Diagnostic Stewardship of Respiratory Cultures Using Clinical Decision Support in the PICU

Anna C Sick-Samuels; Charlotte Z Woods-Hill; Daniel P Kelly; Danielle W Koontz; Nora Elhaissouni; Urmi Kumar; Anping Xie; Troy Richardson; Sreejata Dutta; Jill A Marsteller; Elizabeth Colantuoni; Aaron M Milstone; and the BrighT STAR Respiratory Authorship Group; Zachary Aldewereld; Michael J Auth; Ritu Banerjee; Jennifer A Blumenthal; Katharine Boyle; Cara Cecil; Samantha Dallefeld; Brian F Flaherty; Charles B Foster; Keshava M N Gowda; Kelly A Hardy; Sarmistha B Hauger; Andrea Green Hines; Sue J Hong; Nicholas D Hysmith; Andrew Kiragu; Aileen L Kirby; Christina Koutsari; Gitte Y Larsen; Caitlin Naureckas Li; John C Lin; Matthew H M Marx; Bridget M Norton; Gregory P Priebe; Glenn J Rapsinski; Rebecca G Same; Hayden T Schwenk; Katherine M Steffen; Sachin D Tadphale; Philip Toltzis; Lorne Walker

doi:10.1001/jamanetworkopen.2026.26547

. 2026 Jul 31;9(7):e2626547. doi: 10.1001/jamanetworkopen.2026.26547

Diagnostic Stewardship of Respiratory Cultures Using Clinical Decision Support in the PICU

Anna C Sick-Samuels ^1,^2,^✉, Charlotte Z Woods-Hill ^3,⁴, Daniel P Kelly ^5,⁶, Danielle W Koontz ¹, Nora Elhaissouni ¹, Urmi Kumar ¹, Anping Xie ^7,⁸, Troy Richardson ⁹, Sreejata Dutta ⁹, Jill A Marsteller ^8,¹⁰, Elizabeth Colantuoni ¹¹, Aaron M Milstone ^1,²; and the BrighT STAR Respiratory Authorship Group, Zachary Aldewereld ^12,¹³, Michael J Auth ¹⁴, Ritu Banerjee ¹⁵, Jennifer A Blumenthal ^16,¹⁷, Katharine Boyle ^18,¹⁹, Cara Cecil ^20,^21,²², Samantha Dallefeld ¹⁴, Brian F Flaherty ²³, Charles B Foster ²⁴, Keshava M N Gowda ^25,²⁶, Kelly A Hardy ²⁷, Sarmistha B Hauger ²⁸, Andrea Green Hines ²⁹, Sue J Hong ^20,^30,^31,³², Nicholas D Hysmith ³³, Andrew Kiragu ³⁴, Aileen L Kirby ³⁵, Christina Koutsari ^36,³⁷, Gitte Y Larsen ³⁸, Caitlin Naureckas Li ³⁹, John C Lin ⁴⁰, Matthew H M Marx ⁴¹, Bridget M Norton ^42,⁴³, Gregory P Priebe ^16,^17,⁴⁴, Glenn J Rapsinski ¹³, Rebecca G Same ^45,⁴⁶, Hayden T Schwenk ⁴⁷, Katherine M Steffen ⁴⁸, Sachin D Tadphale ^49,⁵⁰, Philip Toltzis ⁴¹, Lorne Walker ⁵¹

¹Department of Pediatrics, Johns Hopkins School of Medicine, Baltimore, Maryland

²Department of Hospital Epidemiology and Infection Control, Johns Hopkins Hospital, Baltimore, Maryland

³Division of Critical Care Medicine, Children’s Hospital of Philadelphia, University of Pennsylvania Perelman School of Medicine, Philadelphia

⁴The Leonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia

⁵Division of Medical Critical Care, Department of Pediatrics, Boston Children’s Hospital, Boston, Massachusetts

⁶Department of Pediatrics, Harvard Medical School, Boston, Massachusetts

⁷Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland

⁸Armstrong Institute of Patient Safety and Quality, Johns Hopkins University School of Medicine, Baltimore, Maryland

⁹Children’s Hospital Association, Lenexa, Kansas

¹⁰Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland

¹¹Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland

¹²Division of Pediatric Critical Care Medicine, Department of Critical Care Medicine, University of Pittsburgh School of Medicine, UPMC Children’s Hospital of Pittsburgh, Pittsburgh, Pennsylvania

¹³Division of Pediatric Infectious Diseases, Department of Pediatrics, University of Pittsburgh School of Medicine, UPMC Children’s Hospital of Pittsburgh, Pittsburgh, Pennsylvania

¹⁴Division of Pediatric Critical Care, Department of Pediatrics, Dell Medical School, Austin, Texas

¹⁵Division of Infectious Diseases, Department of Pediatrics, Vanderbilt University Medical Center, Nashville, Tennessee

¹⁶Division of Critical Care Medicine, Department of Anesthesiology, Critical Care and Pain Medicine, Boston Children’s Hospital, Boston, Massachusetts

¹⁷Division of Infectious Diseases, Department of Pediatrics, Boston Children’s Hospital, Boston, Massachusetts

¹⁸Division of Critical Care Medicine, Department of Pediatrics, Vanderbilt University Medical Center, Nashville, Tennessee

¹⁹Now with: Division of Critical Care Medicine, Department of Pediatrics, University of Maryland School of Medicine, Baltimore, Maryland

²⁰Division of Critical Care, Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, Illinois

²¹Department of Pediatrics, Northwestern University Feinberg School of Medicine, Chicago, Illinois

²²Now with: Department of Anesthesiology and Critical Care, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania

²³Division of Critical Care, Department of Pediatrics, University of Utah, Salt Lake City

²⁴Section of Pediatric Infectious Diseases, Cleveland Clinic Children’s Hospital, Cleveland, Ohio

²⁵Department of Pediatric Critical Care Medicine, Cleveland Clinic Children’s Hospital, Cleveland, Ohio

²⁶Now with: Department of Pediatrics, Henry Ford St. John Children’s Hospital, Detroit, Michigan

²⁷Division of Pediatric Critical Care, Monroe Carell, Jr. Children’s Hospital at Vanderbilt, Vanderbilt School of Nursing, Nashville, Tennessee

²⁸Division of Pediatric Infectious Disease, Dell Medical School, University of Texas at Austin

²⁹Division of Infectious Diseases, Department of Pediatrics, Children’s Nebraska, University of Nebraska Medical Center, Omaha

³⁰Division of Neurology, Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, Illinois

³¹Department of Pediatrics, Northwestern University Feinberg School of Medicine, Chicago, Illinois

³²Now with: Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland

³³Department of Pediatrics, University of Tennessee Health Science Center, Le Bonheur Children’s Hospital, Memphis, Tennessee

³⁴Division of Critical Care Medicine, Department of Pediatrics, Children’s Minnesota, Children’s Respiratory and Critical Care Specialist’s P.A., Minneapolis

³⁵Division of Pediatric Critical Care, Doernbecher Children’s Hospital, Oregon Health & Science University, Portland

³⁶Department of Pharmacy, Antimicrobial Stewardship Program, Children’s Minnesota, Minneapolis

³⁷Now with: Allina Health Group Pharmacy Services, Allina Health, Minneapolis, Minnesota

³⁸Division of Critical Care Medicine, Department of Pediatrics, Intermountain Primary Children’s Hospital and University of Utah, Salt Lake City

³⁹Division of Infectious Diseases, Lurie Children’s Hospital of Chicago, Northwestern University Feinberg School of Medicine, Chicago, Illinois

⁴⁰Division of Pediatric Critical Care Medicine, Washington University School of Medicine, St. Louis Children’s Hospital, St. Louis, Missouri

⁴¹Division of Pediatric Critical Care, Rainbow Babies and Children’s Hospital, Cleveland, Ohio

⁴²Division of Pediatric Critical Care Medicine, Children’s Nebraska, University of Nebraska College of Medicine, Omaha

⁴³Now with: current affiliation Division of Cardiac Critical Care, Department of Cardiology, Arkansas Children’s Hospital, University of Arkansas for Medical Sciences, Little Rock

⁴⁴Department of Anesthesia, Harvard Medical School, Boston, Massachusetts

⁴⁵Division of Infectious Diseases, Department of Pediatrics, Washington University School of Medicine, St. Louis Children’s Hospital, St. Louis, Missouri

⁴⁶Now with: Division of Infectious Diseases, Department of Pediatrics, Children’s Hospital of Philadelphia, Perelman School of Medicine, University of Pennsylvania, Philadelphia

⁴⁷Division of Infectious Diseases, Department of Pediatrics, Lucile Packard Children’s Hospital, Stanford School of Medicine, Stanford, California

⁴⁸Division of Pediatric Critical Care Medicine, Department of Pediatrics, Stanford University, Stanford, California

⁴⁹Division of Pediatric Critical Care & Cardiology, Department of Pediatrics, Le Bonheur Children’s Hospital, University of Tennessee Health Science Center, Memphis

⁵⁰Now with: Division of Pediatric Intensive Care, Cook Children’s Medical Center, Fort Worth, Texas

⁵¹Division of Infectious Diseases, Department of Pediatrics, Doernbecher Children’s Hospital, Oregon Health and Sciences University, Portland

Accepted for Publication: June 9, 2026.

Published: July 31, 2026. doi:10.1001/jamanetworkopen.2026.26547

Open Access: This is an open access article distributed under the terms of the CC-BY-NC-ND License, which does not permit alteration or commercial use, including those for text and data mining, AI training, and similar technologies. © 2026 Sick-Samuels AC et al. JAMA Network Open.

^✉

Corresponding Author: Anna C. Sick-Samuels, MD, MPH, Department of Pediatrics, Johns Hopkins University School of Medicine, 200 N Wolfe St, Ste 3093, Baltimore, MD 21287 (asick1@jhmi.edu).

Author Contributions: Dr Sick-Samuels and Mrs Elhaissouni had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Sick-Samuels, Woods-Hill, Kelly, Koontz, Kumar, Xie, Richardson, Colantuoni, Milstone, Aldewereld, Auth, Blumenthal, Boyle, Cecil, Larsen, Walker.

Acquisition, analysis, or interpretation of data: Sick-Samuels, Woods-Hill, Kelly, Koontz, Elhaissouni, Xie, Richardson, Dutta, Marsteller, Colantuoni, Milstone, Aldewereld, Banerjee, Blumenthal, Boyle, Cecil, Dallefeld, Flaherty, Foster, Gowda, Hardy, Hauger, Green Hines, Hong, Hysmith, Kiragu, Kirby, Koutsari, Larsen, Naureckas Li, Lin, Marx, Norton, Priebe, Rapsinski, Same, Schwenk, Steffen, Tadphale, Toltzis.

Drafting of the manuscript: Sick-Samuels, Kelly, Colantuoni, Milstone, Auth, Larsen.

Critical review of the manuscript for important intellectual content: Sick-Samuels, Woods-Hill, Kelly, Koontz, Elhaissouni, Kumar, Xie, Richardson, Dutta, Marsteller, Colantuoni, Milstone, Aldewereld, Banerjee, Blumenthal, Boyle, Cecil, Dallefeld, Flaherty, Foster, Gowda, Hardy, Hauger, Green Hines, Hong, Hysmith, Kiragu, Kirby, Koutsari, Larsen, Naureckas Li, Lin, Marx, Norton, Priebe, Rapsinski, Same, Schwenk, Steffen, Tadphale, Toltzis, Walker.

