A modern pediatric intensive care unit (PICU) is a complex system, operating in conjunction with other complex systems (inpatient units, operating rooms, and emergency departments) within and between hospitals.
Incorporating principles of high reliability is necessary to bring about system changes that will lead to continued improvements in PICU outcomes.
As the multidisciplinary PICU team leader, pediatric intensivists should understand and collaborate in managing all system aspects, not just clinical care.
PICU structure, processes, and outcomes are key system components requiring management for success.
Anticipated future changes in reimbursement methodology will likely influence the value proposition of current diagnostic and treatment models in an increasingly high-technology PICU world.
Despite inherent challenges, severity-adjusted comparative data for PICU outcomes are available and useful for assessing outcomes within and between hospitals.
An often-repeated quotation states, “Every system is perfectly designed to achieve the results it gets.” Improved outcomes and higher quality of care in pediatric intensive care units (PICUs) derives from system change. The Institute of Medicine report Crossing the Quality Chasm raised significant concerns about healthcare quality in the United States and called for fundamental redesign of the US healthcare delivery system. Since that time, the healthcare industry has begun to adopt high-reliability practices of industries such as nuclear power, commercial aviation, and aircraft carrier flight deck operations. These industries operate within complex and inherently dangerous environments. Despite these challenging work environments, they have achieved near flawless outcomes by applying many of the high-reliability principles outlined by Weick et al., earning them the designation high-reliability organizations (HROs). We posit that, to dramatically improve outcomes, ICUs must become HROs ( Table 2.1 ).
|LOW Reliability (Generally More Basic and Inconsistent)||←||RELIABILITY||→||High Reliability (Generally More Robust and Effective)|
Studying HROs yields five interrelated principles that characterize how individuals in these organizations behave and think and how the systems where they work adapt to these behaviors, decreasing the likelihood of error and enhancing an organization’s ability to quickly recover when an error occurs ( Table 2.2 ). , “Preoccupation with failure” means treating even minor errors as potential catastrophes and learning opportunities. “Sensitivity to operations” implies that leaders pay attention to operations at all levels—especially at the front line, relying heavily on transparent, nonpunitive information dissemination to all parties. “Reluctance to simplify” recognizes the dangers of oversimplification in complex systems and looks for differing viewpoints when analyzing events. “Deference to expertise” means that in traditionally hierarchical organizations such as hospitals, understanding of operational details for any process usually resides within frontline staff. When possible, shifting operational responsibility to the front line (and, recently, even to parents and families) may optimize results. “Commitment to resilience” specifies that when unforeseen events occur, HROs adapt swiftly, communicate rapidly, and problem solve creatively.
To date, high-reliability care in the US healthcare system remains elusive. Data from pediatric hospitals suggest that recommended care is delivered only 55% of the time. Multiple pediatric hospital systems report significant harm reduction by applying HRO principles as part of robust quality improvement (QI) programs. In addition, Berry et al. describe a link between PICU harm reduction and safety and teamwork culture as part of a larger comprehensive hospital initiative focused on high-reliability principles. Further, pediatric hospitals have banded into statewide and national collaboratives using HRO principles and QI science to lower rates of serious safety events and hospital acquired conditions (HACs). ,
PICUs are an essential, rapidly expanding component of hospital systems. Critically ill patients usually receive care in the PICU but, in some cases, they receive care by the PICU team outside the PICU. Niedner and colleagues describe the current state of US PICUs and their journey toward high reliability. They conclude that, while progress has been made and many PICUs are on this journey, currently no hospital or PICU truly fulfills all of the HRO criteria. Furthermore, leaders in any PICU system must pay attention to aspects of the system well beyond clinical care, which adds challenges that must be managed. This chapter’s purpose is to equip pediatric intensivists, as leaders of multidisciplinary PICU teams, with an enhanced understanding of the PICU as a system within the greater hospital system, including management and evaluation of PICU operations and outcomes. Integrated into this discussion are specific challenges to attaining high-reliability care delivery. It is through implementation of administrative and quality knowledge, in combination with clinical expertise, that highly reliable PICUs could soon flourish.
