Clinical Monograph: Aggression & Violence

An Evidence-Based Medicine (EBM) Monograph
for Psychosomatic Medicine Practice

View the entire Aggression & Violence monograph as PDF

Evidenced-Based Medicine Summary

On this page:		Introduction  •  Treatment  •  Quality Indicators
On other pages:		References  •  Appendixes:  A  B  C  D  E

INTRODUCTION

Objective and methods

This monograph summarizes current knowledge related to the diagnosis, epidemiology, and management of aggression and violence in the general hospital. The monograph is modeled after previous guidelines provided by the Evidence-Based Medicine Subcommittee of the Academy of Psychosomatic Medicine and the European Association of Psychosomatic Medicine (EAPM). It is based on a systematic review of clinical trials and expert opinion. The quality of the evidence discussed in this monograph is graded as ‘high’, ‘moderate’, ‘low’ or ‘very low’, following the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system, which was developed by the Cochrane Collaborative (www.gradeworkinggroup.org/intro.htm) Readers are encouraged to consult the recommended readings for more detailed information (Appendix A).

Definition and symptoms

The terms aggression and violence and the related concepts of agitation and hostility have no universally accepted definitions. Being concerned with overt behavior that might have the consequence of causing damage and harm, the terms describe complex behaviors that have dimensions with biological, psychological and social determinants. The following definitions pertaining to these overt behaviors are based on a consensus from the extant literature:

• Impulsivity: A behavioral state characterized by diminished restraint to action in response to some stimulus.
• Agitation: A state of emotional arousal or motor restlessness associated with purposeless activity.
• Aggression: Physical violence toward people or objects or verbal threats and intimidation (1).
• Violence: Overt physical aggression that has the potential consequence of physical harm to another person or object.

Risk factors

Aggression and violence are often classified according to whether the behavior is considered to be impulsive (associated with a high degree of affective arousal) or predatory (planned and not associated with high levels of affect).

A variety of psychiatric diagnoses are associated with either repetitive or acutely elevated risk for aggression and violence. When confronted with episodes of aggression or violence within the general hospital, the consultation psychiatrist may be called upon to make a specific diagnosis as a basis for understanding and intervention.

Prevalence and incidence

Violence is a major public health problem worldwide. Each year, millions of people die as the result of injuries due to violence. Many more survive their injuries, but live with permanent disabilities. Violence is a leading cause of death among people aged 15-44 years worldwide, accounting for 14% of deaths among males and 7% of deaths in females (2). The Centers for Disease Control (CDC) maintains a comprehensive surveillance system for monitoring violent deaths in the US. The National Violent Death Reporting System (NVDRS) is operative in 18 states. The CDC estimated that there were 18,500 homicides in the US in 2006 (3). While affecting individuals at each stage of life, homicide is the third leading cause of death among those ages 15-24, after accidents and suicide (4).

Workplace violence is an occupational hazard drawing increasing attention. Healthcare facilities are among the more frequent sites for violence. Between 1993 and 1999, there were 1.7 million episodes of workplace-related violence each year in the United States, and in 12% of these episodes, the victim was a health care worker (4). The occurrence rate is such that surveys in general hospitals reveal that half of surveyed employees have witnessed patient or visitor violence within the prior year, 11% within the prior week (5).

Consultation psychiatrists may be called upon to assist with recommending methods to predict or manage such events.

There are no available surveys of the incidence of aggression or violence among patients in general hospitals.

Diagnosis and risk factors

In epidemiological studies, aggression is often associated with psychiatric diagnoses (6). Longitudinal data collected during the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) indicates that mental illness alone does not predict future violent behavior. Future violence instead is associated with various clinical factors (substance abuse and perceived threats), demographic factors (younger age, male sex, less than high school education, and lower income), historical factors (past violence, juvenile detention, physical abuse, and parental arrests), and social factors (recent divorce, unemployment, history of victimization). The strongest predictor of future violence is the co-occurrence of serious mental illness and substance abuse (7). Sheridan, et al (8,9) describe that patient and staff conflicts, especially those that involve enforcement of rules, denial of requests, and involuntary admission or transfer can precipitate aggressive acts and violence.

