Children's Health Council

Patient safety event reporting systems are widely used, yet organisations face challenges analysing the high volume of incident reports. While low-harm events represent the majority of submissions, they are rarely examined systematically due to the time and resources required to manually review the complex, lengthy, narrative data. Emerging technology like large language models (LLMs) offers new opportunities to address this gap. This study aimed to develop and evaluate an artificial intelligence-driven approach to identify patient safety issues, uncover system-level trends and assess readiness for implementation within a US healthcare system. We quantitatively evaluated OpenAI's GPT-4o model accuracy in analysing patient safety event reports and qualitatively assessed pre-implementation outcomes. The LLM extracted safety problems from 9357 free-text narratives and then generated a taxonomy of 'parent' and 'child' subcategories. The model labelled every report's problem list using the taxonomy, and dashboards were developed to visualise trends. Patient safety experts reviewed two separate subsets of reports (n=100 and n=219) to validate the LLM's accuracy. We conducted 10 stakeholder interviews to assess the dashboards' acceptability, appropriateness and adoption. The LLM had scores of 94% mean agreement among reviewers in identifying problems, and 91.5% and 83.3% agreement in assigning 'parent' and 'child' category labels, respectively. The model identified previously hidden patterns of patient safety issues. Stakeholders described LLM-generated insights as clear, appropriate and valuable for their work. They perceived few barriers to adoption and believed the model could expedite manual report reviews and support system-level quality improvement. LLMs offer an approach to capture and analyse multidimensional concepts in patient safety reports. By extracting problem summaries and categorising them into an accepted taxonomy, they can expose previously unidentified trends and provide visibility into system-level risks. LLMs could effectively augment and expedite manual reviews of safety events, while guiding prioritisation of quality and safety interventions.

Abstract Background Pediatric blood cultures are frequently ordered but have low positivity rates (< 4%) in emergency departments (EDs), highlighting the need for better-targeted testing. Accurate prediction can reduce unnecessary cultures, conserve resources, and support stewardship—particularly during the global blood culture bottle shortage. Models developed for adults perform poorly in children due to physiological and clinical differences; in prior work, applying an adult model to pediatric data yielded an AUC of 0.61. We excluded infants < 90 days, who have distinct risk factors (e.g., perinatal history), and developed machine learning models to predict bacteremia in children aged > 90 days to ≤ 18 years using electronic health record (EHR) data.Table 1:PedsBactoScore Point-Based Scoring System Derived from Logistic Regression CoefficientsEach feature contributes a fixed number of points based on clinically meaningful thresholds. The total score is used to stratify risk of bacteremia at the point-of-care.Table 2:Performance Metrics of Pediatric Bacteremia Prediction ModelsComparison of PedsBactoRisk and PedsBactoScore models on the pediatric test set. Metrics include AUC with 95% confidence intervals, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) at pre-specified thresholds. Methods We analyzed 26,829 blood culture orders from 9,362 pediatric emergency department (ED) encounters at Stanford Medicine | Children's Health. To preserve temporal validity, data were split chronologically, with the most recent encounters used as the test set. We developed two models: PedsBactoRisk, a logistic regression model, and PedsBactoScore, a simplified point-based tool derived from the most influential PedsBactoRisk predictors. The PedsBactoScore rubric is shown in Table 1. Results Table 2 summarizes model performance. We evaluated sensitivity, specificity, PPV, and NPV, focusing on thresholds achieving 90% sensitivity. PedsBactoRisk achieved an AUC-ROC of 0.75; PedsBactoScore, 0.64. While PedsBactoRisk showed superior performance, PedsBactoScore allows easier implementation via its interpretable scoring system. PedsBactoScore performance is shown across thresholds to illustrate sensitivity–specificity trade-offs. Conclusion PedsBactoRisk demonstrated the highest overall performance (AUC: 0.75), but PedsBactoScore offers a pragmatic, interpretable bedside tool with strong sensitivity. Both models support more judicious blood culture use by identifying low-risk patients with high sensitivity. Future work will focus on integrating provider notes using large language models to enhance predictive accuracy and extending this approach to infants < 90 days by incorporating maternal and delivery data. Disclosures Jonathan H. Chen, MD, PhD, Reaction Explorer: Ownership Interest

Identifying patient portals (PP) activation disparities, especially in electronic health record (EHR) activation workflows, can help facilitate equitable health care access.Our study aimed to assess whether the parent/guardian's preferred language was associated with being offered, activating, and using the PP and the methods used to offer activation codes.This retrospective cohort study examined PP offer, activation, and usage rates at a large freestanding children's hospital. Patients <12 years old with ambulatory visits from July 1, 2022, to June 30, 2023, without prior active proxy PP accounts were included. The primary independent variable was the self-reported parent/guardian preferred language (English/Spanish). Outcomes included the probability of being offered, overall and by specific offer methods, activation, and usage. Zou's modified multivariate Poisson regression models examined the association between preferred language and offer/activate/use status.Among 39,578 patients, 85.1% were patients with English as preferred language (PEPL) and 14.9% had Spanish as preferred language (PSPL). PSPL had a lower probability of being offered (adjusted relative risk ratio [aRR]: 0.65, 95% confidence interval [CI]: 0.63-0.67), activated (aRR: 0.72, 95% CI: 0.70-0.75), and used (aRR: 0.68, 95% CI: 0.65-0.72) a PP compared to PEPL. Specifically, PSPL had a lower probability of activating if ever offered via instant activation (aRR: 0.72, 95% CI: 0.69-0.75), parent/guardian with existing account (aRR: 0.73, 95% CI: 0.69-0.76), and had equal probability of activating if ever offered via letter (aRR: 0.42, 95% CI: 0.19-0.94) and clinician-assisted method (aRR: 0.99, 95% CI: 0.86-1.16), compared to PEPL.PSPL at a large, free-standing pediatric health system had a lower probability of PP offer, activation, and usage than PEPL. Activation methods were not universally effective across language groups, emphasizing the need for equitable workflow optimization. This study highlights an approach to analyzing health disparities in activation workflows to inform targeted interventions to improve equitable PP access.

