Outcomes of a Quality Improvement Program to Reduce Hospital-acquired Pressure Ulcers in Pediatric Patients

Hospital-acquired pressure injuries (PIs) present a significant challenge to pediatric providers. Purpose: The purpose of this quality improvement program was to develop and implement a debrief protocol and to evaluate compliance with and the implementation of a comprehensive prevention bundle to decrease the overall incidence and severity of pediatric pressure ulcers (PUs)/PIs in a free-standing children’s hospital. Methods: As a member of the Children’s Hospitals Solution for Patients Safety national network, a PU Hospital Acquired Conditions (HAC) team was created in 2013, followed by the development and implementation of a PU occurrence debrief tool and discussion guide and implementation of multiple staff educational strategies and a comprehensive prevention bundle. The PU occurrence debriefing occurred within 24 to 48 hours of a PU. Incidence data were collected annually from 2014 until 2017. Results: Compliance on implementation and documentation of bundle elements ranged from 88% to 94%, and PU/PI incidence decreased by 30% from 2014 to 2016 and by 40% in 2017. The overall PU rate was 0.0057 in 2014, 0.0050 in 2015, 0.0036 in 2016, and 0.0023 in 2017; 65% of all PUs were device-related. Of those, >50% were related to respiratory devices, 25% to peripheral intravenous catheters/central lines, 10% to tracheostomies, and 15% to other devices. Respiratory device-related PUs decreased by 50% in the pediatric intensive care unit, by 80% in the neonatal unit, and eliminated completely in extracorporeal membrane oxygenation patients. Conclusion: The debriefing process, debriefing tool, educational programs, and prevention bundle reduced the rate of hospital-acquired PIs in pediatric patients and propagated a culture of safety.

Hospital-acquired Stage 3 and Stage 4 pressure ulcers (PUs) or pressure injuries (PIs), including those in pediatric patients, are reportable and considered “never events” by several national benchmarking organizations.¹ Reductions in reimbursements for health care-acquired PUs have been implemented by the Centers for Medicare and Medicaid Services (CMS) since 2012.¹ The reported prevalence of pediatric PUs ranges from 0% to 43%^2-4; Baharestani and Ratliff¹ reported an incidence of 23% in neonatal units. In a multisite study of >5000 children in 9 pediatric critical care units, the overall PU incidence was 10.2% (range 0.8%–17.5%) across sites.⁵ Curley et al⁶ found a 27% incidence of PUs in a study of 3 critical pediatric care units; most patients (97%) had Stage 1 or Stage 2 injuries.⁶ In a  survey involving >1000 children in a pediatric hospital, McLane et al⁷ reported a 4% rate of hospital-acquired PUs. Razmus et al³reported a PU prevalence of 1.4%, 1.1% of which were hospital-acquired.

New medical technologies have increased the survival rates of extremely preterm newborns and infants with critical congenital conditions such as central nervous system (CNS) injury, cardiac diseases, or mechanical ventilation; respiratory devices, central lines, internal catheters, and tracheostomies contribute to cutaneous injuries such as PI.² Intrinsic factors such as immobility, altered sensation, sedation, inactivity, nutritional deficiencies, and immaturity of the skin also are known to heighten the PI risk.² Few risk assessment tools have been modified for pediatric patients and none have been validated that include medical devices as a risk factor for PI. Although adult practitioners are fairly proficient at managing PI, medical and nursing education on pediatric PIs is generally considered inadequate.¹

So far, no specific approach/tool has been shown to have an impact on staff education/management of pediatric PI. The purpose of this quality improvement project was to develop and implement a debrief protocol and evaluate compliance with and the implementation of a comprehensive prevention bundle to decrease the overall incidence and severity of pediatric PUs/PIs in a free-standing children’s hospital.

Setting. Cohen Children’s Medical Center (New Hyde Park, NY) is a 200-bed, free-standing, quaternary care academic facility that serves patients from around the region, nation, and globe. The 57-bed, neonatal intensive care unit (NICU) and 34-bed pediatric intensive care unit (PICU) are state-of-the-art facilities serving children with the most complex conditions, offering extracorporeal membrane oxygenation (ECMO), therapeutic hypothermia, and innovative cardiac, neurologic, and orthopedic surgery. Extensive bone marrow transplant, cancer, organ transplant, and level 1 trauma patients are routinely treated on the medical floors and intensive care units (ICUs).