Statistical analysis: Sick-Samuels, Kelly, Elhaissouni, Richardson, Dutta, Colantuoni, Aldewereld, Cecil, Lin, Walker.

Obtained funding: Sick-Samuels, Woods-Hill, Milstone.

Administrative, technical, or material support: Sick-Samuels, Kelly, Koontz, Kumar, Aldewereld, Blumenthal, Boyle, Cecil, Flaherty, Foster, Gowda, Hardy, Hauger, Green Hines, Hong, Hysmith, Naureckas Li, Lin, Marx, Norton, Rapsinski, Same, Steffen, Tadphale, Walker.

Supervision: Sick-Samuels, Kelly, Marsteller, Milstone, Blumenthal, Boyle, Cecil.

Conflict of Interest Disclosures: Dr Aldewereld reported receiving grants from the National Institute of Allergy and Infectious Diseases (grant No. 1K23AI190128) outside the submitted work. Dr Hauger reported receiving grants from the National Institute of Allergy and Infectious Diseases and National Institutes of Health (NIH) outside the submitted work. Dr Priebe reported receiving grants from the NIH; having a patent for techniques for measuring frequencies of antimicrobial resistance mutations in clinical samples (No. 62/309,368), pending to Boston Children’s Hospital; and having a patent for vaccines for Pseudomonas aeruginosa (No. 63/521,635) pending to Boston Children’s Hospital outside the submitted work. No other disclosures were reported.

Funding/Support: This work was funded by the Donoghue Foundation Greater Value Portfolio, the National Institutes of Health (NIH; grant No. K23HL161449 to Dr. Sick-Samuels, grant No. K24AI141580 to Dr. Milstone, and grant No. K23HL151381 to Dr. Woods-Hill), and the Agency for Healthcare Research and Quality (AHRQ; grant No. R01HS028634 to Dr. Milstone).

Role of the Funder/Sponsor: The sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

BrighT STAR Respiratory Authorship Group: Zachary Aldewereld, MD; Michael J. Auth, DO; Ritu Banerjee, MD, PhD; Jennifer A. Blumenthal, MD; Katharine Boyle, MD; Cara Cecil, MD, MS; Samantha Dallefeld, MD; Brian F. Flaherty, MD; Charles B. Foster, MD; Keshava M. N. Gowda, MBBS; Kelly A. Hardy, PhD, CPNP-PC/AC; Sarmistha B. Hauger, MD; Andrea Green Hines, MD; Sue J. Hong, MD; Nicholas D. Hysmith, MD; Andrew Kiragu, MD; Aileen L. Kirby, MD; Christina Koutsari, PharmD, PhD, BCPS, BCIDP; Gitte Y. Larsen, MD, MPH; Caitlin Naureckas Li, MD, MHQS; John C. Lin, MD; Matthew H. M. Marx, MD, MS; Bridget M. Norton, MD, MBA; Gregory P. Priebe, MD; Glenn J. Rapsinski, MD, PhD; Rebecca G. Same, MD; Hayden T. Schwenk, MD, MPH; Katherine M. Steffen, MD, MHS; Sachin D. Tadphale, MBBS, MPH; Philip Toltzis, MD; Lorne Walker, MD, PhD.

Affiliations of BrighT STAR Respiratory Authorship Group: Division of Pediatric Critical Care Medicine, Department of Critical Care Medicine, University of Pittsburgh School of Medicine, UPMC Children’s Hospital of Pittsburgh, Pittsburgh, Pennsylvania (Aldewereld); Division of Pediatric Infectious Diseases, Department of Pediatrics, University of Pittsburgh School of Medicine, UPMC Children’s Hospital of Pittsburgh, Pittsburgh, Pennsylvania (Aldewereld, Rapsinski); Division of Pediatric Critical Care, Department of Pediatrics, Dell Medical School, Austin, Texas (Auth, Dallefeld); Division of Infectious Diseases, Department of Pediatrics, Vanderbilt University Medical Center, Nashville, Tennessee (Banerjee); Division of Critical Care Medicine, Department of Anesthesiology, Critical Care and Pain Medicine, Boston Children’s Hospital, Boston, Massachusetts (Blumenthal, Priebe); Division of Infectious Diseases, Department of Pediatrics, Boston Children’s Hospital, Boston, Massachusetts (Blumenthal, Priebe); Division of Critical Care Medicine, Department of Pediatrics, Vanderbilt University Medical Center, Nashville, Tennessee (Boyle); Now with: Division of Critical Care Medicine, Department of Pediatrics, University of Maryland School of Medicine, Baltimore, Maryland (Boyle); Division of Critical Care, Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, Illinois (Cecil, Hong); Department of Pediatrics, Northwestern University Feinberg School of Medicine, Chicago, Illinois (Cecil); Now with: Department of Anesthesiology and Critical Care, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania (Cecil); Division of Critical Care, Department of Pediatrics, University of Utah, Salt Lake City (Flaherty); Section of Pediatric Infectious Diseases, Cleveland Clinic Children’s Hospital, Cleveland, Ohio (Foster); Department of Pediatric Critical Care Medicine, Cleveland Clinic Children’s Hospital, Cleveland, Ohio (Gowda); Now with: Department of Pediatrics, Henry Ford St. John Children’s Hospital, Detroit, Michigan (Gowda); Division of Pediatric Critical Care, Monroe Carell, Jr. Children’s Hospital at Vanderbilt, Vanderbilt School of Nursing, Nashville, Tennessee (Hardy); Division of Pediatric Infectious Disease, Dell Medical School, University of Texas at Austin (Hauger); Division of Infectious Diseases, Department of Pediatrics, Children’s Nebraska, University of Nebraska Medical Center, Omaha (Green Hines); Division of Neurology, Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, Illinois (Hong); Department of Pediatrics, Northwestern University Feinberg School of Medicine, Chicago, Illinois (Hong); Now with: Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland (Hong); Department of Pediatrics, University of Tennessee Health Science Center, Le Bonheur Children’s Hospital, Memphis, Tennessee (Hysmith); Division of Critical Care Medicine, Department of Pediatrics, Children’s Minnesota, Children’s Respiratory and Critical Care Specialist’s P.A., Minneapolis (Kiragu); Division of Pediatric Critical Care, Doernbecher Children’s Hospital, Oregon Health & Science University, Portland (Kirby); Department of Pharmacy, Antimicrobial Stewardship Program, Children’s Minnesota, Minneapolis (Koutsari); Now with: Allina Health Group Pharmacy Services, Allina Health, Minneapolis, Minnesota (Koutsari); Division of Critical Care Medicine, Department of Pediatrics, Intermountain Primary Children’s Hospital and University of Utah, Salt Lake City (Larsen); Division of Infectious Diseases, Lurie Children’s Hospital of Chicago, Northwestern University Feinberg School of Medicine, Chicago, Illinois (Naureckas Li); Division of Pediatric Critical Care Medicine, Washington University School of Medicine, St. Louis Children’s Hospital, St. Louis, Missouri (Lin); Division of Pediatric Critical Care, Rainbow Babies and Children’s Hospital, Cleveland, Ohio (Marx, Toltzis); Division of Pediatric Critical Care Medicine, Children’s Nebraska, University of Nebraska College of Medicine, Omaha (Norton); Now with: current affiliation Division of Cardiac Critical Care, Department of Cardiology, Arkansas Children’s Hospital, University of Arkansas for Medical Sciences, Little Rock (Norton); Department of Anesthesia, Harvard Medical School, Boston, Massachusetts (Priebe); Division of Infectious Diseases, Department of Pediatrics, Washington University School of Medicine, St. Louis Children’s Hospital, St. Louis, Missouri (Same); Now with: Division of Infectious Diseases, Department of Pediatrics, Children’s Hospital of Philadelphia, Perelman School of Medicine, University of Pennsylvania, Philadelphia (Same); Division of Infectious Diseases, Department of Pediatrics, Lucile Packard Children’s Hospital, Stanford School of Medicine, Stanford, California (Schwenk); Division of Pediatric Critical Care Medicine, Department of Pediatrics, Stanford University, Stanford, California (Steffen); Division of Pediatric Critical Care & Cardiology, Department of Pediatrics, Le Bonheur Children’s Hospital, University of Tennessee Health Science Center, Memphis (Tadphale); Now with: Division of Pediatric Intensive Care, Cook Children’s Medical Center, Fort Worth, Texas (Tadphale); Division of Infectious Diseases, Department of Pediatrics, Doernbecher Children’s Hospital, Oregon Health and Sciences University, Portland (Walker).

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the above funding agencies.

Data Sharing Statement: See Supplement 2.

Additional Contributions: We would like to acknowledge many additional contributors to this work who have provided coordination support, facilitated data collection or supported local quality improvement efforts: Slade Decker, BS (Johns Hopkins University); Alexandra Minor, BS (Johns Hopkins University); Ann Bailey, MBA, BSN, (Dell Children’s Medical Center); Susan Beedle, MSN, RN (Children’s Nebraska); Morgan Benedict, BSN, RN (Rainbow Babies & Children’s Hospital); Amanda Hinkel, DNP (Children’s Nebraska); Matthew Hudkins, MD, MS (Oregon Health & Science University); Larry Kociolek, MD, MSCI (Ann & Robert H. Lurie Children’s Hospital of Chicago); Molly Lewen, MD (Children’s Hospital of Pittsburgh); Erik Mikkelsen, MD, MEd (Children's Hospitals and Clinics of Minnesota); Dimple Mirchandani, MPH (Boston Children’s Hospital); Nicole Newman, MD (Children’s Hospitals and Clinics of Minnesota); Alanalyn Pinaula-Toves, BA (Children’s Hospitals and Clinics of Minnesota); Nicole Saliba, BS (Cleveland Clinic Children’s); John Sarmiento, MSN, RN, CCRN (Ann & Robert H. Lurie Children’s Hospital of Chicago); Lucy Sommerdorf, BSN, RN, CIC (Dell Children’s Medical Center); Luke Starnes, PhD, RN (Siteman Cancer Center at Washington University in St. Louis); Andrea Talukdar, MD (Children’s Nebraska); Gale Thomas, MMHC (Vanderbilt University Medical Center); Kathleen Zani, MD (Arkansas Children’s Hospital); Holli Stanford, BSN (Dell Children’s Medical Center). These individuals did not receive compensation for their contribution and have provided permission to be named in the acknowledgements.

^✉

Corresponding author.

PMCID: PMC13428280 PMID: 42536370

This cohort study evaluates the association of diagnostic stewardship of endotracheal aspirate cultures using clinical decision support with culture rates, antibiotic use, and patient outcomes across a multicenter collaborative of pediatric intensive care units (PICUs).

Key Points

Question

What is the association of diagnostic stewardship of endotracheal aspirate cultures using clinical decision support with culture rates, antibiotic use, and patient outcomes in the pediatric intensive care unit (PICU)?

Findings

In this cohort study, a collaborative of 15 PICUs observed a statistically significant reduction in endotracheal aspirate culture rates. There were no significant changes in antibiotic utilization or balancing measures such as bronchoalveolar lavage culture rates, length of stay, readmissions, and ventilation duration, nor was patient harm detected.

Meaning

These findings suggest that diagnostic stewardship using clinical decision support to standardize testing indications and optimize endotracheal aspirate culture practices in patients without suspected ventilator-associated infection may reduce culture use in critically ill children without safety concerns.

Abstract

Importance

Endotracheal aspirate culture (EAC) practices for evaluation of ventilator-associated infections (VAI) vary widely across pediatric hospitals, and overuse can contribute to overdiagnosis and overtreatment for VAI. Diagnostic stewardship strategies to optimize EAC testing practices may reduce overtesting and unnecessary antibiotic treatment.

