To Name :
To Email :
From Name :
From Email :
Comments :

Evidence-Based Reviews

Assessing potential for harm: Would your patient injure himself or others?

Questions to ask, steps to take when evaluating tendencies toward suicide and violence

Vol. 8, No. 7 / July 2009

Comment on this article

Police take Ms. L, age 23, to the emergency room (ER) after her fiancé called them. He told the police that after a “night of drinking” they argued about a girl he had flirted with. Ms. L took out a loaded gun and threatened to shoot herself. She eventually handed the gun over to the police.

In the ER, Ms. L’s blood alcohol level is 0.20%. She tells the admitting emergency room nurse, “I would never hurt myself. I drank too much and was acting stupid. I just want to go home and sleep it off. I promise not to harm myself.” Emergency room staff observe Ms. L smile and giggle while waiting for a psychiatric evaluation.

What would you do? Hospitalize Ms. L for safety, or accept her promise not to hurt herself and send her home? What criteria would you use?

Knowing how to assess patients such as Ms. L is an essential psychiatric skill, whether or not you trained in forensic psychiatry. This article includes case reports that illustrate techniques for evaluating patients who may harbor suicidal or homicidal thoughts.

Evaluating danger to self

Your role is to weigh an individual’s risk factors for suicide against potential protective factors and to make a judgment call. Potential risk factors for suicide include (but are not limited to):

  • current suicidal thoughts
  • prior suicide attempts
  • presence of a comorbid psychiatric disorder (particularly depression)
  • intoxication or ongoing substance use
  • feelings of hopelessness
  • marked anxiety
  • recent stressors
  • family history of suicide
  • lack of psychosocial support.1

Not all factors will be present or relevant in every individual. See questions in Table 1 to further evaluate suicidality in patients who report suicidal thoughts.

Also review information that is reasonably available. In the ER, records from other facilities or private psychiatric treatment notes may not be accessed easily. However, in addition to conducting a suicide risk assessment and mental status examination, consider reviewing the collateral information outlined in Table 2.

Do not rely solely on an intoxicated patient’s word that he or she will not self-harm because such statements may not represent the person’s sober state of mind. Use caution in basing a release decision on an intoxicated person’s statements.

Furthermore, do not rely on “no-suicide” contracts. They do not guarantee that a person won’t attempt suicide, and they will not provide legal protection if the patient commits suicide after being released from your care.2

What to document. After completing your evaluation, specifically document:

  • that a suicide risk assessment was conducted
  • what risk factors were present
  • interventions to address those risk factors
  • the level of risk determined (minimal, moderate, or high)
  • factors that may protect the patient against suicide.

Protective factors include a desire to live for their family or children, strong psychosocial support in the patient’s life, and the removal of an acute stressor associated with suicidal thinking.3

Table 1

Evaluating suicidality: Sample questions to ask

Do you wish you were dead?

Do you have thoughts about harming yourself?

Do you have an actual plan as to how you would harm yourself? If yes: Have you taken any steps to enact that plan? If so, what were they?

Have you ever attempted suicide before? If yes: What stopped you from enacting this plan? What do you think would keep you from acting on this plan in the future?

Has anyone in your family or close to you committed suicide?

How close have you come to killing yourself?

Table 2

Sources of information to assess suicidality in the ER

Police reports on circumstances that led the patient to come to the ER

ER nursing/physician notes

Family members’/friends’ statements regarding reasons for the patient’s visit

Observations by anyone assigned to monitor the patient while awaiting your consult

Laboratory and physical exam findings, particularly related to substance use and/or self-injurious behavior

Psychiatric records at that facility

Concerns expressed by anyone responsible for the patient if he/she is released from the hospital

ER: emergency room

CASE REPORT: Paranoid and armed

Mr. J, age 21, is brought involuntarily to the psychiatric ER by police. His mother reports he was locked in his room with a gun, claiming “the FBI is going to kill me.”

Mr. J’s mother tells the ER psychiatrist that her son has schizophrenia, paranoid type, and stopped taking risperidone, 3 mg/d, 4 weeks ago. She explains that Mr. J sometimes “hears voices” whispering to him that his medications are poison and to not trust his family. She states that Mr. J also abuses alcohol and methamphetamine and has 2 prior arrests for assault with a deadly weapon.

She adds that Mr. J now believes his family is working with the FBI to have him placed in a “secret detention camp.” Mr. J’s mother found a loaded pistol in his room and is “scared of what might happen.” During your psychiatric interview, Mr. J appears frightened and paranoid and provides only minimal answers to your questions. He clenches his teeth while staring intently at you.

Evaluating danger to others

There is good reason to be concerned that Mr. J might behave violently, and you likely have sufficient information to hospitalize him. When creating a long-term violence risk prevention plan, divide the concept of dangerousness into 5 components:

  • magnitude of potential harm
  • likelihood that harm will occur
  • imminence of harm
  • frequency of dangerous behavior
  • situational variables that promote or protect against aggressive behavior.

Review a patient’s history of violence because this is the single best predictor of future violent behavior.4 Criminal and court records are particularly useful in evaluating the person’s history of violence. Table 3 provides sample questions for eliciting information about a person’s history of violence when records are not readily available.

