SCCM Resource Library-32-Improving Pediatric Sepsis Outcomes for All Children Together (IPSO FACTO): Interim Results

Video Transcription

Good afternoon, everybody. Today we'll be speaking about Improving Pediatric Sepsis Outcomes for All Children Together, the Ipso Facto Study, our interim results.  My name is Dr. Paul, and I am one of the co-chairs for the collaborative, and I'm currently  the Director of Quality Improvement for Advocate Children's Hospital's Emergency Department. I have no disclosures to report.  The Ipso Facto Study is a quality improvement sepsis collaborative among children's hospitals across the United States. It represents 58 sites from 2016 until current.  Our goal is to improve recognition and treatment of pediatric sepsis to reduce mortality and prevent hospital onset of septic shock based on up-to-date evidence using QI data-driven  science. The hospitals span the country and range in beds from 43 beds up to Quaternary Children's Hospitals of 604 beds.  They're both children's hospitals that are freestanding as well as those that are not. Our primary aim on the left, highlighted in yellow, is to reduce pediatric sepsis mortality  and hospital onset incidents. We wish to do this in a variety of ways.  appropriate and timely treatment of infection before it becomes severe sepsis or septic  shock, specific, efficient, and timely recognition of sepsis, diagnosis of sepsis to understand  where we are on the continuum of septic shock, timely and effective treatment of sepsis once  it's been identified, appropriate and timely de-escalation of inpatient care, patient and  family engagement, and optimizing performance using the robust and numerous change strategies indicated on the right. IHPSA defined two main cohorts.  The first, not further described in this presentation, are those labeled IHPSA-suspected infection, those that had a blood culture, antibiotics, and admission.  These are patients with perhaps a urinary tract infection or pneumonia who did not present with severe sepsis or septic shock at the outset.  The second group, IHPSA-sepsis, are those that met one of eight IHPSA inclusion criteria described further on the next slide.  This was further broken down into a sicker cohort approximating septic shock labeled IHPSA-critical sepsis and is delineated by red.  Those that did not meet this additional criteria are delineated by the blue and labeled IHPSA-suspected sepsis.  The orange components on the left represent the IHPSA-suspected infection group that we will not be speaking about further.  The central purple components represent the cascading IHPSA inclusion criteria. If a patient met one of these criteria, a sepsis screen, a bedside huddle, an order  set alone, or ICU admission, lactate, or a presser, or vasoactive agents beyond with a predefined treatment criteria, they met the IHPSA-sepsis definition.  If the patient further received a third bolus or a vasoactive agent, they met the criteria for IHPSA-critical sepsis as delineated by the red components on the right side.  If they did not meet IHPSA-critical definitions, they were labeled IHPSA-suspected sepsis as delineated by the blue.  The results presented will refer to the red IHPSA-critical sepsis and the blue IHPSA-sepsis categories. The results will be presented as statistical process control charts.  The month is plotted on the x-axis. The red dotted lines represent the upper and lower control limit lines.  On the left, we see the rate of critical sepsis cases per 1,000 admissions, which has remained stable over time.  On the left, we see the rate of IHPSA-suspected sepsis cases, which has increased over time.  This is likely due, however, to increased detection of cases with better screens, bedside huddles, and order sets as the collaborative is ongoing.  As such, this cohort possibly represents a cohort of slightly less sick patients over time as we are using optimized identification strategies to identify them earlier.  When we analyzed our interim results of 40,000 patients, we noted that utilization of some form of screening tool, whether it be an EHR, electronic health record, embedded screen,  or a paper screen, a bedside huddle, or a sepsis order set, this improved the timeliness of care delivered.  So what one can see here is that the time to antibiotics and the time to bolus improved for the IHPSA-critical cohort.  Similarly, for the IHPSA-suspected sepsis cohort, utilization of a screen or a huddle or an order set improved the time to antibiotics and also improved the time to first bolus  delivery. On the bottom, one can see that utilization of any of these recognition methods, whether  it be a screen, a huddle, or an order set, improved the time to delivery of an antibiotic or a bolus for both cohorts.  So if we know that utilization of a screen or a huddle or an order set improves our timeliness of therapeutic delivery, how are we doing as a collaborative over time?  One can see for the IHPSA-critical sepsis cohort on the left and the IHPSA-suspected sepsis cohort on the right, there has been an improvement in utilization of a screen,  huddle, or order set, approaching our target goal of close to 85%. So if we are utilizing these screens, huddles, and order sets effectively, are we improving  timely delivery of therapeutics? One can see on the left for the IHPSA-critical sepsis cohort, we have seen a decrease in  the time to first bolus in minutes from about 37 minutes to 34 minutes. There's a slight uptick one can see coinciding with the onset of COVID-19 in the United States.  