Pediatric intensive care unit as a system
Industrial engineering and operations research would describe hospitals and PICUs as emergent systems. Emergence is a phenomenon in which larger systems arise from combined interactions of smaller systems in such a way that the whole is greater than the sum of the individual components. The PICU is a system with inputs and outputs. PICU inputs include critically ill patients admitted from all units within the healthcare system. Outputs include patients transferred back to those same units, including home. Therefore overall quality of care in the PICU cannot be viewed in isolation. Indeed, the whole system (critical care) is greater than the sum of the individual components (e.g., emergency department, operating room, hospital inpatient unit, interhospital transport, and rehabilitation unit). Improvements at the whole-system level will primarily occur if it is viewed in its entirety rather than as a collection of multiple individual components. This concept has fostered a relatively new way of thinking about the modern ICU as an “ICU without walls” rather than as a distinct geographic unit. As McQuillan et al. stated, “the greatest impact on the outcome for intensive care units may come from improvements in the input to intensive care, particularly in the quality of acute care. . . .” The “PICU without walls” concept is perhaps best understood in the context of reduced morbidity and mortality associated with implementation of rapid response systems, watch-stander or “watcher” programs, and enhanced focus on situation awareness. Intensivists—and often pediatric critical care medicine (PCCM) fellows—are expected, with increasing frequency, to provide consultation for children outside the ICU, often in conjunction with hospital medicine colleagues. Additionally, as children recover from critical illness, long-term sequelae often persist. Thus intensivists are recognizing the importance of early and frequent involvement of rehabilitation specialists, both while the patient is still in the ICU and after transfer out of the PICU. ,
Another example of the PICU without walls concept is telemedicine. The American College of Critical Care Medicine Task Force on Models of Critical Care specifically mentions provision of tele-ICU services as key elements for the ideal critical care delivery model. Berrens et al. describe use of telemedicine at a satellite facility with rapid response and code team activations staffed remotely by pediatric intensivists. They note a similar standard of response team activation and response at their satellite facility when compared with the main institution using telemedicine. However, they also note that an effective system must be in place at the main institution for full telemedicine effectiveness. Telemedicine is likely an effective and important tool for improving critical care delivery, especially when used in conjunction with many of the other high-reliability design principles discussed in this chapter.
Models of critical care delivery
Avedis Donabedian, an early systems thinker in healthcare, stated that healthcare quality should be based on three dimensions: structure, processes, and outcomes ( Fig. 2.1 ). In this model, processes are effective (1) when the right structures are in place to support them and (2) when outcomes are measured, so that these processes can be evaluated for effectiveness and modified to produce better results. Structure refers to the setting in which care is delivered. Structure elements include patients, providers, technology, and therapy. Providers include the entire PICU team. When organized according to HRO principles, the team is optimally led jointly by a physician and nurse and is inclusive of all disciplines that touch the patient ( Fig. 2.2 ). A recent trend is greater subspecialization of PICUs based on specific subpopulations of critically ill children (e.g., neurointensive care, cardiac intensive care). These subspecialty PICUs are discussed elsewhere in this textbook; however, the structure and processes to achieve optimal outcomes apply regardless of subspecialization of a particular ICU. Process refers specifically to how care is provided, including incorporation of high-reliability principles into daily activities. Outcomes refer to end points of care, including commonly used quality and safety measures, and other key outcomes, such as length of stay (LOS), patient/family experience, and cost/value.
The critical care team is perhaps the most important structural component driving care quality and high-reliability processes in the PICU. The exact number of physicians practicing pediatric critical care medicine is unknown. A 2017 survey by the American Board of Pediatrics reported 2603 board-certified pediatric intensivists in the United States (up from 1881 in 2011) but did not account for physicians no longer practicing critical care or others (cardiologists, anesthesiologists) staffing PICUs. Pediatric intensivists are aging—the average age is 50 years. Many PICUs now use a 24/7 in-house attending coverage model. One recent PICU survey revealed that 53% of responding hospitals had full in-house attending coverage, 21% had mixed coverage (some nights in-house, some nights home coverage), and only 26% had exclusive home coverage. Perhaps in response to the aging intensivist population and the increased staffing needs associated with increased in-house attending coverage, PCCM training programs are growing rapidly. The number of PCCM fellows increased from 271 in 2001 to 548 in 2017. The field of PCCM is also becoming increasingly female, with the number of first-year female PCCM fellows increasing from 41 in 2001 to 120 in 2017. , Staffing models that incorporate more advanced practice nurses, physician assistants, and hospitalists—as well as telemedicine and care regionalization—have been proposed as solutions to the potential critical care physician shortage. Critical care services regionalization can be both a problem and a solution. Halpern et al. report that 91% of hospitals with intensivists are located in metropolitan areas and 93% of all ICU beds are located in metropolitan locations, raising concern for a shortage of intensivists and ICU beds in rural areas and potentially making care more difficult to access for patients living away from metropolitan areas.
