Hospital pediatrics最新文献

Objective: Pediatric somatic symptom and related disorders (SSRDs) negatively impact patients' lives and can lead to hospitalization. No national U.S. studies have examined inpatient SSRD prevalence and health care utilization. We aimed to report demographic characteristics and health care utilization of patients admitted for pediatric SSRDs using the Pediatric Health Information System (PHIS) and identify characteristics associated with high utilization.

Methods: We included inpatient encounters from 2016 through 2023 for patients 5 to 21 years old with an admitting or primary SSRD diagnosis. We used descriptive statistics to summarize patient- and encounter-level variables and multivariable logistic regression to identify factors independently associated with high utilization (readmission or >75th percentile length of stay [LOS] or cost).

Results: There were 6820 encounters and 6297 patients from 48 hospitals. Most patients were 12 to 18 years old (n = 4889, 78%), female (74%, n = 4666), and non-Hispanic white (55%, n = 3449). Median LOS was 2 days (IQR, 1-4), and median encounter cost was $7946 (IQR, 4737-13 701). Four hundred sixteen patients (7%) had multiple admissions. Carrying 7 or more diagnoses and Northeast hospitalization were associated with higher odds of prolonged LOS, high cost, and readmission. Odds of prolonged LOS alone were higher with commercial insurance (odds ratio [OR], 1.19 [95% CI, 1.03-1.36]) and comorbid anxiety (OR, 1.29 [95% CI, 1.10-1.51]). Hispanic patients had higher odds of high-cost admission (OR, 1.57 [95% CI, 1.24-1.99]) compared with non-Hispanic white patients.

Conclusions: An average of 853 admissions for SSRDs occur annually across 48 US tertiary care children's hospitals and affect a sociodemographically diverse population. Further investigation into drivers of disparate health care utilization is needed.

Background: Elevated blood lead levels are still prevalent among children in the United States, and there continue to be published reports of severe childhood lead poisoning across the United States. However, the epidemiology of hospital encounters for severe lead poisoning is still unknown.

Methods: We included encounters for children with a primary diagnosis of lead poisoning using Pediatric Hospital Information System data from 2016 to 2023. We used descriptive statistics to characterize patient sociodemographic factors, clinical characteristics, and hospital utilization, as well as Poisson regression accounting for hospital state-level variation, to estimate lead poisoning rates by patient race and ethnicity, insurance payor, urbanicity, and Childhood Opportunity Index (COI) level.

Results: There were 845 inpatient hospitalizations and 1137 emergency department visits for lead poisoning across 47 hospitals, resulting in over $4.5 million mean cumulative annual billed charges. Compared with patients who identified as white, non-Hispanic, patients who identified as Asian, non-Hispanic (incidence rate ratio [IRR], 1.75); Black, non-Hispanic (IRR, 1.44); and multiracial, non-Hispanic (IRR, 2.26) had significantly higher rates of lead poisoning encounters. Those with public insurance had higher rates of encounters compared with those with commercial insurance (IRR, 2.14). Patients in rural zip codes had significantly higher rates of encounters than patients in urban zip codes (IRR, 1.51). Patients in low- (IRR, 1.79) and very low-COI (IRR, 3.22) zip codes had significantly higher rates of encounters compared with those in very high-COI zip codes.

Conclusions: This study finds that severe lead poisoning remains a substantial health concern. We found significant disparities in the rate of encounters by sociodemographic factors that warrant further action.

Background: Asthma exacerbations are a leading cause of pediatric hospitalization, and systemic corticosteroids are a mainstay of inpatient treatment. This study describes hospital-level variability and trends in systemic corticosteroid prescribing during acute asthma exacerbation hospitalizations and examines hospital-level associations between prescribed corticosteroid and hospitalization outcomes.

Methods: This retrospective cross-sectional study used the Pediatric Health Information System database to examine encounters of patients aged 2 to 18 years who were hospitalized with an acute asthma exacerbation between January 1, 2016, and December 31, 2023 and were administered dexamethasone, prednisone, prednisolone, or methylprednisolone. We analyzed trends and hospital-level variation in systemic corticosteroid prescribing. We used generalized estimating equations to analyze the association of annual hospital-level dexamethasone use with hospitalization outcomes-length of stay, ED revisit, and readmission rates, with models adjusted for relevant clinical and demographic factors.

Results: We identified 122 856 asthma hospitalizations across 38 children's hospitals. From 2016 to 2023, the proportion of hospital-level dexamethasone use increased from 42% to 77%. The proportion of hospitals prescribing dexamethasone for over 80% of hospital encounters rose from 18% in 2016 to 66% in 2023. There was no difference in hospitalization outcomes based on annual hospital-level dexamethasone use, including a subanalysis also based on annual hospital-level dexamethasone use focusing on exclusive dexamethasone or exclusive prednisone/prednisolone use (P > .05).