A variety of psychiatric disorders are associated with either repetitive or acutely elevated risk for aggression and violence, such as psychosis, mania, delirium, and some personality disorders.

Psychotic Disorders:

Schizophrenia
There is an extensive literature documenting increased risk for violent behavior among individuals diagnosed with schizophrenic disorder. Moderate quality evidence from epidemiological studies links violent episodes with the presence of some positive psychotic symptoms. There is low quality evidence suggesting that some symptoms, such as persecutory delusions (10) and command hallucinations (11) increase risk of violence. The presence of negative symptoms may decrease risk of violence (12). There is moderate quality evidence that the majority of risk for serious violence in this group is due to the frequent co-occurrence of substance use disorders (13). Patients with high levels of hostility and low levels of insight are at greater risk for violent offending (14, 15).

Bipolar Disorder:

There is low quality evidence that the presence of bipolar disorder is a predisposing factor to violence (with odds ratios reported between 2.0 to 9.0). While severity of illness or symptom profile does not predict aggressive and violent behavior the largest amount of variance is due to presence of co-occurring substance use disorder (16, 17).

Personality Disorders:

Antisocial personality disorder
There is high quality evidence for linking violent episodes to disorders associated with a lack of empathy or concern for others. Individuals with these characteristics are overrepresented in cases of purposeful and directed violence (18, 19).

Borderline personality disorder
The diagnostic criteria for Borderline Personality include the presence of impulsivity, affective instability, inappropriate anger or deficits in emotional regulation. Paranoid ideation and brief psychotic episodes have been described in this group of patients (20). There is moderate quality evidence that a diagnosis of BPD is a risk factor for elevated risk for violence in clinical and forensic populations. The evidence for violent predisposition in community samples is low (21).

Impulse Control Disorders:

Intermittent Explosive Disorder (IED)
IED is an impulse control disorder characterized by repeated episodes of serious assaultive or destructive acts out of proportion to precipitating stressors. Since DSM-IV TR includes an exclusion criterion for the co-occurrence of another disorder that would better account for this symptom picture, some have argued that this has led to under-diagnosis of this condition because of frequent co-occurrence of personality disorder (22). The DSM-5 criteria for IED does not include such exclusionary criteria. Epidemiological studies suggest that this condition appears early in life (age 14-20) and is underrepresented in treatment samples. Estimates of lifetime prevalence are 5-7% in community samples. (22). There is no equivalent diagnosis present in ICD-10.

Substance Use Disorders:

Aggression and violence are frequently associated with patients who have alcohol and drug use disorders. There is moderate evidence that violent behavior attributed to serious mental illness is accounted for by the coincident presence of substance use.

Alcohol
There is high quality evidence supporting an association between alcohol use and violent injuries reported in many countries (23). There is moderate quality evidence linking the onset of aggressive and violent behavior to the amount of drinking preceding such events and blood alcohol levels at the time of presentation. There is high quality evidence for the relationship between alcohol use disorders and admission to general hospitals for multiple causes (24). Alcohol use has been long recognized to play a role in criminal violence (25).

Stimulants
Use of stimulant drugs, including cocaine, is associated with a propensity for irritability, hostility and aggression, resulting in high rates of interpersonal violence, emergency department and trauma center visits, assault, weapons charges and increased public health and safety burdens. (26,27). Up to a third of such users presenting to emergency departments require sedation for agitation and aggression. High rates of aggressive behavior have been reported in community samples of amphetamine users. (28).

CNS Depressants
There is moderate quality evidence that aggressive behavior may follow the use of benzodiazepine drugs in persons with impulse control or personality disorders (29).

Anabolic Steroids
There is low quality evidence for development of irritability, aggression and violent behavior in individuals using anabolic steroids to enhance athletic performance or in body-building (30,31).

Developmentally Disabled:

There is moderate quality evidence demonstrating increased prevalence of significant aggressive and challenging behavior in persons with developmental disability. Over half of this population engages in some form of aggression. Only a small number is responsible for frequent or severe acts (32). There is moderate quality evidence suggesting that half of patients are involved in overt aggressive acts. Such behavior is often directed at healthcare personnel and may lead to considerable caretaker burden in all settings (33).

