ rights advocacy organizations)
   • small units housed in police depart­ments and prosecutors’ offices.¹⁷

Note: When victims seek services at criminal justice agencies, they may be feel­ing particularly unsafe and distressed. This underscores the importance of co-locating victim service providers and criminal jus­tice agencies.¹⁶

Stalking victims might benefit from multi-disciplinary team consultation, including input from psychiatric, psy­chotherapeutic, and law enforcement or security professionals. Key priorities for practitioners to address with stalking vic­tims are given in Table 2.¹⁹

Stalking behavior does not significantly decrease when victims are in contact with victim services.16 Practitioners can integrate this prospect into their understanding of stalking when they work with victims: That is, it is likely that the problem will not go away quickly, even with intervention.

Victims’ needs remain great and broad-based. Spence-Diehl et al¹⁷ conducted a survey of service providers for stalking victims, evaluating the needs of those vic­tims and the response of their communities. Some of their recommendations for better meeting victims’ needs are in Table 3.¹⁶

At the heart of a client-centered model of victim assistance is the provider’s abil­ity to listen to a victim’s story and respond in a nonjudgmental manner. This approach honors the victim’s circumstances and her personal understanding of risk.²¹
 

Bottom Line
Stalking victims are a distinctive population, experiencing numerous emotional, physical, and social effects of their stalking over an extended period. Services to treat this underserved population need to be further developed. A multifaceted approach to treating victims incorporates psychological, somatic, and practical interventions, and a victim-centered approach is associated with better outcomes.
 

Related Resources
• Harmon RB, O’Connor M. Forcier A, et al. The impact of anti-stalking training on front line service providers: using the anti-stalking training evaluation protocol (ASTEP). J Forensic Science. 2004;49(5):1050-1055.
• Spitzberg BH, Cupach WR. The state of the art of stalking: tak­ing stock of the emerging literature. Aggression and Violence Behavior. 2007;12(1):64-86.
 

Disclosure
The author reports no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

The obsessive pursuit of another has long been described in fiction and the scientific literature, but was conceptualized as “stalking” only rela­tively recently—first, under the guise of celebrity stalking and, later, as a public health issue recognized as affecting the general population. A useful working definition of stalking is “… the willful, malicious, and repeated fol­lowing of and harassing of another person that threatens his/her safety.”¹

Stalking victims report numerous, severe, life-changing effects from being stalked, including physical, social, and psychological harm. They typically experience mood, anxi­ety, and posttraumatic stress symptoms that require prompt evaluation and treatment.

Prevalence and other characteristics
Stalking and its subsequent victimization are common. Here are statistics:
   • in the United States, approximately 1 million women and 370,000 men are stalked annually
   • women are 3 times more likely to be stalked than raped²
   • lifetime prevalence of stalking victimization is 20% (women, 23.5%; men, 10.5%)
   • 75% of stalking victims are women
   • 77% of stalking emerges from a prior acquaintance, including 49% that originated in a romantic relationship
   • 33% of stalking encounters eventually lead to physical violence; slightly >10% of encounters lead to sexual violence
   • stalking persists for an extended period; on average, almost 2 years.³

Penalties. Stalking can result in interven­tion by the criminal justice system. Legal sanctions levied on the perpetrator vary, depending on (among other variables) the severity of stalking; type of stalking; motive of the stalker; and the strength of incriminating evidence. Surprisingly, the outcome of the perpetrator’s prosecution (arrest, conviction, length of sentence) is unrelated to whether the victim reported continued stalking at follow-up.^4,5

What are the symptoms and the dam­age? Given the intrusive nature of stalking behaviors and the extended period during which stalking persists, victims typically experience harmful psychological effects that range from subclinical symptoms to overt psychiatric disorders.

Stalking can have a profound impact on the victim and result in numerous psycho­logical symptoms that become the focus of clinical attention. The typically chronic nature of stalking probably plays a signifi­cant role in its contributions to its victims’ psychological distress.⁶ Melton⁷ found that the most common adverse effect of stalk­ing was related to the emotional impact of being stalked—with victims feeling scared, depressed, humiliated, embarrassed, dis­trustful of others, and angry or hateful.

