Mitigating Pediatric Inpatient Aggression: A Quality Improvement Initiative

An interdisciplinary quality improvement team consisting of unit leadership, nursing staff, medical providers, psychology, a Psychiatry Consultation Liaison Team, the Department of Public Safety, and members of the quality and safety team met to discuss our SMART aim, determine the scope of the problem, and discuss possible next steps and plan-do-study-act (PDSA) cycles. The team found that the risk for aggressive behavior within the pediatric inpatient population was widespread and was not confined to 1 particular unit or perceived more strongly by any particular discipline. To assess the efficacy of our project, we aimed to increase staff event reporting of safety events related to patient aggression.

Our organization is a member of the Children’s Hospitals’ Solution for Patients Safety (SPS) national network. As part of our SPS work, a multidisciplinary employee safety team was created.¹¹  We partnered our quality improvement effort with our unit-based performance program. Our institutional review board does not require approval for quality improvement projects because the activity is not research. Our organization is a children’s hospital within a larger adult teaching hospital in the Northeast.

Our quality improvement team created a key driver diagram (Fig 1) to display our SMART aim, drivers, and improvement plans. PDSA cycles were used on the basis of the Model for Improvement framework.¹²  Through action-oriented learning practices, the impact of each test of change and PDSA cycle was evaluated to increase the success of each change in practice.¹³ 

The interdisciplinary quality improvement team convened to address the need for resources to safely care for aggressive patients. Given the complexity of the care, the Director of the Psychiatry Consultation Liaison Team was a key expert consultant on the team. The team met on a monthly basis and defined 3 main tools to meet our aim: a risk screening tool, a clinical best practice guideline, and the behavioral personal protective equipment cart for staff. The team felt the development of the cart was a direct result of staff reporting the need for improved and readily accessible safety equipment. The team also learned from the staff who report safety events that equipment improvements increased staff confidence that solutions really do come directly from staff reports. Event reporting really drives and informs our quality improvement. In addition to interventions that targeted more reporting of safety events directly, we also used quality improvement methodology to improve the response in how our center worked to make interaction with patients with high risks of aggression safer and more reliable.

The Pediatric Aggressive Behavior Risk Screen tool was developed by the Psychiatry Consultation Liaison team by using a literature search to recommend the best screening questions. Feedback received from bedside RNs revealed discomfort with the use of words in the questions such as “aggressive” and “violent” due to concern that they may offend parents. The questions were edited several times by members of the interdisciplinary team to address any negative connotations in the language and support nursing comfort in asking the questions.

The Pediatric Aggressive Behavior Risk Screening tool was then vetted by family navigators associated with the developmental and behavioral pediatrics clinic. The family navigators supported the questions but recommended that they be asked in a different order. The tool was piloted with paper documentation on 1 general care unit. Subsequent PDSA cycles involved expanding the pilot screening questions to other units. The quality improvement team received written feedback from the nurses. The team optimized the entire behavioral health and wellness section of the electronic medical record admission navigator to ensure a streamlined approach and consistency with the language style used.

The risk screen tool was built into the electronic medical record and was accompanied by an educational tip sheet on how to use the screening tool. A notification banner was then created as a visual cue for staff caring for patients who screened at risk for aggression. Screening tool completion is mandatory for all patients >2 years old on admission. Reeducation for nurses was provided with a comprehensive online educational module for the use of the aggression risk screening tool, and target compliance was tracked monthly. Data on compliance with the screening tool was collected via electronic medical record audit reports.

The mobile behavioral personal protective equipment cart was designed with a focus on preventing serious staff injury from targeted behaviors. Contents of the behavioral personal protective equipment cart were determined through benchmarking with institutions also engaged in the SPS cohort. The cart contents are aimed to match the anticipated patient behaviors with the concern for what might cause injury to the employee.

The bin on the top of the cart provided a safe storage location and a reminder to remove loose items that could be used to pull or grab the employee. Loose items were identified as hospital identification tags, stethoscopes, neckties, and necklaces. The splash guard was a disposable, full face shield to protect the employee from patient spitting. Baseball hats were included to provide additional protection from hair pulling. Industrial strength Kevlar sleeves and gloves were available to prevent serious employee injury from patient biting, scratching, pinching, and/or grabbing.

The mobile behavioral personal protective equipment carts were deployed in the 2 highest-risk units: the pilot unit and the pediatric emergency department. A third roving cart was housed with the 24-hour clinical resource nurse team for rapid deployment if a need was identified on another clinical unit. The carts were placed outside of an at-risk patient’s room when a safety need was identified.