Objective

To evaluate the association of diagnostic stewardship of EACs using clinical decision support with culture rates, antibiotic use, and patient outcomes across a multicenter collaborative of pediatric intensive care units (PICUs).

Design, Setting, and Participants

This was a multicenter cohort study with a pre-post study design among the BrighT STAR (Testing Stewardship for Antibiotic Reduction) Quality Improvement (QI) Collaborative involving PICUs across the US between 2019 and 2023. Data were collected from the participating sites and from the Children’s Hospital Association Pediatric Health Information System and were analyzed from August to December 2025.

Exposure

Participating PICUs conducted local QI programs focused on optimizing EAC practices, facilitated by the BrighT STAR collaborative.

Main Outcomes and Measures

The primary outcome was the monthly rate of EACs per 100 ventilator-days. Secondary outcomes included rates of antibiotic initiations and antibiotic days of therapy, bronchoalveolar lavage cultures, readmissions, length of stay, ventilation duration, ventilation-free days, sepsis, and septic shock. Analysis included adjustment for seasonality.

Results

Across 15 sites (median [IQR] unit size, 30 [25-38] beds), the study captured 106 967 ventilator-days preimplementation and 92 167 ventilator-days postimplementation. Comparing 24 months in the preimplementation period with the 18 months in the postimplementation period, the mean monthly EAC rate declined by 16% from a preimplementation to postimplementation rate of 7.80 to 6.55 cultures per 100 ventilator-days (relative rate [RR], 0.84; 95% CI, 0.78-0.90). The rate of antibiotic initiations remained stable (RR, 0.98; 95% CI, 0.89-1.08), as well as the antibiotic days of therapy rate (RR, 1.03; 95% CI, 0.95-1.11). There were no significant changes in the rates of bronchoalveolar lavage cultures, PICU length of stay, PICU or hospital readmissions, sepsis, septic shock, ventilation duration, or ventilator-free days.

Conclusions and Relevance

In this multicenter cohort study, diagnostic stewardship of EACs using clinical decision support led by multidisciplinary teams was associated with reduced EAC use in the PICU without safety concerns. Future work will determine optimal implementation strategies, assess sustainability and the cost impact of EAC stewardship.

Introduction

Endotracheal aspirate cultures (EACs) from patients with artificial airways are commonly used diagnostic tests in the evaluation of ventilator-associated infections (VAI) such as pneumonia or tracheobronchitis.^1,2 These cultures have limited specificity for infection because the respiratory tract is not sterile and detection of bacteria does not distinguish colonization from infection.^3,4 Due to the lack of definitive diagnostic parameters for VAI,⁵ associated antibiotic treatment is a top contributor of antibiotic overuse in the pediatric intensive care unit (PICU).⁶ Bacterial growth in EACs drives antibiotic overtreatment.^{7,8,9,10,11,12} Downstream impacts of excess testing and treatment include extended lengths of stay, delayed recognition of symptom causes, increased cost, antibiotic adverse effects, and selection for antibiotic-resistant bacteria.^13,14 Further, in national surveys, EAC practices for testing, specimen collection, and laboratory processes vary substantially,^2,15 and higher testing rates were associated with clinicians having higher likelihood to obtain EACs in response to nonrespiratory clinical changes.² Diagnostic stewardship can standardize patient treatment, and reduce overtesting and associated antibiotic treatment, leading to improved clinical outcomes and health care value.¹³

Diagnostic stewardship quality improvement (QI) programs employing clinical decision support (CDS) tools have reduced EACs and associated antibiotic use without identified patient harms in single-center adult and pediatric ICU studies.^16,17,18,19 However, it is not known if applying CDS to optimize testing indications¹³ would be effective across diverse settings. Previously, the BrighT STAR (Testing Stewardship for Antibiotic Reduction) Collaborative²⁰ evaluated whether diagnostic stewardship programs effectively reduced blood culture overuse. In that study,²¹ sites implemented guidance for the decision to order and methods to collect blood cultures, leading to a 33% reduction in blood cultures and a 12% reduction in broad-spectrum antibiotic use without safety concerns. Following a similar model, the BrighT STAR Collaborative employed diagnostic stewardship focusing on indications for testing and specimen collection to standardize practices and reduce overuse of EACs from PICU patients with artificial airways. Our objectives were to measure the association of implementing diagnostic stewardship using CDS with EAC rates, antibiotic use, and patient outcomes across geographically diverse PICUs.

Methods

This cohort study was approved by the coordinating center’s (Johns Hopkins) institutional review board with a waiver of informed consent due to use of aggregate deidentified data in accordance with the Common Rule. Sites obtained local institutional review board approval as per their institutional regulations.

Study Design and Setting: The BrighT STAR Respiratory Collaborative

We conducted a multicenter cohort study with a pre-post study design in 15 US PICUs (within 14 hospital systems) who participated in the BrighT STAR Respiratory (BSR) Quality Improvement Collaborative from 2019 to 2023. Participating PICUs executed local QI programs to improve EAC practices, facilitated by the collaborative coordinating center. Sites included 10 PICUs previously engaged with BrighT STAR blood culture stewardship and 5 PICUs new to the collaborative. All participating PICUs were part of academic pediatric hospitals and were geographically dispersed with mixed patient populations. The BSR Steering Committee, with expertise in pediatric critical care medicine, pediatric infectious diseases, implementation science, human factors engineering, patient safety and QI science, and biostatistics, facilitated the project. Figure 1 diagrams the study timeline. Individual site implementation dates were staggered based on site readiness (eTable 1 in Supplement 1). Data are reported according to Standards for Quality Improvement Reporting Excellence (SQUIRE) reporting guideline.²²

Flowchart of pre, during, and post implementation tasks for sites and a coordinating center. Three vertical columns span the page, titled at the top Before implementation, Implementation, and After implementation, each in a shaded header band. A dashed vertical line separates the first and second columns. Along the left margin, two stacked row labels read Sites in the upper section and Coordinating center in the lower section. In the Sites row under Before implementation, a series of right-pointing arrow bullets precede text items: Project champions identified with parenthetical engage staff slash leadership; Established core support team with parenthetical engage staff slash leadership; Obtained local leadership support and local I R B s if necessary with parenthetical engage staff slash leadership; Engage key stakeholders with parenthetical engage stakeholders; Collected and reviewed baseline data with parenthetical organizational readiness; Staff completed practices survey with parenthetical organizational readiness; Q I teams completed O R C A survey with parenthetical organizational readiness; Completed Q I charter and created implementation plan with parenthetical execute slash implement; Developed C D S tool with parenthetical execute slash implement; Established data collection plan with parenthetical monitoring slash evaluation. In the Sites row under Implementation, right-pointing arrow bullets list: Educated staff about evidence supporting intervention with parenthetical educate; Implemented C D S tool with parenthetical execute slash implement; Standardized work process or workflow with parenthetical execute slash implement; Iterative improvements slash adjustments to clinical tool or workflow with parenthetical execute slash implement. In the Sites row under After implementation, right-pointing arrow bullets list: Incorporated feedback from stakeholders with parenthetical engage stakeholders; Adapted C D S tools with parenthetical execute slash implement; Created sustainability plan with parenthetical execute slash implement; Audited monthly culture rate and sent to coordinating center with parenthetical monitoring slash evaluation; Conducted monthly safety surveys with parenthetical monitoring slash evaluation; Evaluated adherence with parenthetical monitoring slash evaluation. In the Coordinating center row under Before implementation, right-pointing arrow bullets list Confirmed enrollment, Conducted orientation, and Reviewed O R C A survey results. In the Coordinating center row under Implementation, a right-pointing arrow bullet introduces Met with sites 3 times followed by three indented lines labeled 1 Orientation, 2 Practices survey results, and 3 Reviewed implementation plan. In the Coordinating center row under After implementation, right-pointing arrow bullets list Met with sites quarterly, Conducted focus groups discussing implementation, and Met with sites to review sustainability plans. — The implementation stage is listed in parentheses. CDS indicates clinical decision support; IRB, institutional review board; ORCA, organizational readiness to change assessment; QI, quality improvement.

Program Implementation: Practice Changes Driven by Local Multidisciplinary QI Teams

The overall goals for the QI programs were to standardize and reduce EAC testing among patients who were unlikely to have a VAI (eg, new fever without respiratory changes). Sites followed a general implementation framework adapted from the Translating Evidence Into Practice Model²³ and the Clinical Sustainability Assessment Tool²⁴ with the following stages:

Engaging staff and leadership: QI teams were led by pediatric critical care medicine and pediatric infectious diseases physician champions and included other roles at their discretion (eg, data analyst, quality and safety specialist, antibiotic stewardship personnel, registered nurse, respiratory therapist, advanced practice clinician, or trainees). Sites sought support from local leadership to conduct QI projects and completed a QI charter including local QI goals and an implementation worksheet (eFigure 1 in Supplement 1).
Engaging stakeholders: Teams considered engagement from relevant stakeholders (eg, registered nurse, respiratory therapist, microbiology, or pulmonology).
Organizational readiness: Teams obtained EAC rate data, completed an organizational readiness to change assessment,²⁵ and conducted a survey of PICU staff to assess local EAC practices and identify improvement goals.²
Execute and implement: Teams developed and implemented CDS guidance to standardize EAC indications adapted from existing literature and tools,^16,17 and all BSR sites later participated in developing a consensus approach.²⁶ Sites formatted and adapted their CDS based on identified local needs² and stakeholder feedback. All sites used a checklist or algorithm on paper and/or local intranet; none were embedded into the electronic medical record (EMR) within the first 18 months. An example CDS tool is in eFigure 2 in Supplement 1.
Education: Teams educated PICU staff via staff meetings, email notification, and/or in person.
Monitoring and evaluation: Sites monitored EAC rates in U-charts, assessed adherence to CDS, sought feedback regarding safety concerns, and shared data with their unit staff.

Outcome and Process Measures

The primary outcome was the monthly rate of EACs (including tracheostomy or endotracheal tube), expressed as the number of cultures per 100 ventilator-days, and alternatively, per 1000 ICU patient-days. The primary outcome was reported by all 15 sites for the 24-month preimplementation and 18-month postimplementation periods. For all outcomes, site-level monthly summaries were expressed as rates, proportions or medians, as appropriate.

Secondary clinical outcomes included antibiotic use, length of ICU stay, 7-day ICU readmission within hospitalization, 7-day hospital readmission, sepsis or septic shock during admission, duration of mechanical ventilation among alive patients, and ICU ventilator-free days measured among patients who were mechanically ventilated during their ICU admission. (eTable 2 in Supplement 1). Antibiotic use was measured as total antibiotic days of therapy (DOT) and new antibiotic initiations (≥2 calendar days without an antibiotic from that grouping) administered on ICU day 3 onwards. We excluded the first 2 days because many patients receive empiric sepsis treatment, a practice that would not be impacted by this intervention. Antibiotics were grouped into 3 categories: those commonly used for VAI or pneumonia (primary antibiotic outcome), broad-spectrum antibiotics used for sepsis or bacteremia, and all commonly used antibiotics (eTable 3 in Supplement 1). We obtained data for the secondary outcomes from the Children’s Hospital Association Pediatric Health Information System (PHIS) database (queried on December 10, 2024) for the 12 sites that contributed data to PHIS. PHIS is an administrative database including clinical and resource utilization data from nearly 50 US children’s hospitals.²⁷

All 15 sites also tracked balancing, process, and safety measures. Balancing measures included bronchoalveolar lavage culture rates per 100 ventilator-days (because they could increase), and 3 sites routinely captured ventilator-associated events (VAEs), according to National Healthcare Safety Network criteria.²⁸ Sites assessed adherence to CDS guidance at postimplementation months 1, 6, and 12 by manual record review. Sites tracked EACs repeated within 3 days as an indirect adherence measure or process measure because most sites included guidance not to repeat cultures within 3 days. To characterize EACs, sites assessed whether cultures were obtained from an endotracheal tube or tracheostomy, and if cultures were obtained at time of intubation (if endotracheal tube) or admission (if tracheostomy) for postimplementation months 1 and 12. To monitor safety throughout the postimplementation period, QI teams encouraged colleagues to report any concerns in real time and actively solicited patient safety concerns regarding the new CDS guidance from PICU attendings recently on service by sending an email or asking during in-person meetings. Responses were categorized as (1) no concerns, (2) concern for a specific patient safety harm event, or (3) general feedback and aggregated as monthly summaries (eFigure 3 in Supplement 1).

