Working in the medical field does not come without its hazards. Unfortunately, even in the pursuit to provide aid to those in need, those same patients can become agitated and violent. The Occupational Safety and Health Administration (OSHA) stated that 75% of annual assaults in the workplace occur in the healthcare and social service fields. As reported in the National Crime Victimization Survey, healthcare workers face a 20% higher chance of being victimized in the workplace when compared to other workers. Because patient violence can occur in any clinical setting, it is imperative for providers to be prepared in order to minimize the risk of injury to the patient and caretakers. This document will discuss guidelines for harm reduction in the healthcare setting with a focus on patient restraint and seclusion.
The impetus to administer restraint and seclusion protocol is to obviate potential violence and potentiate harm reduction. Hazards to be avoided include both harm to the patient and the caretaker. This danger encompasses both nonviolent and violent risks. Healthcare providers are encouraged to always remain vigilant. However, data suggest that incidents involving the following risk factors may be associated with increased patient agitation diathesis:
- Interpersonal communication
- Waiting time
- Drug and alcohol intoxication or withdrawal
- Metabolic conditions (e.g., hypoglycemia, hypoxia)
- Neurologic conditions (e.g., infection, dementia, stroke)
- Psychiatric disorders (e.g., schizophrenia, psychosis, personality disorder)
Issues of Concern
As many as 50% of providers report being the victim of violence at one point in their careers. In a survey specifically for Emergency Medicine residents and attendings, 78% reported being involved in a violent workplace act within the past year. Furthermore, 4-8% of patients presenting to a psychiatric emergency department are armed.
General Prevention Measures
General prophylactic safety measures to best obviate workplace violence include and prevent the use of seclusion and restraint:
- Security — A survey of 250 Emergency Departments in the United States showed that nearly 77% of hospitals did not have 24-hour security, and only 1.6% used metal detectors. Although healthcare workers are among the highest risk groups of workplace violence, this survey suggests that they remain vulnerable to agitated patients and visitors partly due to limited or absent security. A 2008 study of 3518 surveys from 65 Emergency Department (ED) locations noted that guns or knives were consistently brought into the ED daily or weekly. Security personnel with sufficient training and experience are a principal aspect of violence prevention.
- The efficiency of operation — Long waiting times have been associated with a predisposition to violence in Emergency Departments. A census of approximately 50,000 patients showed that a waiting time of two hours is significantly associated with an increased incidence of violence.
- Warning sign recognition — Training personnel to recognize disruptive patients at the onset of a visit can help raise staff awareness and increase set precautions. A 1997 survey of 517 psychiatric residents indicated that 36% were physically assaulted, and nearly 66% described themselves as undertrained or without training in managing violent patients.
- Access control — A study of emergency department security in 250 hospitals concluded that solely 21% of hospitals controlled ED access during high-risk time periods.
- Alarm systems — Alarms can be used for enhanced patient and staff safety. Some examples of alarm types include pressure-sensitive bed alarms, patient room alarms, and staff panic alarms.
Negative Consequences of Restraint and Seclusion
Healthcare workers should be aware that restraint and seclusion can have significant adverse implications on patients and should be deemed a last resort. It is the duty of healthcare professionals to follow the four basic ethical healthcare principles: autonomy, justice, beneficence, and non-maleficence. In accordance with both beneficence - the act of doing good - and non-maleficence - do no harm, healthcare providers must ensure the administration of restraint is implemented as a last resort. A 2019 meta-analysis on the effects of restraint and seclusion estimated that the precipitation of posttraumatic stress disorder following restraint interventions ranged from 25% to 47%.
Healthcare workers must also be sure to appropriately monitor the patient following restraint and seclusion to avoid deleterious effects such as pressure ulcers, skin breakdown, abrasions, asphyxia, strangulation, incontinence, depression, social isolation, and drug overdose or interaction.
Prior to the interview, all patients need to be searched and disarmed. Metal detectors, as well as the routine practice of gowning the patient, act as non-confrontational strategies to uncover weapons. Once the patient has been appropriately searched, the interview can commence. The evaluation should occur in privacy, but not in isolation. The common practice is for the interviewer to locate themselves between the patient and the door. Ideally, in the case of a high-risk patient, the provider should have proximal access to a panic button.
