RESUMO
Introduction: Most providers have routinely performed universal lumbar puncture (LP) on well-appearing, febrile infants 22 to 28 days old. In 2021, the American Academy of Pediatrics recommended clinicians should perform an LP in this age group if inflammatory markers are abnormal. This quality improvement project aimed to decrease LP rates in febrile infants 22 to 28 days old in the emergency department (ED) within 1 year, regardless of race/ethnicity, from a baseline of 87%. Methods: We used our institution's quality improvement framework to perform multiple Plan-Do-Study-Act cycles. A multidisciplinary team reviewed the febrile infant literature, local epidemiology, and identified key drivers. We provided departmental education, updated our clinical pathway, and used clinical decision support. We analyzed baseline (January 2017-March 2022) and intervention data (April 2022-March 2024) and tracked data using statistical process control charts. Our primary outcome measure was rates of LP in the ED for this cohort. Process measures included rates of infants with procalcitonin results. ED length of stay, rates of first LP attempt after hospitalization, and missed bacterial meningitis were balancing measures. Results: The baseline LP rate of 87% decreased to 44% during the intervention period, resulting in a downward centerline shift. There were no significant differences when LP rates were analyzed by race/ethnicity. There was an upward centerline shift in the process measure of infants with procalcitonin results. There was no observed special cause variation in our balancing measures. Conclusion: Quality improvement efforts, including education, clinical pathway updates, and clinical decision support, safely reduced rates of LPs in febrile infants 22 to 28 days old.
RESUMO
OBJECTIVES: To determine the incidence of return visits (RVs), types of RVs, and factors associated with RVs to a pediatric emergency department (ED). METHODS: : Retrospective cohort study of patients seen in an urban, tertiary care pediatric ED. MAIN OUTCOME: RV within 48 hours, identified from a computerized log. RESULTS: The total RV rate was 3.5% (95% confidence interval, 3.3-3.6), similar to rates (2.4% to 3.4%) reported in general EDs. Most (78.5%) RVs were unscheduled, 17% were scheduled, and 4% were called back to the ED. Infectious disease (45%), respiratory (16%), and trauma (16%) accounted for most RV diagnoses. When compared with the overall ED population, RV patients were more likely to be younger than 2 years [relative risk, 1.3 (1.2-1.4)], to be admitted to the hospital [relative risk, 1.3 (1.2-1.5)], and to be triaged as acute [relative risk, 1.1 (1.0-1.2)]. Patients called back to the ED were younger, more likely to be triaged as acute, and more likely to be admitted than other RV patients. Significant diagnoses were made at RV in 7 (0.4%; 95% confidence interval, 0.1-0.7) patients, half of whom were called back to the ED or had a scheduled RV. CONCLUSION: Similarities between our pediatric ED RV rate and other published research implies that benchmarking and quality improvement tools for RV can be used and compared in both pediatric and general EDs. Focusing on systems to call patients back to the ED when necessary may be an efficient way to reduce medical error and adverse patient outcomes.