Statistical Analysis

Following an a priori determined statistical plan, the analysis was completed in August to December 2025. Analysis of the primary outcome, monthly EAC rate, included estimating the relative and absolute difference between postimplementation and preimplementation using a Poisson regression model with an indicator for postimplementation (vs preimplementation), ventilator-days as an offset, and adjustment for seasonality (indicators for fall, winter, and spring vs summer). To account for clustering at the site level, the model included a random intercept for site, and we obtained SEs using a robust variance estimate. Planned secondary analyses of the primary outcome included (1) using an alternative denominator of ICU patient-days that included nonventilated days and was more readily available for long-term monitoring; (2) extending the Poisson model to include a 3-month wash-in period postimplementation; (3) fitting the Poisson model excluding data prior to June 2020 to adjust for effects of the COVID-19 pandemic; (4) fitting the Poisson model with adjustment for patient population complexity by including a hospital resource intensity score for kids as a pediatric case-mix index,²⁹ proportion of admissions with complex chronic conditions,³⁰ and proportion of admissions with a COVID-19 diagnostic code; (5) fitting an interrupted time-series analysis using a Poisson regression model. Additional post hoc sensitivity analysis considered implementation factors including (1) excluding the 3 months lead-in prior to implementation, recognizing planning and stakeholder engagement could begin to influence EAC testing; (2) stratifying sites that had previously participated with the BrighT STAR blood culture stewardship collaborative; and (3) stratifying sites with highest preimplementation EAC testing rates in the top quartile.

For analysis of the secondary outcomes, based on our prior work demonstrating PHIS data may lack precision to identify unit-level cohorts,³¹ we developed a curated cohort. Nine sites provided Children’s Hospital Association with a list of patients and dates of admission to and discharge from the participating PICU during the study period. These patients formed the curated PHIS cohort, reflecting confirmed PICU patients. We also conducted a sensitivity analysis using the full PHIS cohort for all 12 PHIS-participating sites. Secondary outcomes were analyzed akin to the primary outcome. Statistical significance was considered if the 95% CI of relative rates (RRs) did not cross 1.0 or the 2-sided P was less than .05. Data were analyzed using R version 4.5.2 (R Foundation for Statistical Computing) and STATA version18.0 (StataCorp LLC).

Results

Cohort Characteristics

Among the 15 sites, the median (IQR) unit size was 30 (25-38) beds with roughly 20% of admissions receiving mechanical ventilation (median [IQR] number of admissions with ventilation, 328 [292-743]); all sites had antibiotic stewardship support, 6 sites (40%) included cardiac surgery patients, and 10 sites (67%) included hematopoietic stem cell transplant patients (Table 1). The total number of ventilator-days was 106 967 preimplementation and 92 167 postimplementation. The median (IQR) crude preimplementation monthly EAC rate was 8.4 (5.7-12.9) per 100 ventilator-days (eTable 4 in Supplement 1).

Table 1. Site Baseline Characteristics.

Characteristic	Sites, No. (%) (N = 15)
Size of unit
No. of ICU beds, median (IQR)	30 (25-38)
Annual admissions 2019, median (IQR)	1597 (1283-2077)
Annual admissions with ventilation 2019, median (IQR)	328 (292-743)
Geographic region
Northeast	2 (13)
Midwest	7 (47)
South	3 (20)
West	3 (20)
Antibiotic stewardship or infection control support
Conduct surveillance for ventilator–associated event	3 (20)
Antimicrobial stewardship team supports ICU	15 (100)
Treatment guidelines for ventilator-associated infections	4 (27)
Prior education for indications of respiratory cultures in unit	2 (13)
Prior algorithm or guidelines for respiratory cultures in unit	2 (13)
ICU participates in national quality improvement collaboratives	11 (73)
Improving Pediatric Sepsis Outcomes	6 (40)
Solutions for Patient-Safety	9 (60)
Clinical staff and training programs
Pediatric residents	13 (87)
Emergency medicine residents	13 (87)
Pediatric critical care fellows	11 (73)
Critical care attendings have overnight in-house call	15 (100)
Advanced practice clinician (nurse practitioners and physician assistants)	14 (93)
Respiratory therapists specifically assigned to unit	11 (73)
Open unit	5 (33)
Patient populations
Medical patients	15 (100)
Surgical patients	15 (100)
Neurosurgical patients	15 (100)
Cardiac surgery patients	6 (40)
Nonconventional ventilation modes	15 (100)
Extracorporeal membrane oxygenation	13 (87)
Hematopoietic stem cell transplants	10 (67)
Solid organ transplants	11 (73)
Congenital diaphragmatic hernia	6 (40)
Microbiology laboratory restrictions of respiratory cultures
No rejection criteria or restrictions	6 (40)
Laboratory has some type of stewardship criteria
Any	9 (60)
If specimens are consistent with saliva, no further work-up	8 (53)
Rejects specimens repeated in certain number of days^a	5 (33)
Do not repeat antibiotic susceptibility testing within certain number of days^b	7 (47)

Open in a new tab

Abbreviation: ICU, intensive care unit.

^{^a}

Responses included rejecting if repeated in same day (3 sites), 2 days (1 site), or 3 days (1 site).

^{^b}

Responses included: did not repeat susceptibilities within 3 days (5 sites) or 7 days (2 sites).

Primary Outcome: Respiratory Culture Rates

The adjusted mean monthly EAC rate declined 16% from 7.80 to 6.55 cultures per 100 ventilator-days (RR, 0.84; 95% CI, 0.78 to 0.90; P < .001), with an absolute reduction of 1.25 (95% CI, −1.87 to −0.63) cultures per 100 ventilator-days per month (P < .001) (Figure 2 and Table 2). The crude median EAC rate declined 19% (eTable 4 in Supplement 1). The sensitivity analyses had consistent findings for the monthly EAC rates; there was a 17% reduction using PICU patient-days as the denominator (RR, 0.83; 95% CI, 0.76 to 0.92; P < .001), a 15% reduction excluding a 3-month wash-in period (RR, 0.85; 95% CI, 0.78 to 0.91; P < .001), a 13% reduction excluding baseline data through height of the COVID-19 pandemic (RR, 0.87; 95% CI, 0.81 to 0.93; P < .001), and a 13% reduction with adjustment for patient population complexity (RR, 0.87; 95% CI, 0.79 to 0.95; P = .002) (eTable 5 in Supplement 1). In the interrupted time-series analysis, the monthly EAC rate was stable during the preimplementation and postimplementation periods, with no significant difference between the preimplementation and postimplementation monthly trends. However, there was a significant 10% reduction associated with timing of CDS implementation (RR, 0.90; 95% CI, 0.81 to 1.00; P = .04).

Line chart of respiratory cultures per 100 V D over months, with many site lines and a mean trend. Horizontal axis labeled Time before and after implementation, mo, with tick marks from minus 24 at the left to 18 at the right. Vertical axis labeled Respiratory cultures per 100 V D, ranging from 0 at the bottom to 40 at the top, with horizontal gridlines at approximately 10, 20, 30, and 40. Multiple colored jagged lines represent individual sites, each varying month to month; several lines intermittently rise above 20, with the highest spikes reaching roughly 32 to 33 near the far left around minus 24 and again near plus 6, and another peak near minus 8 around 32. Many other site lines fluctuate mostly between about 3 and 15, with occasional dips near 1 to 2 and occasional peaks near 18 to 22. A thick black solid line runs across the plot as a smoothed mean, starting near 11 at minus 24, declining gradually to around 9 by about minus 12, remaining near 9 to 8 through the period around minus 6 to 0, then continuing a gentle downward slope to roughly 7 to 8 by plus 18. A vertical dashed black line at 0 months marks the implementation time, extending from the top to the bottom of the plotting area. — Monthly respiratory cultures per 100 VD for each of the 15 participating sites in the 24 months before implementation and 18 months after implementation of clinical decision support tools. The mean monthly average rate over time was estimated with a natural cubic smoothing spline with 4 degrees of freedom (black solid line). Individual site monthly respiratory culture rates are presented in colored lines. Indicator for timing of implementation is represented with the vertical dashed black line.

Table 2. Primary and Secondary Outcomes Before and After Implementation of Respiratory Culture Clinical Decision Support in 15 PICUs.

Outcome	Mean monthly rate (95% CI)^a		Postimplementation vs preimplementation comparison^a
Outcome	Preimplementation	Postimplementation	Relative rate (95% CI)	Absolute rate difference (95% CI)	P value
Primary outcome: respiratory culture rate^b	7.8 (5.55 to 10.96)	6.55 (4.63 to 9.28)	0.84 (0.78 to 0.90)	−1.25 (−1.87 to −0.63)	<.001
Secondary outcomes: antibiotic use
VAI antibiotic days of therapy rate^b^,^c^,^d	62.41 (53.60 to 72.66)	61.3 (51.62 to 72.78)	0.98 (0.89 to 1.08)	−1.11 (−7.17 to 4.95)	.72
VAI new antibiotic initiation rate^b^,^c^,^d	4.89 (3.86 to 6.18)	5.03 (4.09 to 6.20)	1.03 (0.95 to 1.11)	0.15 (−0.24 to 0.54)	.47
Secondary outcomes: clinical outcomes
PICU length of stay, median No. of d^d	3.22 (2.95 to 3.49)	3.19 (3.05 to 3.34)	NA	−0.04 (−0.29 to 0.24)	.84
7-d PICU readmission, %^d	3.74 (2.45 to 5.04)	3.39 (2.19 to 5.59)	0.90 (0.76 to 1.04)	−0.36 (−0.92 to 0.20)	.20
7-d Hospital readmission, %^d	2.00 (1.00 to 3.02)	1.70 (0.78 to 2.61)	0.85 (0.63 to 1.06)	−0.31 (−0.80 to 0.17)	.20
Sepsis, %^d^,^e	14.24 (10.89 to 17.59)	14.79 (12.22 to 17.35)	1.04 (0.95 to 1.13)	0.55 (−0.64 to 1.74)	.39
Septic shock, %^d^,^e	10.32 (9.03 to 11.61)	10.83 (9.66 to 11.99)	1.05 (0.95 to 1.15)	0.51 (−0.53 to 1.54)	.34
PICU ventilator-free days, median days^d	0.93 (0.48 to 1.24)	0.89 (0.48 to 1.30)	NA	−0.04 (−0.25 to 0.18)	.74
Mechanical ventilation duration, median days^d^,^f	3.46 (3.185 to 3.74)	3.33 (2.85 to 3.80)	NA	−0.14 (−0.56 to 0.29)	.53
Secondary outcomes: process and balancing measures^g
Culture repeated within 72 h, %	13.08 (8.21 to 17.94)	10.26 (5.88 to 14.64)	0.78 (0.68 to 0.89)	−2.81 (−4.34 to–1.29)	.001
Bronchoalveolar lavage culture rate^b	0.67 (0.43 to 1.04)	0.72 (0.46 to 1.11)	1.07 (0.85 to 1.34)	0.05 (−0.11 to 0.20)	.57
Ventilator-associated event rate^b^,^g	0.21 (0.15 to 0.31)	0.25 (0.13 to 0.48)	1.18 (0.87 to 1.58)	0.04 (−0.05 to 0.12)	.28

Open in a new tab

Abbreviations: PICU, pediatric intensive care unit; NA, not applicable; VAI, ventilator-associated infection.