A person who has used weapons against others may pose a serious risk of future violence. Ask patients whether they own or have ever owned a weapon. In our experience, the recent movement of a weapon—such as transferring a gun from a closet to a nightstand—is particularly ominous in a paranoid person. The greater the psychotic fear, the more likely a paranoid person is to kill someone he misperceives as a persecutor.

Drugs and alcohol are strongly associated with violent behavior.5 Most persons involved in violent crimes are under the influence of alcohol or drugs at the time of their aggression.6 Stimulants such as cocaine, crack, amphetamines, and phencyclidine are of special concern. These drugs often are associated with feelings of disinhibition, a sense of power, and paranoia. The violence linked with cocaine use differs by gender: men are more likely to perpetrate violent crimes, whereas women are more likely to be the victims of violence.7

Table 3

10 questions to ask patients about a history of violence

What is the most violent thing you have ever done?

What types of violent behavior have you engaged in?

What is your understanding of why this violence occurred?

Who was involved in prior violent incidents?

Have you ever been arrested for any type of violent act?

Have you ever been intoxicated at the time you were violent?

Were you experiencing mental health symptoms when violent?

What is the greatest degree of injury you inflicted on someone else?

What weapons have you used when violent?

Have you ever been a victim of violence?

Mental illness and violence

Studies examining whether individuals with mental illness are more violent than the non-mentally ill have yielded mixed results.8,9 In a study of civilly committed psychiatric patients released into the community, most mentally ill individuals were not violent.10 Although researchers noted a weak relationship between mental illness and violence, violent conduct was greater only when the person was experiencing acute psychiatric symptoms. Subsequent research suggests that individuals with schizophrenia may have increased rates of violence even when not experiencing active signs of their illness.11

Psychosis. In paranoid psychotic patients, violence often is well planned and in line with their false beliefs.12 These patients usually direct the violence at a specific person they perceive as a persecutor. Paranoid individuals often target relatives or friends. In addition, community-dwelling paranoid persons are more likely to be dangerous because they have greater access to weapons than institutionalized patients.12

Carefully inquire about hallucinations—particularly auditory ones—to determine whether their presence increases the person’s risk to commit a violent act. Patients with schizophrenia are more likely to be violent if their auditory hallucinations generate negative emotions (anger, anxiety, or sadness) and if the patients have not developed successful coping strategies.13 Although most patients ignore violent command hallucinations to harm others, the presence of command hallucinations may increase the likelihood of behaving violently,14 particularly if:

  • the voice is familiar to the person,15 and
  • the person has delusional beliefs associated with the hallucinations.16

Depression. Individuals who are depressed may strike out against others in despair. After committing a violent act, a depressed person may attempt suicide. Depression is the most common psychiatric diagnosis in murder-suicides.17 Patients with mania often engage in assaultive or threatening behavior, but serious physical violence is rare.12 Patients with mania most commonly exhibit violent behavior when they are restrained or have limits set on their behavior.18

Antisocial personality disorder (APD). Violence by those with APD often is motivated by revenge or occurs during a period of heavy drinking. Violent behavior by these persons frequently is cold, calculated, and lacks emotionality.19

In addition to DSM-IV-TR personality disorders, be familiar with the psychological construct known as psychopathy. Cleckley20 used the term psychopath to describe a person who is superficially charming, lacks close relationships, is impulsive, and is primarily concerned with self-gratification. Hare et al21 developed the Psychopathy Checklist-Revised as a validated measure of psychopathy in adults. Psychopathy is a strong predictor of criminal behavior and violence among adults.22

Affect. Individuals who are angry and lack empathy for others are at increased risk for violent behavior.23 Also observe the patient for physical signs and symptoms of changes indicating incipient violence. Berg et al24 noted that signs of imminent violence can include:

  • chanting
  • clenched jaw
  • flared nostrils
  • flushed face
  • darting eyes
  • close proximity to the clinician
  • clenched or gripping hands.

Asking patients if they are experiencing homicidal ideations may not always elicit important information regarding a patient’s potential thoughts about harming others. For example, in persons who report feeling persecuted, ask what they would do if they came face-to-face with the individual they fear. Some patients may report that they would attempt to avoid all contact to minimize their personal risk. Others might feel a need to make a preemptive strike for protective purposes. In neither situation would the patient have reported experiencing homicidal thoughts.

Static vs dynamic factors. When organizing strategies to decrease risk factors for violence, distinguish static from dynamic factors. Static factors include demographic information and history of violence. Dynamic factors, which are subject to change with intervention, include access to weapons, psychotic symptoms, active substance use, and a person’s living situation.

Organizing a chart that outlines known risk factors, interventions to address dynamic risk factors, and the status of each risk factor/intervention may be helpful. Table 4 provides an example of such a chart for Mr. J. This approach can help you develop a violence prevention plan that addresses each patient’s combination of risk factors.

Table 4

Sample violence risk management chart for Mr. J

Did you miss this content?
Botulinum toxin for depression? An idea that’s raising some eyebrows