Do Emergency Transfers Reflect Hospital Quality? Preventable Harm and the Next 100 000 Lives

In 2004, the Institute for Healthcare Improvement launched the 100 000 Lives Campaign to reduce preventable harm through widespread adoption of hospital safety practices.1 One of its 6 key interventions was the implementation of rapid response teams to identify and address clinical deterioration before cardiac arrests occurred outside intensive care units (ICUs). The campaign introduced phrases like "a code is a sign of a system failure," reflecting the growing recognition that many such events are not inevitable but preventable. In pediatrics, the Children's Health Corporation of America collaborative helped put these concepts into practice by standardizing rapid response systems across member hospitals.2 Over time, the field has broadened its focus to include more proactive risk mitigation. Although tools and strategies vary across institutions, the pursuit of early, shared situation awareness3—through standardized communication, consistent policies, and, recently, predictive technology—has become a defining feature of effective care escalation.A major advance in this work was the development of standardized measures for deterioration-related harm. Pediatric inpatient mortality remains rare, and commonly used process metrics—like rapid response team activation or ICU transfer—often reflect institutional behavior more than patient outcomes. To address this gap, Brady et al introduced the concept of the UNSAFE Transfer in 2013, defined as an unplanned ICU admission requiring life-sustaining intervention (eg, vasopressors, intubation, or high-volume fluid resuscitation) within 1 hour of ICU arrival.4 Now widely referred to as emergency transfers, this measure has been validated across various pediatric settings and is increasingly recognized as a patient-centered outcome for evaluating deterioration-related harm.5–9In this issue of Hospital Pediatrics, West et al contribute meaningfully to this growing body of work through their institutional evaluation of emergency transfers in the pediatric cardiac population.10 Using a 5-year retrospective cohort design with propensity score matching, they demonstrate a strong association between emergency transfers and in-hospital mortality in this high-risk population. Their findings further validate the emergency transfer metric and highlight its value as a hospital-level quality indicator.Notably, they report a mortality rate of 25% among patients who experienced an emergency transfer, compared with 10% in the matched comparison group. This rate slightly exceeds mortality rates reported in general pediatric populations and likely reflects the acuity and complexity of the cardiac population. The study also shows that patients who received the same life-sustaining interventions in a nonemergent fashion (ie, within 1–12 hours after ICU transfer) had significantly lower mortality (odds ratio: 3.6; 95% CI: 1.40–11). These findings echo similar patterns in general pediatrics: although the clinical need for intervention may be the same, the timing and context of escalation play a critical role in outcomes. We see from West et al that emergency transfers in the cardiac population represent meaningful clinical events with direct implications for mortality. Their frequency makes them a particularly high-yield target for improvement, especially when compared with more heterogeneous or lower-acuity inpatient settings.West et al report an emergency transfer rate of 13.25 per 10 000 non-ICU patient days, substantially higher than in general pediatric populations. This rate aligns with anecdotal experience at many institutions and underscores the cardiac population as both high risk and high opportunity for targeted intervention. Their work reinforces the growing recognition that emergency transfers are neither isolated nor inevitable, but are instead measurable and actionable signals of system-level gaps in recognition and escalation of care. Efforts to close these gaps are already underway in pediatric cardiology through programs like HeartWatch and others,11,12 which aim to proactively identify and mitigate deterioration risk. By applying a validated, patient-centered outcome measure to their specific clinical context, West et al exemplify the kind of locally driven, data-informed safety work that is both feasible and urgently needed. With this strong data foundation, they are now well positioned to pursue meaningful improvement efforts that build on these findings.Across pediatric hospital medicine, consensus is growing that emergency transfers represent an important and modifiable outcome. These events are not rare, isolated incidents, but rather they occur with regularity and carry significant consequences for morbidity and mortality. Journals such as Hospital Pediatrics have published extensively on the topic, reflecting its relevance across care settings. Thanks to work like this, we are gaining insight into how the emergency transfer measure functions in diverse clinical populations and how it can guide targeted improvement efforts. Several institutions have already reduced emergency transfer rates through initiatives focused on shared situation awareness, early risk identification, and timely escalation.4,13–16 Still, most of this progress remains localized at individual sites. Although single-institution work is valuable, it does not address the persistent variation in how hospitals design rapid response systems, define escalation pathways, or prioritize this aspect of patient safety.Despite rising interest, emergency transfers are not yet a universally adopted safety metric, and no consensus-driven, evidence-based strategies exist to guide field-wide improvement. Traditional research methods have struggled to account for the wide variation in infrastructure, staffing, and clinical culture across pediatric hospitals. What works in one setting may not generalize easily to another. As a result, individual site successes, although important, are unlikely to drive the broad, sustained change needed to improve outcomes for all deteriorating pediatric patients.How, then, should we move forward? One option is to continue refining traditional translational research pipelines to test and disseminate best practices. Experts in pediatric patient safety have identified research on clinical deterioration as a top national priority.17 But variation in hospital resources, workflows, and populations often limits the generalizability of these efforts. A more scalable approach may lie in multisite quality improvement collaboratives that enable shared learning, benchmarking, rapid-cycle testing, and local adaptation, all while maintaining a focus on common goals and meaningful measurement.Statewide and national initiatives offer promising models. For example, the Ohio Solutions for Patient Safety collaborative has adopted emergency transfer definitions and embedded them into shared safety improvement strategies.18 Ohio institutions—with emergency transfer rates now at some of the lowest levels reported—contribute significantly to the growing evidence base. National networks such as the Solutions for Patient Safety collaborative19,20 and the Children's Hospital Association21 offer platforms for peer learning, common measurement, and coordinated improvement with demonstrated success in other safety domains. In parallel, national data networks exist that could provide opportunities for benchmarking, risk-adjusted tracking, and deeper understanding of deterioration trends across systems.A collaborative, safety-driven approach has the potential to advance this work in a way that single institutions, acting alone, may struggle to achieve. Shared learning across hospitals allows best practices to emerge naturally while still leaving room for local adaptation based on site-specific needs and resources. It supports transparency, accountability, and, increasingly, alignment with payment models that reward harm reduction. Just as important, this approach also allows us to move faster, accelerating the translation of evidence into practice. By testing and refining interventions in real-world settings, we can learn what works and iterate continuously rather than wait for perfect evidence before acting. This model aligns with the principles of a learning health system and brings us closer to the kind of responsive, data-driven innovation that hospitalized children deserve.The article by West et al demonstrates how meaningful progress can begin at the local level. It reminds us that emergency transfers are not simply process failures, but serious outcomes highly associated with mortality. By building a robust data foundation in a high-risk population, West et al have laid critical groundwork for future safety efforts, and we look forward to seeing where that work leads. This commentary reflects perspectives shaped by our ongoing quality improvement work and early efforts to lean into multisite collaboration. As a field, we are united in our commitment to identifying deterioration early and escalating care effectively. Now is the time to harness that shared commitment through collaboration—to ensure that emergency transfers, like other forms of preventable harm, become increasingly rare in the hospitals where we care for children every day.