^{^a}

Results derived from a Poisson regression model for the log monthly number of cultures with a main term for postimplementation vs preimplementation and season (fall, winter, and spring vs summer) and an offset for the log monthly number of ventilator-days. Binomial regression models were used for secondary outcomes measured as proportions and linear regression models were used for outcomes expressed as monthly mean or median. The binomial and linear models were fit using generalized estimating equations assuming an exchangeable correlation for month nested within PICU. SEs for all models were estimated using robust variance estimates.

^{^b}

Rate per 100 ventilator-days.

^{^c}

Antibiotics administered on PICU day 3 or later; antibiotics included are described in eTable 2 in Supplement 1.

^{^d}

Data from 9 sites in the curated cohort who participated in the Children’s Hospital Association Pediatric Health Information System (PHIS), including patients identified by the site to have been admitted to the PICU during the study period and matched to this PHIS database.

^{^e}

Proportion of events that were out-of-hospital admissions.

^{^f}

Cumulative mechanical ventilation-days among patients who were discharged alive.

^{^g}

Data from 3 hospitals with ventilator-associated events data for the full study period.

Although most sites observed a lower postimplementation EAC rate, there was heterogeneity across sites (Figure 3). The relative change in EAC rates varied from a 2% increase to a 53% reduction, with 9 of 15 sites achieving statistically significant adjusted rate reductions. In the post hoc analysis (eTable 5 in Supplement 1), excluding the 3 months lead-in to implementation, the relative reduction was similar at 17% (RR, 0.83; 95% CI, 0.77 to 0.90; P < .001). Stratifying by site characteristics, sites previously participating in BrighT STAR blood culture stewardship (ie, prior sites) experienced larger and statistically significant EAC rate reductions (20%; RR, 0.80; 95% CI, 0.75 to 0.86; P < .001) compared with new sites (10%; RR, 0.90; 95% CI, −2.84 to 0.16; P = .05), and sites with the highest (top quartile) preimplementation EAC rates had smaller relative reductions (13%; RR, 0.87; 95% CI, 0.77 to 0.99; P = .03) than sites with lower baseline rates (18%; RR, 0.82; 95% CI, 0.76 to 0.99; P < .001).

Data figure with table and forest plot of I R R by site and pooled estimate. Left panel table with four columns labeled Site No., Preimplementation rate (95% C I), Postimplementation rate (95% C I), and I R R (95% C I). Rows list sites 1 through 15 plus a final row labeled Pooled. Row values: Site 1, pre 8.0 (7.0-9.2), post 8.2 (7.1-9.4), I R R 1.02 (0.90-1.15). Site 2, pre 3.8 (3.2-4.5), post 3.8 (3.2-4.5), I R R 0.98 (0.85-1.14). Site 3, pre 14.6 (13.4-16.0), post 13.8 (12.7-15.1), I R R 0.95 (0.87-1.03). Site 4, pre 10.2 (8.9-11.8), post 9.6 (8.3-11.1), I R R 0.93 (0.81-1.08). Site 5, pre 1.3 (1.0-1.7), post 1.2 (0.9-1.5), I R R 0.89 (0.69-1.15). Site 6, pre 15.5 (12.2-19.7), post 13.5 (10.9-16.6), I R R 0.87 (0.70-1.07). Site 7, pre 5.2 (4.5-6.1), post 4.3 (3.6-5.1), I R R 0.82 (0.71-0.96). Site 8, pre 15.6 (13.0-18.8), post 12.7 (10.6-15.3), I R R 0.81 (0.70-0.95). Site 9, pre 8.4 (7.3-9.6), post 6.8 (5.8-7.9), I R R 0.81 (0.72-0.91). Site 10, pre 6.6 (5.9-7.3), post 5.3 (4.7-5.8), I R R 0.80 (0.72-0.88). Site 11, pre 11.0 (9.9-12.3), post 8.7 (7.8-9.7), I R R 0.78 (0.71-0.87). Site 12, pre 15.5 (13.8-17.5), post 12.0 (10.7-13.4), I R R 0.77 (0.69-0.86). Site 13, pre 9.2 (8.2-10.4), post 6.5 (5.7-7.4), I R R 0.70 (0.62-0.79). Site 14, pre 6.3 (5.1-7.8), post 4.4 (3.4-5.5), I R R 0.70 (0.56-0.87). Site 15, pre 6.1 (4.9-7.6), post 2.9 (2.2-3.8), I R R 0.47 (0.37-0.60). Pooled, pre 7.8 (5.6-11.0), post 6.6 (4.6-9.3), I R R 0.84 (0.78-0.90). Right panel forest plot aligned to the same rows. Horizontal axis label I R R (95% C I) with tick marks at 0.4, 0.6, 0.8, and 1.0. A vertical dotted reference line at 1.0. Text above the plot reads Reduction in culture rate on the left of the dotted line and Increase in culture rate on the right. Each site row contains a dark teal square marker at the I R R value with a horizontal line spanning its 95% C I; the pooled row contains a tan square marker with — Pediatric intensive care unit–specific relative rate of change with 95% CI comparing the mean monthly rate of respiratory cultures per 100 ventilator-days postimplementation vs preimplementation of clinical decision support tools. The individual site numeric label on the left corresponds to the ranking of respiratory culture rate reductions. IRR indicates incidence rate ratio.

Secondary Outcomes

In the subgroup of 9 sites that contributed data to the curated PHIS cohort, there was no change in the antibiotic DOT commonly used for VAI (RR, 0.98; 95% CI, 0.89 to 1.08) or initiations of VAI antibiotics (RR, 1.03; 95% CI, 0.95 to 1.11) (Table 2). The sensitivity analysis using the 12 sites in the full PHIS cohort also showed no change in VAI antibiotic DOT (RR, 0.93; 95% CI, 0.86 to 1.01) or VAI antibiotic initiations (RR, 1.01; 95% CI, 0.95 to 1.08). eFigure 4 in Supplement 1 shows the distribution of individual site VAI antibiotic DOT ranked by the sites’ EAC rate reduction; 3 of 9 sites in the curated cohort and 5 of 12 sites in the full PHIS cohort observed significant antibiotic DOT reductions. We found no changes when evaluating a narrower group of broad-spectrum antibiotics or comprehensive group of all antibiotics (eTable 6 in Supplement 1).

We found no differences in the monthly PICU length of stay, 7-day PICU readmissions, 7-day hospital readmissions, the percentage of patients with sepsis or septic shock, PICU ventilator-free days among living patients, and the cumulative mechanical ventilation duration across the study periods (Table 2). Sensitivity analysis using the full PHIS cohort of all ventilated ICU patients had similar findings (eTable 7 in Supplement 1).

Balancing, Site Safety, Adherence, and Process Measures

The mean monthly rate of bronchoalveolar lavage cultures per 100 ventilator-days did not differ across the study periods (Table 2). The VAE rate among the 3 sites conducting routine VAE surveillance was low at 0.2 per 100 ventilator-days preintervention and did not change. All sites submitted survey data for a median (IQR) of 15 (12-16) months with a median (IQR) total of 162 (49-247) attending survey inquiries per site. In total, there were 2525 safety survey inquiries, with a 93% response rate (2348 responses), of which 2 (<1%) reported delays to culture in patients from different sites. Both cases had evolving clinical symptoms, and a culture was obtained within the day of clinical changes; there were no concerns for delay in treatment or harm to the patients.

The adjusted monthly proportion of EACs repeated within 3 days declined from 13% to 10% (RR, 0.78; 95% CI, 0.68 to 0.89). The median (IQR) percentage adherence with CDS guidance postimplementation was 63% (50%-70%) at 1 month, 68% (60%-80%) at 6 months, and 67% (53%-81%) at 12 months. Comparing practices at postimplementation months 1 and 12, most cultures were obtained via endotracheal tubes (median [IQR], 75% [65%-81%] and 77% [67%-92%], respectively), and EACs obtained at time of intubation increased from a median (IQR) of 36% (25%-44%) to 43% (36%-59%) (P = .002) (eTable 8 in Supplement 1).

Discussion

In this multicenter cohort study, we observed an overall 16% decline in the rate of EAC use across 15 PICUs that implemented CDS to standardize and reduce overuse of EACs among children with artificial airways. The sensitivity analyses, which included an interrupted time-series analysis with a stable baseline preimplementation rate, exclusion of wash-out periods, analysis excluding COVID-19 pandemic data, and adjustment for patient population case mix, COVID-19 rates, and patient complexity support that EAC rate reduction was associated with implementing the CDS tools rather than representative of secular trends. Importantly, this study found no signal of patient harm across the 15 sites, including no increases in ventilation duration, ventilator-free days, length of stay, VAE, and clinician-reported concerns. Because this study did not enroll individual patients, active feedback was pursued at all sites, and there were 2 reports of delayed cultures among patients who were cultured within a day of infectious symptoms without resulting harm. The findings from this study build upon prior studies^{16,17,18,19,32,33} supporting reducing EACs among patients without signs of VAIs and avoiding closely repeated cultures as a safe and reasonable practice.

In this multicenter study, the overall EAC rate reduction was lower than the 34% to 48% reductions in prior single-center studies,^16,17,18 and although most sites observed EAC rate reductions, there was heterogeneity, ranging from no change to a 53% reduction. Different than prior studies, we adjusted for seasonality; however, local contextual factors may have contributed to the variability. First, sites did not follow a strict protocol; instead, they followed a structured QI framework with consensus on the premise to avoid EACs in patients without signs of respiratory infection but made adaptations to specific tools and implementation choices that could have variable effects on clinician testing behaviors. Second, the survey of EAC practices² indicated that sites had differing local EAC practices and preimplementation EAC rates. All sites, even those with low baseline rates, identified goals to improve practices, but baseline rates may have influenced potential for EAC reductions. Interestingly, sites with the highest baseline EAC rates observed lower relative reductions, suggesting that habituated or default testing practices may be a barrier to EAC reduction.^34,35,36 Additionally, there was variability in EACs obtained at intubation; because QI programs focused on EACs obtained for assessment of VAI (>48 hours of ventilation), as these declined and the proportion obtained at intubation increased, sites with more intubation cultures would have lower potential for EAC rate reductions. Third, sites that had previously participated in the BrighT STAR blood culture stewardship program had larger EAC reductions than new sites, suggesting that prior diagnostic stewardship efforts may have facilitated readiness for a second program. Finally, organizational factors may have influenced implementation because the postimplementation period followed the COVID-19 pandemic and many PICUs faced staffing and operational challenges that may have limited engagement with QI efforts leading to less EAC reduction. We are conducting an in-depth implementation assessment to fully characterize implementation and contextual factors that facilitated or hindered CDS adoption and sustainability.

