Impact of a relocation to a new critical care building on pediatric safety events.

BACKGROUND: Cincinnati Children's Hospital Medical Center (CCHMC) relocated the pediatric, cardiac, and neonatal intensive care units (PICU, CICU and NICU) to a newly constructed critical care building (CCB) in November 2021. Simulation and onboarding sessions were implemented before the relocation, aimed at mitigating latent safety threats. OBJECTIVE: To evaluate the impact of ICU relocation to the CCHMC CCB on patient safety as measured by the quantity, rate, severity score, and category of safety reports. METHODS: This retrospective, cross-sectional, observational study compared safety reports filed in a 90-day period before and following the CCB relocation. The primary outcome was pre- and postrelocation safety report rates per 100 patient-days. Secondary outcomes included safety report severity, category, and rate of hospital acquired conditions (HACs). RESULTS: Total safety report incidence increased by 16% across all ICUs postrelocation with no difference in post- versus prerelocation odds ratio between ICUs. Three isolated instances of special cause variation were found, one in NICU and two in CICU. No special cause variation was found in the PICU. There were no statistical differences in assigned safety report severity pre- to postrelocation for all ICUs, and only lab specimen/test related safety reports showed a statistically significant increase postrelocation. Overall rates of HACs were low, with six occurring prerelocation and eight postrelocation. CONCLUSIONS: All three ICUs were relocated to the new CCB with minimal changes in the incidence, severity, or category of safety reports filed, suggesting staff training and preparations ahead of the relocation mitigated latent safety threats.

Usability Testing of Situation Awareness Clinical Decision Support in the Intensive Care Unit.

OBJECTIVE: Our objective was to evaluate the usability of an automated clinical decision support (CDS) tool previously implemented in the pediatric intensive care unit (PICU) to promote shared situation awareness among the medical team to prevent serious safety events within children's hospitals. METHODS: We conducted a mixed-methods usability evaluation of a CDS tool in a PICU at a large, urban, quaternary, free-standing children's hospital in the Midwest. Quantitative assessment was done using the system usability scale (SUS), while qualitative assessment involved think-aloud usability testing. The SUS was scored according to survey guidelines. For think-aloud testing, task times were calculated, and means and standard deviations were determined, stratified by role. Qualitative feedback from participants and moderator observations were summarized. RESULTS: Fifty-one PICU staff members, including physicians, advanced practice providers, nurses, and respiratory therapists, completed the SUS, while ten participants underwent think-aloud usability testing. The overall median usability score was 87.5 (interquartile range: 80-95), with over 96% rating the tool's usability as "good" or "excellent." Task completion times ranged from 2 to 92 seconds, with the quickest completion for reviewing high-risk criteria and the slowest for adding to high-risk criteria. Observations and participant responses from think-aloud testing highlighted positive aspects of learnability and clear display of complex information that is easily accessed, as well as opportunities for improvement in tool integration into clinical workflows. CONCLUSION: The PICU Warning Tool demonstrates good usability in the critical care setting. This study demonstrates the value of postimplementation usability testing in identifying opportunities for continued improvement of CDS tools.

Implementation Science Research in Pediatric Critical Care Medicine.

OBJECTIVES: Delay or failure to consistently adopt evidence-based or consensus-based best practices into routine clinical care is common, including for patients in the PICU. PICU patients can fail to receive potentially beneficial diagnostic or therapeutic interventions, worsening the burden of illness and injury during critical illness. Implementation science (IS) has emerged to systematically address this problem, but its use of in the PICU has been limited to date. We therefore present a conceptual and methodologic overview of IS for the pediatric intensivist. DESIGN: The members of Excellence in Pediatric Implementation Science (ECLIPSE; part of the Pediatric Acute Lung Injury and Sepsis Investigators Network) represent multi-institutional expertise in the use of IS in the PICU. This narrative review reflects the collective knowledge and perspective of the ECLIPSE group about why IS can benefit PICU patients, how to distinguish IS from quality improvement (QI), and how to evaluate an IS article. RESULTS: IS requires a shift in one's thinking, away from questions and outcomes that define traditional clinical or translational research, including QI. Instead, in the IS rather than the QI literature, the terminology, definitions, and language differs by specifically focusing on relative importance of generalizable knowledge, as well as aspects of study design, scale, and timeframe over which the investigations occur. CONCLUSIONS: Research in pediatric critical care practice must acknowledge the limitations and potential for patient harm that may result from a failure to implement evidence-based or professionals' consensus-based practices. IS represents an innovative, pragmatic, and increasingly popular approach that our field must readily embrace in order to improve our ability to care for critically ill children.

