Cost analysis of hospitals performing continuous albuterol in non-intensive care settings.

OBJECTIVE: To compare hospital costs and resource utilization for pediatric asthma admissions based on the hospitals' availability of continuous albuterol aerosolization administration (CAA) in non-intensive care unit (ICU) settings. METHODS: We conducted a retrospective cohort study of children ages 2-17 years admitted in 2019 with a principal diagnosis of asthma using the Pediatric Health Information System. Hospitals and hospitalizations were categorized based on location of CAA administration, ICU-only versus general inpatient floors. Hospitals preforming CAA in an intermediate care unit were excluded. We calculated total cost, standardized unit costs and rates of interventions. Groups were compared using Chi-Square, t-test and Wilcoxon rank-sum test as indicated. A log linear mixed model was created to evaluate potential confounders. RESULTS: Twenty-one hospitals (7084 hospitalizations) allowed CAA on the floor.Twenty-four hospitals (6100 hospitalizations) allowed CAA in the ICU-only. Median total cost was $4639 (Interquartile Range (IQR) $3060-$7512) for the floor group and $5478 (IQR $3444-$8539) for the ICU-only group (p < 0.001) (mean cost difference of $775 per patient). Hospitalization costs were $4,726,829 (95% CI $3,459,920-$5,993,860) greater for the children treated at hospitals restricting CAA to the ICU. We observed higher standardized laboratory, imaging, clinical and other unit costs, along with higher use of interventions in the ICU-only group. After adjustment, we found that ICU stay and hospital LOS were the main drivers of cost difference between the groups. CONCLUSIONS: There was cost savings and decreased resource utilization for hospitals that performed CAA on the floor. Further studies exploring variations in asthma management are warranted.

The Deterioration Risk Index: Developing and Piloting a Machine Learning Algorithm to Reduce Pediatric Inpatient Deterioration.

OBJECTIVES: Develop and deploy a disease cohort-based machine learning algorithm for timely identification of hospitalized pediatric patients at risk for clinical deterioration that outperforms our existing situational awareness program. DESIGN: Retrospective cohort study. SETTING: Nationwide Children's Hospital, a freestanding, quaternary-care, academic children's hospital in Columbus, OH. PATIENTS: All patients admitted to inpatient units participating in the preexisting situational awareness program from October 20, 2015, to December 31, 2019, excluding patients over 18 years old at admission and those with a neonatal ICU stay during their hospitalization. INTERVENTIONS: We developed separate algorithms for cardiac, malignancy, and general cohorts via lasso-regularized logistic regression. Candidate model predictors included vital signs, supplemental oxygen, nursing assessments, early warning scores, diagnoses, lab results, and situational awareness criteria. Model performance was characterized in clinical terms and compared with our previous situational awareness program based on a novel retrospective validation approach. Simulations with frontline staff, prior to clinical implementation, informed user experience and refined interdisciplinary workflows. Model implementation was piloted on cardiology and hospital medicine units in early 2021. MEASUREMENTS AND MAIN RESULTS: The Deterioration Risk Index (DRI) was 2.4 times as sensitive as our existing situational awareness program (sensitivities of 53% and 22%, respectively; p < 0.001) and required 2.3 times fewer alarms per detected event (121 DRI alarms per detected event vs 276 for existing program). Notable improvements were a four-fold sensitivity gain for the cardiac diagnostic cohort (73% vs 18%; p < 0.001) and a three-fold gain (81% vs 27%; p < 0.001) for the malignancy diagnostic cohort. Postimplementation pilot results over 18 months revealed a 77% reduction in deterioration events (three events observed vs 13.1 expected, p = 0.001). CONCLUSIONS: The etiology of pediatric inpatient deterioration requires acknowledgement of the unique pathophysiology among cardiology and oncology patients. Selection and weighting of diverse candidate risk factors via machine learning can produce a more sensitive early warning system for clinical deterioration. Leveraging preexisting situational awareness platforms and accounting for operational impacts of model implementation are key aspects to successful bedside translation.

The Diagnostic Error Index: A Quality Improvement Initiative to Identify and Measure Diagnostic Errors.

OBJECTIVE: To develop a diagnostic error index (DEI) aimed at providing a practical method to identify and measure serious diagnostic errors. STUDY DESIGN: A quality improvement (QI) study at a quaternary pediatric medical center. Five well-defined domains identified cases of potential diagnostic errors. Identified cases underwent an adjudication process by a multidisciplinary QI team to determine if a diagnostic error occurred. Confirmed diagnostic errors were then aggregated on the DEI. The primary outcome measure was the number of monthly diagnostic errors. RESULTS: From January 2017 through June 2019, 105 cases of diagnostic error were identified. Morbidity and mortality conferences, institutional root cause analyses, and an abdominal pain trigger tool were the most frequent domains for detecting diagnostic errors. Appendicitis, fractures, and nonaccidental trauma were the 3 most common diagnoses that were missed or had delayed identification. CONCLUSIONS: A QI initiative successfully created a pragmatic approach to identify and measure diagnostic errors by utilizing a DEI. The DEI established a framework to help guide future initiatives to reduce diagnostic errors.

Trends in adult patients presenting to pediatric emergency departments.