History of preventive concern (2012-2013). Children’s Hospitals Solution for Patients Safety (CHSPS) national network was launched in 2011. The goal of this initiative was to prevent hospital-acquired harmful conditions through the use of standard definitions; training in the Model for Improvement and Plan/Do/Study/Act cycles8,9; the creation, implementation, and measurement of event prevention bundles; data analysis; and transparency across the collaborative. In 2012, Cohen Children’s Hospital became the first pediatric hospital in New York state to join CHSPS. The program was awarded a multiyear contract by the CMS as part of its Partnership for Patients initiative, a priority project designed to reduce hospital inpatient harm by 40% over a 3-year period. A new safety program, “Commit to Zero,” that addresses reducing errors to none, was introduced; it includes implementing daily safety briefings, safety behavior education, leadership rounding, a “Great Catch” employee recognition program, development of hospital-acquired condition (HAC) teams, and a safety coach program. In this framework, a PU HAC team was created and a PU Prevention Bundle was developed and propagated throughout the hospital.

Before the team’s formation, surveillance of PUs and data collection were inconsistent. Five (5) random, 1-day studies revealed a PU prevalence of 3% on medical floors, 15% in critically ill pediatric patients, and 8% in the neonatal unit. Prevalence was calculated by dividing the total number of hospital-acquired PUs by the total number of patients surveyed times 100. Total rate of PUs (total PU per year/total number of admissions) was 0.0069 for 2013 (likely underreported because data collection was inconsistent).

PU Prevention Bundle. The PU Prevention Bundle focused on the National Pressure Ulcer Advisory Panel (NPUAP) recommended elements10: skin assessment, repositioning, device rotation, bed support surfaces, moisture management, and nutrition. All nursing staff were required to complete mandatory education modules presented as didactic lectures and computerized webinars with postlecture tests that addressed skin assessment scales, PU staging, risk, prevention, management, and bundle element documentation. Random patients’ computerized records were reviewed for each bundle element. The bundle review sheet, completed and assessed by PU HAC co-leads, revealed inconsistencies in implementation and documentation of bundle elements stemming from suboptimal knowledge of PU prevention strategies, staging, and treatment. According to the subjective opinion of the co-leads after interviewing staff, staff were hesitant to report PUs; thus, reports were not timely or accurate. As a result, opportunities for prevention often were missed.

Evaluation of gaps (2013). In order to accomplish the primary goal of PU/PI reduction, the members of the PU HAC group developed a key driver diagram (see Figure 1), identifying 4 primary drivers: prevention process, accurate recognition, timely and accurate reporting, and change of culture. Didactic lectures and mandatory computer presentations were recognized by the PU HAC co-leads and nursing education leadership as ineffective in bridging theoretical knowledge with practical care. Most nurses did not feel comfortable implementing preventive measures, staging, diagnosing, and treating PUs, despite the initial wave of mandatory education. Many staff members, including nurses and clinicians, were complacent about existing injuries. A culture of prevention, urgency of reporting, and preoccupation with harm was not prevalent. Some staff did not feel comfortable reporting PUs, fearing punitive reaction and blame.

The team’s evaluation of existing practices as part of subjective discussion between nursing education leadership, quality departments, and the PU HAC co-leads suggested a need for a different approach that tied all driver elements together in a safe but realistic environment.