Use of design thinking and human factors approach to improve situation awareness in the pediatric intensive care unit.

BACKGROUND: Optimal design of healthcare spaces can enhance patient care. We applied design thinking and human factors principles to optimize communication and signage on high risk patients to improve situation awareness in a new clinical space for the pediatric ICU. OBJECTIVE: To assess the impact of these tools in mitigating situation awareness concerns within the new clinical space. We hypothesized that implementing these design-informed tools would either maintain or improve situation awareness. DESIGN, SETTINGS, AND PARTICIPANTS: A 15-week design thinking process was employed, involving research, ideation, and refinement to develop and implement new situation awareness tools. The process included engagement with interprofessional clinical teams, scenario planning, workflow mapping, iterative feedback collection, and collaboration with an industry partner for signage development and implementation. INTERVENTION: Improved and updated communication devices and bedside mitigation plans. MAIN OUTCOME AND MEASURES: Process metrics included individual and shared situation awareness of PICU care teams and our patient outcome metric was the rate of cardiopulmonary resuscitation (CPR) events pre- and post-transition. RESULTS: When evaluating all patients, shared situation awareness for accurate high-risk status improved from 81% pre-transition to 92% post-transition (p = .006). When assessing individual care team roles, accuracy of patient high-risk status improved from 88% to 95% (p = .05) for RNs, 85% to 96% (p = .003) for residents, and 88% to 95% (p = .03) for RTs. There was no change in the rate of CPR events following the transition.

How do we detect and respond to clinical deterioration in hospitalized children? Results of the Pediatric Care BefOre Deterioration Events (CODE) survey.

Systems to detect and respond to deteriorating hospitalized children are common despite little evidence supporting best practices. Our objective was to describe systems to detect/respond to deteriorating hospitalized children at Pediatric Resuscitation Quality Collaborative (pediRES-Q) institutions. We performed a cross-sectional survey of pediRES-Q leaders. Questionnaire design utilized expert validation and cognitive interviews. Thirty centers (88%) responded. Most (93%) used ≥1 system to detect deterioration: most commonly, early warning scores (83%), watcher lists (55%), and proactive surveillance teams (31%). Most (90%) had a team to respond to deteriorating patients and the majority of teams could be activated by clinician or family concerns. Most institutions (90%) collect relevant data, including number of rapid responses (88%), arrests outside intensive care units (100%), and serious safety events (88%). In conclusion, most pediRES-Q institutions utilize systems to detect/respond to deteriorating hospitalized children. Heterogeneity exists among programs. Rigorous evaluation is needed to identify best practices.

The Need for a Standard Outcome for Clinical Deterioration in Children's Hospitals.

Unrecognized clinical deterioration is a common and significant source of preventable harm to hospitalized children. Yet, unlike other sources of preventable harm, clinical deterioration outside of the ICU lacks a clear, "gold standard" outcome to guide prevention efforts. This gap limits multicenter learning, which is crucial for identifying effective and generalizable interventions for harm prevention. In fact, to date, no coordinated safety/quality initiative currently exists targeting prevention of harm from unrecognized clinical deterioration in hospitalized pediatric patients, which is startling given the morbidity and mortality risk patients incur. In this article, we compare existing outcomes for evaluating clinical deterioration outside of the ICU, highlighting sources of variation and vulnerability. The broader aim of this article is to highlight the need for a standard, consensus outcome for evaluating clinical deterioration outside of the ICU, which is a critical first step to preventing this type of harm.

Improving Usability of the Pediatric Code Cart by Combining Lean and Human Factors Principles.

