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

Charlotte Z Woods-Hill; Elizabeth A Colantuoni; Danielle W Koontz; Annie Voskertchian; Anping Xie; Cary Thurm; Marlene R Miller; James C Fackler; Aaron M Milstone; and the Bright STAR Authorship Group; Asya Agulnik; J Elaine-Marie Albert; Michael J Auth; Erin Bradley; Jason A Clayton; Susan E Coffin; Samantha Dallefeld; Chidiebere P Ezetendu; Nina A Fainberg; Brian F Flaherty; Charles B Foster; Sarmistha B Hauger; Sue J Hong; Nicholas D Hysmith; Aileen L Kirby; Larry K Kociolek; Gitte Y Larsen; John C Lin; William M Linam; Jason G Newland; Dawn Nolt; Gregory P Priebe; Thomas J Sandora; Hayden T Schwenk; Craig M Smith; Katherine M Steffen; Sachin D Tadphale; Philip Toltzis; Joshua Wolf; Danielle M Zerr

doi:10.1001/jamapediatrics.2022.1024

. 2022 May 2;176(7):690–698. doi: 10.1001/jamapediatrics.2022.1024

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

Charlotte Z Woods-Hill ^1,^2,^✉, Elizabeth A Colantuoni ³, Danielle W Koontz ⁴, Annie Voskertchian ⁴, Anping Xie ^5,⁶, Cary Thurm ⁷, Marlene R Miller ^8,⁹, James C Fackler ⁵, Aaron M Milstone ^4,¹⁰; and the Bright STAR Authorship Group, Asya Agulnik ¹¹, J Elaine-Marie Albert ¹², Michael J Auth ¹³, Erin Bradley ¹⁴, Jason A Clayton ¹⁵, Susan E Coffin ¹⁶, Samantha Dallefeld ¹³, Chidiebere P Ezetendu ¹⁷, Nina A Fainberg ¹⁸, Brian F Flaherty ¹⁹, Charles B Foster ²⁰, Sarmistha B Hauger ²¹, Sue J Hong ²², Nicholas D Hysmith ²³, Aileen L Kirby ²⁴, Larry K Kociolek ²⁵, Gitte Y Larsen ²⁶, John C Lin ²⁷, William M Linam ²⁸, Jason G Newland ²⁷, Dawn Nolt ²⁹, Gregory P Priebe ^30,^31,³², Thomas J Sandora ³¹, Hayden T Schwenk ³³, Craig M Smith ^25,³⁴, Katherine M Steffen ³⁵, Sachin D Tadphale ³⁶, Philip Toltzis ⁸, Joshua Wolf ^37,³⁸, Danielle M Zerr ³⁹

¹Division of Critical Care Medicine, Children’s Hospital of Philadelphia, University of Pennsylvania Perelman School of Medicine, Philadelphia

²Leonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia

³Department of Biostatistics, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland

⁴Division of Infectious Diseases, Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore, Maryland

⁵Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland

⁶Armstrong Institute for Patient Safety and Quality, Johns Hopkins University School of Medicine, Baltimore, Maryland

⁷Children’s Hospital Association, Lenexa, Kansas

⁸Rainbow Babies and Children’s Hospital, Cleveland, Ohio

⁹Case Western Reserve University School of Medicine, Cleveland, Ohio

¹⁰Department of Hospital Epidemiology and Infection Control, Johns Hopkins Hospital, Baltimore, Maryland

¹¹Department of Global Pediatric Medicine, Division of Critical Care, St Jude Children’s Research Hospital, Memphis, Tennessee

¹²Division of Pediatric Critical Care Medicine, Department of Pediatrics, Seattle Children’s Hospital, University of Washington, Seattle

¹³Division of Pediatric Critical Care, Department of Pediatrics, Dell Medical School, University of Texas at Austin

¹⁴Children’s Healthcare of Atlanta, Atlanta, Georgia

¹⁵University Hospitals, Rainbow Babies and Children’s Hospital, Cleveland, Ohio

¹⁶Children’s Hospital of Philadelphia, University of Pennsylvania Perelman School of Medicine, Philadelphia

¹⁷Cleveland Clinic Children’s Hospital, Cleveland, Ohio

¹⁸Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania

¹⁹University of Utah, Department of Pediatrics, Division of Critical Care, Salt Lake City

²⁰Center for Pediatric Infectious Diseases, Cleveland Clinic Children’s Hospital, Cleveland, Ohio

²¹Division of Pediatric Infectious Disease, Dell Medical School, University of Texas at Austin

²²Division of Critical Care, Ann & Robert H. Lurie Children’s Hospital of Chicago, Illinois

²³Department of Pediatrics, University of Tennessee Health Science Center, Le Bonheur Children’s Hospital, Memphis

²⁴Division of Pediatric Critical Care, Doernbecher Children’s Hospital, Oregon Health & Science University, Portland

²⁵Department of Pediatrics, Northwestern University Feinberg School of Medicine, Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, Illinois

²⁶Division of Critical Care Medicine, Department of Pediatrics, Intermountain Primary Children’s Hospital, University of Utah, Salt Lake City

²⁷Washington University School of Medicine, St Louis, Missouri

²⁸Emory School of Medicine, Children’s Healthcare of Atlanta, Atlanta, Georgia

²⁹Doernbecher Children’s Hospital, Oregon Health & Science University, Portland

³⁰Division of Critical Care Medicine, Department of Anesthesiology, Critical Care and Pain Medicine, Boston Children’s Hospital, Boston, Massachusetts

³¹Division of Infectious Diseases, Department of Pediatrics, Boston Children’s Hospital, Boston, Massachusetts

³²Department of Anesthesia, Harvard Medical School, Boston, Massachusetts

³³Stanford University School of Medicine, Lucile Packard Children’s Hospital Stanford, Stanford, California

³⁴Department of Neurology, Northwestern University Feinberg School of Medicine, Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, Illinois

³⁵Stanford University Department of Pediatrics, Division of Pediatric Critical Care Medicine, Stanford, California

³⁶Division of Pediatric Critical Care and Cardiology, Department of Pediatrics, Le Bonheur Children’s Hospital, University of Tennessee Health Science Center, Memphis

³⁷Department of Infectious Diseases, St Jude Children’s Research Hospital, Memphis, Tennessee

³⁸Department of Pediatrics, University of Tennessee Health Science Center, Memphis

³⁹Seattle Children’s Hospital, Department of Pediatrics, University of Washington, Seattle

Group Information: Additional members of the Bright STAR Authorship Group appear at the end of this article.

Accepted for Publication: January 15, 2022.

Published Online: May 2, 2022. doi:10.1001/jamapediatrics.2022.1024

^✉

Corresponding Author: Charlotte Z. Woods-Hill, MD, MSHP, Division of Critical Care Medicine, Children’s Hospital of Philadelphia, 3401 Civic Center Blvd, Wood Building, Sixth Floor, Ste 6026A, Philadelphia, PA 19104 (woodshillc@chop.edu).

Bright STAR Authorship Group: Asya Agulnik, MD; J. Elaine-Marie Albert, MD, MHA; Michael J. Auth; Erin Bradley, MD, MS; Jason A. Clayton, MD, PhD; Susan E. Coffin, MD, MPH; Samantha Dallefeld, MD; Chidiebere P. Ezetendu, MD; Nina A. Fainberg, MD; Brian F. Flaherty, MD; Charles B. Foster, MD; Sarmistha B. Hauger, MD; Sue J. Hong, MD; Nicholas D. Hysmith, MD; Aileen L. Kirby, MD; Larry K. Kociolek, MD, MSCI; Gitte Y. Larsen, MD, MPH; John C. Lin, MD; William M. Linam, MD, MS; Jason G. Newland, MD, MEd; Dawn Nolt, MD, MPH; Gregory P. Priebe, MD; Thomas J. Sandora, MD, MPH; Hayden T. Schwenk, MD, MPH; Craig M. Smith, MD; Katherine M. Steffen, MD, MHS; Sachin D. Tadphale, MBBS, MPH; Philip Toltzis, MD; Joshua Wolf, MBBS, PhD; Danielle M. Zerr, MD, MPH.

