Aggression and Prevention of Use of Seclusion and Restraint in Inpatient Psychiatry

Nationally, an increase in violence has been seen in hospital inpatient psychiatry units and in general medical settings (1, 2). Identification and skilled management of aggressive patients are major safety concerns in psychiatric facilities. Among civilly committed patients, it is likely a confluence of factors, such as staff variables, structure (rules and routines), features of the physical environment, mental health legislation, and external influences (e.g., family stressors) that contribute to patient aggression (3).

It is important to note that patient–staff interactions provide a complex set of circumstances that can trigger patient aggression and are associated with 39% of violent and aggressive incidents (4, 5). Although their causes are complex, aggressive incidents are often evaluated from one dimension and dichotomized as being present or absent, rather than being seen as situational variables that can affect the outcome of an aggressive incident.

Although not new in the mental health care setting, de-escalation techniques in general health care settings have been proposed. An educative program aimed at renewing knowledge and skills in de-escalation is a timely project (6). Current de-escalation strategies include using patient-chosen interventions for managing aggression, empowering the patient to work with staff to decrease aggression, using creative techniques and humor, learning to listen and identifying triggers for each patient, learning cultural differences, and extending empathy for the individual (7).

In 2007–2008, the Joint Commission released rules specifying training requirements for personnel who could order seclusion and restraint, requirements for reporting seclusion- and restraint-related deaths, and interpretive guidelines for the use of seclusion and restraint (8). In 2009, the American Psychiatric Association’s Committee on Patient Safety, among other groups, identified the use of seclusion and restraint as an area of high priority in SAFE MD: Practical Applications and Approaches to Safe Psychiatric Practice, a handbook for psychiatric safety, after some of its members had noted published accounts of seclusion-related deaths (9). As well, the American Psychiatric Nurses Association made a commitment to the reduction and ultimate elimination of the use of seclusion and restraint in 2014. Their position statement “recognizes that the ultimate responsibility for maintaining the safety of both individuals and staff in the treatment environment and for maintaining standards of care in the day-to-day treatment of individuals rests with nursing and the hospital leadership or behavioral health care organization leadership that supports the unit” (10). Increasingly, in accordance with the current standards, hospitals committed to reducing or eliminating the use of seclusion and restraint as a way of managing patient aggression have to examine the organization of personnel and regular assessment of aggressive patients, followed by use of less restrictive measures (11).

Before looking at tools and evaluations of patients and staff practice, we should ask ourselves as mental health professionals, What are we concerned about? Keeping staff and patients and the environment safe? Promoting safer and quicker discharges? Preventing the use of seclusion, restraint, and unsafe holds? Promoting a culture of safety and multidisciplinary teamwork, transitions, and handoffs? Developing tools that are useful for the location, milieu, and case mix of the institution?

Assessment Tools and Practices

The literature on reducing patient aggression on adult inpatient units emphasizes two factors: first, the importance of early assessment and identification of patient characteristics that may be indicative of aggression and, second, strategies to reduce the use of seclusion and restraint on these units using systems measures or protocols. The literature does not report, to my knowledge, the use of a documentation tool that is systematic and proactive, avoids recommending the use of seclusion and restraint, gathers research data, and satisfies regulatory practices, other than the Phipps Aggression Screening Tool described later in this article (12).

Research suggests that the use of seclusion or restraint prevents staff and patient injury and that the attitudes of the staff and hospital administration influence rates of use. However, the use of seclusion or restraint can lead to longer stays that result in more expensive hospital costs that result from the staffing of seclusion rooms and observation of patients (13). No mental health initiatives to reduce adverse events exist in psychiatry that are universally embraced, understood, or taught (14). In the United Kingdom, hospitals are far behind those in the United States in implementing better and more uniform methods to contain aggression and avoid use of seclusion and restraint. Stewart et al. found, for example, wide variation among hospitals in seclusion and restraint use and that time-outs were as effective as seclusion for verbal aggression (15).