On the right side, we have not moved the needle on time to first bolus in minutes. You can see that our target line is 30 minutes.  This is aligning with the surviving sepsis campaign guidelines, which recommended a slightly more liberalized delivery of bolus for these septic shock patients.  Therefore, we see our target line being 30 minutes as opposed to our previous 20 minutes, which we will go forward in the next slides and discuss.  Our time to first antibiotics has also improved in the IHPSA-critical sepsis cohort from 90  minutes to 84 minutes, approaching our target line of 60 minutes for these sicker patients.  And on the right side, we see for the IHPSA-suspected sepsis, we are below target, going from 135 minutes to 150 minutes.  Again, you see this target line is 180 minutes, again, representing the more liberalized target  for those undifferentiated sepsis patients as delineated by their surviving sepsis campaign in February of this year.  When we first began the collaborative, we utilized the available evidence to discuss and identify an initial bundle that was optimal.  This included delivery of a bolus within 20 minutes, antibiotics within 60, and utilization  of a recognition method, whether it be a sepsis screen, a sepsis huddle, or sepsis order set.  And we looked to see whether these actually did improve outcomes if we were actually adherent and compliant with utilizing these bundles, so all components of the bundle.  And one can see here for the IHPSA-critical sepsis cohort that if one was compliant with the entire bundle, the sepsis attributable mortality decreased.  Now, this was just a trend, not statistically significant. However, we did see if one was compliant with this initial bundle, that the median hospital  days did decrease, and this was statistically significant. Similarly, we saw significant decreases in sepsis attributable mortality and median hospital  days for the IHPSA-suspected sepsis cohort. As we progressed through the collaborative and analyzed our 40,000 patients as part of  our interim results, in addition to looking at the new evidence that has emerged this year, we recognized a more liberalized bundle was actually improving outcomes more.  So this liberalized bundle is now delivery of a bolus within 60 minutes, as opposed to 20, antibiotics within 180, as opposed to 60, and some recognition method, whether it  be a screen, a huddle, or an order set. Again, you can see the results are magnified. And if one was compliant with all components of this liberalized bundle in the IHPSA-critical  sepsis cohort, the sepsis attributable mortality went down, and the median hospital days decreased as well.  Similarly, for the IHPSA-suspected sepsis cohort, one can see that sepsis attributable mortality went down when one was adherent to the liberalized bundle, and median hospital  days went down when one was compliant with this liberalized bundle. So how are we doing with this bundle over time, if we know that this does improve outcomes  if you adhere to all three components of the bundle? So for the IHPSA-critical sepsis cohort, we have seen a slight uptick in bundle compliance.  Similarly, for the IHPSA-suspected sepsis cohort, we've also seen a slight uptick in compliance. You can see, however, that we are nowhere near our target.  We are not performing in a highly reliable way of 80 or higher percent. Therefore, this collaborative is still ongoing, and we are identifying where we are having fallouts.  At this point, it does seem to be in delivery of the fluid bolus, where we seem to be failing the most.  We are doing better with antibiotic delivery, and we are doing better with utilization of a recognition method.  So because we have not moved the needle on bundle compliance, we have not expected to see us move the needle for sepsis attributable mortality or major outcome.  So on the left-hand side, our sepsis attributable mortality for the IHPSA-critical sepsis group has remained relatively stable at 3.7%.  And on the right-hand side, you do see for the suspected sepsis group, we have seen a slight decrease in attributable mortality from 1.9% to 1.3%, with a slight uptick coinciding  with COVID-19. Now, potentially, this is related to the increased incidence of suspected sepsis that we showed  on that first control chart slide and a potential dilution of that denominator.  But nonetheless, this does represent a decrease in attributable mortality. Therefore, in conclusion, we have refined pediatric sepsis definitions for quality improvement  on a large scale to accurately capture meaningful cohorts. We've had successful and sustained automated data capture across 58 hospitals.  We've demonstrated improvement in process measures, and have identified an ideal and optimized bundle of care in order to improve outcomes.  We've shown small improvements in compliance to this bundle, and we are hoping for more highly reliable care as we move forward.  Thank you very much for listening to our results with regards to the IHPSA Sepsis Collaborative, and I will take any questions now. Thank you. ♪♪♪