Affiliations of Bright STAR Authorship Group: Rainbow Babies and Children’s Hospital, Cleveland, Ohio (Toltzis); Department of Global Pediatric Medicine, Division of Critical Care, St Jude Children’s Research Hospital, Memphis, Tennessee (Agulnik); Division of Pediatric Critical Care Medicine, Department of Pediatrics, Seattle Children’s Hospital, University of Washington, Seattle (Albert); Division of Pediatric Critical Care, Department of Pediatrics, Dell Medical School, University of Texas at Austin (Auth, Dallefeld); Children’s Healthcare of Atlanta, Atlanta, Georgia (Bradley); University Hospitals, Rainbow Babies and Children’s Hospital, Cleveland, Ohio (Clayton); Children’s Hospital of Philadelphia, University of Pennsylvania Perelman School of Medicine, Philadelphia (Coffin); Cleveland Clinic Children’s Hospital, Cleveland, Ohio (Ezetendu); Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania (Fainberg); University of Utah, Department of Pediatrics, Division of Critical Care, Salt Lake City (Flaherty); Center for Pediatric Infectious Diseases, Cleveland Clinic Children’s Hospital, Cleveland, Ohio (Foster); Division of Pediatric Infectious Disease, Dell Medical School, University of Texas at Austin (Hauger); Division of Critical Care, Ann & Robert H. Lurie Children’s Hospital of Chicago, Illinois (Hong); Department of Pediatrics, University of Tennessee Health Science Center, Le Bonheur Children’s Hospital, Memphis (Hysmith); Division of Pediatric Critical Care, Doernbecher Children’s Hospital, Oregon Health & Science University, Portland (Kirby); Department of Pediatrics, Northwestern University Feinberg School of Medicine, Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, Illinois (Kociolek, Smith); Division of Critical Care Medicine, Department of Pediatrics, Intermountain Primary Children’s Hospital, University of Utah, Salt Lake City (Larsen); Washington University School of Medicine, St Louis, Missouri (Lin, Newland); Emory School of Medicine, Children’s Healthcare of Atlanta, Atlanta, Georgia (Linam); Doernbecher Children’s Hospital, Oregon Health & Science University, Portland (Nolt); Division of Critical Care Medicine, Department of Anesthesiology, Critical Care and Pain Medicine, Boston Children’s Hospital, Boston, Massachusetts (Priebe); Division of Infectious Diseases, Department of Pediatrics, Boston Children’s Hospital, Boston, Massachusetts (Priebe, Sandora); Department of Anesthesia, Harvard Medical School, Boston, Massachusetts (Priebe); Stanford University School of Medicine, Lucile Packard Children’s Hospital Stanford, Stanford, California (Schwenk); Department of Neurology, Northwestern University Feinberg School of Medicine, Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, Illinois (Smith); Stanford University Department of Pediatrics, Division of Pediatric Critical Care Medicine, Stanford, California (Steffen); Division of Pediatric Critical Care and Cardiology, Department of Pediatrics, Le Bonheur Children’s Hospital, University of Tennessee Health Science Center, Memphis (Tadphale); Department of Infectious Diseases, St Jude Children’s Research Hospital, Memphis, Tennessee (Wolf); Department of Pediatrics, University of Tennessee Health Science Center, Memphis (Wolf); Seattle Children’s Hospital, Department of Pediatrics, University of Washington, Seattle (Zerr).

Author Contributions: Drs Woods-Hill and Milstone 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: Woods-Hill, Colantuoni, Koontz, Voskertchian, Xie, Miller, Fackler, Milstone, Coffin, Hauger, Smith.

Acquisition, analysis, or interpretation of data: Woods-Hill, Colantuoni, Koontz, Voskertchian, Xie, Thurm, Miller, Milstone, Agulnik, Albert, Auth, Bradley, Clayton, Coffin, Dallefeld, Ezetendu, Fainberg, Flaherty, Foster, Hauger, Hong, Hysmith, Kirby, Kociolek, Larsen, Lin, Linam, Newland, Nolt, Priebe, Sandora, Schwenk, Steffen, Tadphale, Toltzis, Wolf, Zerr.

Drafting of the manuscript: Woods-Hill, Colantuoni, Koontz, Xie, Milstone, Hauger, Schwenk, Wolf.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Woods-Hill, Colantuoni, Thurm, Miller.

Obtained funding: Woods-Hill, Voskertchian, Milstone.

Administrative, technical, or material support: Koontz, Voskertchian, Miller, Clayton, Coffin, Ezetendu, Fainberg, Flaherty, Hong, Kirby, Kociolek, Lin, Nolt, Schwenk, Steffen, Tadphale, Toltzis, Zerr.

Supervision: Miller, Fackler, Milstone, Ezetendu, Foster, Newland, Wolf.

Conflict of Interest Disclosures: Dr Woods-Hill reported grants from Agency for Healthcare Research and Quality (AHRQ) (the grant that funded the Bright STAR project) and grants from National Institutes of Health (NIH) (a training grant and then a career development award) during the conduct of the study. Drs Colantuoni, Koontz, Voskertchian, Xie, and Fackler reported grants from AHRQ during the conduct of the study. Dr Coffin reported grants from Merck (investigator-initiated study) and personal fees from Merck (member of data and safety monitoring board) outside the submitted work. Dr Kociolek reported grants from Merck outside the submitted work. Dr Priebe reported grants from AHRQ during the conduct of the study and grants from US Department of Defense W81XWH-19-1-0208 on Pseudomonas vaccines outside the submitted work. Dr Tadphale reported grants from Johns Hopkins University during the conduct of the study. Dr Wolf reported nonfinancial support from Karius Inc and in-kind support for investigator-initiated research outside the submitted work. Dr Zerr reported grants from a subcontract from the primary site during the conduct of the study. No other disclosures were reported.

Funding/Support: The study was funded by the Agency for Healthcare Research and Quality (1R18HS025642). Dr Woods-Hill also receives support from the National Heart, Lung, and Blood Institute of the NIH (K23HL151381). Dr Milstone also receives support from the National Institute of Allergy and Infectious Diseases of the NIH (K24AI141580).

Role of the Funder/Sponsor: The Agency for Healthcare Research and Quality (through grant 1R18HS025642) directly supported 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. In addition, the National Institutes of Health supported the time spent by Drs Woods-Hill and Milstone on this study’s design, data collection, data analysis, and manuscript preparation through their K23 and K24 awards.

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Additional Contributions: The Bright STAR collaborative study team thanks the following people for their contributions to this project: Adam Hersh, MD, PhD, Primary Children’s Hospital, Intermountain Healthcare, University of Utah; Jennifer Blumenthal, MD, Renee Lehane, and Dimple Mirchandani, Boston Children’s Hospital; Morgan Benedict, BSN, CCRN, Rainbow Babies and Children’s Hospital; Sheneka Bell and Jose Ferrolino, St Jude Children’s Research Hospital; Amanda Adler, Seattle Children’s Hospital; and Ximin Li; Lena Warrak, BS, BA; Yacoba Zwennes, BA; and Anne King, RN, Johns Hopkins University. They did not receive compensation for their contributions outside of their salaries, if applicable.

Additional Information: The coordinating center for the study was Johns Hopkins Children’s Center in Baltimore, Maryland.

^✉

Corresponding author.

PMCID: PMC9062771 PMID: 35499841

This quality improvement study examines data for blood culture rates, antibiotic use, and patient outcomes from 14 pediatric intensive care units that implemented a program intended to reduce overuse of blood cultures.

Key Points

Question

What is the association of diagnostic stewardship for blood cultures with culture rates, antibiotic use, and patient outcomes in the pediatric intensive care unit?

Findings

In this quality improvement study, a collaborative of 14 pediatric intensive care units found significant decreases in blood culture utilization and broad-spectrum antibiotic use. Balancing measures (mortality, readmission, length of stay, sepsis, and severe sepsis/septic shock) remained stable.

Meaning

Multidisciplinary diagnostic stewardship to optimize blood culture practices in patients without suspected sepsis can safely decrease blood cultures and antibiotic use in critically ill children.

Abstract

Importance

Blood culture overuse in the pediatric intensive care unit (PICU) can lead to unnecessary antibiotic use and contribute to antibiotic resistance. Optimizing blood culture practices through diagnostic stewardship may reduce unnecessary blood cultures and antibiotics.

Objective

To evaluate the association of a 14-site multidisciplinary PICU blood culture collaborative with culture rates, antibiotic use, and patient outcomes.

Design, Setting, and Participants

This prospective quality improvement (QI) collaborative involved 14 PICUs across the United States from 2017 to 2020 for the Bright STAR (Testing Stewardship for Antibiotic Reduction) collaborative. Data were collected from each participating PICU and from the Children’s Hospital Association Pediatric Health Information System for prespecified primary and secondary outcomes.

Exposures

A local QI program focusing on blood culture practices in the PICU (facilitated by a larger QI collaborative).

