Pediatric hospital medicine (PHM) continues to evolve as a specialty, fueled by growing recognition of the distinct needs of hospitalized children and an expanding focus on quality, safety, and systems-based care. This growth is mirrored in an increasingly robust body of literature, making regular review critical to advancing clinical practice. In keeping with tradition, we present a selection of the top PHM articles from the past year, originally presented at SHM Converge 2026 in Nashville, Tenn.
This year, we modeled our review after a previously published selection process and used a web-based collaboration software platform that streamlines the production of systematic and other literature reviews.1,2 After screening more than 17,000 titles and abstracts, 255 full-text articles were further narrowed to approximately 50, which were scored across four domains using a 40-point rubric. Each domain was awarded a maximum of 10 points based on quality and credibility, methodological rigor, novelty, and impact on PHM practice. The final selection sought to represent the breadth of PHM, including clinical management, quality improvement, health equity, and emerging technologies. Internationally based studies were considered for inclusion if the study question and findings were considered broadly applicable by the reviewers.
The following review highlights the top articles of 2025.
1. Clesrovimab for Prevention of RSV Disease in Healthy Infants and Nirsevimab Against Hospitalizations and Emergency Department Visits for Lower Respiratory Tract Infection in Infants: A Meta-Analysis
Two landmark studies defined 2025 as a breakthrough year for respiratory syncytial virus (RSV) prevention and are presented together here, given their complementary contributions to the evidence base.3,4 Clesrovimab is a monoclonal antibody with an extended half-life that binds to site IV of the RSV fusion protein, allowing single-dose, single-season protection. The phase 3 trial of clesrovimab demonstrated 60% efficacy against RSV-associated, medically attended, lower respiratory infection and 84% efficacy against RSV-associated hospitalization in healthy preterm and full-term infants. In a post-hoc analysis, clesrovimab showed 88% efficacy in preventing medically attended lower respiratory infection when using the same criteria used in the nirsevimab melody trial.5 The safety profile in the clesrovimab group was similar to that of the placebo group, with no increase in adverse events.
Where clesrovimab provides new phase 3 efficacy data, a meta-analysis of 15 real-world studies (comprising more than 236,000 infants) offers validation of nirsevimab’s real-world effectiveness, showing a 62% reduction in lower respiratory tract infection (LRTI) hospitalizations, a 48% reduction in LRTI emergency department (ED) visits, and 76% effectiveness against RSV-specific LRTI ED visits.
These findings support the inclusion of monoclonal antibodies in infant immunization programs and herald a new era in RSV prevention with broader, simpler, and more durable protection strategies.
2. Comparison of Procedural Sedation Outcomes in Children With and Without Autism Spectrum Disorder
This prospective observational study from the Pediatric Sedation Research Consortium analyzed 64,708 children undergoing procedural sedation, including 4,421 (6.8%) with autism spectrum disorder (ASD).6 Using 2:1 propensity matching, investigators found that children with ASD did not have higher rates of critical adverse events such as cardiac arrest or death (odds ratio [OR], 0.63; not significant). However, they did have significantly higher rates of airway complications: complete airway obstruction (OR, 1.89; P=0.002), partial obstruction (OR, 1.43; P <0.001), and hypoxia (OR, 1.31; P=0.002).
The clinical implication is clear: this study should change where we sedate children with ASD, not whether we sedate them. These patients should be sedated in settings equipped for advanced airway management, with personnel trained in pediatric airway rescue and appropriate backup plans in place.
3. Infant Outcomes, Risk Factors, and Diagnostic Yield After a Brief Resolved Unexplained Event: A Systematic Review and Meta-Analysis
Nearly a decade after the American Academy of Pediatrics (AAP) redefined apparent life-threatening events, or ALTEs, as brief resolved unexplained events, or BRUEs, clinical uncertainty persists. This systematic review and meta-analysis of 24 studies (6,603 infants) provides the most comprehensive data to date on outcomes following BRUE.7
Key findings include: 6% prevalence of serious underlying diagnosis (with seizure or epilepsy at 1.5% and serious apnea at 0.9% being the most common); 14% event recurrence within three months; and mortality of 1.8 per 10,000 infants. Several factors were independently associated with a serious underlying diagnosis, including the AAP higher-risk classification (OR, 2.11), multiple events (OR, 2.01), and abnormal medical history (OR, 1.78). Notably, higher-risk AAP classification did not predict mortality within three months (OR, 0.68; 95% confidence interval [CI], 0.07 to 6.25).
Diagnostic yields of labs and ancillary testing in identifying a serious underlying diagnosis were notably low: electroencephalography had the highest yield at 5.5%, with pertussis testing at 4.3%, while electrolytes had a yield of 0.4%, complete blood count at 1.65%, and metabolic panels at 0% yield in identifying a serious underlying diagnosis. These findings support a shift toward risk-informed diagnostic testing rather than routine screening and underscore the importance of caregiver counseling and close follow-up.