Stalking victims report traumatic stress, hypervigilance, excessive fear, and anxiety coupled with disruptions in employment and social interactions.⁸ Many report hav­ing become highly distrustful or suspicious (44%); fearful (42%); nervous (31%); angry (27%); paranoid (36%); and depressed (21%). In general, victims have elevated scores on the Trauma Symptom Checklist.⁹

Stalking in the setting of intimate part­ner abuse is associated with harmful out­comes for the victim. These include repeat physical violence, psychological distress, and impaired physical or mental health, or both.^3,7,10

Stalking victims who are female; had a prior relationship with the stalker; have experienced a greater variety of stalking behaviors; are divorced or separated; and have received government assistance were found to be more likely to experience mul­tiple negative outcomes from stalking.¹¹

Effects on mental health. Stalking victims have a higher incidence of mental disorders and comorbid illnesses compared with the general population,¹² with the most robust associations identified between stalking victimization, major depressive disorder, and panic disorder. Stalking contributes to symptoms of posttraumatic stress disor­der,¹³ and there is an association between posttraumatic stress and poor general health.¹⁴ Stalking victims report higher cur­rent use of psychotropic medications.¹²

Victims who blame themselves for being stalked report a significantly higher severity of depression, anxiety, and post­traumatic stress symptoms. Those who ruminate more about the stalking expe­rience, or who explicitly emphasize the terror of stalking to a greater extent, also report a significantly higher severity of symptoms.¹⁵

Spitzberg³ reported that stalking victim­ization has several possible effects on vic­tims (Table 1).

The degree of a victim’s symptoms corre­lates partially with the severity of stalking. However, other variables play a crucial role in explaining the level of distress among stalking victims¹⁵; these include the types of coping strategies adopted by victims. Self-blame, catastrophizing, and rumination are significantly associated with malad­justment; on the other hand, positive reap­praisal—thoughts of attaching a positive meaning to the event, in terms of personal growth—is associated with greater psycho­logical adjustment.

The more stalking a victim experiences (and, presumably, experiences greater dis­tress), the greater the variety of coping strat­egies she (he) employs.¹⁶

How should stalking victims be treated?
Stalking victims are an underserved popu­lation. Practitioners often are unsure how to address stalking; furthermore, available treatments can be ineffective.

There is a great deal of variability in what professionals who work with stalk­ing victims believe is appropriate practice. Services provided to victims vary widely,¹⁷ and the field has not yet come to a consen­sus on best practices.¹⁶

Proceed case by case. Practitioners must understand the nuances of each case to con­sider what might work at a particular point in time, and information from victims can help guide decision-making.16 Evidence suggests that stalking victims can feel frus­trated in their attempt to seek help, particu­larly from the criminal justice system; it is possible that such bad experiences may dis­suade them from seeking help later.^5,8,18 It is worth noting that, as the frequency of stalk­ing decreases for any given victim, her (his) perception of safety increases and distress diminishes.¹⁶

Few communities have attempted to address systemically the problem of stalk­ing. Existing anti-stalking programs have focused on the criminal justice aspects of intervention,⁸ with less emphasis on treat­ing victims.

Some stalking victims rely on friends and family for support and assistance, but research shows that most reach out to agen­cies for assistance and, generally, seek help from multiple sources.¹⁸ Typically, stalking victims are served by 2 types of victim ser­vice organizations:
   • specialized, small, private and non­profit agencies (eg, domestic violence shelters, rape crisis centers, victims’ rights advocacy organizations)
   • small units housed in police depart­ments and prosecutors’ offices.¹⁷

Note: When victims seek services at criminal justice agencies, they may be feel­ing particularly unsafe and distressed. This underscores the importance of co-locating victim service providers and criminal jus­tice agencies.¹⁶

Stalking victims might benefit from multi-disciplinary team consultation, including input from psychiatric, psy­chotherapeutic, and law enforcement or security professionals. Key priorities for practitioners to address with stalking vic­tims are given in Table 2.¹⁹

Stalking behavior does not significantly decrease when victims are in contact with victim services.16 Practitioners can integrate this prospect into their understanding of stalking when they work with victims: That is, it is likely that the problem will not go away quickly, even with intervention.

Victims’ needs remain great and broad-based. Spence-Diehl et al¹⁷ conducted a survey of service providers for stalking victims, evaluating the needs of those vic­tims and the response of their communities. Some of their recommendations for better meeting victims’ needs are in Table 3.¹⁶

At the heart of a client-centered model of victim assistance is the provider’s abil­ity to listen to a victim’s story and respond in a nonjudgmental manner. This approach honors the victim’s circumstances and her personal understanding of risk.²¹
 

Bottom Line
Stalking victims are a distinctive population, experiencing numerous emotional, physical, and social effects of their stalking over an extended period. Services to treat this underserved population need to be further developed. A multifaceted approach to treating victims incorporates psychological, somatic, and practical interventions, and a victim-centered approach is associated with better outcomes.
 

Related Resources
• Harmon RB, O’Connor M. Forcier A, et al. The impact of anti-stalking training on front line service providers: using the anti-stalking training evaluation protocol (ASTEP). J Forensic Science. 2004;49(5):1050-1055.
• Spitzberg BH, Cupach WR. The state of the art of stalking: tak­ing stock of the emerging literature. Aggression and Violence Behavior. 2007;12(1):64-86.
 

Disclosure
The author reports no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.