To identify predictors within the first day of admission or prior to admission, one must address the following questions: Who is the patient? What are the patient’s personality dimensions? Is the patient an unstable extravert? Will rewards help more than disincentives? What diagnosis was the patient given? If the diagnosis was for a disease, what behavior or behaviors impact this disease? Is the patient going through withdrawal from illicit substances that could be more aggressively treated? What has the patient experienced? Has the patient been a victim of physical or sexual trauma? (16).

Another set of questions to address has to do with self-assessment of the milieu, nursing and physician practice, methods and their refinement, and protocols and their measurement. The team must ask, How do we know we are operating more safely now than we have in the past? How often do we harm patients? How often do patients receive the appropriate interventions? How do we know that we have learned from defects in systems of care? How well have we created a culture of safety? How satisfied are our patients with our interactions with them?

To improve care within institutions, caregivers must be engaged, must participate in the selection and development of measures, and must receive feedback regarding their performance. If safety tools are developed, the team must consider the following attributes: Measures must be important to the organization, valid (represent what they intend to measure), reliable (produce similar results when used repeatedly), feasible (affordable and expedient to collect data), usable for the people expected to improve safety, and have universal applicability within the entire institution.

In the United Kingdom, results from a survey of wards identified a number of factors predisposing a ward to experiencing violence: poor staffing, lack of resources, lack of engagement of patients, and patient boredom. Also, poor physical design and occupancy level problems compromised the safety of patients (17).

Factors and Practices Triggering Violence Leading to Seclusion and Restraint

Factors and practices triggering the violent incidents that lead to the use of seclusion and restraint can be broadly grouped. Multiple techniques to reduce the use of seclusion and restraint by diffusing violence were well described by Jayaram and others (13), ranging from “staff-related factors, such as organization, deployment, training, and education. These factors can be managed using strategies such as increased staff-to-patient ratio, communication, collaboration among staff members, postevent debriefings to understand process flow, verbal de-escalation techniques, and mindful distribution of staff on the unit with respect to patient load. Improvements in communication could involve sharing details of hand-offs, talking with and listening to patients, and examining both successful and failed interventions.” Overall goals include improving the staff’s ability to detect precursors of violence and redirect them using diversion techniques and alternative coping methods, engaging in collaborative problem solving by increasing patient participation in the process, and improving medication management with multidisciplinary input (18–22).

Precipitants that set off the behaviors fall into three broad categories: nondirectable behavior (e.g., conflicts with visitors, family members, or peers; issues surrounding smoking privileges, food choices, or the treatment plan), acute psychotic symptoms, or behavior problems accompanying cognitive limitations and disagreements with nursing staff (in that order).

The application of a violence assessment tool that can be quickly and efficiently used, with good interrater reliability and predictive ability, enables staff to implement preventative interventions as quickly as possible to avoid violence and therefore the use of seclusion and restraint.

Studying and debriefing patients using forms can be done by using a coping questionnaire, conducting a risk screening, assessing patient preferences for dealing with agitation, and completing postseclusion evaluations as well as restraint forms focusing on altering preventative treatment plans to suit individual patients. Forms such as the Brøset-Violence-Checklist and the Violence Risk Appraisal Guide are two such instruments (23, 24).

Finally, as solutions are extracted from many models, in the interests of patient and staff safety, programmatic efforts must focus on training staff in the accurate recognition of potential users of seclusion and restraint in the milieu; minimizing the use of these practices by identifying and systematically promoting less restrictive interventions; and debriefing staff, patients, or family members to minimize the negative emotional consequences of use of seclusion or restraint. Staff training for doctors and nurses must be repeated annually to include new recruits and to refresh protocols.

These measures are described by nursing teams from Johns Hopkins University and others from California (25, 26). The Hopkins nursing protocol changes were based on a prevention model to improve public health that emphasized a significant culture change for unit staff (25). This included new primary prevention (i.e., early identification of coping skills, creating a comforting environment), secondary prevention (i.e., using comfort carts and increased staff communication about patients who are having a difficult time), and tertiary prevention (i.e., a formal witnessing program after every seclusion and restraint event) strategies. These practice changes were quite successful, leading to a 75% reduction in seclusion and restraint use from 2005 to 2006 (25).