4. Evaluation of Influenza Vaccine Clinical Decision Support Systems Bundle for Hospitalized Children
With baseline inpatient influenza vaccination rates at only 14%, this quasi-experimental study evaluated three sequentially implemented Clinical Decision Support (CDS) interventions across a three-hospital pediatric system over four flu seasons (46,706 encounters).8
The opt-out order group (flu vaccine pre-checked in admission order sets) showed the strongest effect (adjusted odds ratio [aOR], 2.28), followed by the conditional discharge alert (aOR, 1.82), while the passive status banner showed the smallest effect (aOR, 1.19). Overall vaccination rates increased from a baseline of 14% to a peak of 24.7% following sequential CDS implementation. This aligns with behavioral economics principles: making the right thing the easy thing drives the greatest impact.
However, despite these interventions, vaccination rates waned over time, down to 11% by the final season from a peak of 24.7% (aOR, 0.34), coinciding with the COVID-19 pandemic. The lesson: CDS works, but sustained improvement requires ongoing human engagement and maintenance.
5. Association Between Initial Antibiotic Route and Outcomes for Children Hospitalized With Pneumonia
Antibiotic route is an important but understudied area in the treatment of community-acquired pneumonia (CAP). This multicenter retrospective cohort study of 1,147 children aged three months to 18 years hospitalized with CAP aimed to evaluate the association between initial antibiotic route (oral versus IV) and outcomes such as length of stay for children hospitalized with moderately severe CAP.9 After robust confounder adjustment, including factors related to illness severity or likelihood of bacterial etiology that may influence antibiotic route, children receiving initial oral antibiotics had 8% shorter length of stay and 14% lower hospital costs, with no significant differences in oxygen duration, care escalation, or seven-day readmission rates.
The study also revealed antibiotic route momentum: 77% of children who began oral treatment continued with it throughout, suggesting that the initial choice of treatment creates inertia. While further prospective research is needed, this work suggests that oral antibiotics may be a safe and cost-effective alternative to IV antibiotics for children hospitalized with moderately severe CAP.
6. Impact of the 2022 AAP Guidelines on Neonatal Hyperbilirubinemia Admissions: A PHIS Study
The 2022 AAP hyperbilirubinemia guidelines significantly raised phototherapy thresholds for healthy term infants.10 This PHIS study using interrupted time series analysis evaluated the real-world impact across freestanding U.S. children’s hospitals.
Results showed a 25% reduction in hospitalizations (5,051 to 3,778, P <0.001), with the decrease beginning just one month after guideline publication, demonstrating remarkably rapid adoption. Critically, safety outcomes were unchanged: no increases in intravenous immunoglobulin use, exchange transfusions, or kernicterus diagnoses.
The study noted an acuity shift: infants still being admitted had higher Hospitalization Resource Intensity Scores for Kids, or H-RISK scores, suggesting the guideline successfully filtered out unnecessary admissions while appropriately capturing sicker infants who genuinely need intervention.
7. Derivation and Validation of Predictive Models for Early Pediatric Sepsis
In recent years, artificial intelligence has been applied to various fields in healthcare, such as health systems management, predictive modeling, and clinical decision making. This study developed machine learning models using multicenter electronic health record data from five health systems affiliated with the Pediatric Emergency Care Applied Research Network (PECARN) to predict sepsis with organ dysfunction within 48 hours of ED arrival.11 Using inputs including the emergency severity index, vital signs, chronic conditions, and prior healthcare utilization, the model achieved an area under the received operating characteristics of 0.936 for sepsis and 0.926 for septic shock. These values indicate that the predictive model has excellent discrimination between patients who will develop sepsis or septic shock and those who will not.
Strengths include early prediction of a rare outcome, exclusion of children already presenting with sepsis, and demonstration of model equity across sites and demographic groups. Limitations include low positive predictive value (reflecting the challenge of predicting rare outcomes) and implementation complexity.
The takeaway: there is an opportunity to combine machine learning models with clinical judgment to improve sepsis prediction, but human partnership remains essential.
8. Prediction of Bacteremia and Bacterial Meningitis Among Febrile Infants Aged 28 Days or Younger
This pooled analysis of four prospective cohorts from six countries (1,537 well-appearing febrile infants 28 days and younger) validated updated PECARN low-risk criteria: negative urinalysis, procalcitonin 0.5 ng/mL or less, and absolute neutrophil count (ANC) 4,000/mm3 or less.12
The rule identified 41% of infants as low risk with 94.2% sensitivity and 99.4% negative predictive value (NPV). Most importantly, zero cases of bacterial meningitis were missed among low-risk infants: 0/11 in primary analysis and 0/22 in secondary analysis (NPV for meningitis was greater than 99.96%). The four missed IBI cases were all bacteremia without meningitis, and one was likely a contaminant.
This challenges decades of routine lumbar puncture for all febrile neonates practice and supports shared decision-making with families. Importantly, the rule does not address herpes simplex virus risk, which requires separate evaluation.