PICU physician workforce challenges are clearly multifactorial, including career path opportunities, work environment, and job satisfaction. PICU physicians report the highest number of work hours per week as compared with all other pediatric medical subspecialties. , They also report the smallest percentage of a part-time workforce as compared with other pediatric subspecialties. Given the work environment (long work hours and high stress), the prevalence of burnout among pediatric critical care physicians, as reported in a Critical Care Societies Statement, was 71%. This rate is higher than critical care physicians as a whole and more than twice the rate of general pediatricians. Methods to mitigate burnout are multifactorial and involve both individual and institutional/organizational buy-in. , As Perlo et al. discuss in their Institute for Healthcare Improvement white paper entitled “Joy in Work,” a focus on factors that bring enjoyment may be more effective than a focus on what causes burnout (see also Chapter 22 ). Importantly, 94% of pediatric critical care physicians report that they would encourage a resident to pursue a career in PCCM and 87% report that they would choose a career in PCCM again. Finally, burnout scores were not different between those who provide in-house coverage versus those that take home calls. , When surveyed, 36% of intensivists planned to reduce their workload in the next 5 years, with nearly 10% planning to retire. These physicians cite long work hours and stress as contributing factors to their decisions. In sum, the impact of burnout and work environment on PCCM physician staffing remains unclear but certainly warrants attention intended to minimize factors that contribute to the stress of working in an ICU. Another key element that pertains to PICU staffing is ICU strain. ICU strain is “a discrepancy between the availability of ICU resources and demand to admit and provide high-quality care for patients with critical illness.” Rewa et al. identified several indicators of ICU strain, including ICU census (capacity to admit new patients), ICU queuing (delay in time to physical admission from decision to admit), ICU readmission, early ICU discharge, after-hours discharges, and surgery cancellations. All of these factors have potential impacts on patient safety and outcomes. Opgenorth et al. surveyed providers and found that ICU strain increased stress levels and contributed to burnout among providers, reduced perceived quality of patient care, and contributed to high staff turnover. Strategies to mitigate ICU strain include increasing ICU bed capacity; increasing nonacute care beds, which can improve ICU throughput; improving care transitions; better ICU staff training; and increased nurse-to-patient ratios.
Optimal physician coverage and direct presence in the PICU remains unclear. As discussed previously, 24/7 in-house attending coverage is increasingly common among PICUs. A 2013 survey indicated that 86% of pediatric intensivists believed in-house PICU attending coverage improved patient care and more than 80% thought patient care was safer and timelier with in-house attending coverage. Several published studies comparing ICUs with and without 24/7 attending critical care physician coverage suggest mixed results. Demonstrating a mortality difference between 24/7 in-house attending coverage and home coverage is difficult, especially in the context of a low overall mortality rate in PICUs. However, Gupta et al. demonstrated a significantly lower mortality rate, lower cardiac arrest incidence, and decreased mortality after cardiac arrest with a 24/7 staffing model. They also demonstrated shorter PICU LOS and decreased mechanical ventilation duration. Another study by Gupta, examining children after cardiac surgery, revealed a lower incidence of cardiac arrest, decreased use of extracorporeal membrane oxygenation, earlier tracheal extubation, and shorter central venous line duration. Nishisaki and colleagues reported that 24/7 in-hospital intensivist coverage decreased duration of mechanical ventilation and PICU LOS but did not affect risk-adjusted mortality. In sum, while the optimal PICU attending staffing model remains unclear, recent data suggest improved outcomes with a 24/7 in-house attending staffing model.
Optimal ICU physician staffing ratios are also unclear. A study conducted in adult ICUs demonstrated a relationship between staffing ratios and mortality. Other studies demonstrated a higher nursing workload, which can negatively impact nurse retention, burnout, and patient safety. , The Society of Critical Care Medicine published guidelines regarding how to address this important issue but did not recommend specific physician-patient ratios. The society did note: “In academic medical ICUs, there is evidence that intensivist/patient ratios more than 1:14 negatively impacted perceptions of teaching quality, stress, patient care, and workforce stability.”