Understanding outcomes of spinal fusion (SF) in children with cerebral palsy (CP) beyond the short term is important to determine efficacy and durability. This study examined complications, unplanned returns to the operating room (UPROR), and radiographic and clinical outcomes after SF in children with CP. Patients with GMFCS IV or V CP who had been followed for a minimum of 5 years after SF were identified in a prospective multicenter database and analyzed. The major Cobb angle and pelvic obliquity (PO) were recorded. Data regarding complications and any UPROR were collected prospectively. The Caregiver Priorities and Child Health Index of Life with Disabilities (CPCHILD) questionnaire was completed preoperatively and at 2 and 5 years postoperatively. Data were analyzed using a generalized estimating equation (GEE), repeated-measures analysis of variance (RM-ANOVA), and the Kaplan-Meier method. One hundred and eighty-nine patients (mean age, 13.4 ± 2.7 years; 94 male and 95 female; 45% White, 24% Hispanic, 16% Black, and 4% Asian as self-reported) were included in the study. The preoperative Cobb angle (82.8° ± 23.0°) was improved at 2 years (29.9° ± 16.2°; p < 0.001), and the improvement was maintained at 5 years (30.2° ± 17.0°; p = 0.284). The preoperative PO (27.4° ± 15.6°) was improved at 2 years (9.7° ± 9.1°; p < 0.001), and the improvement was maintained at 5 years (9.9° ± 9.8°; p = 0.997). There were 46 patients (24.3%) with major complications and 25 patients (13.2%) who required UPROR, mostly within the first year. The probability of remaining free of major complications and of having no UPROR at 5 years was >75% and >87%, respectively. Improvements in CPCHILD scores were observed at 2 years and remained improved at 5 years compared with baseline. RM-ANOVA demonstrated no significant differences in the change in scores over time in patients with complications or UPROR compared with those who did not. Those with complications or UPROR showed no evidence of a major decline in CPCHILD scores. SF resulted in durable radiographic correction and sustained improvements in caregiver-reported quality of life over a minimum of 5 years. Patients with major complications or UPROR showed no deterioration in CPCHILD scores. These findings support SF as an effective treatment option for appropriately selected patients with GMFCSIV or V CP. Therapeutic Level IV . See Instructions for Authors for a complete description of levels of evidence.

Pediatric palliative care, including end-of-life care, remains a relatively new area of interdisciplinary clinical practice and research. Improving the multifaceted and complex care of children and their families involves research that (1) documents the experiences of children with serious illness, their families, and clinicians; (2) evaluates relationships between contextual factors and health outcomes; and (3) establishes a stronger foundation for child- and family-focused interventions to improve care. Partnership among stakeholders in family-focused research begins from design through conduct of the study. This partnership is the foundation of a dynamic research process that illuminates critical perspectives. We present a hypothetical pediatric palliative study; a qualitative descriptive study of the perspectives of adolescents and young adults with life-limiting illnesses and their parents after a discussion about an end-of-life decision. Pediatric palliative care researchers, institutional review board leaders, and the parent of a child who died comment on how to balance the obligations to improve clinical care and to protect participants in research. Their recommendations include recruiting a wide range of participants, differentiating emotional responses from harm, approaching potential participants as individuals, and seeking feedback from family advisory boards and designated reviewers with content expertise.

Ghana has a large and growing burden of injury morbidity and mortality. There is a substantial unmet need for trauma surgery, highlighting a need to understand gaps in care. We conducted 8 in-depth interviews with trauma care providers (surgeons, nurses, and specialists) at a large teaching hospital to understand factors that contribute to and reduce delays in the provision of adequate trauma care for severely injured patients. The study aimed to understand whether providers thought factors differed between patients that were enrolled in the National Health Insurance Scheme (NHIS) and those that were not. Findings were presented for the third delay (provision of appropriate care)in the Three Delays Framework. Key findings included that most factors contributing delays in the provision of adequate care were related to the costs of care, including for diagnostics, medications, and treatment for patients with and without NHIS subscription. Other notable factors included conflicts between providers, resource constraints, and poor coordination of care at the facility. Factors which reduce delays included advocacy by providers and informal processes for prioritizing critical injuries. We recommend facility-level changes including increasing equity in access to trauma and elective surgery through targeted system strengthening efforts (e.g., a scheduled back-up call system for surgeons, anesthetists, other specialists, and nurses; designated operating theatres and staff for emergencies; training of staff), policy changes to simplify the insurance renewal and subscription processes, and future research on the costs and benefits of including diagnostics, medications, and common trauma services into the NHIS benefits package.

SynopsisMenstrual health hygiene equity involves increasing awareness about the global environmental impact of disposable menstrual products and creating equitable opportunity for healthy hygiene products.