Main Outcomes and Measures

The primary outcome was blood culture rates (per 1000 patient-days/mo). Secondary outcomes included broad-spectrum antibiotic use (total days of therapy and new initiations of broad-spectrum antibiotics ≥3 days after PICU admission) and PICU rates of central line–associated bloodstream infection (CLABSI), Clostridioides difficile infection, mortality, readmission, length of stay, sepsis, and severe sepsis/septic shock.

Results

Across the 14 PICUs, the blood culture rate was 149.4 per 1000 patient-days/mo preimplementation and 100.5 per 1000 patient-days/mo postimplementation, for a 33% relative reduction (95% CI, 26%-39%). Comparing the periods before and after implementation, the rate of broad-spectrum antibiotic use decreased from 506 days to 440 days per 1000 patient-days/mo, respectively, a 13% relative reduction (95% CI, 7%-19%). The broad-spectrum antibiotic initiation rate decreased from 58.1 to 53.6 initiations/1000 patient-days/mo, an 8% relative reduction (95% CI, 4%-11%). Rates of CLABSI decreased from 1.8 to 1.1 per 1000 central venous line days/mo, a 36% relative reduction (95% CI, 20%-49%). Mortality, length of stay, readmission, sepsis, and severe sepsis/septic shock were similar before and after implementation.

Conclusions and Relevance

Multidisciplinary diagnostic stewardship interventions can reduce blood culture and antibiotic use in the PICU. Future work will determine optimal strategies for wider-scale dissemination of diagnostic stewardship in this setting while monitoring patient safety and balancing measures.

Introduction

Up to one-third of US health care spending represents medical overuse: care in which net benefits do not exceed the net harms and which is associated with excess cost, worse patient outcomes, and death.^1,2,3,4,5,6 While deadoption of unnecessary clinical practices is now prioritized by professional societies, governments, and funding agencies, unnecessary care remains prevalent.^1,7,8 Diagnostic stewardship, optimizing the use of diagnostic tests to improve treatment decisions, is emerging as a potentially high-impact strategy to combat overuse.¹ Successful diagnostic stewardship programs have reduced urine cultures, respiratory tract cultures, and Clostridioides difficile testing.^{9,10,11,12,13} Bacterial cultures are often coupled with starting antibiotics, so focused stewardship of microbiology testing may be an important strategy to reduce antibiotic overuse and resistance.

Ordering blood cultures to evaluate for bloodstream infection in critically ill children includes complex decision-making, recognizing that severe sepsis in hospitalized children has an 8% prevalence and a 25% mortality rate.¹⁴ Early recognition and appropriate antibiotics improve sepsis outcomes, placing urgent attention on rapid recognition, diagnosis, and empirical treatment of possible sepsis.^15,16 However, among pediatric intensive care unit (PICU) patients, many clinical symptoms like fever, leukocytosis, and hypotension are neither sensitive nor specific for infection.¹⁷ No single test, biomarker, or decision rule can perfectly discriminate the presence of serious bacterial infection, so the diagnosis of sepsis relies on clinical judgment and decision-making.^18,19,20 Blood cultures are the gold-standard diagnostic test for sepsis due to bloodstream infection in children but can be collected excessively in PICU patients with nonspecific symptoms.^21,22,23,24 Clinicians typically initiate empirical broad-spectrum antibiotics when ordering blood cultures. More than half of PICU patients receive antibiotics, often for nonspecific symptoms such as fever.²⁵ Preliminary work found that reducing unnecessary blood cultures in PICU patients, when clinical data do not suggest sepsis, may be feasible and safe.^26,27,28

We created a national quality improvement collaborative called Bright STAR (Testing Stewardship for Antibiotic Reduction), intending to reduce overuse of blood cultures in patients with little suggestion of sepsis and to measure the association of this diagnostic stewardship collaborative with blood culture rates, broad-spectrum antibiotic use, and other patient outcomes.

Methods

We conducted multicenter analyses of prespecified outcomes from 14 PICUs using data from 24 months before and 18 months after each site’s implementation. No individual patients were enrolled. The institutional review board at Johns Hopkins University acknowledged the Bright STAR collaborative as quality improvement and the collection and analysis of summary-level deidentified data as nonhuman subjects research, with a waiver for informed consent (eTable 1 in the Supplement). Data are reported according to Standards for Quality Improvement Reporting Excellence (SQUIRE) guidelines.²⁹

Bright STAR Quality Improvement Collaborative

The Bright STAR steering committee comprises experts in pediatric infectious diseases, critical care, quality improvement (QI), human factors engineering, and statistics based at Johns Hopkins University. In brief, 14 sites from across the United States joined the Bright STAR collaborative in 2017. Participating sites were diverse with respect to institution size, patient population, and geographic area (Table 1 and eTable 1 in the Supplement). Guided by the Bright STAR committee, each site created a core project team of pediatric critical care and pediatric infectious disease physicians, with additional team members (QI experts, nurses, data analysts) at each site’s discretion. With a staggered time line reflecting individual site readiness, sites participated in preimplementation assessment of current blood culture practices, partnered with key stakeholders within their institution, developed a clinical decision support tool, developed an implementation plan, and then launched their local QI program (Figure 1).²⁶ The coordinating team held regular calls with individual sites and the larger collaborative throughout all steps of the project.

Table 1. Baseline Characteristics of 14 Enrolled PICUs.

Metric	Median (IQR) or No. (%)
Site-level metrics
Average No. of admissions per month^a	152 (128-177)
No. of beds	37 (23-43)
Proportion of admissions with oncologic diagnoses, %^b	7 (5-13)
Performs stem cell transplants	11 (79)
Performs solid organ transplants	11 (79)
Cardiac surgical patients within PICU	6 (43)
ECMO	13 (93)
Pediatric critical care fellowship program	13 (93)
Advanced practice nurses	14 (100)
Bedside nurses perform peripheral venipuncture	11 (79)
Phlebotomy team performs peripheral venipuncture	12 (86)
Participates in IPSO sepsis collaborative	6 (43)
Freestanding children’s hospital	13 (93)
Academic/university-affiliated hospital	14 (100)
PICU location, US geographic region
Northeast	2 (14)
Southeast	3 (21)
Southwest	1 (7)
Midwest	4 (29)
West	4 (29)
Clinical metrics, monthly rate^c
Blood cultures per 1000 patient-days/mo	145.6 (101.1-187.4)
Total broad-spectrum antibiotic days of therapy on PICU day ≥3 per 1000 patient-days/mo^d	487.2 (406.6-630.2)
Sepsis per 100 admissions/mo^d	6.4 (5.2-8.0)
Severe sepsis/septic shock per 100 admissions/mo^d	4.7 (3.8-5.3)
PICU readmission per 100 admissions/mo^d	3.4 (2.4-4.5)
Length of stay per 100 admissions/mo, d^d^,^e	4.6 (4.0-5.1)
Deaths per 100 admissions/mo^d	1.8 (1.5-2.0)

Open in a new tab

Abbreviations: ECMO, extracorporeal membrane oxygenation; IPSO, Improving Pediatric Sepsis Outcomes, Children’s Hospital Association; PICU, pediatric intensive care unit.

^{^a}

For each site, the average of the monthly admissions was computed for the 18-month period before implementation.

^{^b}

Available for 13 of 14 sites.

^{^c}

The monthly rate was computed by a weighted mean of the monthly rate (eg, number of blood cultures divided by patient-days), weighted by the proportion of patient-days represented by each month.

^{^d}

Metrics available for 11 of 14 sites enrolled in the Pediatric Health Information System. Antibiotic use is defined in Table 2 and eTable 2 in the Supplement.

^{^e}

Number of days in the PICU per number of new PICU admissions per month.

Figure 1. — JHCC indicates Johns Hopkins Children’s Center.

Intervention: Site-Specific Blood Culture Practice Change Driven by Local QI Teams

Based on results of the preimplementation assessment survey, each site created unique clinical decision support tools to guide blood culture practice.²⁶ While the exact composition of the clinical decision support tools varied, each tool sought to (1) standardize practice and reduce variability in the decision to order blood cultures, the source of a blood culture, or frequency of repeat blood cultures; and (2) highlight patient safety considerations in the decision to order a blood culture. Each PICU had discretion to determine if certain patient populations, such as immunocompromised patients, were excluded; however, each clinical decision support tool stated that patients with suspicion for sepsis should receive usual care, including blood cultures if appropriate. With the support of the Bright STAR team, each PICU devised their own multifaceted implementation plan, including such strategies as educational sessions, electronic clinical pathways, audit and feedback of culture rates to PICU clinicians, and targeted emails to PICU clinicians. Blood culture rates were tracked and reported monthly. Sites audited blood cultures postimplementation for compliance with the new recommended practices and to monitor patient safety (described below).