9. Optimizing Management of Febrile Young Infants Without Serum Procalcitonin
The 2021 AAP guidelines recommend procalcitonin (PCT) for febrile infant risk stratification, but PCT is not available everywhere. When PCT is unavailable, the 2021 AAP guidelines recommend an alternative low-risk criteria set: ANC, 5,200 or less; C-Reactive Protein (CRP), 20 mg/L or less; and temperature, 38.5 °C or less, which achieves 100% sensitivity but only 50.7% specificity, meaning more than half of low-risk babies are incorrectly classified as high-risk.
This retrospective study (1,987 febrile infants aged eight to 60 days) used classification and regression tree (CART) analysis to derive optimized thresholds: ANC, 4,500 or less; CRP, 22.2 mg/L or less; and temperature, 39.0 °C or less.13 The result: 100% sensitivity maintained while specificity improved to 83.8%, a 33-percentage-point improvement translating to 33% fewer unnecessary lumbar punctures and admissions.
These are promising derivation findings, but prospective multicenter validation is needed before clinical implementation. Until then, this work offers a practical, data-informed starting point for centers navigating PCT unavailability in real time.
10. A Quality Improvement Initiative to Improve the Discharge Process for Spanish-Speaking Patients
Children whose families speak languages other than English face increased risk of medication errors, missed follow-up, and higher healthcare utilization after discharge. This quality improvement initiative implemented a multi-component intervention including a discharge navigator, provider education, nurse and interpreter training, a discharge checklist, and refresher sessions.14
Results showed a centerline shift from 65.5% to 88% in Spanish-speaking patients receiving translated discharge documents. Most strikingly, physician-placed translation orders increased from 11% to 75%, nearly a sevenfold improvement.
The success drivers were interdisciplinary teamwork, stakeholder buy-in, and EHR modifications. Hospital leadership should pursue similar opportunities to narrow health inequities, with translated discharge documentation serving as a pivotal starting point.
Dr. Gupta
Dr. Tran
Dr. Gupta is a pediatric hospital medicine fellow at Vanderbilt University Medical Center in Nashville, Tenn. She will join the division of pediatric hospital medicine there as an assistant professor of pediatrics in July 2026. Dr. Tran is an assistant professor of clinical pediatrics at the University of Illinois College of Medicine and a pediatric hospitalist at the Children’s Hospital of Illinois, both in Peoria, Ill. She is also the pediatric editor for The Hospitalist and a deputy editor of digital media for the Journal of Hospital Medicine.
References
- McDaniel CE, Russell CJ. Top articles in pediatric hospital medicine: July 2019 to June 2020. Hosp Pediatr. 2020;10(10):906-912. doi:10.1542/hpeds.2020-001651.
- Covidence software. The world’s #1 systematic review tool. Covidence website. www.covidence.org. Accessed May 28, 2026.
- Zar HJ, et al. Clesrovimab for prevention of RSV disease in healthy infants. N Engl J Med. 2025;393(13):1292-1303. doi:10.1056/NEJMoa2502984.
- Sumsuzzman DMd, et al. Nirsevimab against hospitalizations and emergency department visits for lower respiratory tract infection in infants: a meta-analysis. JAMA Pediatr. 2026;180(2):152. doi:10.1001/jamapediatrics.2025.5280.
- Hammitt LL, et al. Nirsevimab for prevention of RSV in healthy late-preterm and term infants. N Engl J Med. 2022;386(9):837-846. doi:10.1056/NEJMoa2110275.
- Kannikeswaran N, et al. Comparison of procedural sedation outcomes in children with and without autism spectrum disorder. Hosp Pediatr. 2025;15(5):398-406. doi:10.1542/hpeds.2024-008153.
- Nama N, et al. Infant outcomes, risk factors, and diagnostic yield after a brief resolved unexplained event: a systematic review and meta-analysis. JAMA Pediatr. 2026;180(3):250. doi:10.1001/jamapediatrics.2025.5858.
- Kandaswamy S, et al. Evaluation of influenza vaccine clinical decision support systems bundle for hospitalized children. Hosp Pediatr. 2025;15(4):351-359. doi:10.1542/hpeds.2024-008099.
- Cotter JM, et al. Association between initial antibiotic route and outcomes for children hospitalized with pneumonia. J Hosp Med. 2025;20(3):238-247. doi:10.1002/jhm.13516.
- Jameel A, et al. Impact of the 2022 AAP guidelines on neonatal hyperbilirubinemia admissions: a PHIS study. Hosp Pediatr. 2025;15(7):537-544. doi:10.1542/hpeds.2024-008205.
- Alpern ER, et al. Derivation and validation of predictive models for early pediatric sepsis. JAMA Pediatr. 2025;179(12):1318-1325. doi:10.1001/jamapediatrics.2025.3892.
- Burstein B, et al. Prediction of bacteremia and bacterial meningitis among febrile infants aged 28 days or younger. JAMA. 2026;335(5):425. doi:10.1001/jama.2025.21454.
- Burstein B, et al. Optimizing management of febrile young infants without serum procalcitonin. Pediatrics. 2025;155(2):e2024068200. doi:10.1542/peds.2024-068200.
- Kupelian C, et al. A quality improvement initiative to improve the discharge process for Spanish-speaking patients. Hosp Pediatr. 2025;15(9):721-729. doi:10.1542/hpeds.2024-008227.