Cardiac arrests are common in hospitalized children. Well-organized code carts are needed during these events to help staff efficiently find supplies and medications for the patient. This study aimed to improve the efficiency and utilization of the code cart at a major academic pediatric medical center. Methods: This quality improvement project used a phased approach to redesign the code cart. A multidisciplinary team used Lean and Human Factors principles to improve the efficiency and intuitiveness of the redesigned cart. Nurses and respiratory therapists participated in simulations asking for certain supplies with the original and redesigned code cart and filled out surveys for feedback on each code cart. Facilitators measured retrieval times during each simulation. Results: We performed 10 simulations with the original code cart and 13 with the redesigned code cart. Staff could find intraosseous access equipment more quickly (23.9 versus 46.4 seconds; P = 0.003). In addition, staff reported they were less likely to open the wrong drawer or grab the wrong equipment and that the redesigned code cart was overall more well organized than the original code cart. Finally, the redesigned code cart reduced the cost by over 800 dollars per full cart restock. Conclusion: Revising the code cart using Lean and Human Factors improves efficiency and usability and can contribute to cost savings.

CPR quality and outcomes after extracorporeal life support for pediatric In-Hospital cardiac arrest.

AIM OF STUDY: To determine outcomes in pediatric patients who had an in-hospital cardiac arrest and subsequently received extracorporeal cardiopulmonary resuscitation (ECPR). Our secondary objective was to identify cardiopulmonary resuscitation (CPR) event characteristics and CPR quality metrics associated with survival after ECPR. METHODS: Multicenter retrospective cohort study of pediatric patients in the pediRES-Q database who received ECPR after in-hospital cardiac arrest between July 1, 2015 and June 2, 2021. Primary outcome was survival to ICU discharge. Secondary outcomes were survival to hospital discharge and favorable neurologic outcome at ICU and hospital discharge. RESULTS: Among 124 patients included in this study, median age was 0.9 years (IQR 0.2-5) and the majority of patients had primarily cardiac disease (92 patients, 75%). Survival to ICU discharge occurred in 61/120 (51%) patients, 36/61 (59%) of whom had favorable neurologic outcome. No demographic or clinical variables were associated with survival after ECPR. CONCLUSION: In this multicenter retrospective cohort study of pediatric patients who received ECPR for IHCA we found a high rate of survival to ICU discharge with good neurologic outcome.

Inappropriate Shock Delivery Is Common During Pediatric In-Hospital Cardiac Arrest.

OBJECTIVES: To characterize inappropriate shock delivery during pediatric in-hospital cardiac arrest (IHCA). DESIGN: Retrospective cohort study. SETTING: An international pediatric cardiac arrest quality improvement collaborative Pediatric Resuscitation Quality [pediRES-Q]. PATIENTS: All IHCA events from 2015 to 2020 from the pediRES-Q Collaborative for which shock and electrocardiogram waveform data were available. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We analyzed 418 shocks delivered during 159 cardiac arrest events, with 381 shocks during 158 events at 28 sites remaining after excluding undecipherable rhythms. We classified shocks as: 1) appropriate (ventricular fibrillation [VF] or wide complex ≥ 150/min); 2) indeterminate (narrow complex ≥ 150/min or wide complex 100-149/min); or 3) inappropriate (asystole, sinus, narrow complex < 150/min, or wide complex < 100/min) based on the rhythm immediately preceding shock delivery. Of delivered shocks, 57% were delivered appropriately for VF or wide complex rhythms with a rate greater than or equal to 150/min. Thirteen percent were classified as indeterminate. Thirty percent were delivered inappropriately for asystole (6.8%), sinus (3.1%), narrow complex less than 150/min (11%), or wide complex less than 100/min (8.9%) rhythms. Eighty-eight percent of all shocks were delivered in ICUs or emergency departments, and 30% of those were delivered inappropriately. CONCLUSIONS: The rate of inappropriate shock delivery for pediatric IHCA in this international cohort is at least 30%, with 23% delivered to an organized electrical rhythm, identifying opportunity for improvement in rhythm identification training.

Preventing Cardiac Arrest in the Pediatric Cardiac Intensive Care Unit Through Multicenter Collaboration.