Outcome Measures and Data Sources

The primary outcome was monthly blood culture rate. Secondary outcomes are defined in eTable 2 in the Supplement and included (1) clinical metrics and (2) balancing measures. The clinical metrics were rates of central line–associated bloodstream infections (CLABSI), hospital-onset C difficile infection (CDI), broad-spectrum antibiotic use (Table 2), and new initiation of broad-spectrum antibiotics. The balancing measures were audited episodes of positive blood cultures; rates of mortality, sepsis, and severe sepsis/septic shock; as well as mean PICU length of stay and PICU and hospital readmissions. Rates of antibiotic use and new antibiotic initiations only included broad-spectrum antibiotics administered 3 or more days after PICU admission. All 14 sites reported data on blood cultures, CLABSI, and C difficile infection. For the remaining secondary outcomes, data were provided for 11 sites from the Children’s Hospital Association Pediatric Health Information System (PHIS). The 3 non-PHIS sites were not able to provide all secondary outcomes.

Table 2. Primary and Secondary Outcomes Before and After Implementation of the Bright STAR Collaborative in 14 PICUs.

Outcome	Mean monthly rate (95% CI)^a		Postimplementation vs preimplementation		P value^a
Outcome	Preimplementation	Postimplementation	Relative rate (95% CI)^a	Absolute rate difference (95% CI)^a	P value^a
Primary outcome
Blood cultures^b	149.37 (119.33 to 186.97)	100.50 (78.00 to 129.51)	0.67 (0.61 to 0.74)	−48.86 (−62.76 to −34.97)	<.001
Secondary outcomes: clinical metrics
Central line–associated bloodstream infection^c	1.79 (1.35 to 2.38)	1.14 (0.76 to 1.70)	0.64 (0.51 to 0.80)	−0.65 (−0.94 to −0.36)	<.001
Clostridioides difficile infection	0.38 (0.27 to 0.55)	0.36 (0.22 to 0.61)	0.94 (0.59 to 1.49)	−0.02 (−0.19 to 0.15)	.80
Broad-spectrum antibiotic use^d^,^e	505.97 (446.94 to 572.80)	440.35 (386.65 to 501.51)	0.87 (0.81 to 0.93)	−65.62 (−97.23 to −34.01)	<.001
New initiation of broad-spectrum antibiotics^f^,^e	58.14 (53.49 to 63.20)	53.59 (49.32 to 58.24)	0.92 (0.89 to 0.96)	−4.55 (−6.62 to −2.48)	<.001
Secondary outcomes: balancing measures
Mortality^g^,^h	1.79 (1.56 to 2.06)	1.88 (1.58 to 2.24)	1.05 (0.97 to 1.14)	0.09 (−0.07 to 0.25)	.25
PICU length of stay, d^g^,ⁱ	4.37 (3.90 to 4.90)	4.46 (3.97 to 5.00)	1.02 (0.99 to 1.04)	0.09 (−0.01 to 0.19)	.07
PICU readmission^g^,^h	3.09 (2.31 to 4.13)	3.33 (2.50 to 4.44)	1.08 (0.99 to 1.17)	0.25 (−0.02 to 0.52)	.07
Hospital readmission^g^,^h	2.12 (1.68 to 2.67)	2.06 (1.61 to 2.64)	0.97 (0.89 to 1.07)	−0.06 (−0.25 to 0.14)	.56
Sepsis^g^,^h	6.64 (5.57 to 7.91)	7.07 (5.48 to 9.12)	1.06 (0.89 to 1.28)	0.43 (−0.87 to 1.73)	.50
Severe sepsis/septic shock^g^,^h	4.79 (3.96 to 5.79)	4.99 (4.08 to 6.11)	1.04 (0.86 to 1.27)	0.20 (−0.75 to 1.16)	.67

Open in a new tab

Abbreviation: PICU, pediatric intensive care unit.

^{^a}

Results derived from a generalized linear mixed model assuming the Poisson distribution for the log rate as a function of an intercept plus a main term for postimplementation vs preimplementation. The model included a random intercept for PICU, and CIs are derived using robust variance estimates. The rate preimplementation and postimplementation and the absolute rate difference are derived from appropriate linear or nonlinear functions of the model intercept and coefficient for the main term for postimplementation vs preimplementation.

^{^b}

Rate per 1000 patient-days.

^{^c}

Rate per 1000 central venous line days.

^{^d}

Days of broad-spectrum antibiotics for PICU days ≥3.

^{^e}

Antibiotics included aztreonam, cefepime, cefotaxime, ceftazidime, ceftazidime-avibactam, ceftriaxone, ciprofloxacin, imipenem, meropenem, piperacillin-tazobactam, tigecycline, and vancomycin (parenteral therapy only).

^{^f}

Days of new initiations of broad-spectrum antibiotics for PICU days ≥3.

^{^g}

Data from 11 of 14 sites that are Children’s Hospital Association Pediatric Health Information System participating hospitals.

^{^h}

Rate per 100 PICU admissions.

^ⁱ

Number of days in the PICU per number of new PICU admissions per month.

Balancing Measures and Event Monitoring

The primary balancing measure was delay in diagnosing bloodstream infection in PICU patients. Sites reviewed each positive blood culture episode in their PICU after program implementation using a standardized form to identify any delay in the collection of the blood culture and to determine whether the delay was related or unrelated to program guidance for blood culture collection (eFigure 1 in the Supplement). Positive blood culture episodes were discussed during postimplementation all-site calls and site-specific calls.

Statistical Analysis

In a series of preplanned analyses, we compared the primary and secondary outcomes before and after implementation. All outcome data were explored visually by plotting the monthly rates vs month of implementation for each site. For the primary analysis, we fit a generalized linear mixed model assuming a Poisson distribution for the monthly number of blood cultures with the monthly number of patient-days as an offset and the main effect of project implementation. The effect of the Bright STAR project is expressed as relative change in blood culture rate comparing postimplementation vs preimplementation. To account for the correlation of blood cultures over time within each PICU, the model included a random intercept for PICU with SE derived from a robust variance estimate. In a secondary analysis, a quasi-experimental interrupted time series analysis was implemented to estimate (1) relative change in blood culture rates per quarter, ie, 3 months preimplementation; (2) main effect of the Bright STAR project, reported as relative change in blood culture rate comparing the first month of the project with the month preceding project implementation; and (3) subsequent effect of the project, reported as relative change in blood culture rate per quarter postimplementation. The primary and secondary analyses described above were applied to each secondary outcome.

We performed sensitivity analyses for the primary outcome to account for (1) participation in the Improving Pediatric Sepsis Outcomes project, (2) a 3-month preimplementation “wash-in” period in which blood culture practices may have changed because of program awareness prior to official launch, and (3) seasonal variation in case mix (eTable 3 in the Supplement). Further, we conducted post hoc sensitivity analyses for all outcomes to account for the impact of COVID-19 by excluding data collected after February 2020 (eTable 3 in the Supplement). Next, we used funnel plots to visualize site-specific outcomes and conducted post hoc exclusions of a limited number of sites with potentially high influence in our primary analyses (eFigure 2A-2C in the Supplement). Further, we conducted a post hoc analysis to compare antibiotic use in Bright STAR and non–Bright STAR (contemporaneous control) PHIS sites (eResults in the Supplement). Analyses were conducted using Stata (version 15) and R (version 4.1.1).^30,31

Results

Primary Outcome

Across the 14 sites, there were 37 527 total blood cultures in the preimplementation period and 20 340 total cultures in the postimplementation period. Across sites, the median preimplementation blood culture rate was 146 per 1000 patient-days/mo (IQR, 101-187), and the postimplementation rate was 99 per 1000 patient-days/mo (IQR, 70-120). Overall, 13 of 14 sites reduced their blood culture rates in the postimplementation period (eFigure 2A in the Supplement). Across the 14 sites, there were a mean 149.4 blood cultures per 1000 patient-days/mo before implementation and 100.5 blood cultures per 1000 patient-days/mo after implementation, corresponding to a 33% relative reduction in the rate (95% CI, 26% to 39%) and an absolute reduction of 48.9 blood cultures per 1000 patient-days/mo (95% CI, 35.0 to 62.8) (Table 2 and Figure 2). The primary outcome results were qualitatively unchanged in the sensitivity analyses (eTable 3 in the Supplement). The secondary interrupted time series analysis estimated a 1% relative reduction in blood cultures per 1000 patient-days per quarter (3 months) (95% CI, −4% to 2%) before implementation, a 29% relative reduction at the time of implementation (95% CI, 19% to 38%) and subsequently no additional temporal change after implementation (relative reduction in rate per quarter, 0; 95% CI, −3% to 3%) (eTable 4 in the Supplement). Further, there was no evidence that the temporal trends differed before or after implementation (P = .50).

