Adoption of a Pediatric Acute Care Cardiology Inpatient Model.

Importance: Despite widespread adoption and investment of resources nationally, the pediatric acute care cardiology (ACC) model of care has not been previously evaluated prospectively. Objective: To test the hypothesis that adoption of an ACC model will be associated with improved clinical outcomes. Design, Setting, and Participants: This single-center prospective quality improvement study was conducted in a 26-bed ACC unit of a high surgical volume, freestanding children's hospital. The baseline period was May 15 to October 31, 2023, and the intervention period was November 1, 2023, to November 30, 2024. All ACC unit encounters during the baseline and intervention periods were included. Data sources were hospital administrative data, local Pediatric Acute Care Cardiology Collaborative registry, and Patient and Family Experience (PFE) scores. Exposure: Full-scale change in the model of care: transitioned unit leadership from hospital pediatrics to cardiology, changed attending of record for medical patients to cardiologist, hired nurse practitioners as frontline clinicians, integrated residents into the team, implemented multidisciplinary family-centered rounding, updated communication processes, and transitioned cardiology fellows to in-house overnight call. Main Outcomes and Measures: Complication rate and back transfer to the intensive care unit (ICU) were outcome measures, discharge time was a process measure, 7-day unplanned readmissions and length of stay (LOS) were balancing measures. Standard rules for identifying special cause variation (SCV) were applied. The percentage of patients and families with positive PFE scores (defined as scores of 9 or 10) before and after the intervention were compared using an independent t test. Hypothesis was formulated prior to data collection. Results: There were 483 encounters (45.2% among children aged 1-18 years) in the baseline period and 973 (52.7% among children aged 1-18 years) in the intervention period. Outcome and process measures significantly improved showing SCV following adoption of the ACC model (mean complications: baseline, 23.6% vs intervention, 16.0%; mean back transfer to ICU: baseline, 11.4% vs intervention, 6.9%; mean patient discharge time: baseline, 15.37 hours vs intervention, 14.43 hours). LOS and 7-day unplanned readmissions were unchanged, suggesting no major inadvertent negative consequences of the ACC model. Mean LOS for medical patients decreased (7.83 vs 4.97 days). PFE improved after the intervention (median [SD], preintervention: 76.9% [3.7] vs postintervention: 82.9% [4.3]; P = .04). Conclusions and Relevance: In this quality improvement study of an ACC model, multiple outcomes improved without evidence of negative consequences. These clinical improvements may justify necessary investment of resources to support ACC models. Adaptation of this model for other subspecialties may help address pediatric resident workforce changes. Ongoing evaluation of resource utilization, sustainability of improvement, and newly embedded improvement efforts is underway.