Conclusions: Dexamethasone use during asthma hospitalizations increased during the study period, without differences in hospitalization outcomes between hospitals that used a higher proportion of dexamethasone vs those that used less.

Objective: Pediatric emergency transfers (ETs), unplanned intensive care unit (ICU) transfers in which a child needs intubation, vasopressor initiation, or at least 60 mL/kg fluid resuscitation within 1 hour, are associated with longer stays and higher mortality, yet their financial burden is unknown. Thus, we compared post-transfer financial charges for ETs vs non-ETs.

Patients and methods: We conducted a retrospective cohort study of 2034 ICU transfers between 2015 and 2019 at a freestanding children's hospital. We compared charges between ETs and non-ETs, including aggregate post-transfer ICU charges (transfer through ICU discharge), aggregate total post-transfer hospital charges (transfer through 100 days post-transfer), and average daily post-transfer charges over the first 100 days. Charge comparisons were adjusted for age, presence of complex chronic conditions, pretransfer length of stay, originating service, and deterioration type using regression models with generalized estimating equations.

Results: Compared to non-ETs, ETs had higher unadjusted post-transfer charges (ICU: 108% [95% CI 51-188], P < .01; total: 91% [95% CI 50-143], P < .01; daily: 61% [95% CI 35-91], P < .01). After adjustment, ETs remained associated with higher post-transfer charges (ICU: 65% [95% CI 22-123], P < .01); total: 49% [95% CI 17-90], P < .01; daily: 20% [95% CI 3-98], P = .02). ET-associated post-transfer charge increases varied significantly by originating service (general pediatrics: 104% [95% CI 30-221] vs surgical services: -19% [95% CI -55 to 47], P < .01) and deterioration type (respiratory: 177% [52%-407%] vs circulatory: 2% [-28% to 47%], P < .01).

Conclusions: ETs are associated with significantly higher post-transfer charges for hospitalized children. This financial impact highlights the economic imperative, alongside clinical benefits, for investing in systems aimed at preventing delayed escalation and reducing ETs.

Objective: Fentanyl is an opioid analgesic commonly used for pain management in children who are critically ill. However, fentanyl use is discouraged during extracorporeal membrane oxygenation (ECMO) because of its lipophilic and protein-binding properties. These properties have been shown to increase the risk of opioid withdrawal and other poor health outcomes. Our objectives were to evaluate institutional variation and the factors associated with fentanyl use for children on ECMO.

Patients and methods: This retrospective study included children aged less than 18 years who received ECMO between 2016 to 2023 at children's hospitals in the Pediatric Health Information System. Multivariable hierarchical logistic regression evaluated factors associated with fentanyl prescribing during ECMO.

Results: Overall, 7731 children (54.5% male, 53.8% white, 48.6% neonatal) were included, with 91.9% receiving fentanyl during ECMO. Significant institutional variability was observed, with percent days of fentanyl use during ECMO ranging from 13.4% to 100%. Overall, 20.3% of the total variation in fentanyl use on ECMO was attributable to differences among hospitals. On multivariable regression, compared with adolescents, neonates (odds ratio [OR], 2.35; 95% CI, 1.61-3.43) had the highest likelihood of receiving fentanyl. Additionally, children with cardiovascular disease (OR, 1.63; 95% CI, 1.35-1.95) had an increased likelihood of receiving fentanyl, and children with a history of prematurity (OR, 0.50; 95% CI, 0.38-0.66) had a decreased likelihood.

Conclusion: Despite its known risks, fentanyl remains widely used for children on ECMO. Additionally, significant institutional variation exists, with neonates having the highest risk of fentanyl exposure. Our findings underscore the need for enhanced guidelines for sedation and pain management for children receiving ECMO.

Background: The American Academy of Pediatrics (AAP) published guidelines in 2021 recommending the use of inflammatory markers (IMs), including procalcitonin (PCT), to evaluate febrile infants and identify those at low risk for bacterial infection for whom clinicians may forego antibiotics. This quality improvement (QI) project aimed to safely decrease antibiotic administration in febrile infants 22 to 56 days old presenting to the emergency department from a baseline of 46% to 36% within 1 year, inclusive of race and ethnicity and preferred language.

Methods: A multidisciplinary team identified interventions including clinical pathway updates, clinical decision support (CDS), and education, which were implemented in 4 Plan-Do-Study-Act cycles. Statistical process control methodology was used to analyze the primary outcome measure (percentage of patients receiving antibiotics) and the process measure (percentage of infants with PCT result). The balancing measure was missed bacteremia or bacterial meningitis.