Figure 2. — Monthly blood cultures per 1000 patient-days for each of the 14 Bright STAR sites before and after implementation (24 and 18 months, respectively). The mean monthly average rate over time was estimated using a smoothing spline with 4 degrees of freedom.

Secondary Outcomes

After program implementation, there was a reduction in the rate of CLABSI across the 14 sites by 36% (95% CI, 20%-49%), from 1.79 to 1.14 per 1000 central-line days/mo preimplementation and postimplementation, respectively. There was no change in C difficile infection rates comparing the 2 periods (0.38 vs 0.36 infections per 1000 patient-days/mo; P = .80) (Table 2).

In the 11 PHIS sites, there were 506.0 vs 440.3 total days of broad-spectrum antibiotic use per 1000 patient-days/mo in the preimplementation vs postimplementation period, respectively, representing a 13% relative reduction (95% CI, 7%-19%) and an absolute reduction of 65.6 days (95% CI, 34.0-97.2) of antibiotic therapy per 1000 patient-days/mo. For new initiation of broad-spectrum antibiotics on PICU day 3 or later, there were 58.1 vs 53.6 initiations per 1000 patient-days/mo preimplementation vs postimplementation, respectively, an 8% relative reduction (95% CI, 4%-11%) and an absolute reduction of 4.5 initiations (95% CI, 2.5-6.6) per 1000 patient-days/mo (Figure 3 and Table 2). The PICU mortality rates, length of stay, and readmission as well as hospital readmissions were similar before and after implementation (Table 2). Similar to the secondary interrupted time series analysis of the primary outcome, after allowing for a change in the rate of secondary outcomes at the start of implementation, there were no significant temporal changes in the secondary outcomes before or after implementation. The secondary clinical metric results were unchanged in the post hoc sensitivity analyses (eTable 3 in the Supplement) and supported the earlier finding of no significant change in the secondary balancing measures (eFigure 2A-2C in the Supplement).

Figure 3. — Rates of antibiotic use and new antibiotic initiation 3 or more days after pediatric intensive care unit admission, before and after implementation (24 and 18 months, respectively). The mean monthly average rate over time was estimated using a smoothing spline with 4 degrees of freedom.

Site leads examined 793 episodes of positive blood cultures postimplementation using the review form, presented results on Bright STAR collaborative calls, and determined that 792 episodes (99.9%) were collected at the appropriate time and without any delay that may have resulted from recommendations to avoid blood cultures in certain settings. In 1 case, a child who was medically complex developed tachycardia that was attributed to noninfectious reasons, but 24 hours later, the child developed fever and was found to have blood culture positive for Staphylococcus aureus. One of the 2 site leads concluded that there may have been a delay in obtaining the blood culture. The child was treated and discharged home.

Post Hoc Contemporaneous Control Analysis

Compared with matched contemporaneous non–Bright STAR PHIS sites, Bright STAR PHIS sites had similar reductions in rate of broad-spectrum antibiotic use, but greater reductions in rate of new initiations of broad-spectrum antibiotics after program implementation (broad-spectrum antibiotic use: 15% vs 8% relative reduction, respectively; P = .17; new initiations of broad-spectrum antibiotics: 8% relative reduction vs 9% relative increase, respectively; P < .001) (eTable 5 in the Supplement).

Discussion

Diagnostic stewardship is an emerging strategy to reduce testing overuse in hospitals. Given the strong link between testing and treatment, early studies have found that reducing microbiology testing may lead to decreases in antibiotic use.³² To our knowledge, Bright STAR is the first multicenter collaborative to apply diagnostic stewardship to reduce unnecessary blood culture use. Over 18 months, 14 participating PICUs reduced blood culture rates by 33%. In addition, participating PICUs realized a reduction in antibiotic use, including new initiation of broad-spectrum antibiotics. Prespecified balancing measures were similar in the periods before and after implementation. Of 793 bloodstream infections reviewed, in all but 1 episode blood cultures were collected at the appropriate time without delay. These findings suggest that multidisciplinary efforts to standardize blood culture collection and avoid unnecessary testing in the PICU can be done successfully and safely in diverse settings and that reducing blood culture use is associated with a reduction in broad-spectrum antibiotic use.

Diagnostic stewardship offers important benefits distinct from antimicrobial stewardship. Most antimicrobial stewardship efforts include prior authorization to start broad-spectrum antibiotics or after prescription review to reduce antibiotic duration of therapy.^33,34 Diagnostic stewardship focuses “upstream” on the decision to send a diagnostic test, when clinicians are still considering if infection is present. Because clinicians often start antibiotics at the time bacterial cultures are ordered, diagnostic stewardship is a complementary strategy to further combat antibiotic overuse. This strategy may be especially useful in environments like the PICU where children often have nonspecific signs and symptoms of infection, such as fever. Second, more than half of PICU patients receive antibiotics, often empirically without definitive evidence of a bacterial infection.²⁵ This population is a high-risk group for antibiotic overuse and emergence of antibiotic resistance. Our study focused on new initiations of broad-spectrum antibiotics in children admitted to the PICU for more than 3 days, recognizing that most blood cultures and empirical antibiotics started within 48 hours of admission are indicated for suspected sepsis or infection. The intended target of the program was relatively stable patients with fever but without additional signs or symptoms of sepsis, for example, avoiding cultures when another fever source was more likely, avoiding cultures from multiple catheter lumens, or avoiding surveillance cultures in asymptomatic patients. Bright STAR sites successfully implemented clinical decision support tools to help clinicians identify patients at low risk of sepsis to reduce blood culture and antibiotic use and sustained these improvement programs for 18 months. Data from contemporaneous controls show no decrease in antibiotic initiations, supporting the hypothesis that reducing blood cultures in patients where data did not suggest sepsis may reduce concurrent initiation of broad-spectrum antibiotics.

Perhaps no question is more important than whether a program to reduce blood culture use in critically ill children was safe or led to delayed recognition and treatment of bacterial sepsis. Importantly, early recognition and treatment of sepsis is complementary, not in conflict, with Bright STAR’s approach. Both Bright STAR as an overall program, and each sites’ clinical tools, intentionally make no recommendations to avoid blood cultures in patients with signs or symptoms of suspected sepsis. Rather, Bright STAR’s focus is on clinically stable patients in whom bloodstream infection may be, after thoughtful consideration and patient evaluation, considered unlikely and who can be monitored without a blood culture. When monitoring balancing measures, preimplementation and postimplementation rates of mortality, PICU readmission, PICU length of stay, sepsis, and severe sepsis/septic shock were similar. These measures are frequently confounded by changes in (1) patient populations, (2) case-mix index, and (3) coding. We therefore paired this high-level evaluation with detailed review of nearly 800 distinct positive blood culture episodes. In only 1 episode did a site lead suggest a possible delay in obtaining a blood culture due to program guidance. These results suggest that as units worked to change blood culture practices, clinicians remained vigilant to appropriately identify and test patients for potential bloodstream infections; however, ongoing monitoring for harm is needed to ensure this program continues to safely do less.

Limitations

There are important limitations to consider. First, the review of blood cultures was limited to episodes of positive cultures, and this approach may have missed instances in which antibiotics were initiated without collecting blood cultures, an unintended consequence not seen in a prior single-center study.³⁵ Second, enrolled sites demonstrated diversity in geographic region, size, and patient population, but all were experienced in quality improvement and were resourced to participate in the project, limiting generalizability. Third, the analyses do not account for variation in site implementation. Fourth, only the 11 PHIS sites contributed data to some secondary outcomes. Prior studies have validated PHIS antibiotic data, but fewer data are available on validity of mortality and other secondary balancing measures.^36,37,38 Fifth, blood culture data from PHIS have not been validated and were not available for contemporaneous controls. However, the interrupted time series analysis for the primary outcome did not show evidence of different temporal trends before and after implementation, supporting our finding of significantly reduced blood culture rates in our sites. Furthermore, pre-post studies are subject to confounding by temporal trends and unmeasured confounding. We were able to obtain contemporaneous control data for antibiotic use, which supported our finding of a significant decrease in new initiations of broad-spectrum antibiotics in the Bright STAR vs non–Bright STAR PHIS sites.