As evidenced by the lack of information in the literature prior to 2014, quality improvement is a relatively new concept for medicine in general and very new to PI specifically. In the past, quality was generally defined as achieving best clinical outcomes and safety was defined as not harming patients; however, these concepts have converged over the last decade and the idea of “aiming for zero harm” has emerged.^9,11 According to a review of descriptive studies discussing core concepts of quality improvement and patient safety,¹¹ 3 key principles are required for achieving zero harm: development of safety culture, staff accountability, and transparency. Simulation-based or event-based medical education has been shown in reviews of descriptive core concepts and studies involving on medical simulation^12-14to be useful for creating a safety culture in events of critical resuscitations, trauma, and negative surgical scenarios. Appropriate feedback (including debriefing) has been shown to be central to this process.^13-15Because no literature was available that described the role of debriefing in PU reduction, our group borrowed the concept of simulation and debriefing from industries such as aviation, automotive, and trauma where it had been used successfully.^12,13,16

Conceptual framework — debriefing. Feedback is a post-experience analytic process. It involves discussion and analysis of an experience and reviewing and integrating lessons learned into one’s knowledge bank. A review of the literature¹² shows that in adult teaching, “active” participation is an important factor in effective learning. Critical learning is achieved by reflecting on experiences, recognizing strengths and weaknesses, and reviewing alternative choices. Feedback is an integral part of medical debriefing.¹²

Debriefing is defined by Merriam-Webster as questioning someone about a completed mission with an aim of obtaining useful information. Generally, it is a process of inquiry and evaluation. First developed as a formal process in the military during World War II when troops were gathered after missions to reconstruct and describe what happened in an effort to reduce psychological stress,¹² post combat discussions or “performance critiques” eventually became a fundamental component of battle simulation exercises for soldiers in training.^12,13 The focus of after-action reviews shifted over time from subjective emphasis on error identification to nonpunitive guided group discussions and self-reflection. Debriefing also has deep roots in aviation, mass transportation, and nuclear power, areas in which overt or latent human and system weaknesses can lead to loss of life.¹³ As a result, it appears these industries have developed cultures of safety that are far less tolerant of conditions that place human lives at risk than are prevalent in the health care industry.

A relevant review found effective communication is crucial to patient safety.¹⁶ The Joint Commission found communication failures are the root cause of 60% to 70% of sentinel events. In health care, it is important to find ways to open lines of communication. Post-event debriefings in medicine are defined as a “discussion of actions and thought processes after an event to promote reflective learning and improve clinical performance”; they are facilitated discussions of a clinical event focusing on learning and performance improvement.¹⁶ Essential elements of post-event debriefings include active self-learning, a primary intent for improvement, reflection on specific events, and the inclusion of input from multiple team members.¹⁴ Post-event debriefings are a foundational behavior of high-performing teams.¹⁶

The American Heart Association endorses debriefing as a strategy to improve cardiopulmonary resuscitation quality.^15,16 A recent meta-analysis^11,17 found organizations can improve individual and team performance by up to 25% by conducting effective debriefings; in these simulation-based studies, debriefing has been associated with enhancements in team performance and improvements in both technical and behavioral skills. In clinical medicine, post-event debriefings have been shown to increase overall performance, reduce the frequency of equipment-related problems, and improve communication and teamwork.

Debrief Tool (2013–2014). The PU HAC lead (an intensive care attending physician)  developed the “PU Occurrence Debrief” tool, a 1-page summary to guide a discussion after PU events in real time led by PU HAC leads (see Figure 2). A frequent attendee of post event debriefs, she searched the literature for information on safety approaches, adult learning, and debriefing specific to medicine and applied the concept to PU staff education and management. She presented her instrument to the group, her co-lead, and a quality partner, and the tool was trialed in a few pilot instances where the concept was well-received. Staff were supportive and appreciative of the framework for discussion. Concomitantly, staff were re-educated on the reporting process via an electronic medical database, with emphasis on time (goal to report within 12 to 24 hours), accuracy, and completeness.

Debrief process. The PU implemented debrief process is triggered by assistant nurse managers on each unit, who report each event via a centralized computerized system. These reports are transmitted to a quality department member and to the PU HAC lead. The PU HAC lead then conducts the debrief with the team taking care of the affected patient within 24 to 48 hours of the occurrence. This short gap serves to minimize recall bias, allowing timely patient care and potentially addressing global issues such as equipment failure, gaps in communication, lack of support, or staffing problems. To improve team performance, all members of the team who are physically present participate in the post-event debriefing. The goal is to include all staff, patient, and/or family if appropriate, because each offers a unique perspective and each perspective is important to understanding the individual and team strengths and weaknesses.