Recognition of Violence Escalation
Violence history remains the best predictor for future violence. The classic escalation of patient violence progresses from anger, resistance, and finally to confrontation. Signs of impending violent behavior include provocative behavior, posturing, pacing, angry demeanor, and aggressive acts. It is also plausible that violent behavior erupts impulsively and without warning, especially in the setting of organic disease. Alternative tools to assess violence include the following batteries:
- Coburn and Mycyk describe three phases of violence escalation: (1) anxiety, (2) defensiveness, (3) physical aggression. As individuals move through these stages, they typically get closer to violence. It is a useful tool for clinicians to initiate de-escalation techniques at earlier phases to prevent progression quickly.
- The STAMP (Staring, Tone, and volume of voice, Anxiety, Mumbling, and Pacing) tool for use in ED details behavior that can be observed in patients, family, and friends that can be indicative of impending violence.
- The Overt Aggression Scale (OAS) is a rating scale to measure the aggressive behavior of inpatient children and adults. It is divided into four sections, including verbal aggression and physical aggression against objects, self, and others.
Once an agitated patient has been identified, staff must give the patient the opportunity to calm down before physical intervention. Often, agitated but cooperative patients will be amenable to verbal de-escalation. Guidelines recommend an honest and straightforward approach with the implementation of friendly gestures proves most beneficial in the setting of an agitated patient. 10 key features for verbal de-escalation as provided by the American Association for Emergency Psychiatry De-escalation Workgroup include:
- Maintain a distance of two arm's lengths
- Maintain a relaxed and non-confrontational posture
- Establish verbal contact
- Use simple and concise language.
- Identify requests and feelings.
- Actively listen to what the patient is saying.
- Do not be afraid to agree to disagree.
- Set clear boundaries
- Attempt to offer choices
- Debrief the staff and patient
Indication for Emergency Seclusion and Restraint
Following fruitless de-escalation techniques, emergency seclusion and restraint can be indicated. The following list suggests incidences when such measures should be administered.
- Imminent danger to others
- Imminent danger to the patient
- Profound disruption of treatment or damage
Types of Restraints
Always, the least restrictive method necessary to correct the issue should be used.
Physical restraints encompass hand mitts, soft cloth limb restraints, leather limb restraints, enclosed beds, belts, and vests. Ideally, a restraint team should include at least five people, including the team leader. If the patient is female, at least one member should be female to minimize the potential of sexual assault allegations.
Chemical restraint (sedatives/hypnotics/neuroleptics/dissociatives) can be administered alone or with physical restraints. The medication used must have a rapid onset with minimal side effects. The three primary drug classes used are benzodiazepines, first-generation antipsychotics, and second-generation antipsychotics. If the patient is only minimally responsive following the administration of the chemical restraint, it may be fruitful to administer a subsequent psychotropic from an alternate class. The pharmacological agent administered can vary upon the scenario:
Severely violent (rapid tranquilization)
- First-generation or second-generation antipsychotics (e.g., haloperidol [2.5 to 10 mg IM], olanzapine [5-10 mg IM])
- Benzodiazepines (e.g., lorazepam [0.5 to 2 mg IV/IM], midazolam [2.5 to 5 mg IV/IM])
- Combination of first-generation antipsychotics and a benzodiazepines
Agitation from drug intoxication or withdrawal
- Benzodiazepines (contraindicated in the setting of intoxication with CNS depressant)
Agitation due to an unknown cause
- Benzodiazepines (preferred) or
- First-generation antipsychotics
Agitation in a patient with a psychiatric condition
- First-generation antipsychotics or
- Second-generation antipsychotics 
Lorazepam can be given in doses of 0.5 to 2mg IV or IM every 10 to 30 minutes, depending on the severity of agitation. The half-life is 10 to 20 hours. Midazolam works more rapidly than lorazepam; however, it lasts a shorter time period (one to two hours). The dose for midazolam is 2.5 to 5 mg IV or IM and can be given every three to five minutes, depending on severity. The dose for haloperidol is 2.5 to 10 mg IM or IV (although the FDA does not approve IV administration) every 15 to 30 minutes as needed. Haloperidol has an onset of action of about 28 minutes IM and 3 to 20 minutes IV. Olanzapine can be given in a dose of 5 to 10 mg IM with an onset of action of 15 to 45 minutes and a half-life of two to four hours.