Importance: Preventing in-hospital cardiac arrest (IHCA) likely represents an effective strategy to improve outcomes for critically ill patients, but feasibility of IHCA prevention remains unclear. Objective: To determine whether a low-technology cardiac arrest prevention (CAP) practice bundle decreases IHCA rate. Design, Setting, and Participants: Pediatric cardiac intensive care unit (CICU) teams from the Pediatric Cardiac Critical Care Consortium (PC4) formed a collaborative learning network to implement the CAP bundle consistent with the Institute for Healthcare Improvement framework; 15 hospitals implemented the bundle voluntarily. Risk-adjusted IHCA incidence rates were analyzed across 2 time periods, 12 months (baseline) and 18 months after CAP implementation (intervention) using difference-in-differences (DID) regression to compare 15 CAP and 16 control PC4 hospitals that chose not to participate in CAP but had IHCA rates tracked in the PC4 registry. Patients deemed at high risk for IHCA, based on a priori evidence-based criteria and empirical hospital-specific criteria, were selected to receive the CAP bundle. Data were collected from July 2018 to December 2019, and data were analyzed from March to August 2020. Interventions: CAP bundle included 5 elements developed to promote increased situational awareness and communication among bedside clinicians to recognize and mitigate deterioration in high-risk patients. Main Outcomes and Measures: Risk-adjusted IHCA incidence rate across all CICU admissions (IHCA events divided by all admissions). Results: The bundle was activated in 2664 of 10 510 CAP hospital admissions (25.3%); admission characteristics were similar across study periods. There was a 30% relative reduction in risk-adjusted IHCA incidence rate at CAP hospitals (intervention period: 2.6%; 95% CI, 2.2-2.9; baseline: 3.7%; 95% CI, 3.1-4.0), but no change at control hospitals (intervention period: 2.7%; 95% CI, 2.3-2.9; baseline: 2.7%; 95% CI, 2.2-3.0). DID analysis confirmed significantly reduced odds of IHCA among all admissions at CAP hospitals compared with control hospitals during the intervention period vs baseline (odds ratio, 0.72; 95% CI, 0.56-0.91; P = .01). DID odds ratios were 0.72 (95% CI, 0.53-0.98) for the surgical subgroup, 0.74 (95% CI, 0.48-1.14) for the medical subgroup, and 0.72 (95% CI, 0.50-1.03) for the high-risk admission subgroup at CAP hospitals after intervention. All-cause risk-adjusted mortality rate did not change after intervention. Conclusions and Relevance: Implementation of this CAP bundle led to significant IHCA reduction across multiple pediatric CICUs. Future studies may determine if this bundle can be effective in other critically ill populations.

Weight-Based Versus Flat Dosing of Epinephrine During Cardiac Arrest in the PICU: A Multicenter Survey.

OBJECTIVES: Pediatric Advanced Life Support (PALS) guidelines include weight-based epinephrine dosing recommendations of 0.01 mg/kg with a maximum of 1 mg, which corresponds to a weight of 100 kg. Actual practice patterns are unknown. DESIGN: Multicenter cross-sectional survey regarding institutional practices for the transition from weight-based to flat dosing of epinephrine during cardiopulmonary resuscitation in PICUs. Exploratory analyses compared epinephrine dosing practices with several institutional characteristics using Fisher exact test. SETTING: Internet-based survey. SUBJECTS: U.S. PICU representatives (one per institution) involved in resuscitation systems of care. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Of 137 institutions surveyed, 68 (50%) responded. Most responding institutions are freestanding children's hospitals or dedicated children's hospitals within combined adult/pediatric hospitals (67; 99%); 55 (81%) are academic and 41 (60%) have PICU fellowship programs. Among respondents, institutional roles include PICU medical director (13; 19%), resuscitation committee member (23; 34%), and attending physician with interest in resuscitation (21; 31%). When choosing between weight-based and flat dosing, 64 respondents (94%) report using patient weight, 23 (34%) patient age, and five (7%) patient pubertal stage. Among those reporting using weight, 28 (44%) switch at 50 to less than 60 kg, 17 (27%) at 60 to less than 80 kg, five (8%) at 80 to less than 100 kg, and eight (12%) at greater than or equal to 100 kg. Among those reporting using age, four (17%) switch at 14 to less than 16 years, five (22%) at 16 to less than 18, and six (26%) at greater than or equal to 18. Twenty-nine respondents (43%) report using ideal body weight when dosing epinephrine in obese patients. Using patient age in choosing epinephrine dosing is more common in institutions that require Advanced Cardiac Life Support (ACLS) certification for some/all code team responders compared with institutions that do not require ACLS certification (52% vs 22%; p = 0.02). CONCLUSIONS: The majority of PICUs surveyed report epinephrine dosing practices that are inconsistent with PALS guidelines.