Three (3) stages of debriefing process are described in literature¹²:
Stage 1: The reaction phase: Allow for initial responses. Summarize what happened. Ensure common understanding.
Stage 2: The understanding phase: the heart of the debriefing inquiry and analysis. Inquire about assumptions/actions.
Stage 3: The summary phase: Lessons learned. Provide take-away points for the future as soon as possible after the event to review the facts, identify system errors, and improve the process with goals to reduce future recurrences.

We found debriefing to be the most effective when structured and facilitated. This structure is based on the “Gather, Analyze, Summarize” approach to debriefing endorsed by the American Heart Association and incorporated in its life support courses.¹⁷ We begin by reviewing the details of the event, examining the patient and discussing staging accuracy. This is followed by discussion of the risk factors, mitigating circumstances, presence or absence of preventive measures, and the nature of treatment to minimize sequelae. Finally, a summary or a simulation for the next time is generated, with emphasis on “What would I do next time?” We try to empower staff to speak frankly and offer their opinions and suggestions. The atmosphere of psychological safety, where team members feel secure in critically analyzing their own performance, is best achieved when the debriefing proceeds in a nonpunitive fashion. We focus on high-value issues such as adherence to prevention guidelines, equipment or assessment issues, and appropriateness of treatment. Debriefing is kept brief, taking no more than 10 minutes.

Bundle implementation and tracking. The consistency of bundle element implementation and documentation has been tracked since 2014. Random audits of electronic PU/PI prevention documentation are performed every quarter (5 patients are selected from every unit) by PU HAC champions and documented on a PU HAC Bundle Audit sheet; the audits focus on completeness, accuracy, meaningful interventions based on the assessed risk, and (if the patient develops a PU/PI) accuracy and timing of the reporting. Audit questions require Yes/No responses (eg, Performed vs. Not performed, Interventions documented vs. Not documented, based on the patient’s clinical risk). When the audit is complete, the patient PU/PI assessment score as well as clinical risk is determined by the auditors first and then compared to the documented data. The PU/PI champions are well-trained and are able to assess the accuracy of the bedside nurses’ clinical judgment.

Data collection and analyses. The primary outcome was total  number of hospital-acquired PUs/PIs per year. New PUs/PIs occurring after admission were counted and prevalence was calculated as the number of PUs/PIs divided by the total admissions per year. Key process measures (bundle compliance, preventive measures implementation, and timeliness and accuracy of the PU/PI incident report) were assessed quarterly via random patient audits and documented in a hospital-based Health Insurance Portability and Accountability Act-compliant database  via a computerized data collection tool.

Since implementation of the program and debrief protocol, compliance on implementation and documentation of bundle elements has ranged between 88% and 94%, with an ongoing aim to maintain >90%.

Following the initiation of the debrief process and coupled with continuous education efforts, we have seen a significant decrease in hospital-acquired PIs. The total number of PUs decreased 30% from 2014, and as much as 40% in 2017 (from 2014; see Figure 3). The overall rate of PUs/PIs (total PUs/PIs per year/total number of admissions) was 0.0057 in 2014, 0.0050 in 2015, 0.0036 in 2016, and 0.0023 in 2017. While 60% of hospitalizations were short stay (2 to 3 days), 70% of PUs occurred in longer term admissions, with an average time to PU occurrence of 8 ± 2 days. In 2017, PICU patients accounted for 40% of PIs; 37% occurred on medicine/postsurgical floors, 13% in the NICU, 5% in operating rooms, and 5% in emergency room/other (see Figure 4). Device-related PUs/PIs accounted for 65% of all PUs/PIs, consistent with previous reports2-5; of those, >50% were related to respiratory devices, 25% to peripheral intravenous/central lines, 10% to tracheostomies, and 15% to other devices (see Figure 5).

As of the writing of this article, most PUs/PIs reported in the hospital have been in the early stages (Stage 1 and Stage 2), suggesting our approach has resulted in improved surveillance, prevention, and timely recognition in the 3 years of study (2014–2017) (see Figure 3). PU/PI occurrence has been reduced in ECMO patients from 40% in 2013 to 0% in 2016–2017), and PICU respiratory device-related PIs have decreased by 50% over the last 2.5 years. No neonatal respiratory device-related injuries have occurred in the last 2 years. The trajectory for 2018 follows the same pattern.