Prior studies have suggested EACs are a driver of overtreatment for respiratory infections in ventilated PICU patients.^6,8 In contrast with the single-center studies that observed significant reductions (range, 4%-71%) in antibiotic use in ventilated patients with reduction in EACs,^16,17,18,19 we did not observe a clear overall reduction in antibiotic use. It is possible that the 16% EAC reduction was not a large enough reduction to impact overall antibiotic use, particularly if partially driven by reduced repeated cultures. We observed decreased repeated EACs; clinicians may have treated based on the first culture but have comparable antibiotic use. Particularly for antibiotic initiations, EACs obtained as part of pan cultures during sepsis evaluations may have reduced,³⁷ but patients would still be initiated on antibiotics for empiric sepsis treatment. This study did not capture individual patient EACs, antibiotic indications, or timing; thus, there may be individual reductions that were not captured. Importantly, prior single-center studies had antibiotic treatment guidance for VAI,^16,17,18 whereas this program focused on optimizing sample collection and EAC testing decisions without dedicated treatment guidance, which may be valuable to include in VAI management guidance. For all these reasons, this study may be insufficient to conclude that reducing EACs does not impact antibiotic treatment decisions. The overall goals to standardize testing and to reduce testing overuse remain important because each culture is associated with costs, materials, and staff effort to collect and process samples. Further, many BSR sites also standardized specimen collection practices, which were not measured, but also improve quality and validity of culture results.

The diagnostic stewardship strategy used in this study was CDS in the form of algorithms or checklists, requiring engagement from clinicians. There remain opportunities to optimize EAC utilization including improving adherence to guidance and considering intubation culture practices. Programs did not initially implement EMR-based changes because paper and intranet-based tools were more rapid to implement, could be modified, and required less institutional resources, although some sites incorporated CDS into their EMR after 18 months. Other diagnostic stewardship strategies include ordering restrictions embedded into the EMR, changes in specimen quality criteria by the laboratory, 2-step conditional testing (eg, urinalysis with reflex culture), and modification of result reports interpretation guidance.¹³ System-based changes built into the workflow that do not require active engagement of individual clinicians may have greater impact on testing utilization, but these approaches must be balanced with necessary resources, clinician autonomy, and individual patient treatment.³⁸ Future studies should assess the effectiveness and acceptability of these stewardship interventions.

Limitations

This study should be interpreted with some limitations. First, individual patient-level data were not available from sites and precluded adjustment for patient-level confounders; instead, we adjusted for patient population complexity over time. Similarly, individual patients were not enrolled to track safety outcomes; instead, there was a reliance on clinical teams to report concerns, which could allow for unrecognized events, especially very rare events. Reassuringly, safety concerns associated with avoiding cultures in patients without signs of respiratory infection have not been substantiated in other studies.^{16,17,18,19,32,33} Second, the source of the secondary clinical outcomes was administrative billing data, which can vary from EMR data and potentially include patients with ICU billing charges who were not cared for in the participating unit.³¹ These administrative billing discrepancies are expected to be consistent over time, permitting temporal comparisons. We created a specific curated cohort to narrow the administrative database to the patients cared for in the ICUs that participated in the study to address this concern. Regarding generalizability, all the sites were within academically-affiliated hospitals; however, the majority of pediatric ICU patients in the US are admitted to academically-affiliated children’s hospitals.³⁹ Third, due to reliance on ventilation codes to identify the secondary outcomes, patients with tracheostomies without ventilation were not captured.

Conclusions

In this multicenter cohort study, we observed a significant reduction in EACs among ventilated patients across 15 PICUs after implementation of CDS to standardize practices and reduce overtesting among patients unlikely to have VAIs without detection of patient harm. Implementation of CDS supports individual patient treatment but may have heterogenous uptake across settings. Future studies should consider optimal implementation strategies, additional opportunities to optimize testing and treatment for patients with artificial airways, the sustainability of interventions, and the cost impact of EAC stewardship.

Supplement 1.

eTable 1. Sites Participating in BrighT STAR Respiratory and Their Time Periods

eFigure 1. BrighT STAR Respiratory Quality Improvement Charter and Implementation Worksheet

eFigure 2. Example Clinical Decision Support Tool

eTable 2. Detailed Outcome Definitions

eTable 3. Antibiotics Evaluated and Syndromic Categorization

eFigure 3. BrighT STAR Respiratory Safety Monitoring Survey

eTable 4. Unadjusted Crude Primary and Secondary Outcomes Before and After Implementation of Respiratory Culture Clinical Decision Support in 15 PICUs

eTable 5. Sensitivity Analysis for Primary Outcome- Respiratory Culture Rate Before and After Implementation of Respiratory Culture Clinical Decision Support in 15 PICUs

eFigure 4. Funnel Plot Depicting Individual Site Relative Rate of Change of the Ventilator-Associated Infection Antibiotic Days of Therapy (DOT) Rate

eTable 6. Secondary Antibiotic Outcomes Comparing Post-Implementation vs. Pre-Implementation Rates Across the Primary and Sensitivity Analysis Cohorts

eTable 7. Secondary Clinical Outcomes Sensitivity Analysis Comparing Post-Implementation vs. Pre-Implementation Rates in the Full PHIS Cohort Data of Mechanically Ventilated Patients in the ICU

eTable 8. Process Measures After Implementation of Respiratory Culture Clinical Decision Support in 15 PICUs

jamanetwopen-e2626547-s001.pdf^{(883.5KB, pdf)}

Supplement 2.

Data Sharing Statement

jamanetwopen-e2626547-s002.pdf^{(13.5KB, pdf)}

References

1.Willson DF, Kirby A, Kicker JS. Respiratory secretion analyses in the evaluation of ventilator-associated pneumonia: a survey of current practice in pediatric critical care. Pediatr Crit Care Med. 2014;15(8):715-719. doi: 10.1097/PCC.0000000000000213 [DOI] [PubMed] [Google Scholar]
2.Sick-Samuels AC, Koontz DW, Xie A, et al. ; BrighT STAR Authorship Group . A survey of PICU clinician practices and perceptions regarding respiratory cultures in the evaluation of ventilator-associated infections in the BrighT STAR Collaborative. Pediatr Crit Care Med. 2024;25(1):e20-e30. doi: 10.1097/PCC.0000000000003379 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Willson DF, Conaway M, Kelly R, Hendley JO. The lack of specificity of tracheal aspirates in the diagnosis of pulmonary infection in intubated children. Pediatr Crit Care Med. 2014;15(4):299-305. doi: 10.1097/PCC.0000000000000106 [DOI] [PubMed] [Google Scholar]
4.Tsukahara K, Johnson B, Klimowich K, et al. Comparison of tracheal aspirate and bronchoalveolar lavage samples in the microbiological diagnosis of lower respiratory tract infection in pediatric patients. Pediatr Pulmonol. 2022;57(10):2405-2410. doi: 10.1002/ppul.26049 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Venkatachalam V, Hendley JO, Willson DF. The diagnostic dilemma of ventilator-associated pneumonia in critically ill children. Pediatr Crit Care Med. 2011;12(3):286-296. doi: 10.1097/PCC.0b013e3181fe2ffb [DOI] [PubMed] [Google Scholar]
6.Chiotos K, Blumenthal J, Boguniewicz J, et al. Antibiotic indications and appropriateness in the pediatric intensive care unit: a 10-center point prevalence study. Clin Infect Dis. 2023;76(3):e1021-e1030. doi: 10.1093/cid/ciac698 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Willson DF, Hoot M, Khemani R, et al. ; Ventilator-Associated INfection (VAIN) Investigators and the Pediatric Acute Lung Injury and Sepsis Investigator’s (PALISI) Network . Pediatric ventilator-associated infections: the ventilator-associated infection study. Pediatr Crit Care Med. 2017;18(1):e24-e34. doi: 10.1097/PCC.0000000000001001 [DOI] [PubMed] [Google Scholar]
8.Karsies T, Tarquinio K, Shein SL, et al. ; Pediatric Acute Lung Injury and Sepsis Investigator’s (PALISI) Network . Compliance With an antibiotic guideline for suspected ventilator-associated infection: the ventilator-associated infection (VAIN2) study. Pediatr Crit Care Med. 2021;22(10):859-869. doi: 10.1097/PCC.0000000000002761 [DOI] [PubMed] [Google Scholar]
9.Morrison JM, Kono N, Rush M, et al. ; Pediatric Research in Inpatient Settings Network . Factors associated with tracheostomy-associated infection treatment: a multicenter observational study. Pediatr Pulmonol. 2024;59(11):2761-2771. doi: 10.1002/ppul.27117 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Prinzi A, Parker SK, Thurm C, Birkholz M, Sick-Samuels A. Association of endotracheal aspirate culture variability and antibiotic use in mechanically ventilated pediatric patients. JAMA Netw Open. 2021;4(12):e2140378. doi: 10.1001/jamanetworkopen.2021.40378 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Albin OR, Saravolatz L, Petrie J, Henig O, Kaye KS. Rethinking the “pan-culture”: clinical impact of respiratory culturing in patients with low pretest probability of ventilator-associated pneumonia. Open Forum Infect Dis. 2022;9(6):ofac183. doi: 10.1093/ofid/ofac183 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Nussenblatt V, Avdic E, Berenholtz S, et al. Ventilator-associated pneumonia: overdiagnosis and treatment are common in medical and surgical intensive care units. Infect Control Hosp Epidemiol. 2014;35(3):278-284. doi: 10.1086/675279 [DOI] [PubMed] [Google Scholar]
13.Fabre V, Davis A, Diekema DJ, et al. Principles of diagnostic stewardship: a practical guide from the Society for Healthcare Epidemiology of America Diagnostic Stewardship Task Force. Infect Control Hosp Epidemiol. 2023;44(2):178-185. doi: 10.1017/ice.2023.5 [DOI] [PubMed] [Google Scholar]
14.Sick-Samuels AC, Woods-Hill C. Diagnostic stewardship in the pediatric intensive care unit. Infect Dis Clin North Am. 2022;36(1):203-218. doi: 10.1016/j.idc.2021.11.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Prinzi AM, Parker SK, Curtis DJ, Ziniel SI. The Pediatric Endotracheal Aspirate Culture Survey (PETACS): examining practice variation across pediatric microbiology laboratories in the United States. J Clin Microbiol. 2021;59(3):e02232-20. doi: 10.1128/JCM.02232-20 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Sick-Samuels AC, Linz M, Bergmann J, et al. Diagnostic stewardship of endotracheal aspirate cultures in a PICU. Pediatrics. 2021;147(5):e20201634. doi: 10.1542/peds.2020-1634 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Ormsby J, Conrad P, Blumenthal J, et al. Practice improvement for standardized evaluation and management of acute tracheitis in mechanically ventilated children. Pediatr Qual Saf. 2020;6(1):e368. doi: 10.1097/pq9.0000000000000368 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Chiotos K, Lee G, Sydney G, et al. Impact of a tracheal aspirate culture diagnostic test stewardship intervention in a tertiary care PICU. Infect Control Hosp Epidemiol. 2024;45(8):943-951. doi: 10.1017/ice.2024.105 [DOI] [PubMed] [Google Scholar]
19.Albin OR, Troost JP, Saravolatz L II, et al. A quasi-experimental study of a bundled diagnostic stewardship intervention for ventilator-associated pneumonia. Clin Microbiol Infect. 2024;30(4):499-506. doi: 10.1016/j.cmi.2023.12.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Johns Hopkins Children’s Center. BrighT STAR: a learning health community. Accessed May 29, 2024. https://www.hopkinsmedicine.org/johns-hopkins-childrens-center/what-we-treat/specialties/infectious-diseases/programs-centers/bright-star
21.Woods-Hill CZ, Colantuoni EA, Koontz DW, et al. ; Bright STAR Authorship Group . Association of diagnostic stewardship for blood cultures in critically ill children with culture rates, antibiotic use, and patient outcomes: results of the BrighT STAR Collaborative. JAMA Pediatr. 2022;176(7):690-698. doi: 10.1001/jamapediatrics.2022.1024 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Smeraglio A, Pittenger B, DiVeronica M, et al. Adapting SQUIRE 2.0 to create a quality improvement evidence-based medicine critical appraisal tool (QI-EBM-CAT) for graduate medical education trainees. J Grad Med Educ. 2022;14(6):704-709. doi: 10.4300/JGME-D-22-00033.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Pronovost PJ, Berenholtz SM, Needham DM. Translating evidence into practice: a model for large scale knowledge translation. BMJ. 2008;337:a1714. doi: 10.1136/bmj.a1714 [DOI] [PubMed] [Google Scholar]
24.Malone S, Prewitt K, Hackett R, et al. The Clinical Sustainability Assessment Tool: measuring organizational capacity to promote sustainability in healthcare. Implement Sci Commun. 2021;2(1):77. doi: 10.1186/s43058-021-00181-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Helfrich CD, Li YF, Sharp ND, Sales AE. Organizational readiness to change assessment (ORCA): development of an instrument based on the Promoting Action on Research in Health Services (PARIHS) framework. Implement Sci. 2009;4:38. doi: 10.1186/1748-5908-4-38 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Sick-Samuels AC, Kelly DP, Woods-Hill CZ, et al. ; BrighT STAR (Testing STewardship for Antibiotic Reduction) Respiratory Consensus Authorship group . Diagnostic stewardship of endotracheal aspirate cultures in hospitalized children with artificial airways: expert consensus statements from the BrighT STAR (Testing Stewardship for Antibiotic Reduction) Respiratory Collaborative. Pediatr Crit Care Med. 2025;26(4):e569-e582. doi: 10.1097/PCC.0000000000003695 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Children’s Hospital Association . Leverage your data with CHA’s pediatric health information system. Accessed January 23, 2026. https://www.childrenshospitals.org/content/analytics/product-program/pediatric-health-information-system
28.National Healthcare Safety Network . Pneumonia (ventilator-associated [VAP] and non-ventilator-associated pneumonia [PNEU]) event. Centers for Disease Control and Prevention . Published January 2026. Accessed June 24, 2026. https://www.cdc.gov/nhsn/pdfs/pscmanual/6pscvapcurrent.pdf
29.Richardson T, Rodean J, Harris M, Berry J, Gay JC, Hall M. Development of hospitalization resource intensity scores for kids (H-RISK) and comparison across pediatric populations. J Hosp Med. 2018;13(9):602-608. doi: 10.12788/jhm.2948 [DOI] [PubMed] [Google Scholar]
30.Feinstein JA, Hall M, Davidson A, Feudtner C. Pediatric complex chronic condition system version 3. JAMA Netw Open. 2024;7(7):e2420579. doi: 10.1001/jamanetworkopen.2024.20579 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Xiao S, Woods-Hill CZ, Koontz D, et al. ; Bright STAR Authorship Group . Comparison of administrative database-derived and hospital-derived data for monitoring blood culture use in the pediatric intensive care unit. J Pediatric Infect Dis Soc. 2023;12(7):436-442. doi: 10.1093/jpids/piad048 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Sick-Samuels AC, Booth LD, Milstone AM, Schumacher C, Bergmann J, Stockwell DC. A novel comprehensive algorithm for evaluation of PICU patients with new fever or instability. Pediatr Crit Care Med. 2023;24(8):670-680. doi: 10.1097/PCC.0000000000003256 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Tripathi RK, Kenaa B, Claeys KC, et al. Improving antibiotic use for ventilator-associated pneumonia through diagnostic stewardship: a proof-of-concept mixed methods study. Open Forum Infect Dis. 2024;11(9):ofae500. doi: 10.1093/ofid/ofae500 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Chiotos K, Marshall D, Kellom K, et al. Mixed-methods process evaluation of a respiratory-culture diagnostic stewardship intervention. Infect Control Hosp Epidemiol. 2023;44(2):191-199. doi: 10.1017/ice.2022.299 [DOI] [PubMed] [Google Scholar]
35.Woods-Hill CZ, Koontz DW, Xie A, et al. ; BrighT STAR authorship group . Diagnostic stewardship for blood cultures in the pediatric intensive care unit: lessons in implementation from the BrighT STAR Collaborative. Antimicrob Steward Healthc Epidemiol. 2024;4(1):e148. doi: 10.1017/ash.2024.416 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Helfrich CD, Majerczyk BR, Nolen E. De-Implementing Low-Value Practices in Healthcare and Public Health. Practical Implementation Science: Moving Evidence Into Action. Springer Publishing Co; 2022. [Google Scholar]
37.Linz MS, Booth LD, Milstone AM, Stockwell DC, Sick-Samuels AC. Evaluation of a comprehensive algorithm for PICU patients with new fever or instability: association of clinical decision support with testing practices. Pediatr Crit Care Med. 2024;25(11):998-1004. doi: 10.1097/PCC.0000000000003582 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Advani S, Vaughn VM. Quality improvement interventions and implementation strategies for urine culture stewardship in the acute care setting: advances and challenges. Curr Infect Dis Rep. 2021;23(10):16. doi: 10.1007/s11908-021-00760-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Killien EY, Keller MR, Watson RS, Hartman ME. Epidemiology of intensive care admissions for children in the US From 2001 to 2019. JAMA Pediatr. 2023;177(5):506-515. doi: 10.1001/jamapediatrics.2023.0184 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