Take caution when giving a patient first-generation antipsychotics as they have a propensity to induce extrapyramidal side effects, as well as QT prolongation, which can potentially lead to dysrhythmias like torsades de pointes. If possible, obtain an EKG before administering the medication; otherwise, an EKG should be obtained once the patient has been sedated. Per the US Food and Drug Administration, higher doses and IV administration of haloperidol increases the incidence of QT prolongation. In the setting of a lowered seizure threshold (e.g., alcohol or benzodiazepine withdrawal, epilepsy, anticholinergic toxicity), first-generation neuroleptics should be avoided, as they can further increase seizure diathesis. Regarding benzodiazepines, the clinician should be wary of CNS depression and, less commonly, paradoxical disinhibition.
Generally, heuristic models suggest that the provider moves to an alternate class when the initial attempt is ineffective. An alternative to benzodiazepines and antipsychotics is ketamine. Ketamine is a dissociative anesthetic with minimal adverse effects. Off label uses include management in the setting of excited delirium, acute-on-chronic substance abuse, and when first-line measures have been ineffective.
Active Restraint Monitoring
Document appropriate clinical indication and have a standardized checklist prepared for staff to monitor and supply patient needs effectively. Numerous deaths and adverse patient outcomes have been reported due to inappropriate restraint placement and negligent monitoring. After restraint placement, patients should be reevaluated every hour and moved at regular intervals to prevent sequelae such as pressure ulcers, rhabdomyolysis, and paresthesias. The patient should also be evaluated for medical causes of agitation once it is safe to do so. Generally, patients above the age of 40 experiencing new-onset psychiatric symptoms are more likely to be suffering from an organic pathology, whereas elderly patients are more likely to experience delirium secondary to a medical illness or iatrogenic etiology.
Seclusion is also known as a type of environmental restraint that is used to prevent free movement of the patient and decrease environmental stimulation. It can be used involuntarily or voluntarily, depending on the indication.
There are circumstances when restraint and seclusion are contraindicated. Seclusion is inappropriate if a patient requires constant monitoring. Examples include patients presenting with suicidal ideation, self-injurious behavior, hemodynamic instability, or overdose. Restraint and seclusion should not be used as a means of punishment or convenience.
Generally, restraints and seclusion cannot be administered longer than 4 hours for adults (> 18 years), 2 hours for children and adolescents (9 - 17 years), or 1 hour for children (<9 years) unless state laws are more restrictive.
Removal of Restraints
When the patient is no longer a danger to themselves or others, the restraints should be removed immediately.
Typically patients sign a document, which legally details their agreement to consent to treatment. When the decision to treat becomes involuntary as is the case in most instances of patient restraint and seclusion, the physician is put at risk of common legal ramifications pertaining to topics such as false imprisonment, battery, duty to do no harm, duty to warn, and competence. For this reason, physicians must be sure that their decision to restrain or seclude is fully justified and thoroughly documented to explain the details behind the decision.
To illustrate this topic, picture a scenario where a patient is being emergently evaluated for a serious condition that can alter his level of competency. During the evaluation, the patient is lethargic and unaware of where he is located; he proceeds to become agitated and express a desire to leave. At this point, it is the physician's decision to restrain the patient because the patient is now a danger to himself. All aspects of the decision-making process should be documented to demonstrate the need for restraint. Documentation should include how the situation was emergent, why consent was not able to be obtained, and why the treatment was of benefit to the patient.
The clinician should not permit the patient to leave if he is incompetent and a danger to himself as this becomes an issue of negligence, which is significantly more difficult to defend in court. Additionally, the physician must also be aware of any statement from the patient regarding the impending harm of a third party as there is a duty to warn that party. If the patient leaves without the staff notifying the third party and that third party is harmed, the physician can face legal ramifications. Knowledge of how certain cases can lead to legal consequences can help a physician to appropriately document and take action to avoid liability.
Enhancing Healthcare Team Outcomes
It is of paramount importance to effectively train staff on how to appropriately care for agitated and at-risk patients requiring restraint and/or seclusion. If staff is not appropriately trained, patients and others are placed at a higher risk of adverse outcomes. In addition to staff training, healthcare facilities must be equipped with means to protect individuals such as continuous security presence, efficient operation of the business, access control, alarm systems, availability of means to restrain and seclude. Restraint and seclusion must be thought of as last resort methods to use if de-escalation techniques are ineffective. The patient should be included as much as possible in the decision-making process. It is important for staff to review restraint and seclusion situations to evaluate performance and provide feedback to ensure efficient, effective, and safe medical practice.