Delirium:

There is high quality evidence that the presence of the attention deficits, paranoia, illusions and hallucinations can result in reactive aggression in patients with hyperactive delirium. (34).

Dementia (Neurocognitive Disorders – moderate and severe):

There is high quality evidence that psychotic symptoms such as hallucinations and delusions are common among those suffering with dementing illnesses with rates of 30 to 50% in those with Alzheimer’s and other dementias (35). Aggressive behaviors are frequent symptoms among patients with dementia. Symptoms such as hitting, kicking, biting, pushing are very commonly associated with cognitive decline and occur more frequently (70-100%). These symptoms are particularly common in patients with moderate to severe neurocognitive disorders in inpatient settings. Aggressive symptoms are the cause of considerable caregiver burden and risk of injury to patients and staff (36).

Brain Disorders:

Moderate evidence suggests that lesions that affect frontal lobe function including tumors, vascular disorders or traumatic injuries are associated with disinhibition resulting in aggression and violence (37).

Traumatic Brain Injury:

While there is moderate quality evidence for agitation and aggression in the acute phase of traumatic brain injury (10-34%), it is predominantly verbal, characterized by anger and threats of violence rather than physical aggression. Violent acting-out is not common. Significant correlates for violence include impaired psychosocial functioning and new-onset major depression (38). In case studies, lesions of the hypothalamus and amygdala have been associated with rage attacks (39,40).

Epilepsy:

While agitated and aggressive behavior has been observed among patients with seizure disorder there is low quality of evidence supporting an association between idiopathic epilepsy and elevated levels of aggression or violent behavior (41).

TREATMENT

Acute agitation and violence when diagnosis is unclear

Intervention in situations involving aggression and violence begins with an understanding of the nature of any provocation and a diagnostic assessment of factors that might lead to altered arousal or impaired impulse control. A detailed assessment of provocative factors that produce frustration or anger can often suggest methods of intervention. Evidence for effectiveness in reducing aggression and violence in the acute care setting has been limited by the obvious difficulty in obtaining informed consent to participate in such studies.

Non-pharmacological treatment
Recommendations provided here are based on expert consensus in the management of acute agitation (42).

• De-escalation of agitated patients

  Clinicians who have worked with patients who present with agitation to emergency departments have developed the following principles published as guidelines for management of patients recognized at risk for aggressive or violent acting out:

  • Ensure the safety of patients, staff and others in the area.
  • Assist the patient in managing emotional arousal and distress while helping to regain control of behavior.
  • Avoid use of coercion or restraint that tend to escalate agitation.

  If such approaches fail to reduce agitation and threat then the following are suggested:

  • Offer of pharmacologic assistance
  • Offer of a “quiet room” if available
  • Show of force (show of concern)

• Coercive interventions

  There is moderate quality evidence that application of the above methods can result in dramatic reductions in the use of seclusion and restraints within inpatient hospital settings. If trial of these methods fails to result in return of behavioral control coercive methods such as seclusion and restraint may become necessary to prevent harm to patients, staff and others. There is no evidence to support any coercive intervention having advantage over any other.

Pharmacological symptomatic treatment

The aim of pharmacologic treatment is to calm a patient to allow complete evaluation of the causes of a behavioral disturbance. Efficacy for use of agents to achieve “rapid tranquilization” is supported by low quality evidence. This should be undertaken only after consideration of risks of such intervention balanced against risk of harm to patients and others in attendance (43).

Three categories of drugs have been customarily used and studied for control of agitation and aggression in these circumstances.

First-Generation Antipsychotics
Haloperidol is a highly potent antagonist of the D2 receptor approved in oral and intramuscular forms for treatment of schizophrenia. Haloperidol has a long record of efficacy and safety, though the quality of evidence for its use in treating acute agitation or aggression is low (44,45,46). The FDA has issued a safety alert highlighting the risk of prolonged QTc interval associated with the aggressive use of haloperidol in parenteral formulations. There have been case reports of development of Torsade de Pointes, some with cardiac arrest. While these cases have generally involved intravenous use clinicians should be aware of the risk of potentially fatal arrhythmias when parenteral haloperidol is used. Cardiac monitoring and assessment of electrolyte status may reduce risk. Haloperidol should be used with greater caution in patients with QTc greater than 450 milliseconds (47).