Results: Antibiotic use decreased from 46% to 33%, with the greatest decrease in infants 22 to 28 days old (from 86% to 43%). We decreased antibiotic use in non-Hispanic white infants but not non-Hispanic Black infants. PCT use increased from 4% to 97%, and there were no disparities in obtaining IMs. There were no missed cases of bacteremia or bacterial meningitis during the intervention.

Conclusions: Using a QI framework, we aligned clinical pathway updates, CDS, and education with AAP guidelines to safely reduce unnecessary antibiotic exposure in febrile infants as young as 22 days old. Future work should focus on ensuring equitable decreases in antibiotic exposure.

Objectives: Secure messaging is an increasingly common mode of communication among hospital-based clinicians. The use of these systems in pediatric settings is poorly understood. We sought to describe secure message volume, message response time, and emoji reaction use in a large children's hospital.

Methods: We extracted 6 months of secure message data from staff working on 3 inpatient units of an urban tertiary care children's hospital. Using assignment data, we isolated secure message activity during clinical shifts on these 3 units to report message volume, message response time, and emoji reaction use by unit and clinical role. We performed statistical analysis to compare message metrics within units and secondary analysis for key role comparisons.

Results: A total of 2493 clinicians sent and read 2 848 677 secure messages during the study period. Physicians and advanced practice providers (APPs) providing direct front-line care had the highest message use, a median of 13.4 (IQR 9.9, 17.3) combined sent and received messages per hour. We found significant differences in message volume and time to read, respond, and react to messages across clinical roles within units (all P<0.001). Front-line clinicians across units had significant differences in secure message volume and response times. (P<0.001).

Conclusions: Front-line physicians and APPs experienced the highest message burden per person across all clinical roles. We demonstrated nearly immediate message read and response times. Additional study is needed to understand the effects of secure messaging on clinician well-being and patient safety.

Objectives: Medical emergency teams (METs) are activated in response to signs and symptoms, or triggers, of clinical deterioration in acute care settings. However, the patterns in which triggers manifest and impact outcomes are poorly understood. We identified and described the patterns in which multiple triggers cluster to activate pediatric METs and examined the associations between these clusters and outcomes.

Methods: Pediatric MET events from January 2015 to December 2019 in the Get With The Guidelines®-Resuscitation national registry focused on METs (N = 4289) were grouped into MET trigger clusters using cluster analyses based on triggers used to activate the MET. Differences in patient characteristics across MET trigger clusters were compared using Pearson χ2 and analysis of variance (ANOVA) tests. Hierarchical logistic regressions tested associations between trigger clusters and outcomes.

Results: A total of 4 MET trigger clusters were identified. The triggers that predominantly defined each cluster were as follows: Cluster 1, decreased oxygen saturation and mental status changes; Cluster 2, tachypnea, tachycardia, and staff concern; Cluster 3, new onset difficulty in breathing and staff concern; and Cluster 4, the reference cluster, tachypnea, new onset difficulty in breathing, and decreased oxygen saturation. Patients in Cluster 1 were more likely to experience acute respiratory compromise (need for emergent assisted ventilation), and patients in Clusters 1 and 3 were more likely to be transferred to critical care.

Conclusions: A total of 4 MET trigger clusters were identified and have varying associations with outcomes. MET trigger clusters could guide bedside care and triage in clinical emergencies and help develop more accurate predictive models for detecting clinical deterioration.

Pediatricians caring for newborns during the birth hospitalization are tasked with identifying and managing newborns with prenatal substance exposure and neonatal opioid withdrawal syndrome (NOWS) and coordinating care plans for affected families. Neonatal toxicology testing (NTT) is a frequently used tool to identify and clinically manage NOWS. At times, however, the harms of testing may outweigh the benefits. Current variability in NTT practices continue to result in negative consequences and harm to affected newborns and their families, adding to perpetuation of bias, inequity in who is tested, and separation of newborns from families. This article highlights well-described ethical principles and frameworks that have the potential to inform clinical practice and policies related to exposed newborns and NTT, which can help to align medical care and legislative requirements, reduce variability and inequity found in current testing approaches, and optimize the health and well-being of affected newborns and families. Pediatricians are well positioned to provide education, influence policy, and challenge the current use of NTT results in legislative definitions of NOWS. Along with hospital leaders, pediatricians should critically examine current toxicology testing practices and hospital policies with the goal of developing supportive, rather than punitive, approaches to care. This article focuses on in utero opioid exposure and may be more generally applied to exposure to other substances.