Structure also encompasses ICU physical design, monitoring and support equipment, and information systems preferably equipped with decision support. , Optimal ICU design may improve patient safety. As Leaf et al. demonstrated, those patients who were in low visibility rooms (i.e., not visible from the nursing station) were at as high a risk of mortality as equally sick patients in high-visibility rooms. Halpern outlines key principles that should guide ICU patient room design, such as single-occupancy rooms and rooms that are similarly designed so that staff can move seamlessly between rooms. Devices in rooms (monitors and ventilators), should be informatics based and have the capability to continuously send and store data to a central location. The PICU is a technical environment in which the interface between technology and humans can both improve care and increase error risk. For example, computerized physician order entry (CPOE) can significantly improve quality of care provided in the ICU setting by decreasing LOS and reducing medication error. However, even with electronic health records (EHRs) and CPOE, risk for error remains (e.g., ordering medication on the wrong patient, relying on the computer for dosage) and some inefficiencies persist (e.g., increased time associated with computer documentation). There is growing interest in using predictive analytics, based on objective data pulled automatically from the medical record, to predict patient outcomes.
Process refers specifically to how care is delivered in the ICU. The aforementioned structural elements interact with key processes to drive improved outcomes. It is important to emphasize that structural elements cannot compensate for the lack of an appropriate institutional climate that supports standardized, reliable, evidence-based processes and uses high-reliability principles. , , Adequate structure, accompanied by rigorous use of high-reliability principles, is essential to drive improved outcomes. For example, optimal nurse and physician staffing structure will not consistently prevent an adverse drug event from causing harm if rigorous medication double-check techniques and the five rights of medication administration are not consistently (reliably) performed with every medication administration. There must also be a robust quality culture—with focus on safety, teamwork, and engagement—emanating from the top and permeating throughout the organization. The role of senior hospital leadership and the board of directors, who should emphasize to everyone the importance of quality and safety, is critical. As Randall et al. state, “increased CEO involvement and accountability are key opportunities to move closer to high reliability.”
Multiple studies demonstrate significant patient harm reduction by implementing standardized reliable processes, including—but not limited to—the use of care bundles, checklists, and clinical pathways ( Table 2.3 ). A care bundle is a relatively short list of standardized, generally evidence-based or best-practice interventions for a patient population or disease that, when implemented consistently, lead to improved outcomes. It is the combination of elements performed consistently and in aggregate that drive improvement. The ABCDEF ICU liberation bundle is a recent example of large-scale bundle use in adult ICUs that, when used reliably, is linked to improved survival, reduced delirium, and decreased ICU readmissions. Large collaborations between children’s hospitals, such as the Children’s Hospitals’ Solutions for Patient Safety, have contributed to the development of pediatric specific best practices and HAC prevention clinical care bundles, including those aimed at surgical site infection, catheter-associated urinary tract infection, and pressure injury reductions. The use of these bundles leads to HAC rate reduction. Finally, sepsis recognition and treatment bundles are increasingly common. Work published by Evans et al. assessing the use of the New York sepsis guidelines demonstrated improved outcomes when a sepsis bundle (antibiotics, cultures, and fluid) was completed within 1 hour. Balamuth et al. demonstrated improved outcomes with use of a sepsis identification bundle in the emergency department in combination with the electronic medical record to generate automatic alerts to identify patients at risk of sepsis. , Checklists have also been used to improve care of critically ill children admitted to the PICU. The NEAR4KIDS group developed an Airway Bundle Checklist to identify tracheal intubation risk factors, improve team situation awareness, improve the use of time-outs, and mitigation plan generation. Clinical pathways are flowcharts or algorithms to guide provider decision-making and offer education to learners about evidence behind clinical care recommendations. Consistent use of pathways decreases hospital LOS, increases adherence to evidence-based practices, and improves flow through emergency departments when applied to care for common pediatric conditions such as asthma and acute gastroenteritis. Finally, structured communication during multidisciplinary rounds and hand-offs are additional key processes driving quality and improved outcomes in the PICU. An example of the efficacy of structured hand-off communication is I-PASS (illness severity, patient summary, action list, situational awareness, and synthesis by receiver). Reliable use of this nursing hand-off bundle demonstrated improvements in hand-off completeness and quality. However, there was no evidence that clinical outcomes improved.