Second generation antipsychotics
Moderate quality evidence supports the view that efficacy of the second generation antipsychotics is equal to Haloperidol in the treatment of acute agitation (48). Antipsychotic agents should be used with caution in situations where a psychotic diagnosis has not been established.

Benzodiazepines
There is moderate quality evidence that benzodiazepines such as diazepam, lorazepam and clonazepam are effective in calming the agitated patient. Risks include central nervous system depression. These drugs are often used together with the first or second-generation antipsychotics though the quality of evidence for enhanced efficacy in combination is low   (49, 50)

Agitation and violence where a psychotic diagnosis (schizophrenia, mania) has been established

When patients present with acute agitation or threatening or violent behavior when a diagnosis of psychotic illness has been established the goal of treatment is twofold. These include the calming of the patient to allow complete evaluation of the causative factors and amelioration of the underlying psychotic process.

When severe agitation, threatening or menacing behaviors occur in patients known to have a diagnosis of schizophrenia or mania, initial pharmacological interventions should include antipsychotic agents. There is no evidence that first-generation antipsychotics are any less effective than second-generation agents in managing agitated behaviors. Increasingly, second-generation agents such as ziprasidone or olanzapine are being utilized in this context (51)

Prevention and Extended Treatment

Psychotic Disorders – Schizophrenia
Antipsychotics are the mainstay of treatment for schizophrenia. High quality evidence supports offering clozapine to patients with schizophrenia who present with persistent symptoms of hostility and/or display persistent violent behaviors.  Clozapine has anti-aggressive properties that go beyond its efficacy for psychotic symptoms.  In the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) cohort, second generation antipsychotics (not including clozapine) were not superior to perphenazine with regards to reducing violence over 6 months (52).  In the European First Episode Schizophrenia Trial (EUFEST), olanzapine was superior in reducing hostility in early course schizophrenia compared to haloperidol, quetiapine, or amisulpride (53)

Bipolar Disorder
There is low quality evidence in support of use of quetiapine to control agitated and aggressive behavior in bipolar manic and mixed states (54).

Personality Disorders
There is low quality evidence suggesting a role for atypical antipsychotics and mood-stabilizing agents in patients with borderline personality disorder who demonstrate repeated aggressive or violent behavior (55).

Intellectually Disabled
Evidence is lacking for the effectiveness of antipsychotic medication in reducing repetitive aggression and challenging behavior in those with intellectual impairment despite their common use. (56). A small series utilizing placebo lead-in provides low quality evidence that buspirone may reduce aggressive behavior in this population (57).

Intermittent Explosive Disorder
There is modest evidence for reduction in impulsive aggressive episodes following use of fluoxetine 20-60 mg per day (58).
Low quality evidence exists from multiple studies for improvement in impulsive aggression with use of several anti-epileptic agents. There is evidence for significant reductions in both frequency and severity of aggression for those taking phenytoin, lithium and carbamazepine and oxcarbazepine, but not valproate or levetiracetam. (59).

Substance Use Disorder
Identification of the presence of substance use disorders can result in referral to specialty care for definitive treatment.

Delirium
Recommendations for treatment of aggressive behavior due to delirium can be found in the EBM Monograph, Delirium (60). The effective treatment of delirium requires identification and management of the medical or toxic factors causing or maintaining the condition. There is expert consensus that orientation strategies are helpful in management of delirium. There is low to moderate quality evidence that antipsychotic agents are effective in the treatment of symptoms of delirium.

Dementia (Neurocognitive Disorders – moderate and severe)
Neuropsychiatric symptoms are commonly associated with Alzheimer’s and other forms of dementia. Evaluation of such complaints requires consideration of precipitants, evaluation of potential painful conditions and the identification of superimposed delirium. Treatment of these conditions can reduce the severity of behavioral symptoms that are challenging. Behavioral interventions should be attempted before pharmacological treatment is considered. There are no FDA approved agents for treatment of agitation, aggression or violence in patients with dementia. While moderate quality evidence exists for modest effects of some antipsychotic agents in controlling agitation and aggression in patients with dementia there is high quality evidence for risks of use of these drugs (61,62). While there is moderate quality evidence for modest effects in improvement of agitation and aggression with risperidone, olanzapine and quetiapine for periods of 6 weeks to 3 months there is moderate quality evidence that advantages can be offset by cognitive decline and increased morbidity associated with use of these drugs (63).