The need for a standardized approach to patients at risk for aggression led the team to begin formulating a guideline for the care of “at risk” patients that would include identifying risk factors, mitigating triggers, and providing resources should behavior escalate. The interdisciplinary team quickly found that, depending on nursing and provider familiarity, comfort, and resources, the approach could vary dramatically. The creation of the standard care bundle was instituted on the basis of evidence-based recommendations to reduce the variability and established expectations.^14–20 

The guidelines for managing aggression emphasize the use of deescalation methods as a way for providers to help minimize threats of harm and maintain the safety of the patient, staff, and others in the area. Creating a safe, calming, and nonthreatening environment is essential in initiating treatment strategies for managing behavioral instability. The deescalating techniques were derived from evidence-based guidelines for preventing and minimizing potential agitation and aggression in patients.²¹ 

Our team created a care guideline (Fig 2) to bring more awareness to partnering with parents and providers who may be more acquainted with the patient and their baseline behavior. If there was not a preexisting behavioral care plan, the care guideline directs staff and providers to use standard deescalation techniques and safety measures while also prompting the primary provider team to work with available resources and consults to establish an individualized behavioral plan.

The team completed a literature search for current practices and publications regarding a psychopharmacologic approach to managing aggression in pediatric patients. Much of the available literature is directed toward youth receiving treatment on an inpatient psychiatric unit. Relevant literature was reviewed in collaboration with a review of pediatric psychopharmacology resources.^14–20 

Pharmacologic agents were selected on the basis of available practice data, safety profile, route of administration, and efficacy. They were further categorized into mild, moderate, or severe aggression with specific explanations of each to help guide the clinician’s decision making. Doses are based on weight, age, the Food and Drug Administration (approval when available), and common practice among child and adolescent psychiatrists. In addition, guidance is provided if there is a need to order an electrocardiogram (for QTc-prolonging agents) and pharmacologic options to treat any dystonic reaction that may arise. The agents and dosing guidelines were reviewed by the pediatric inpatient pharmacists to ensure safety.

These steps were reviewed by the committee until a consensus was reached that optimized operationalization. The goal of the revisions was to make the care guideline most useful in real time at the bedside. We presented the project at our inpatient clinical council meeting and obtained feedback from clinical divisions that would be utilizing the guideline.

The quality improvement team used the standard rules for run charts and controls charts. Our team consulted with an expert in the creation of run charts and control charts. The expert is a fellowship-trained quality improvement scientist. The run chart and control chart rules assisted our team in determining when special cause variation was detected. The expert also advised our quality improvement team to share thoughts on which interventions were likely most associated with the special cause variation we observed in the outcome measure.

Our outcome measure was the number of patient aggression safety events reports related to workplace violence (Fig 3), which was tracked via a prospective time series analysis utilizing a statistical process control chart. Employee reports related to workplace violence mechanisms were tracked on a U-chart. All process measures were tracked on P-charts. Established rules for Shewhart control charts were used to determine if observed changes were due to special cause variation.²²  Our process measure is the compliance of use of the patient aggression risk screening tool (Fig 4). Nursing compliance with the Pediatric Aggressive Behavior Risk Screen tool increased from an initial 27% to compliance consistently >80%. A centerline shift was appreciated in February 2021 (0.83 to 0.89).

The team’s primary outcome measure was improved event reporting, and it was chosen because event reporting systems are well-integrated in all medical centers across the country. Event reports are reviewed daily by the quality team, the nurse manager, and the unit medical director. Our event reporting system software (RL Solutions™) is used by many other academic medical centers in the country. Our team feels the data collected in our reporting system is reliable and valid because of the nurse manager and unit medical director double-checking information details in the event report and following up with reporters as needed. Event reports are seen as an essential type of safety monitoring tool.

With the quality improvement project, we were able to achieve our outcome measure and SMART aim to increase staff reporting of events for verbally and/or physically aggressive patients by 10% over 12 months. The multiple PDSA cycles were successful in increasing awareness, prioritizing workplace safety, and providing resources when caring for aggressive pediatric patients. The process measure that tracked the use of the pediatric patient aggression risk screening tool also improved over the course of the project.

Throughout our project, we met some resistance and hesitation surrounding the topic and labeling of pediatric aggression. We had assumed that, based on the clinical environment on the unit, staff would be accustomed to addressing and calling out patient aggression and staff safety risks. However, during the PDSA cycles focused on developing the screening tool, we discovered that unit staff expressed concerns about approaching patients and families on the topic. Discomfort with discussing the topic of aggressive behavior in pediatric patients may have contributed to the underreporting of related workplace violence events at the outset of this project.