eTable 1. Sites Participating in BrighT STAR Respiratory and Their Time Periods

eFigure 1. BrighT STAR Respiratory Quality Improvement Charter and Implementation Worksheet

eFigure 2. Example Clinical Decision Support Tool

eTable 2. Detailed Outcome Definitions

eTable 3. Antibiotics Evaluated and Syndromic Categorization

eFigure 3. BrighT STAR Respiratory Safety Monitoring Survey

eTable 4. Unadjusted Crude Primary and Secondary Outcomes Before and After Implementation of Respiratory Culture Clinical Decision Support in 15 PICUs

eTable 5. Sensitivity Analysis for Primary Outcome- Respiratory Culture Rate Before and After Implementation of Respiratory Culture Clinical Decision Support in 15 PICUs

eFigure 4. Funnel Plot Depicting Individual Site Relative Rate of Change of the Ventilator-Associated Infection Antibiotic Days of Therapy (DOT) Rate

eTable 6. Secondary Antibiotic Outcomes Comparing Post-Implementation vs. Pre-Implementation Rates Across the Primary and Sensitivity Analysis Cohorts

eTable 7. Secondary Clinical Outcomes Sensitivity Analysis Comparing Post-Implementation vs. Pre-Implementation Rates in the Full PHIS Cohort Data of Mechanically Ventilated Patients in the ICU

eTable 8. Process Measures After Implementation of Respiratory Culture Clinical Decision Support in 15 PICUs

jamanetwopen-e2626547-s001.pdf^{(883.5KB, pdf)}

Supplement 2.

Data Sharing Statement

jamanetwopen-e2626547-s002.pdf^{(13.5KB, pdf)}

[zoi260729r1] 1.Willson DF, Kirby A, Kicker JS. Respiratory secretion analyses in the evaluation of ventilator-associated pneumonia: a survey of current practice in pediatric critical care. Pediatr Crit Care Med. 2014;15(8):715-719. doi: 10.1097/PCC.0000000000000213 [DOI] [PubMed] [Google Scholar]

[zoi260729r2] 2.Sick-Samuels AC, Koontz DW, Xie A, et al. ; BrighT STAR Authorship Group . A survey of PICU clinician practices and perceptions regarding respiratory cultures in the evaluation of ventilator-associated infections in the BrighT STAR Collaborative. Pediatr Crit Care Med. 2024;25(1):e20-e30. doi: 10.1097/PCC.0000000000003379 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r3] 3.Willson DF, Conaway M, Kelly R, Hendley JO. The lack of specificity of tracheal aspirates in the diagnosis of pulmonary infection in intubated children. Pediatr Crit Care Med. 2014;15(4):299-305. doi: 10.1097/PCC.0000000000000106 [DOI] [PubMed] [Google Scholar]

[zoi260729r4] 4.Tsukahara K, Johnson B, Klimowich K, et al. Comparison of tracheal aspirate and bronchoalveolar lavage samples in the microbiological diagnosis of lower respiratory tract infection in pediatric patients. Pediatr Pulmonol. 2022;57(10):2405-2410. doi: 10.1002/ppul.26049 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r5] 5.Venkatachalam V, Hendley JO, Willson DF. The diagnostic dilemma of ventilator-associated pneumonia in critically ill children. Pediatr Crit Care Med. 2011;12(3):286-296. doi: 10.1097/PCC.0b013e3181fe2ffb [DOI] [PubMed] [Google Scholar]

[zoi260729r6] 6.Chiotos K, Blumenthal J, Boguniewicz J, et al. Antibiotic indications and appropriateness in the pediatric intensive care unit: a 10-center point prevalence study. Clin Infect Dis. 2023;76(3):e1021-e1030. doi: 10.1093/cid/ciac698 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r7] 7.Willson DF, Hoot M, Khemani R, et al. ; Ventilator-Associated INfection (VAIN) Investigators and the Pediatric Acute Lung Injury and Sepsis Investigator’s (PALISI) Network . Pediatric ventilator-associated infections: the ventilator-associated infection study. Pediatr Crit Care Med. 2017;18(1):e24-e34. doi: 10.1097/PCC.0000000000001001 [DOI] [PubMed] [Google Scholar]

[zoi260729r8] 8.Karsies T, Tarquinio K, Shein SL, et al. ; Pediatric Acute Lung Injury and Sepsis Investigator’s (PALISI) Network . Compliance With an antibiotic guideline for suspected ventilator-associated infection: the ventilator-associated infection (VAIN2) study. Pediatr Crit Care Med. 2021;22(10):859-869. doi: 10.1097/PCC.0000000000002761 [DOI] [PubMed] [Google Scholar]

[zoi260729r9] 9.Morrison JM, Kono N, Rush M, et al. ; Pediatric Research in Inpatient Settings Network . Factors associated with tracheostomy-associated infection treatment: a multicenter observational study. Pediatr Pulmonol. 2024;59(11):2761-2771. doi: 10.1002/ppul.27117 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r10] 10.Prinzi A, Parker SK, Thurm C, Birkholz M, Sick-Samuels A. Association of endotracheal aspirate culture variability and antibiotic use in mechanically ventilated pediatric patients. JAMA Netw Open. 2021;4(12):e2140378. doi: 10.1001/jamanetworkopen.2021.40378 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r11] 11.Albin OR, Saravolatz L, Petrie J, Henig O, Kaye KS. Rethinking the “pan-culture”: clinical impact of respiratory culturing in patients with low pretest probability of ventilator-associated pneumonia. Open Forum Infect Dis. 2022;9(6):ofac183. doi: 10.1093/ofid/ofac183 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r12] 12.Nussenblatt V, Avdic E, Berenholtz S, et al. Ventilator-associated pneumonia: overdiagnosis and treatment are common in medical and surgical intensive care units. Infect Control Hosp Epidemiol. 2014;35(3):278-284. doi: 10.1086/675279 [DOI] [PubMed] [Google Scholar]

