Sepsis Projects

Principal Investigator: Chanu Rhee, MD, MPH

Hospitals, clinicians, policy makers, and researchers are blind to the full breadth of serious HAIs owing to the current focus by the Centers for Medicare & Medicaid Services on just a limited set of reportable HAIs as well as reliance on traditional manual surveillance methods that are labor-intensive and vary widely in their sensitivity, accuracy, and reproducibility. The recent development of CDC’s electronic hospital-onset Adult Sepsis Event surveillance definition, however, could provide a more complete, efficient, and objective method to identify a fuller array of the most serious HAIs compared to current surveillance methods. This AHRQ-funded project aims to characterize the incidence, outcomes, and hospital-level variation of hospital-onset sepsis compared to current reportable HAIs; elucidate the impact of hospital-onset sepsis on long-term outcomes; and develop electronic algorithms to identify the specific infections leading to hospital-onset sepsis. This project is a collaboration with HCA Healthcare and will utilize CMS data to analyze long-term outcomes.

Funded by an R01 grant from the National Heart, Lung, and Blood Institute (NHLBI), this project aims to advance sepsis quality measurement beyond traditional resuscitation bundles by identifying effective care processes, evaluating hospital-level strategies to improve sepsis care, and developing more accurate measures of hospital performance. Using large-scale electronic health record and administrative data from hundreds of U.S. hospitals, the research will identify care practices associated with improved sepsis outcomes, assess the relationship between hospital sepsis program structures and patient outcomes, and develop and validate an electronic health record-based measure of 30-day readmissions following sepsis hospitalizations for comparison with CMS's claims-based approach. Together, these efforts will inform next-generation sepsis quality measures and help hospitals prioritize strategies to improve patient outcomes.

Funded by the Centers for Disease Control and Prevention (CDC), this project aims to develop, validate, and assess the reliability of risk-adjustment models for two emerging digital quality measures in CDC's National Healthcare Safety Network (NHSN): hospital-onset bacteremia/fungemia (HOB) and healthcare facility-onset, antibiotic-treated Clostridioides difficile infection (HT-CDI). These measures leverage routinely collected electronic health record data to enable standardized, automated surveillance of healthcare-associated infections and comparisons of hospital performance. Using clinical data from diverse U.S. hospitals, the research will identify patient- and hospital-level factors associated with these events, evaluate risk-adjustment model performance, and assess the reliability of hospital-level comparisons. Findings will support the implementation of these digital quality measures within NHSN to strengthen national infection surveillance, hospital performance assessment, and quality improvement efforts.

Principal Investigators: Chanu Rhee, MD, MPH; Michael Klompas, MD, MPH

A major barrier to providing appropriate sepsis care is our limited understanding of which specific clinical presentations are most likely to benefit from immediate antibiotics and which can safely be managed more conservatively. Additionally, we have an imperfect understanding of which care processes are critical to optimizing patient outcomes. To address these gaps, we are researching better ways to define sepsis “time zero,” a concept that embodies both the criteria that should trigger immediate interventions when infection is suspected and the point from which timeliness of care is measured. In this AHRQ-funded project, we are leveraging detailed EHR data from millions of encounters across over 150 hospitals in two datasets in support of the following aims: 1) Develop evidence-based, objective, and electronically computable definitions of sepsis time zero by using electronic health record data to identify the clinical signs that are associated with higher mortality when antibiotics are delayed, 2) Systematically evaluate the associations between each sepsis bundle component and mortality and compare the full bundle to simpler, streamlined versions, and 3) Assess whether and how optimal treatment strategies differ for commonly encountered and easily recognizable sepsis phenotypes based on infection site, comorbidities, and clinical signs. This project is a collaboration with HCA Healthcare and includes data from HCA and Mass General Brigham hospitals.

Principal Investigator: Chanu Rhee, MD, MPH

The goal of this CDC-funded project is to estimate the national burden of pediatric sepsis by optimizing and validating an EHR-based Pediatric Sepsis Event surveillance definition, applying it to multiple large datasets, and creating a toolkit to help hospitals conduct local pediatric sepsis surveillance. This work is conducted in collaboration with Children’s Hospital of Philadelphia, Nemours Children’s Health Delaware, Lurie Children’s Hospital, University of Colorado, HCA Healthcare, and the NIH Clinical Center Critical Care Medicine Department. Data sources include Epic Cosmos, HCA Healthcare, PINC AI, the Pediatric Emergency Care Applied Research Network, and the Novel Pediatric Sepsis Criteria Dataset.

Principal Investigator: Chanu Rhee, MD, MPH

Funded by the CDC Epicenter Program, this project aims to generate data on risk-adjustment, reliability testing, and validity testing of an updated Adult Sepsis Event (ASE) definition in order to support CDC’s submission of a risk-adjusted community-onset ASE mortality metric for consideration as a CMS quality measure. This project builds from the prior work and ASE definition that resulted from the from CDC National Sepsis Survey study. Data sources include Mass General Brigham, HCA Healthcare, and PINC AI.

Principal Investigators: Michael Klompas, MD, MPH; Chanu Rhee, MD, MPH

The increasing national uptake of electronic health record (EHR) systems allows for the possibility of tracking sepsis using objective clinical markers of infection and organ dysfunction rather than subjective and changing administrative code assignments. The Harvard Epicenter developed preliminary EHR-based surveillance definitions for sepsis and demonstrated that they have superior sensitivity, similar positive predictive value, and more stable sensitivity over time compared to definitions based on administrative claims data. In this CDC-funded project, Harvard extended this work by applying an optimized EHR-based definition to a broad national network of hospitals from 7 different datasets in order to generate accurate and generalizable estimates of sepsis burden and epidemiology. This study informed CDC’s estimates of U.S. sepsis incidence and outcomes and and led to CDC’s Adult Sepsis Event toolkit. 

Principal Investigators: Chanu Rhee, MD, MPH; Michael Klompas, MD, MPH

The aims of this CDC-funded multicenter project (conducted in collaboration with University of Washington in St. Louis and Duke University) were to: 

1. Evaluate and compare the clinical characteristics, associated morbidity and mortality, and overlap of the Sepsis-3 clinical definition with the proposed Epicenters’ EHR-based sepsis surveillance definition and sepsis administrative claims definitions
2. Determine the attributable mortality of hospital-onset sepsis and identify the associated clinical characteristics and comorbidities of patients who die from sepsis
3. Quantify interobserver variability in assessing compliance with the CMS sepsis measure