The Phipps Aggression Screening Tool (12) guides the use of hierarchical interventions, promotes early assessment and intervention soon after admission, and can be used in outpatient settings as well. The article cited above gives a detailed description of the instrument, its use, and outcomes determined by nurses and physicians who have used it both prospectively and retrospectively and will not be repeated here.

A number of factors emerge from a review of the pertinent literature. First, there needs to be a systematic and careful standardized assessment of all acutely ill patients for the risk for violence. Second, all staff need to be regularly trained on the use of any standard assessment instrument. Third, although demographics are not always predictive, younger male patients are more likely to be disruptive or violent. Substance use or withdrawal, prior acts of violence inside or outside of the hospital, prior time in seclusion or restraint, recent incidents of aggression, major mental illness with paranoid symptoms, cognitive limitations, delirium, verbal aggression, difficulty following the ward rules of the milieu, and prior experience of sexual or physical trauma all increase the risk of the patient ending up in seclusion and restraint. Fourth, assessment must be done daily and may be needed at each shift change.

Hierarchical interventions, from the least restrictive to the use of seclusion and restraint (as an ultimate unavoidable measure), must be used. Patients must be carefully monitored and moved out of seclusion as soon as is feasible. Family members must be informed daily of the status of their loved ones. Other measures should be individualized, such as providing privacy or a single room, permitting visits by family members when they are able to come, placing the patient’s room closer to the nurses’ station, and having the milieu manager frequently check on the patient’s status and on ward acuity to support patients’ needs. Assigning to the position of milieu manager a senior nurse who pre-emptively addresses the needs of high-risk patients and does not carry a caseload is a good strategy.

Researchers have noted that a small number of patients are repeat offenders who need to be placed in seclusion or restrained (13, 27). However, it is possible to work with such patients, engage them in a discussion, and through better understanding institute more supportive measures that expedite recovery and discharge. In using the Phipps Aggression Screening Tool, researchers have found that delusions or psychotic symptoms were not the only major precipitant for aggression. Also, the increase in the successful use of lower-level interventions by staff when systematically applied was noted by investigators from the United Kingdom who found that de-escalation techniques for verbal aggression were just as successful as the use of seclusion. Training in several aspects of violence management is clearly valuable in reducing the use of seclusion and restraint (28–30).

Conclusions

In psychiatry, serious adverse events are rare. Standard definitions of what constitutes violence or aggression do not exist. For example, some studies measure verbal violence or threats but others do not. The therapeutic relationship that may affect a patient’s sense of comfort or may mitigate threats cannot be measured. There are variations in institutional standards, expectations, and what constitutes a culture of safety in treatment, despite Joint Commission guidelines. Preventing the use of seclusion and restraint reduces the use of staff and unit resources, significantly decreases the length of the patient’s stay and the associated costs, and promotes safe recovery and discharge of even patients with complex illnesses.

Future work lies in finding an instrument that allows medical personnel in immediate management to assign scores to patients who are imminently aggressive. Researchers have not studied milieu acuity to inform number and type of admissions—from the emergency department or those who are electively admitted—to define a safe threshold to prevent violence. Such work would aim to decrease patient and staff injury while permitting access to safe acute treatment services. Future work should indicate the degree of improvement in behavior associated with each intervention.

Ensuring safety in inpatient services has to do with avoiding, preventing, and mitigating adverse outcomes stemming from the process of treating patients. Safety also emerges from the examination of not just individuals but also the interaction among the components of a process of care. Inpatient psychiatric service safety needs to focus on the epidemiology of error, analyses through observation of error mechanisms and their failures, the development of interventions, the reassessment of reporting, and the evaluation of systems. Last, effectiveness is tested by improving performance through efforts to improve the course of care.

When violence and aggression in psychiatric patients is examined, the source of the problem is studied, as is the process of changing the service system for improvement and refinement of practice. As the system is changed, new flaws requiring further study may be detected.