Conclusions

Bright STAR is the first multicenter diagnostic stewardship collaborative to our knowledge to successfully reduce blood cultures and antibiotic use. Our study confirms that diagnostic stewardship is a promising strategy to augment antimicrobial stewardship programs and reduce antibiotic overuse. Implementation was facilitated using components of human factors engineering and implementation science and emphasized balancing local stakeholder priorities with common general principles. Future studies should monitor for effectiveness and unintended consequences as diagnostic stewardship interventions are disseminated to diverse settings and populations.

Supplement.

eTable 1. Sites participating in the Bright STAR Collaborative

eTable 2. Definitions of outcome variables and units of measurement

eFigure 1. Bright STAR review form for site team review of positive blood cultures after Bright STAR implementation

eTable 3. Sensitivity analyses of Bright STAR outcomes

eTable 4. Results of interrupted time series analysis of primary and secondary outcomes in Bright STAR sites

eFigure 2. Bright STAR site-specific and pooled relative rates comparing the post- and pre-implementation periods for primary and secondary outcomes

eResults. Post-hoc contemporaneous control analysis regarding broad-spectrum antibiotic use

eTable 5. Results of post-hoc contemporaneous control analysis

Click here for additional data file.^{(326.1KB, pdf)}

References

1.Morgan DJ, Brownlee S, Leppin AL, et al. Setting a research agenda for medical overuse. BMJ. 2015;351:h4534. doi: 10.1136/bmj.h4534 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Chassin MR, Galvin RW; Institute of Medicine National Roundtable on Health Care Quality . The urgent need to improve health care quality. JAMA. 1998;280(11):1000-1005. doi: 10.1001/jama.280.11.1000 [DOI] [PubMed] [Google Scholar]
3.Berwick DM, Hackbarth AD. Eliminating waste in US health care. JAMA. 2012;307(14):1513-1516. doi: 10.1001/jama.2012.362 [DOI] [PubMed] [Google Scholar]
4.Esserman LJ, Thompson IM Jr, Reid B. Overdiagnosis and overtreatment in cancer: an opportunity for improvement. JAMA. 2013;310(8):797-798. doi: 10.1001/jama.2013.108415 [DOI] [PubMed] [Google Scholar]
5.Fisher ES, Wennberg DE, Stukel TA, Gottlieb DJ, Lucas FL, Pinder EL. The implications of regional variations in Medicare spending, part 2: health outcomes and satisfaction with care. Ann Intern Med. 2003;138(4):288-298. doi: 10.7326/0003-4819-138-4-200302180-00007 [DOI] [PubMed] [Google Scholar]
6.Brownlee S, Chalkidou K, Doust J, et al. Evidence for overuse of medical services around the world. Lancet. 2017;390(10090):156-168. doi: 10.1016/S0140-6736(16)32585-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Levinson W, Born K, Wolfson D. Choosing Wisely campaigns: a work in progress. JAMA. 2018;319(19):1975-1976. doi: 10.1001/jama.2018.2202 [DOI] [PubMed] [Google Scholar]
8.National Institutes of Health . PAR-19-274: dissemination and implementation research in health. Accessed July 13, 2021. https://grants.nih.gov/grants/guide/pa-files/PAR-18-007.html
9.Sperling K, Priddy A, Suntharam N, Feuerhake T. Optimizing testing for Clostridium difficile infection: a quality improvement project. Am J Infect Control. 2019;47(3):340-342. doi: 10.1016/j.ajic.2018.08.027 [DOI] [PubMed] [Google Scholar]
10.Truong CY, Gombar S, Wilson R, et al. Real-time electronic tracking of diarrheal episodes and laxative therapy enables verification of Clostridium difficile clinical testing criteria and reduction of Clostridium difficile infection rates. J Clin Microbiol. 2017;55(5):1276-1284. doi: 10.1128/JCM.02319-16 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Sick-Samuels A, 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]
12.Epstein L, Edwards JR, Halpin AL, et al. Evaluation of a novel intervention to reduce unnecessary urine cultures in intensive care units at a tertiary care hospital in Maryland, 2011-2014. Infect Control Hosp Epidemiol. 2016;37(5):606-609. doi: 10.1017/ice.2016.9 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Mullin KM, Kovacs CS, Fatica C, et al. A multifaceted approach to reduction of catheter-associated urinary tract infections in the intensive care unit with an emphasis on “stewardship of culturing”. Infect Control Hosp Epidemiol. 2017;38(2):186-188. doi: 10.1017/ice.2016.266 [DOI] [PubMed] [Google Scholar]
14.Weiss SL, Fitzgerald JC, Pappachan J, et al. ; Sepsis Prevalence, Outcomes, and Therapies (SPROUT) Study Investigators and Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network . Global epidemiology of pediatric severe sepsis: the Sepsis Prevalence, Outcomes, and Therapies study. Am J Respir Crit Care Med. 2015;191(10):1147-1157. doi: 10.1164/rccm.201412-2323OC [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Weiss SL, Fitzgerald JC, Balamuth F, et al. Delayed antimicrobial therapy increases mortality and organ dysfunction duration in pediatric sepsis. Crit Care Med. 2014;42(11):2409-2417. doi: 10.1097/CCM.0000000000000509 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Singer M, Deutschman CS, Seymour CW, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315(8):801-810. doi: 10.1001/jama.2016.0287 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Hsiao AL, Baker MD. Fever in the new millennium: a review of recent studies of markers of serious bacterial infection in febrile children. Curr Opin Pediatr. 2005;17(1):56-61. doi: 10.1097/01.mop.0000151781.13635.70 [DOI] [PubMed] [Google Scholar]
18.Milcent K, Faesch S, Gras-Le Guen C, et al. Use of procalcitonin assays to predict serious bacterial infection in young febrile infants. JAMA Pediatr. 2016;170(1):62-69. doi: 10.1001/jamapediatrics.2015.3210 [DOI] [PubMed] [Google Scholar]
19.Nijman RG, Moll HA, Vergouwe Y, de Rijke YB, Oostenbrink R. C-reactive protein bedside testing in febrile children lowers length of stay at the emergency department. Pediatr Emerg Care. 2015;31(9):633-639. doi: 10.1097/PEC.0000000000000466 [DOI] [PubMed] [Google Scholar]
20.Lautz AJ, Dziorny AC, Denson AR, et al. Value of procalcitonin measurement for early evidence of severe bacterial infections in the pediatric intensive care unit. J Pediatr. 2016;179:74-81.e2. doi: 10.1016/j.jpeds.2016.07.045 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Kiragu AW, Zier J, Cornfield DN. Utility of blood cultures in postoperative pediatric intensive care unit patients. Pediatr Crit Care Med. 2009;10(3):364-368. doi: 10.1097/PCC.0b013e3181a31bb7 [DOI] [PubMed] [Google Scholar]
22.Darby JM, Linden P, Pasculle W, Saul M. Utilization and diagnostic yield of blood cultures in a surgical intensive care unit. Crit Care Med. 1997;25(6):989-994. doi: 10.1097/00003246-199706000-00016 [DOI] [PubMed] [Google Scholar]
23.Bates DW, Goldman L, Lee TH. Contaminant blood cultures and resource utilization: the true consequences of false-positive results. JAMA. 1991;265(3):365-369. doi: 10.1001/jama.1991.03460030071031 [DOI] [PubMed] [Google Scholar]
24.Doern GV, Carroll KC, Diekema DJ, et al. Practical guidance for clinical microbiology laboratories: a comprehensive update on the problem of blood culture contamination and a discussion of methods for addressing the problem. Clin Microbiol Rev. 2019;33(1):e00009-19. doi: 10.1128/CMR.00009-19 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Versporten A, Sharland M, Bielicki J, Drapier N, Vankerckhoven V, Goossens H; ARPEC Project Group Members . The Antibiotic Resistance and Prescribing in European Children project: a neonatal and pediatric antimicrobial web-based point prevalence survey in 73 hospitals worldwide. Pediatr Infect Dis J. 2013;32(6):e242-e253. doi: 10.1097/INF.0b013e318286c612 [DOI] [PubMed] [Google Scholar]
26.Woods-Hill CZ, Koontz DW, King AF, et al. ; Bright STAR Authorship Group . Practices, perceptions, and attitudes in the evaluation of critically ill children for bacteremia: a national survey. Pediatr Crit Care Med. 2020;21(1):e23-e29. doi: 10.1097/PCC.0000000000002176 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Woods-Hill CZ, Fackler J, Nelson McMillan K, et al. Association of a clinical practice guideline with blood culture use in critically ill children. JAMA Pediatr. 2017;171(2):157-164. doi: 10.1001/jamapediatrics.2016.3153 [DOI] [PubMed] [Google Scholar]
28.Woods-Hill CZ, Lee L, Xie A, et al. Dissemination of a novel framework to improve blood culture use in pediatric critical care. Pediatr Qual Saf. 2018;3(5):e112. doi: 10.1097/pq9.0000000000000112 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Ogrinc G, Davies L, Goodman D, Batalden P, Davidoff F, Stevens D. SQUIRE 2.0 (Standards for QUality Improvement Reporting Excellence): revised publication guidelines from a detailed consensus process. BMJ Qual Saf. 2016;25(12):986-992. doi: 10.1136/bmjqs-2015-004411 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.StataCorp . Stata Statistical Software: Release 15. StataCorp LLC; 2017. [Google Scholar]
31.R Core Team . R: a language and environment for statistical computing. Published 2021. https://www.R-project.org.
32.Brown KA, Chambers A, MacFarlane S, et al. Reducing unnecessary urine culturing and antibiotic overprescribing in long-term care: a before-and-after analysis. CMAJ Open. 2019;7(1):E174-E181. doi: 10.9778/cmajo.20180064 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Klatte JM. Pediatric antimicrobial stewardship programs: current perspectives. Pediatric Health Med Ther. 2020;11(11):245-255. doi: 10.2147/PHMT.S224774 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Godbout EJ, Pakyz AL, Markley JD, Noda AJ, Stevens MP. Pediatric antimicrobial stewardship: state of the art. Curr Infect Dis Rep. 2018;20(10):39. doi: 10.1007/s11908-018-0644-7 [DOI] [PubMed] [Google Scholar]
35.Sick-Samuels AC, Woods-Hill CZ, Fackler JC, et al. Association of a blood culture utilization intervention on antibiotic use in a pediatric intensive care unit. Infect Control Hosp Epidemiol. 2019;40(4):482-484. doi: 10.1017/ice.2019.10 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Chan KH, Moser EA, Cain M, Carroll A, Benneyworth BD, Bell T. Validation of antibiotic charges in administrative data for outpatient pediatric urologic procedures. J Pediatr Urol. 2017;13(2):185-186. doi: 10.1016/j.jpurol.2017.01.009 [DOI] [PubMed] [Google Scholar]
37.Gerber JS, Newland JG, Coffin SE, et al. Variability in antibiotic use at children’s hospitals. Pediatrics. 2010;126(6):1067-1073. doi: 10.1542/peds.2010-1275 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Courter JD, Parker SK, Thurm C, et al. Accuracy of administrative data for antimicrobial administration in hospitalized children. J Pediatric Infect Dis Soc. 2018;7(3):261-263. doi: 10.1093/jpids/pix064 [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplement.