Diversity in Machine Learning: A Systematic Review of Text-Based Diagnostic Applications.

OBJECTIVE: As the storage of clinical data has transitioned into electronic formats, medical informatics has become increasingly relevant in providing diagnostic aid. The purpose of this review is to evaluate machine learning models that use text data for diagnosis and to assess the diversity of the included study populations. METHODS: We conducted a systematic literature review on three public databases. Two authors reviewed every abstract for inclusion. Articles were included if they used or developed machine learning algorithms to aid in diagnosis. Articles focusing on imaging informatics were excluded. RESULTS: From 2,260 identified papers, we included 78. Of the machine learning models used, neural networks were relied upon most frequently (44.9%). Studies had a median population of 661.5 patients, and diseases and disorders of 10 different body systems were studied. Of the 35.9% (N = 28) of papers that included race data, 57.1% (N = 16) of study populations were majority White, 14.3% were majority Asian, and 7.1% were majority Black. In 75% (N = 21) of papers, White was the largest racial group represented. Of the papers included, 43.6% (N = 34) included the sex ratio of the patient population. DISCUSSION: With the power to build robust algorithms supported by massive quantities of clinical data, machine learning is shaping the future of diagnostics. Limitations of the underlying data create potential biases, especially if patient demographics are unknown or not included in the training. CONCLUSION: As the movement toward clinical reliance on machine learning accelerates, both recording demographic information and using diverse training sets should be emphasized. Extrapolating algorithms to demographics beyond the original study population leaves large gaps for potential biases.

Inferior and Superior Vena Cava Saturation Monitoring After Neonatal Cardiac Surgery.

OBJECTIVES: Superior vena cava oxygen saturation (SVC O 2 ) monitoring is well described for early detection of hemodynamic deterioration after neonatal cardiac surgery but inferior vena cava vein oxygen saturation (IVC O 2 ) monitoring data are limited. DESIGN: Retrospective cohort study of 118 neonates with congenital heart disease (52 single ventricle) from February 2008 to January 2014. SETTING: Pediatric cardiac ICU. PATIENTS: Neonates (< 30 d) with concurrent admission IVC O 2 and SVC O 2 measurements after cardiac surgery with cardiopulmonary bypass. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The primary aim was to correlate admission IVC O 2 and SVC O 2 . Secondary aims included: correlate flank or cerebral near-infrared spectroscopy with IVC O 2 and SVC O 2 , respectively, and exploratory analysis to evaluate associations between oximetry data and a composite adverse outcome defined as any of the following: increasing serum lactate or vasoactive support at 2 hours post-admission, cardiac arrest, or mortality. Admission IVC O 2 and SVC O 2 correlated ( r = 0.54; p < 0.001). However, IVC O 2 measurements were significantly lower than paired SVC O 2 (mean difference, -6%; 95% CI, -8% to -4%; p < 0.001) with wide variability in sample agreement. Logistic regression showed that each 12% decrease in IVC O 2 was associated with a 12-fold greater odds of the composite adverse outcome (odds ratio [OR], 12; 95% CI, 3.9-34; p < 0.001). We failed to find an association between SVC O 2 and increased odds of the composite adverse outcome (OR, 1.8; 95% CI, 0.99-3.3; p = 0.053). In an exploratory analysis, the area under the receiver operating curve for IVC O 2 and SVC O 2 , and the composite adverse outcome, was 0.85 (95% CI, 0.77-0.92) and 0.63 (95% CI, 0.52-0.73), respectively. Admission IVC O 2 had strong correlation with concurrent flank near-infrared spectroscopy value ( r = 0.74; p < 0.001). SVC O 2 had a weak association with cerebral near-infrared spectroscopy ( r = 0.22; p = 0.02). CONCLUSIONS: In postoperative neonates, admission IVC O 2 and SVC O 2 correlate. Lower admission IVC O 2 may identify a cohort of postsurgical neonates at risk for low cardiac output and associated morbidity.

The Risk of Coding Racism into Pediatric Sepsis Care: The Necessity of Antiracism in Machine Learning.