All antipsychotics carry a black box warning that elderly patients with dementia-related psychosis treated with antipsychotic drugs are at increased risk of death.  Most deaths occurring in this setting are from cardiovascular, cerebrovascular or infectious causes (64).
Despite the risks associated with use of antipsychotics in this population these drugs are the only agents with evidence of effectiveness in psychotic symptoms associated with Alzheimer’s dementia. US Federal regulations for nursing homes recommend a trial of discontinuation of antipsychotic drugs after three to six months. There is very limited evidence from clinical trials to support this suggestion. One study of a group of patients for whom this advice was followed point to a high risk of relapse in the months after stoppage in patients for whom discontinuation has been pursued (65).

There is low quality evidence for a modest effect in randomized control trials that memantine reduced incidence and severity of agitation in moderate to severe dementia (66).

Recommendation: In patients with dementia who display symptoms of severe agitation, threatening or overtly violent behavior care should be taken to establish the existence of co-existing delirium or disorders that may be producing pain. Use of antipsychotics to reduce the severity of agitation or violence should be confined to periods of up to six weeks. Regular weekly re-evaluation directed at measurement of efficacy and evaluation of side-reactions including metabolic and cardiovascular effects should be part of antipsychotic management with such patients. The majority of patients with moderate or severe neurocognitive disorders lack capacity to provide informed consent. Risks and benefits of pharmacological interventions should be explained to alternate decision makers and consent for specific interventions should be obtained and documented.

Traumatic Brain Injury
There is low quality evidence that use of antipsychotics after brain injury results in declines in cognitive function and slowing of rehabilitation (67,68). Low quality evidence exists for the use of quetiapine for control of agitation that threatens progress in rehabilitation (69). Evidence that carbamazepine or valproate is effective in the management of agitation and/or aggression following brain injury is lacking.

Evidence to support pharmacological intervention to prevent or treat aggressive behavior consequent to traumatic brain injury is lacking. Moderate quality evidence supports aggressive treatment of depressive symptoms using SSRIs (70).

Over thirty studies provide moderate support for the efficacy of propranolol in reducing repetitive violent behavior in patients with a variety of chronic mental illnesses, developmental disability and organic brain disorders such as traumatic brain disorder (71).

TABLE I – Medications used for treatment of agitation (off label) with FDA-approved indications
Name	Dose	FDA Indication
Haloperidol (po, syrup, im, iv)	0.5–2 mg qAM, qHS	• Hyperactive behavior • Problematic behavior in children • Schizophrenia
Chlorpromazine (po, syrup, im, iv, supp)	12.5–50mg bid/tid	• Behavioral syndrome • Hyperactivity • Schizophrenia
Olanzapine (po, wafer, IM	2.5 mg qHS (max 30–40 mg/day)	• Agitation in Bipolar disorder • Agitation in Schizophrenia
Ziprasidone (po, IM)	10mg im q2hr, max 40mg/day	• Agitation in Schizophrenia
Risperidone (po, syrup, tab)	0.5–2 mg q12hrs	• Irritability associated with autistic disorder • Behavioral Syndrome (MR)
Aripiprazole (Po, syrup, IM)	2mg po, 9.5mg im	• Psychomotor agitation in autistic dsd/bipolar dsd/Schizophrenia
Loxapine Inhaler	10mg inh/24 hours	• Agitation with bipolar I disorder

 

QUALITY INDICATORS

Organization of detection and management

The available evidence does not allow for a recommendation of a specific model of intervention for violent and aggressive behavior.

Specific quality indicators

The following indicators provide insight into the quality of the organizational response to potential violent acts.

• The presence of an institutional protocol for response to violent and aggressive behaviors.
• Presence of documentation justifying use of antipsychotics in treatment of aggression and violence in patients without a psychotic diagnosis.