eTable 1. Sites participating in the Bright STAR Collaborative

eTable 2. Definitions of outcome variables and units of measurement

eFigure 1. Bright STAR review form for site team review of positive blood cultures after Bright STAR implementation

eTable 3. Sensitivity analyses of Bright STAR outcomes

eTable 4. Results of interrupted time series analysis of primary and secondary outcomes in Bright STAR sites

eFigure 2. Bright STAR site-specific and pooled relative rates comparing the post- and pre-implementation periods for primary and secondary outcomes

eResults. Post-hoc contemporaneous control analysis regarding broad-spectrum antibiotic use

eTable 5. Results of post-hoc contemporaneous control analysis

Click here for additional data file.^{(326.1KB, pdf)}

[poi220017r1] 1.Morgan DJ, Brownlee S, Leppin AL, et al. Setting a research agenda for medical overuse. BMJ. 2015;351:h4534. doi: 10.1136/bmj.h4534 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r2] 2.Chassin MR, Galvin RW; Institute of Medicine National Roundtable on Health Care Quality . The urgent need to improve health care quality. JAMA. 1998;280(11):1000-1005. doi: 10.1001/jama.280.11.1000 [DOI] [PubMed] [Google Scholar]

[poi220017r3] 3.Berwick DM, Hackbarth AD. Eliminating waste in US health care. JAMA. 2012;307(14):1513-1516. doi: 10.1001/jama.2012.362 [DOI] [PubMed] [Google Scholar]

[poi220017r4] 4.Esserman LJ, Thompson IM Jr, Reid B. Overdiagnosis and overtreatment in cancer: an opportunity for improvement. JAMA. 2013;310(8):797-798. doi: 10.1001/jama.2013.108415 [DOI] [PubMed] [Google Scholar]

[poi220017r5] 5.Fisher ES, Wennberg DE, Stukel TA, Gottlieb DJ, Lucas FL, Pinder EL. The implications of regional variations in Medicare spending, part 2: health outcomes and satisfaction with care. Ann Intern Med. 2003;138(4):288-298. doi: 10.7326/0003-4819-138-4-200302180-00007 [DOI] [PubMed] [Google Scholar]

[poi220017r6] 6.Brownlee S, Chalkidou K, Doust J, et al. Evidence for overuse of medical services around the world. Lancet. 2017;390(10090):156-168. doi: 10.1016/S0140-6736(16)32585-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r7] 7.Levinson W, Born K, Wolfson D. Choosing Wisely campaigns: a work in progress. JAMA. 2018;319(19):1975-1976. doi: 10.1001/jama.2018.2202 [DOI] [PubMed] [Google Scholar]

[poi220017r8] 8.National Institutes of Health . PAR-19-274: dissemination and implementation research in health. Accessed July 13, 2021. https://grants.nih.gov/grants/guide/pa-files/PAR-18-007.html

[poi220017r9] 9.Sperling K, Priddy A, Suntharam N, Feuerhake T. Optimizing testing for Clostridium difficile infection: a quality improvement project. Am J Infect Control. 2019;47(3):340-342. doi: 10.1016/j.ajic.2018.08.027 [DOI] [PubMed] [Google Scholar]

[poi220017r10] 10.Truong CY, Gombar S, Wilson R, et al. Real-time electronic tracking of diarrheal episodes and laxative therapy enables verification of Clostridium difficile clinical testing criteria and reduction of Clostridium difficile infection rates. J Clin Microbiol. 2017;55(5):1276-1284. doi: 10.1128/JCM.02319-16 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r11] 11.Sick-Samuels A, 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]

[poi220017r12] 12.Epstein L, Edwards JR, Halpin AL, et al. Evaluation of a novel intervention to reduce unnecessary urine cultures in intensive care units at a tertiary care hospital in Maryland, 2011-2014. Infect Control Hosp Epidemiol. 2016;37(5):606-609. doi: 10.1017/ice.2016.9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r13] 13.Mullin KM, Kovacs CS, Fatica C, et al. A multifaceted approach to reduction of catheter-associated urinary tract infections in the intensive care unit with an emphasis on “stewardship of culturing”. Infect Control Hosp Epidemiol. 2017;38(2):186-188. doi: 10.1017/ice.2016.266 [DOI] [PubMed] [Google Scholar]

[poi220017r14] 14.Weiss SL, Fitzgerald JC, Pappachan J, et al. ; Sepsis Prevalence, Outcomes, and Therapies (SPROUT) Study Investigators and Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network . Global epidemiology of pediatric severe sepsis: the Sepsis Prevalence, Outcomes, and Therapies study. Am J Respir Crit Care Med. 2015;191(10):1147-1157. doi: 10.1164/rccm.201412-2323OC [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r15] 15.Weiss SL, Fitzgerald JC, Balamuth F, et al. Delayed antimicrobial therapy increases mortality and organ dysfunction duration in pediatric sepsis. Crit Care Med. 2014;42(11):2409-2417. doi: 10.1097/CCM.0000000000000509 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r16] 16.Singer M, Deutschman CS, Seymour CW, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315(8):801-810. doi: 10.1001/jama.2016.0287 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r17] 17.Hsiao AL, Baker MD. Fever in the new millennium: a review of recent studies of markers of serious bacterial infection in febrile children. Curr Opin Pediatr. 2005;17(1):56-61. doi: 10.1097/01.mop.0000151781.13635.70 [DOI] [PubMed] [Google Scholar]

[poi220017r18] 18.Milcent K, Faesch S, Gras-Le Guen C, et al. Use of procalcitonin assays to predict serious bacterial infection in young febrile infants. JAMA Pediatr. 2016;170(1):62-69. doi: 10.1001/jamapediatrics.2015.3210 [DOI] [PubMed] [Google Scholar]

[poi220017r19] 19.Nijman RG, Moll HA, Vergouwe Y, de Rijke YB, Oostenbrink R. C-reactive protein bedside testing in febrile children lowers length of stay at the emergency department. Pediatr Emerg Care. 2015;31(9):633-639. doi: 10.1097/PEC.0000000000000466 [DOI] [PubMed] [Google Scholar]

