Harnessing the Power of Generative AI for Clinical Summaries: Perspectives From Emergency Physicians.

STUDY OBJECTIVE: The workload of clinical documentation contributes to health care costs and professional burnout. The advent of generative artificial intelligence language models presents a promising solution. The perspective of clinicians may contribute to effective and responsible implementation of such tools. This study sought to evaluate 3 uses for generative artificial intelligence for clinical documentation in pediatric emergency medicine, measuring time savings, effort reduction, and physician attitudes and identifying potential risks and barriers. METHODS: This mixed-methods study was performed with 10 pediatric emergency medicine attending physicians from a single pediatric emergency department. Participants were asked to write a supervisory note for 4 clinical scenarios, with varying levels of complexity, twice without any assistance and twice with the assistance of ChatGPT Version 4.0. Participants evaluated 2 additional ChatGPT-generated clinical summaries: a structured handoff and a visit summary for a family written at an 8th grade reading level. Finally, a semistructured interview was performed to assess physicians' perspective on the use of ChatGPT in pediatric emergency medicine. Main outcomes and measures included between subjects' comparisons of the effort and time taken to complete the supervisory note with and without ChatGPT assistance. Effort was measured using a self-reported Likert scale of 0 to 10. Physicians' scoring of and attitude toward the ChatGPT-generated summaries were measured using a 0 to 10 Likert scale and open-ended questions. Summaries were scored for completeness, accuracy, efficiency, readability, and overall satisfaction. A thematic analysis was performed to analyze the content of the open-ended questions and to identify key themes. RESULTS: ChatGPT yielded a 40% reduction in time and a 33% decrease in effort for supervisory notes in intricate cases, with no discernible effect on simpler notes. ChatGPT-generated summaries for structured handoffs and family letters were highly rated, ranging from 7.0 to 9.0 out of 10, and most participants favored their inclusion in clinical practice. However, there were several critical reservations, out of which a set of general recommendations for applying ChatGPT to clinical summaries was formulated. CONCLUSION: Pediatric emergency medicine attendings in our study perceived that ChatGPT can deliver high-quality summaries while saving time and effort in many scenarios, but not all.

Timing of Antibiotic Administration in Pediatric Sepsis.

OBJECTIVES: Antibiotic administration within 1 hour of hypotension has been shown to reduce mortality. It is unknown whether antibiotics before hypotension in children who eventually meet criteria for septic shock improves outcomes. This study assesses whether antibiotic timing from the time of meeting criteria for sepsis in children with septic shock impacts morbidity and mortality. METHODS: This is a retrospective study of children 18 years or younger presenting to a tertiary free-standing children's hospital emergency department with sepsis that subsequently progressed to septic shock and were admitted to an intensive care unit from 2008 to 2012. The time when the patient met criteria for sepsis to the time of first antibiotic administration was assessed and correlated with patient morbidity and mortality. RESULTS: Among 135 children (median age, 13.1 years), 34 (25%) were previously healthy, whereas 49 (36%) had 2 or more medical comorbidities. Twenty-seven children (20%) had positive blood cultures, 17 (13%) had positive urine cultures, and 34 (25%) had chest x-ray findings that were interpreted as pneumonia. Among the 42 (31%) with antibiotics within 1 hour from criteria for sepsis, there was higher mortality (4/42 vs 0/93, P = 0.009), more organ dysfunction, longer time on a vasoactive infusion, and increased intensive care unit and hospital lengths of stay (all P < 0.05). CONCLUSIONS: Children with criteria for sepsis who subsequently progressed to septic shock who received antibiotics within 1 hour of meeting sepsis criteria had increased mortality, length of stay, and organ dysfunction.

Reducing Pediatric Emergency Department Prescription Errors.