Machine learning holds the possibility of improving racial health inequalities by compensating for human bias and structural racism. However, unanticipated racial biases may enter during model design, training, or implementation and perpetuate or worsen racial inequalities if ignored. Pre-existing racial health inequalities could be codified into medical care by machine learning without clinicians being aware. To illustrate the importance of a commitment to antiracism at all stages of machine learning, we examine machine learning in predicting severe sepsis in Black children, focusing on the impacts of structural racism that may be perpetuated by machine learning and difficult to discover. To move toward antiracist machine learning, we recommend partnering with ethicists and experts in model development, enrolling representative samples for training, including socioeconomic inputs with proximate causal associations to racial inequalities, reporting outcomes by race, and committing to equitable models that narrow inequality gaps or at least have equal benefit.

Defining Physiological Decompensation: An Expert Consensus and Retrospective Outcome Validation.

OBJECTIVES: Physiological decompensation of hospitalized patients is common and is associated with substantial morbidity and mortality. Research surrounding patient decompensation has been hampered by the absence of a robust definition of decompensation and lack of standardized clinical criteria with which to identify patients who have decompensated. We aimed to: 1) develop a consensus definition of physiological decompensation and 2) to develop clinical criteria to identify patients who have decompensated. DESIGN: We utilized a three-phase, modified electronic Delphi (eDelphi) process, followed by a discussion round to generate consensus on the definition of physiological decompensation and on criteria to identify decompensation. We then validated the criteria using a retrospective cohort study of adult patients admitted to the Hospital of the University of Pennsylvania. SETTING: Quaternary academic medical center. PATIENTS: Adult patients admitted to the Hospital of the University of Pennsylvania who had triggered a rapid response team (RRT) response between January 1, 2019, and December 31, 2020. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Sixty-nine experts participated in the eDelphi. Participation was high across the three survey rounds (first round: 93%, second round: 94%, and third round: 98%). The expert panel arrived at a consensus definition of physiological decompensation, "An acute worsening of a patient's clinical status that poses a substantial increase to an individual's short-term risk of death or serious harm." Consensus was also reached on criteria for physiological decompensation. Invasive mechanical ventilation, severe hypoxemia, and use of vasopressor or inotrope medication were bundled as criteria for our novel decompensation metric: the adult inpatient decompensation event (AIDE). Patients who met greater than one AIDE criteria within 24 hours of an RRT call had increased adjusted odds of 7-day mortality (adjusted odds ratio [aOR], 4.1 [95% CI, 2.5-6.7]) and intensive care unit transfer (aOR, 20.6 [95% CI, 14.2-30.0]). CONCLUSIONS: Through the eDelphi process, we have reached a consensus definition of physiological decompensation and proposed clinical criteria with which to identify patients who have decompensated using data easily available from the electronic medical record, the AIDE criteria.

Improving CPR Quality by Using a Real-Time Feedback Defibrillator During Pediatric Simulation Training.

OBJECTIVE: The aim of this study was to assess the effectiveness of a defibrillator with real-time feedback during code team training to improve adherence to the American Heart Association (AHA) resuscitation guidelines. METHODS: This is a retrospective cohort study designed to compare pediatric resident adherence to the AHA cardiopulmonary resuscitation guidelines before and after use of real-time feedback defibrillator during code team training simulation. After institution of a real-time feedback defibrillator, first-year resident's adherence to the AHA guidelines for chest compression rate (CCR), fraction, and depth during code team training from January 2017 to December 2018 was analyzed. It was then compared with results of a previously published study from our institution that analyzed the CCR and fraction from January 2015 to January 2016, before the implementation of a defibrillator with real-time feedback. RESULTS: We compared 19 eligible session preintervention and 36 postintervention sessions. Chest compression rate and chest compression fraction (CCF) were assessed preintervention and postintervention. The depth of compression was only available postintervention. There was improvement in the proportion of code team training sessions with mean compression rate (74% preintervention vs 100% postintervention, P = 0.003) and mean CCF (79% vs 97%, P = 0.04) in adherence with the AHA guideline. CONCLUSIONS: The use of real-time feedback defibrillators improved the adherence to the AHA cardiopulmonary resuscitation guidelines for CCF and CCR during pediatric resident simulation.