RESUMEN
OBJECTIVES: To determine if ChatGPT can incorporate patient-specific information to provide high-quality answers to parental questions in the PICU. We hypothesized that ChatGPT would generate high-quality, patient-specific responses. METHODS: In this cross-sectional study, we generated assessments and plans for 3 PICU patients with respiratory failure, septic shock, and status epilepticus and paired them with 8 typical parental questions. We prompted ChatGPT with instructions, an assessment and plan, and 1 question. Six PICU physicians evaluated the responses for accuracy (1-6), completeness (yes/no), empathy (1-6), and understandability (Patient Education Materials Assessment Tool, PEMAT, 0% to 100%; Flesch-Kincaid grade level). We compared answer quality among scenarios and question types using the Kruskal-Wallis and Fischer's exact tests. We used percent agreement, Cohen's Kappa, and Gwet's agreement coefficient to estimate inter-rater reliability. RESULTS: All answers incorporated patient details, utilizing them for reasoning in 59% of sentences. Responses had high accuracy (median 5.0, [interquartile range (IQR), 4.0-6.0]), empathy (median 5.0, [IQR, 5.0-6.0]), completeness (97% of all questions), and understandability (PEMAT % median 100, [IQR, 87.5-100]; Flesch-Kincaid level 8.7). Only 4/144 reviewer scores were <4/6 in accuracy, and no response was deemed likely to cause harm. There was no difference in accuracy, completeness, empathy, or understandability among scenarios or question types. We found fair, substantial, and almost perfect agreement among reviewers for accuracy, empathy, and understandability, respectively. CONCLUSIONS: ChatGPT used patient-specific information to provide high-quality answers to parental questions in PICU clinical scenarios.
RESUMEN
OBJECTIVES: We sought to create a digital application to support clinicians in empiric and pathogen-directed antibiotic ordering based on local susceptibility patterns and evidence-based treatment durations, thereby promoting antimicrobial stewardship. METHODS: We formed a multidisciplinary team that met bimonthly from 2017 to 2018 to design and construct a web-based antimicrobial stewardship platform called Antibiogram + . We used an iterative and agile technical development process with frequent feedback from clinicians. RESULTS: Antibiogram+ is an online tool, accessible via the electronic health record and hospital intranet, which offers institutional antibiotic susceptibilities for major pathogens, recommendations for empiric antibiotic selection and treatment durations for common pediatric conditions, antimicrobial dosing and monitoring guidance, and links to other internal clinical decision support resources. The tool was accessed 11,823 times with 492 average monthly views during the first 2 years after release. Compared with use of a preexisting print antibiogram and dosing card, pediatric residents more frequently reported "often" being sure of antibiotic dosing with Antibiogram+ (58 vs. 15%, p < 0.01). Respondents also reported improved confidence in choice of antibiotic, but this finding did not reach statistical significance (55 vs. 35%, p = 0.26). CONCLUSION: We report the successful development of a digital antimicrobial stewardship platform with consistent rates of access during the first 2 years following release and improved provider comfort with antibiotic management.