Video Summary

The Ipso Facto Study is a quality improvement collaborative among children's hospitals to improve recognition and treatment of pediatric sepsis. The study includes 58 hospitals and aims to reduce mortality and prevent septic shock. The results show that utilization of screening tools, bedside huddles, and order sets improve the timeliness of care delivered. There has been an improvement in the utilization of these methods over time. The study also found that adherence to a bundle of care, including timely delivery of bolus and antibiotics, improves outcomes. However, compliance with the bundle still needs improvement. The sepsis attributable mortality has remained relatively stable, but there has been a slight decrease in suspected sepsis mortality.

Asset Subtitle

Pediatrics, Sepsis, 2021

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2021 Critical Care Congress held virtually from January 31-February 12, 2021.

Raina M. Paul

Introduction/Hypothesis: Timely management of pediatric sepsis is challenging, and previous QI collaboratives have struggled to demonstrate meaningful association between improved care and improved outcomes

Methods: Children's Hospital Association formed a QI collaborative focused on Improving Pediatric Sepsis Outcomes (IPSO). Sepsis research definitions were adapted, enabling automated identification of potential sepsis across the care continuum. IPSO Sepsis (IS) patients were identified in real-time and retrospectively using intent to treat criteria. IS was subdivided into IPSO Non-critical Sepsis (INS) and IPSO Critical Sepsis (ICS) cohorts, the latter likely having septic shock. Changes in measures over time were analyzed using statistical process control. Outcomes of cases compliant vs. non-compliant with care bundles were compared using Pearson Chi Square and Wilcoxon rank sum tests.

Results: Results included 33726 cases, from 40 children's hospitals, between Jan 2017 and Dec 2019. Use of recognition tools (screen, huddle, orderset) increased identification of IS over time (rate/1000 admissions: 23.1 to 29.9), although identification of ICS remained relatively unchanged (rate/1000 admissions: 8.8 to 8.4). Increased IS identification likely influenced IS outcomes, so only results for ICS patients are presented. Use of any recognition method improved from 61.2-67.7%. Time to first bolus and antibiotic improved from 36.9 to 34.5 min and 89.9 to 84.2 min respectively. Cases compliant with initial collaborative goals (Bolus within 20 min, Abx within 60 min and any recognition method) demonstrated similar 30-day sepsis attributable mortality compared with non-compliant cases (3.1% vs. 3.5%, p=0.233) but lower median (IQR) hospital days [6(4,16) vs. 7(4,13), p<0.001]. In cases compliant with relaxed care bundle goals (Bolus within 60 min, Abx within 180 min and any recognition method), both 30-day sepsis attributable mortality (2.4% vs. 4.6%, p<0.001) and median hospital days [6(4,12) vs. 9(4,19), p< 0.001] were lower. There was improvement in adherence to both bundles over time (23 to 27.9% and 52 to 56.7% respectively).

Conclusions: Timely identification of pediatric sepsis and resuscitation is associated with improved sepsis outcomes. Sustained improvement should focus on bundle adherence with higher reliability.

Meta Tag

Content Type Presentation

Knowledge Area Pediatrics

Knowledge Area Sepsis

Knowledge Level Advanced

Membership Level Select

Tag Guidelines

Year 2021

Keywords

Ipso Facto Study

quality improvement collaborative

pediatric sepsis

mortality reduction

septic shock prevention

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.