BACKGROUND: Prescription errors are a significant cause of iatrogenic harm in the health care system. Pediatric emergency department (ED) patients are particularly vulnerable to error. We sought to decrease prescription errors in an academic pediatric ED by 20% over a 24-month period by implementing identified national best practice guidelines. METHODS: From 2017 to 2019, a multidisciplinary, fellow-driven quality improvement (QI) project was conducted using the Model for Improvement. Four key drivers were identified including simplifying the electronic order entry into prescription folders, improving knowledge of dosing by indication, increasing error feedback to prescribers, and creating awareness of common prescription pitfalls. Four interventions were subsequently implemented. Outcome measures included prescription errors per 1000 prescriptions written for all medications and top 10 error-prone antibiotics. Process measures included provider awareness and use of prescription folders; the balancing measure was provider satisfaction. Differences in outcome measures were assessed by statistical process control methodology. Process and balancing measures were analyzed using 1-way analysis of variance and χ2 testing. RESULTS: Before our interventions, 8.6 errors per 1000 prescriptions written were identified, with 62% of errors from the top 10 most error-prone antibiotics. After interventions, error rate per 1000 prescriptions decreased from 8.6 to 4.5 overall and from 20.1 to 8.8 for top 10 error-prone antibiotics. Provider awareness of prescription folders was significantly increased. CONCLUSION: QI efforts to implement previously defined best practices, including simplifying and standardizing computerized provider order entry (CPOE), significantly reduced prescription errors. Synergistic effect of educational and technological efforts likely contributed to the measured improvement.

Improving Discharge Safety in a Pediatric Emergency Department.

BACKGROUND AND OBJECTIVES: Discharge from the emergency department (ED) involves a complex series of steps to ensure a safe transition to home and follow-up care. Preventable, discharge-related serious safety events (SSEs) in our ED highlighted local vulnerabilities. We aimed to improve ED discharge by implementing a standardized discharge process with emphasis on multidisciplinary communication and family engagement. METHODS: At a tertiary children's hospital, we used the model for improvement to revise discharge care. Interventions included a new discharge checklist, a provider huddle emphasizing discharge vital signs, and a scripted discharge review of instructions with families. We used statistical process control to evaluate performance. Primary outcomes included elimination of preventable, discharge-related SSEs and Press Ganey survey results assessing caregiver information for care of child at home. A secondary outcome was number of days between preventable low-level (near-miss, no or minimal harm) events. Process measures included discharge checklist adoption and vital sign acquisition. Balancing measures were length of stay (LOS) and return rates. RESULTS: Over the study period, there were no preventable SSEs and low-level event frequency improved to a peak of >150 days between events. Press Ganey responses regarding quality of discharge information did not change (62%). Checklist use was rapidly adopted, reaching 94%. Vital sign acquisition increased from 67% to 83%. There was no change in the balancing measures of median LOS or return visit rates. CONCLUSIONS: The development and implementation of a standardized discharge process led to the elimination of reported discharge-related events, without increasing LOS or return visits.

Reduction in Resources and Cost for Gastroenteritis Through Implementation of Dehydration Pathway.

OBJECTIVES: Management decisions for patients with gastroenteritis affect resource use within pediatric emergency departments (EDs), and algorithmic care using evidence-based guidelines (EBGs) has become widespread. We aimed to determine if the implementation of a dehydration EBG in a pediatric ED resulted in a reduction in intravenous (IV) fluid administration and the cost of care. METHODS: In a single-center quality improvement initiative between 2010 and 2016, investigators aimed to decrease the percentage of patients with gastroenteritis who were rehydrated with IV fluids. The EBG assigned the patient a dehydration score with subsequent rehydration strategy on the basis of presenting signs and symptoms. The primary outcome was proportion of patients receiving IV fluid, which was analyzed using statistical process control methods. The secondary outcome was cost of the episode of care. Balancing measures included ED length of stay, admission rate, and return visit rate within 72 hours. RESULTS: A total of 7145 patients met inclusion criteria with a median age of 17 months. Use of IV fluid decreased from a mean of 15% to 9% postimplementation. Average episode of care-related health care costs decreased from $599 to $410. For our balancing measures, there were improvements in ED length of stay, rate of admission, and rate of return visits. CONCLUSIONS: Implementation of an EBG for patients with gastroenteritis led to a decrease in frequency of IV administration, shorter lengths of stay, and lower health care costs.