[zoi260729r13] 13.Fabre V, Davis A, Diekema DJ, et al. Principles of diagnostic stewardship: a practical guide from the Society for Healthcare Epidemiology of America Diagnostic Stewardship Task Force. Infect Control Hosp Epidemiol. 2023;44(2):178-185. doi: 10.1017/ice.2023.5 [DOI] [PubMed] [Google Scholar]

[zoi260729r14] 14.Sick-Samuels AC, Woods-Hill C. Diagnostic stewardship in the pediatric intensive care unit. Infect Dis Clin North Am. 2022;36(1):203-218. doi: 10.1016/j.idc.2021.11.003 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r15] 15.Prinzi AM, Parker SK, Curtis DJ, Ziniel SI. The Pediatric Endotracheal Aspirate Culture Survey (PETACS): examining practice variation across pediatric microbiology laboratories in the United States. J Clin Microbiol. 2021;59(3):e02232-20. doi: 10.1128/JCM.02232-20 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r16] 16.Sick-Samuels AC, Linz M, Bergmann J, et al. Diagnostic stewardship of endotracheal aspirate cultures in a PICU. Pediatrics. 2021;147(5):e20201634. doi: 10.1542/peds.2020-1634 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r17] 17.Ormsby J, Conrad P, Blumenthal J, et al. Practice improvement for standardized evaluation and management of acute tracheitis in mechanically ventilated children. Pediatr Qual Saf. 2020;6(1):e368. doi: 10.1097/pq9.0000000000000368 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r18] 18.Chiotos K, Lee G, Sydney G, et al. Impact of a tracheal aspirate culture diagnostic test stewardship intervention in a tertiary care PICU. Infect Control Hosp Epidemiol. 2024;45(8):943-951. doi: 10.1017/ice.2024.105 [DOI] [PubMed] [Google Scholar]

[zoi260729r19] 19.Albin OR, Troost JP, Saravolatz L II, et al. A quasi-experimental study of a bundled diagnostic stewardship intervention for ventilator-associated pneumonia. Clin Microbiol Infect. 2024;30(4):499-506. doi: 10.1016/j.cmi.2023.12.023 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r20] 20.Johns Hopkins Children’s Center. BrighT STAR: a learning health community. Accessed May 29, 2024. https://www.hopkinsmedicine.org/johns-hopkins-childrens-center/what-we-treat/specialties/infectious-diseases/programs-centers/bright-star

[zoi260729r21] 21.Woods-Hill CZ, Colantuoni EA, Koontz DW, et al. ; Bright STAR Authorship Group . Association of diagnostic stewardship for blood cultures in critically ill children with culture rates, antibiotic use, and patient outcomes: results of the BrighT STAR Collaborative. JAMA Pediatr. 2022;176(7):690-698. doi: 10.1001/jamapediatrics.2022.1024 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r22] 22.Smeraglio A, Pittenger B, DiVeronica M, et al. Adapting SQUIRE 2.0 to create a quality improvement evidence-based medicine critical appraisal tool (QI-EBM-CAT) for graduate medical education trainees. J Grad Med Educ. 2022;14(6):704-709. doi: 10.4300/JGME-D-22-00033.1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r23] 23.Pronovost PJ, Berenholtz SM, Needham DM. Translating evidence into practice: a model for large scale knowledge translation. BMJ. 2008;337:a1714. doi: 10.1136/bmj.a1714 [DOI] [PubMed] [Google Scholar]

[zoi260729r24] 24.Malone S, Prewitt K, Hackett R, et al. The Clinical Sustainability Assessment Tool: measuring organizational capacity to promote sustainability in healthcare. Implement Sci Commun. 2021;2(1):77. doi: 10.1186/s43058-021-00181-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r25] 25.Helfrich CD, Li YF, Sharp ND, Sales AE. Organizational readiness to change assessment (ORCA): development of an instrument based on the Promoting Action on Research in Health Services (PARIHS) framework. Implement Sci. 2009;4:38. doi: 10.1186/1748-5908-4-38 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r26] 26.Sick-Samuels AC, Kelly DP, Woods-Hill CZ, et al. ; BrighT STAR (Testing STewardship for Antibiotic Reduction) Respiratory Consensus Authorship group . Diagnostic stewardship of endotracheal aspirate cultures in hospitalized children with artificial airways: expert consensus statements from the BrighT STAR (Testing Stewardship for Antibiotic Reduction) Respiratory Collaborative. Pediatr Crit Care Med. 2025;26(4):e569-e582. doi: 10.1097/PCC.0000000000003695 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r27] 27.Children’s Hospital Association . Leverage your data with CHA’s pediatric health information system. Accessed January 23, 2026. https://www.childrenshospitals.org/content/analytics/product-program/pediatric-health-information-system

[zoi260729r28] 28.National Healthcare Safety Network . Pneumonia (ventilator-associated [VAP] and non-ventilator-associated pneumonia [PNEU]) event. Centers for Disease Control and Prevention . Published January 2026. Accessed June 24, 2026. https://www.cdc.gov/nhsn/pdfs/pscmanual/6pscvapcurrent.pdf

[zoi260729r29] 29.Richardson T, Rodean J, Harris M, Berry J, Gay JC, Hall M. Development of hospitalization resource intensity scores for kids (H-RISK) and comparison across pediatric populations. J Hosp Med. 2018;13(9):602-608. doi: 10.12788/jhm.2948 [DOI] [PubMed] [Google Scholar]

[zoi260729r30] 30.Feinstein JA, Hall M, Davidson A, Feudtner C. Pediatric complex chronic condition system version 3. JAMA Netw Open. 2024;7(7):e2420579. doi: 10.1001/jamanetworkopen.2024.20579 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r31] 31.Xiao S, Woods-Hill CZ, Koontz D, et al. ; Bright STAR Authorship Group . Comparison of administrative database-derived and hospital-derived data for monitoring blood culture use in the pediatric intensive care unit. J Pediatric Infect Dis Soc. 2023;12(7):436-442. doi: 10.1093/jpids/piad048 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r32] 32.Sick-Samuels AC, Booth LD, Milstone AM, Schumacher C, Bergmann J, Stockwell DC. A novel comprehensive algorithm for evaluation of PICU patients with new fever or instability. Pediatr Crit Care Med. 2023;24(8):670-680. doi: 10.1097/PCC.0000000000003256 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r33] 33.Tripathi RK, Kenaa B, Claeys KC, et al. Improving antibiotic use for ventilator-associated pneumonia through diagnostic stewardship: a proof-of-concept mixed methods study. Open Forum Infect Dis. 2024;11(9):ofae500. doi: 10.1093/ofid/ofae500 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r34] 34.Chiotos K, Marshall D, Kellom K, et al. Mixed-methods process evaluation of a respiratory-culture diagnostic stewardship intervention. Infect Control Hosp Epidemiol. 2023;44(2):191-199. doi: 10.1017/ice.2022.299 [DOI] [PubMed] [Google Scholar]

[zoi260729r35] 35.Woods-Hill CZ, Koontz DW, Xie A, et al. ; BrighT STAR authorship group . Diagnostic stewardship for blood cultures in the pediatric intensive care unit: lessons in implementation from the BrighT STAR Collaborative. Antimicrob Steward Healthc Epidemiol. 2024;4(1):e148. doi: 10.1017/ash.2024.416 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r36] 36.Helfrich CD, Majerczyk BR, Nolen E. De-Implementing Low-Value Practices in Healthcare and Public Health. Practical Implementation Science: Moving Evidence Into Action. Springer Publishing Co; 2022. [Google Scholar]

[zoi260729r37] 37.Linz MS, Booth LD, Milstone AM, Stockwell DC, Sick-Samuels AC. Evaluation of a comprehensive algorithm for PICU patients with new fever or instability: association of clinical decision support with testing practices. Pediatr Crit Care Med. 2024;25(11):998-1004. doi: 10.1097/PCC.0000000000003582 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r38] 38.Advani S, Vaughn VM. Quality improvement interventions and implementation strategies for urine culture stewardship in the acute care setting: advances and challenges. Curr Infect Dis Rep. 2021;23(10):16. doi: 10.1007/s11908-021-00760-3 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi260729r39] 39.Killien EY, Keller MR, Watson RS, Hartman ME. Epidemiology of intensive care admissions for children in the US From 2001 to 2019. JAMA Pediatr. 2023;177(5):506-515. doi: 10.1001/jamapediatrics.2023.0184 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Diagnostic Stewardship of Respiratory Cultures Using Clinical Decision Support in the PICU

Anna C Sick-Samuels, MD, MPH

Charlotte Z Woods-Hill, MD, MSHP

Daniel P Kelly, MD, MHQS

Danielle W Koontz, MAA, MAS

Nora Elhaissouni, ScM

Urmi Kumar, MPH

Anping Xie, PhD

Troy Richardson, PhD

Sreejata Dutta, PhD

Jill A Marsteller, PhD, MPP

Elizabeth Colantuoni, PhD

Aaron M Milstone, MD, MHS

Zachary Aldewereld, MD

Michael J Auth, DO

Ritu Banerjee, MD, PhD

Jennifer A Blumenthal, MD

Katharine Boyle, MD

Cara Cecil, MD, MS

Samantha Dallefeld, MD

Brian F Flaherty, MD

Charles B Foster, MD

Keshava M N Gowda, MBBS

Kelly A Hardy, PhD, CPNP-PC/AC

Sarmistha B Hauger, MD

Andrea Green Hines, MD

Sue J Hong, MD

Nicholas D Hysmith, MD

Andrew Kiragu, MD

Aileen L Kirby, MD

Christina Koutsari, PharmD, PhD, BCPS, BCIDP

Gitte Y Larsen, MD, MPH

Caitlin Naureckas Li, MD, MHQS

John C Lin, MD

Matthew H M Marx, MD, MS

Bridget M Norton, MD, MBA

Gregory P Priebe, MD

Glenn J Rapsinski, MD, PhD

Rebecca G Same, MD

Hayden T Schwenk, MD, MPH

Katherine M Steffen, MD, MHS

Sachin D Tadphale, MBBS, MPH

Philip Toltzis, MD

Lorne Walker, MD, PhD

Key Points

Question

Findings

Meaning

Abstract

Importance

Objective

Design, Setting, and Participants

Exposure

Main Outcomes and Measures

Results

Conclusions and Relevance

Introduction

Methods

Study Design and Setting: The BrighT STAR Respiratory Collaborative

Figure 1. Flow Chart of Implementation Stages and Key Steps Conducted by Participating Sites and the Coordinating Center.

Program Implementation: Practice Changes Driven by Local Multidisciplinary QI Teams

Outcome and Process Measures

Statistical Analysis

Results

Cohort Characteristics

Table 1. Site Baseline Characteristics.

Primary Outcome: Respiratory Culture Rates

Figure 2. Line Graph of Monthly Respiratory Cultures Per 100 Ventilator-Days (VD) for Each of 15 Participating Sites Before and After Implementation of Clinical Decision Support Tools.

Table 2. Primary and Secondary Outcomes Before and After Implementation of Respiratory Culture Clinical Decision Support in 15 PICUs.

Figure 3. Individual Site-Adjusted Respiratory Culture Rates and Funnel Plot Depicting Individual Relative Rate of Change Comparing Postimplementation With Preimplementation Rates.

Secondary Outcomes

Balancing, Site Safety, Adherence, and Process Measures

Discussion

Limitations

Conclusions

References

Associated Data

Supplementary Materials

ACTIONS