OBJECTIVE: Adults presenting to pediatric emergency departments (PEDs) include those with complex chronic conditions (CCCs) often still followed by pediatric providers, and those without CCCs (non-CCCs). This paper describes recent trends in adults seen within PEDs, both by age subgroups and CCC status. METHODS: Data were retrospectively reviewed from the Pediatric Health Information System for PED visits between Jan 1, 2013 and Dec 31, 2017. Yearly visit rates were trended for all adult visits, age subgroups (18-21, 22-25, 26-40, and 41-95 years), and by CCC status. The most frequent diagnoses were reviewed for each adult age group and CCC category. RESULTS: Rates of adult PED visits significantly increased from 3.7% in 2013 to 4.2% in 2017 (P < 0.0001). While the overall majority (88% overall and >70% for each age group) of adult PED patients were non-CCC, the rate of CCC patients increased overall (P < 0.0001), especially among older patients (41-95 years) from 8.7% in 2013 to 29% in 2017 (P < 0.0001). The most frequent CCC diagnoses were sickle cell disease (18-21 and 22-25 years), cardiac pathology (26-40 years), and hyperlipidemia (41-95 years). The most frequent non-CCC diagnoses were abdominal pain (18-21 years), pregnancy-related (22-25 and 26-40 years), and chest pain (41-95 years). CONCLUSION: Visits to PEDs by adult patients have increased over time, with the majority of the population being represented by non-CCC adults. The results of this study can be used to guide specifics in adult medicine training during Pediatric Emergency Medicine fellowship and assist in continuing education efforts.

Comparing Outcomes Between Direct and ED Admissions for Neonatal Hyperbilirubinemia.

OBJECTIVES: Pediatric direct admissions (DA) have multiple benefits including reduced emergency department (ED) volumes, greater patient and provider satisfaction, and decreased costs without compromising patient safety. We sought to compare resource utilization and outcomes between patients with a primary diagnosis of neonatal hyperbilirubinemia directly admitted with those admitted from the ED. METHODS: Single-center, retrospective study at a large, academic, free-standing children's hospital (2017-2021). Patients were between 24 hours and 14 days old with a gestational age of ≥35 weeks, admitted with a primary diagnosis of neonatal hyperbilirubinemia. Outcomes included length of stay (LOS), time to clinical care, resource utilization, NICU transfer, and 7-day readmission for phototherapy. RESULTS: A total of 1098 patients were included, with 276 (25.1%) ED admissions and 822 (74.9%) DAs. DAs experienced a shorter median time to bilirubin level collection (1.9 vs 2.1 hours, P = .003), received less intravenous fluids (8.9% vs 51.4%, P < .001), had less bilirubin levels collected (median of 3.0 vs 4.0, P < .001), received phototherapy sooner (median of 0.8 vs 4.2 hours, P < .001), and had a shorter LOS (median of 21 vs 23 hours, P = .002). One patient who was directly admitted required transfer to the NICU. No differences were observed in the 7-day readmission rates for phototherapy. CONCLUSIONS: Directly admitting patients for the management of neonatal hyperbilirubinemia is a preferred alternative to ED admission as our study demonstrated that DAs had a shorter time to clinical care, shorter LOS, and less unnecessary resource utilization with no difference in 7-day readmissions for phototherapy.

Direct Admission Program Implementation: A Qualitative Analysis of Variation Across Health Systems.

OBJECTIVES: Direct admission (DA) to the hospital has the potential to improve family satisfaction and timeliness of care by bypassing the emergency department. Using the RE-AIM implementation framework, we sought to characterize variation across health systems in the reach, effectiveness, adoption, and implementation of a DA program from the perspectives of parents and multidisciplinary clinicians. METHODS: As part of a stepped-wedge cluster randomized trial to compare the effectiveness of DA to admission through the emergency department, we evaluated DA rates across 69 clinics and 3 health systems and conducted semi-structured interviews with parents and clinicians. We used thematic analysis to identify themes related to the reach, effectiveness, adoption, and implementation of the DA program and applied axial coding to characterize thematic differences across sites. RESULTS: Of 2599 hospitalizations, 171 (6.6%) occurred via DA, with DA rates varying 10-fold across health systems from 0.9% to 9.3%. Through the analysis of 137 interviews, including 84 with clinicians and 53 with parents, we identified similarities across health systems in themes related to perceived program effectiveness and patient and family engagement. Thematic differences across sites in the domains of program implementation and clinician adoption included variation in transfer center efficiency, trust between referring and accepting clinicians, and the culture of change within the health system. CONCLUSIONS: The DA program was adopted variably, highlighting unique challenges and opportunities for implementation in different hospital systems. These findings can inform future quality improvement efforts to improve transitions to the hospital.

Effectiveness of Direct Admission Compared to Admission Through the Emergency Department: A Stepped-Wedge Cluster-Randomized Trial.

OBJECTIVE: Direct admission (DA) to hospital can reduce emergency department (ED) utilization by bypassing the ED during the admission process. We implemented a DA program across 3 health systems and compared timeliness of care, family experience of care, and post-admission clinical deterioration among children admitted via DA versus the ED after their clinic was randomized to begin the DA program. METHODS: Using a stepped-wedge design, 69 primary and urgent care clinics were randomized to 1 of 4 time points to begin a voluntary DA program, February 1, 2020 to April 30, 2023. Outcomes in children <18 years admitted with 7 common medical diagnoses were compared using adjusted logistic or linear regression. RESULTS: A total of 2599 children were admitted with eligible diagnoses during the study period , including 145 children admitted directly and 1852 admitted through EDs after program implementation at their clinic. Median age was 2.8 (interquartile range: 1.1-6.8) years, 994 (49.8%) were female, and 1324 (66.3%) were Medicaid-insured. Adjusted regression analyses showed that if each child was admitted via DA versus the ED, average time to initial clinical assessment was 3.1 minute (95% confidence interval: 1.7-4.5) slower, whereas time to initial therapeutic management was 49.6 minutes faster on average (95% confidence interval: 30.3.2-68.9). There were no significant differences in time to initial diagnostic testing or rates of post-admission clinical deterioration. CONCLUSIONS: Compared with ED admission, DA appears equally safe and acceptable to families, and may be associated with a significantly shorter time to initial therapeutic management with modestly longer time to initial clinical assessment.