The staff injury and occupational health reporting system is a shared electronic tool for our enterprise and affiliated university campus. The aim of this project was to support staff reporting to increase total reporting. Staff members have expressed concerns to our team about mandatory reporting fields. The team listened and provided education on why reporting is helpful for system improvements. The team’s aim was to make the reporting process less cumbersome and easier to complete during a clinical shift. The team worked on improving the integrated databases to increase event reporting and help recognize the full scope of workplace violence at our institution.

The safety event reporting rate has steadily increased throughout the project and informed our work to prevent and reduce employee injury. The team aim, theory, and rationale are, when staff members know that we can develop solutions for aggressive patients, event reporting is encouraged and increased. The aim to increase reporting highlights the impact of promoting greater awareness of pediatric aggressive behaviors and the importance of reporting to enhance our safety culture. Another benefit from increased reporting is the ability to schedule debriefings shortly after events are reported. Timely debriefing reduces the loss of recall of event details.

Beyond increasing event reporting, we aimed to increase the resources available for the management of aggressive pediatric patients. During a case review of an aggressive incident, an attending provider expressed discomfort with ordering a psychotropic medication even after consulting with an attending pediatric psychiatrist. This reveals a lack of awareness and comfort among providers with the optimal medication ordering processes to manage aggressive behavior.

Perceived biases and self-preconceptions of ordering psychotropic medications such as Haldol, may have contributed to hesitancy to address aggressive behaviors with medications. This knowledge gap about medication management was consistent with a national survey of pediatric hospitalists.¹⁰  Education around appropriate pharmacological management should be considered and incorporated into training programs and employee onboarding.

Our proactive approach to managing aggressive behavior in pediatric patients has increased staff comfort and confidence in handling these challenging situations. Our findings noted improved compliance with the aggression risk screening tool and increased reporting of safety events. We conclude that the aggression management flowchart was associated with increased awareness and the standardization of the management process.

Additional barriers are present in the pediatric health care setting that further contribute to an inadequate reporting culture. It is not uncommon for pediatric patients to exhibit aggression as a part of an underlying medical condition or developmental age. There may be guilt for reporting aggression from a child. Staff may believe that this behavior is expected when working with some patient populations.²³ 

Most of the reported events involved some harm to the employee. It is likely that events that did not involve harm to the employee were significantly underreported. Efforts should be focused on strengthening a workplace safety culture environment where aggressive events are taken seriously.^24,25 

If events are not reported, we cannot learn from them and make a safer environment for both patients and employees. This project was essential to undertake because of an increasing number of aggressive patients leading to multiple staff injuries. There are often feelings of moral distress and burnout when caring for patients with aggressive behavior. Our interdisciplinary quality improvement team is proud of our approach to the issue of pediatric patient aggression and hopes that it is helpful for other institutions.

The study has 3 main limitations. The first limitation and weakness of our project is the primary outcome measure relies on the self-reporting of events. The second limitation is that our process measure was dependent on documentation and electronic medical record chart review. The screening tool might have been used but not documented. A chart review of the electronic medical record would not detect this. The third limitation is that the team initially used annotated run charts and not control charts that allow the detection of special cause variation. Although we could not control for the first and second limitations, we were able to address the third limitation with the addition of control charts midway through the project.

We plan to gather additional data around the utilization of the resources created in this project and expand additional resources aimed at supporting employees facing workplace violence because of aggressive pediatric patients. Specifically, our children’s hospital organization leadership recommends that our team develop and implement a Behavioral Emergency Response Team to further support our employees. The Behavioral Emergency Response Team will provide infrastructure and standardization of response for behavioral emergencies.

As behavioral events and patient aggression continue to climb within the health care setting, hospitals are tasked with developing a response. There is still much work to be done in the area of how to best manage patient aggression, including both the screening and reporting of the events. The tailored response to these behavioral events based on risk screening and timely reporting is essential. Our team has made progress in the areas of screening at-risk patients, increased reporting, and the development of proactive strategies in our PDSA cycles.

This quality improvement project has included continuous evaluation and feedback from key stakeholders. We have optimized the screening and reporting processes as well as increased support from additional departments such as public safety and psychiatry. Team members now understand the benefits of this work to improve our culture of staff safety, which leads to increased compliance with the patient aggression risk screening tool and improved event reporting.

We thank the Solutions for Patient Safety team for providing support for our interventions. We also thank Daniel Castillo, MILS, and the medical librarians for assistance with references. The team acknowledges Tina Sosa, MD, who was instrumental in creating our control chart and with significant edits to the paper. Although she is new to our organization and was not part of the original unit-based performance team she has been active in our quality scholarship since her arrival in August 2021.