In 1999, the Institute of Medicine released its report, To Err is Human.¹⁸ It stressed that creating a learning environment is crucial for the development of a safe and reliable organization. The report recommended 5 elements. First, simulation should be used in training of staff performing safety-critical functions. Sophisticated gadgets are unnecessary; rather, practice to perfection and effective coaching should be part of training. Second, a culture of reporting errors is required. Third, the reporting of errors must be accountable for both the occurrence and the actual notification of errors but free of shame and blame. Fourth, good communication among disciplines is important. The fifth element is to analyze errors and identify their roots.

As indicated in the Driver diagram (see Figure 1), the debrief process interacts with several of the key drivers and processes in PU/PI prevention. The debrief process brings discussion (of a present case) and simulation (for future cases) to the forefront of learning. This has proven to be effective in fostering small multidisciplinary group learning, facilitated by the PU HAC lead.
The “timely reporting” component was much more difficult to implement. The support of hospital leadership and close collaboration with the hospital quality department were crucial for implementing timely reporting and effecting culture change. The culture had to be changed from perceptions of fear of punitive actions and thoughts that maybe the ulcer will improve on its own and that the ulcer is not life-threatening to positive feelings with regard to recognizing the problem, that this concerns a process not an individual, and that we should feel team pride for our efforts. At the same time, we did not want staff to take PUs/PIs lightly. Multiple guidelines, policies, and safe-guards have been put in place stemming from root cause analyses of PU/PI events. Still, we recognized that certain PUs/PIs are unavoidable and “occurrence debrief” would remain an essential opportunity to reinforce education, review, and plan for “the next time.”

We were able to apply all 5 Institute of Medicine¹⁸ elements in our journey to reduce the rate of HACs in pediatric patients. We nurtured the internal motivation and competency of our staff through education, discussion, availability, and development of necessary tools. Today we are focused on continuing and strengthening our clinical quality and safety program. The future goals are to sustain improvement and to strive for better outcomes.

The biggest limitation of this descriptive study is that data are limited to 1 hospital’s effort to decrease the incidence of PUs/PIs. The quality improvement models we adopted, the campaigns we launched, and the debrief tool we developed worked in our environment. The tool has not been validated outside of our hospital, and further studies with multiple sites would be needed to see if the concept and the tool itself can be generalizable. We cannot sufficiently underscore the importance of the continuous education that took place. It is difficult to analyze if education in itself is a variable that affected the success of the debriefing concept. Having said that, we believe the tool and the approach improved staff ability to retain and to apply the information given.

A quality improvement project was undertaken to develop and implement a debrief protocol and to evaluate compliance with and the implementation of a prevention bundle to decrease the overall incidence and severity of pediatric PUs/PIs in a free-standing children’s hospital. Through the plan’s initiation in 2013 and implementation in 2014, we learned that implementing and sustaining quality improvement initiatives to decrease PUs/PIs is a Herculean task that requires many tools. The strength of this program is its relatively low cost (time commitment from qualified personnel to conduct the debrief), simplicity, and ease of implementation. During the 3 years of implementation, the rate of hospital-acquired PUs/PIs decreased by 40% and compliance with PU/PI prevention bundle is consistently ~90%, the targeted aim. In our experience, the concept of debriefing (utilizing a debrief tool) is an effective way to promote PU/PI education, enhance preventive efforts, reduce hospital-acquired injuries, and propagate a culture of safety. Additional research to examine the effect of this approach on preventing PIs in adults and ascertain staff education needs and the effect of education delivery models is warranted.

Dr. Boyar is Director of Neonatal Wound Services, Division of Neonatal-Perinatal Medicine, Department of Pediatrics, Cohen Children’s Medical Center of New York, New Hyde Park, NY; and Hofstra North Shore-LIJ School of Medicine, Hempstead, NY.

Please address correspondence to: Vita Boyar, MD, 269-01 76 Avenue, Suite 344, New Hyde Park, NY; email: VBoyar@gmail.com.