[poi220017r20] 20.Lautz AJ, Dziorny AC, Denson AR, et al. Value of procalcitonin measurement for early evidence of severe bacterial infections in the pediatric intensive care unit. J Pediatr. 2016;179:74-81.e2. doi: 10.1016/j.jpeds.2016.07.045 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r21] 21.Kiragu AW, Zier J, Cornfield DN. Utility of blood cultures in postoperative pediatric intensive care unit patients. Pediatr Crit Care Med. 2009;10(3):364-368. doi: 10.1097/PCC.0b013e3181a31bb7 [DOI] [PubMed] [Google Scholar]

[poi220017r22] 22.Darby JM, Linden P, Pasculle W, Saul M. Utilization and diagnostic yield of blood cultures in a surgical intensive care unit. Crit Care Med. 1997;25(6):989-994. doi: 10.1097/00003246-199706000-00016 [DOI] [PubMed] [Google Scholar]

[poi220017r23] 23.Bates DW, Goldman L, Lee TH. Contaminant blood cultures and resource utilization: the true consequences of false-positive results. JAMA. 1991;265(3):365-369. doi: 10.1001/jama.1991.03460030071031 [DOI] [PubMed] [Google Scholar]

[poi220017r24] 24.Doern GV, Carroll KC, Diekema DJ, et al. Practical guidance for clinical microbiology laboratories: a comprehensive update on the problem of blood culture contamination and a discussion of methods for addressing the problem. Clin Microbiol Rev. 2019;33(1):e00009-19. doi: 10.1128/CMR.00009-19 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r25] 25.Versporten A, Sharland M, Bielicki J, Drapier N, Vankerckhoven V, Goossens H; ARPEC Project Group Members . The Antibiotic Resistance and Prescribing in European Children project: a neonatal and pediatric antimicrobial web-based point prevalence survey in 73 hospitals worldwide. Pediatr Infect Dis J. 2013;32(6):e242-e253. doi: 10.1097/INF.0b013e318286c612 [DOI] [PubMed] [Google Scholar]

[poi220017r26] 26.Woods-Hill CZ, Koontz DW, King AF, et al. ; Bright STAR Authorship Group . Practices, perceptions, and attitudes in the evaluation of critically ill children for bacteremia: a national survey. Pediatr Crit Care Med. 2020;21(1):e23-e29. doi: 10.1097/PCC.0000000000002176 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r27] 27.Woods-Hill CZ, Fackler J, Nelson McMillan K, et al. Association of a clinical practice guideline with blood culture use in critically ill children. JAMA Pediatr. 2017;171(2):157-164. doi: 10.1001/jamapediatrics.2016.3153 [DOI] [PubMed] [Google Scholar]

[poi220017r28] 28.Woods-Hill CZ, Lee L, Xie A, et al. Dissemination of a novel framework to improve blood culture use in pediatric critical care. Pediatr Qual Saf. 2018;3(5):e112. doi: 10.1097/pq9.0000000000000112 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r29] 29.Ogrinc G, Davies L, Goodman D, Batalden P, Davidoff F, Stevens D. SQUIRE 2.0 (Standards for QUality Improvement Reporting Excellence): revised publication guidelines from a detailed consensus process. BMJ Qual Saf. 2016;25(12):986-992. doi: 10.1136/bmjqs-2015-004411 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r30] 30.StataCorp . Stata Statistical Software: Release 15. StataCorp LLC; 2017. [Google Scholar]

[poi220017r31] 31.R Core Team . R: a language and environment for statistical computing. Published 2021. https://www.R-project.org.

[poi220017r32] 32.Brown KA, Chambers A, MacFarlane S, et al. Reducing unnecessary urine culturing and antibiotic overprescribing in long-term care: a before-and-after analysis. CMAJ Open. 2019;7(1):E174-E181. doi: 10.9778/cmajo.20180064 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r33] 33.Klatte JM. Pediatric antimicrobial stewardship programs: current perspectives. Pediatric Health Med Ther. 2020;11(11):245-255. doi: 10.2147/PHMT.S224774 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r34] 34.Godbout EJ, Pakyz AL, Markley JD, Noda AJ, Stevens MP. Pediatric antimicrobial stewardship: state of the art. Curr Infect Dis Rep. 2018;20(10):39. doi: 10.1007/s11908-018-0644-7 [DOI] [PubMed] [Google Scholar]

[poi220017r35] 35.Sick-Samuels AC, Woods-Hill CZ, Fackler JC, et al. Association of a blood culture utilization intervention on antibiotic use in a pediatric intensive care unit. Infect Control Hosp Epidemiol. 2019;40(4):482-484. doi: 10.1017/ice.2019.10 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r36] 36.Chan KH, Moser EA, Cain M, Carroll A, Benneyworth BD, Bell T. Validation of antibiotic charges in administrative data for outpatient pediatric urologic procedures. J Pediatr Urol. 2017;13(2):185-186. doi: 10.1016/j.jpurol.2017.01.009 [DOI] [PubMed] [Google Scholar]

[poi220017r37] 37.Gerber JS, Newland JG, Coffin SE, et al. Variability in antibiotic use at children’s hospitals. Pediatrics. 2010;126(6):1067-1073. doi: 10.1542/peds.2010-1275 [DOI] [PMC free article] [PubMed] [Google Scholar]

[poi220017r38] 38.Courter JD, Parker SK, Thurm C, et al. Accuracy of administrative data for antimicrobial administration in hospitalized children. J Pediatric Infect Dis Soc. 2018;7(3):261-263. doi: 10.1093/jpids/pix064 [DOI] [PubMed] [Google Scholar]

PERMALINK

Association of Diagnostic Stewardship for Blood Cultures in Critically Ill Children With Culture Rates, Antibiotic Use, and Patient Outcomes

Charlotte Z Woods-Hill, MD, MSHP

Elizabeth A Colantuoni, PhD

Danielle W Koontz, MA, MS

Annie Voskertchian, MPH

Anping Xie, PhD

Cary Thurm, PhD

Marlene R Miller, MD, MSc

James C Fackler, MD

Aaron M Milstone, MD, MHS

Asya Agulnik, MD

J Elaine-Marie Albert, MD, MHA

Michael J Auth

Erin Bradley, MD, MS

Jason A Clayton, MD, PhD

Susan E Coffin, MD, MPH

Samantha Dallefeld, MD

Chidiebere P Ezetendu, MD

Nina A Fainberg, MD

Brian F Flaherty, MD

Charles B Foster, MD

Sarmistha B Hauger, MD

Sue J Hong, MD

Nicholas D Hysmith, MD

Aileen L Kirby, MD

Larry K Kociolek, MD, MSCI

Gitte Y Larsen, MD, MPH

John C Lin, MD

William M Linam, MD, MS

Jason G Newland, MD, MEd

Dawn Nolt, MD, MPH

Gregory P Priebe, MD

Thomas J Sandora, MD, MPH

Hayden T Schwenk, MD, MPH

Craig M Smith, MD

Katherine M Steffen, MD, MHS

Sachin D Tadphale, MBBS, MPH

Philip Toltzis, MD

Joshua Wolf, MBBS, PhD

Danielle M Zerr, MD, MPH

Key Points

Question

Findings

Meaning

Abstract

Importance

Objective

Design, Setting, and Participants

Exposures

Main Outcomes and Measures

Results

Conclusions and Relevance

Introduction

Methods

Bright STAR Quality Improvement Collaborative

Table 1. Baseline Characteristics of 14 Enrolled PICUs.

Figure 1. Key Steps in the Bright STAR (Testing Stewardship for Antibiotic Reduction) Collaborative Time Line.

Intervention: Site-Specific Blood Culture Practice Change Driven by Local QI Teams

Outcome Measures and Data Sources

Table 2. Primary and Secondary Outcomes Before and After Implementation of the Bright STAR Collaborative in 14 PICUs.

Balancing Measures and Event Monitoring

Statistical Analysis

Results

Primary Outcome

Figure 2. Monthly Blood Cultures per 1000 Patient-Days for Each of the 14 Bright STAR (Testing Stewardship for Antibiotic Reduction) Sites.

Secondary Outcomes

Figure 3. Antibiotic Use for Each of the 11 Bright STAR (Testing Stewardship for Antibiotic Reduction) Sites That Participate in the Children’s Hospital Association Pediatric Health Information System.

Post Hoc Contemporaneous Control Analysis

Discussion

Limitations

Conclusions

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles