CDC's Hospital Sepsis Program Core Elements are associated with improved management and outcomes of sepsis.

To the Editor: Sepsis is a leading cause of inpatient mortality in the United States, contributing to over one-third of hospital deaths.1 The Surviving Sepsis Campaign provides evidence-based guidelines for the management of sepsis.2 However, consistent delivery of evidence-based practice is challenging. Sepsis can be difficult to diagnose in real-time,3 and optimal management requires coordination across disciplines and hospital locations.4 To support sepsis performance improvement, the Centers for Disease Control and Prevention (CDC) published the Hospital Sepsis Program Core Elements in 2023.5,6 The Core Elements serve as a high-level guide for running an effective hospital or health system program to improve management and outcomes of sepsis. They recommend 28 key program features mapped to the following 7 domains: hospital leadership commitment, accountability (ie, sepsis program leadership and goals), multidisciplinary expertise, action (ie, tools to support implementation of sepsis management recommendations), tracking, reporting, and education.5 We sought to assess whether these key features of hospital sepsis programs are associated with better management and outcomes of sepsis. The Michigan Hospital Medicine Safety Consortium Sepsis Initiative (HMS-Sepsis) is a multihospital collaborative quality initiative funded by Blue Cross Blue Shield of Michigan. Trained abstractors at each hospital enter data on a random sample of community-onset sepsis hospitalizations into a central registry, including granular data on sepsis management.7 Mortality at 30 days is determined for all patients via chart review, public obituary data, follow-up by telephone, email, and/or SMS messaging, as described previously.8 More information on HMS and HMS-Sepsis is presented in the Supplementary Material (https://www.mi-hms.org/wp-content/uploads/2025/10/Online-Supplement.docx). We surveyed hospitals in spring 2024 to determine presence of the 28 key features of hospital sepsis programs (Table 1) using questions from the National Healthcare Safety Network Annual Survey developed to measure uptake of the Hospital Sepsis Program Core Elements. Surveys were completed by each hospital’s designated HMS representative, who compiled answers from staff at their institution. We calculated a Core Elements score for each hospital (range, 0-28; 1 point per key feature) to yield a summary assessment of the hospital structures to support sepsis quality improvement. Recommended features of Hospital Sepsis Program Core Elements. We examined Core Elements scores by hospital subgroups defined by hospital size, urbanicity, teaching status, profit status, and system status. We additionally examined hospital characteristics by quintile of Core Elements score. We assessed the correlation (Pearson coefficient) of the Core Elements score with sepsis management and outcomes using HMS-Sepsis registry data for patients discharged during January 2022 through August 2024, as well as publicly available performance data from October 2023 to September 2024. Sepsis management was measured via (1) hospital-level compliance with the HMS-Sepsis Early Sepsis Management Bundle used for performance benchmarking in Michigan and (2) performance on the Severe Sepsis and Septic Shock Management Bundle (SEP-1) used for performance benchmarking by the Centers for Medicare and Medicaid Services. The outcome of interest was 30-day standardized mortality ratio, calculated as observed mortality divided by predicted mortality using the HMS-Sepsis mortality model.7 This model uses physiologic, demographic, and baseline health data to predict 30-day mortality among patients hospitalized for sepsis; it was previously validated and shown to have strong discrimination (c-statistic 0.82) and acceptable calibration.7 To examine the independent association of Core Elements score with sepsis mortality, we fit a generalized linear model examining the association of Core Elements score with 30-day standardized mortality ratio, adjusted for hospital size, urbanicity, teaching status, profit status, and system status. The study was deemed not regulated by the University of Michigan’s Institutional Review Board. Among 67 HMS hospitals, the prevalence of key features for hospital sepsis programs ranged from 30% (Expertise 4: “Our program has support from individuals with expertise and formal training in data management, information technology, and quality improvement”) to 100% (Action 1: “Our hospital has implemented a standard process to screen for sepsis on presentation and throughout hospitalization”) (Figure 1). Correlation of Core Elements score with sepsis management and outcomes. (A) Prevalence of 28 key features of hospital sepsis programs. Shown is the proportion of hospitals in the Michigan Hospital Medicine Safety Consortium Sepsis Initiative (HMS-Sepsis) with each key feature. Key features, described as “priority examples” in the Core Elements guidance,6 are mapped to 7 core elements domains of hospital leadership commitment, accountability (sepsis program leadership and goals), multidisciplinary expertise, action (structures to support implementation of sepsis management recommendations), tracking, reporting, and education. Descriptions of each key feature are provided in Table 1. Bars are color-coded to denote features within a particular Core Elements domain. In general, program features were similar to slightly higher than national estimates.9,10 For example, 96% of hospitals in HMS have a committee charged with monitoring and improving the outcomes of sepsis (Accountability 1), compared to 78% nationally,10 and 100% of hospitals in HMS have order set for management of sepsis (Action 3), compared to 86% nationally.10 (B) HMS-Sepsis Early Sepsis Bundle. Each circle represents 1 hospital. Hospitals are plotted by their Core Elements score (x-axis) versus proportion of community-onset sepsis hospitalizations compliant with the HMS-Sepsis Early Management Bundle (y-axis), which is defined in the Supplementary Material (https://www.mi-hms.org/wp-content/uploads/2025/10/Online-Supplement.docx). (C) Severe Sepsis and Septic Shock Management Bundle (SEP-1) performance measure. Each circle represents 1 hospital. Hospitals are plotted by their Core Elements score (x-axis) versus publicly reported SEP-1 score. (D) Thirty-day standardized mortality ratio. Each circle represents 1 hospital. Hospitals are plotted by their Core Elements score (x-axis) versus 30-day standardized mortality ratio calculated as observed 30-day mortality divided by predicted mortality using the HMS-Sepsis mortality model.7 Core Elements scores ranged from 7 to 28 (median, 21) across hospitals. Median scores by subgroup and hospital characteristics by quintile of Core Elements score are presented in the Supplementary Material. Core Elements scores were similar across subgroups defined by hospital size, urbanicity, teaching status, and system status, but were lower among for-profit vs not-for-profit hospitals (median, 14.5 vs 21; P = .005). There were 35 777 patients hospitalized for community-onset sepsis across 67 hospitals during the study period; their data were used to calculate hospital performance on the HMS-Sepsis Early Sepsis Management Bundle and 30-day standardized mortality ratio. Patient characteristics are presented in the Supplementary Material. Core Elements scores were moderately correlated to hospital performance on the HMS-Sepsis Early Sepsis Management Bundle (r = 0.422; P <.001), weakly correlated to hospital performance on the SEP-1 measure (r = 0.303; P = .016), and weakly correlated with hospital-level 30-day standardized mortality ratio (r = 0.261, P = .033) (Figure 1). In a regression model adjusted for hospital size, urbanicity, teaching status, system status, and profit status, Core Elements score was independently associated with 30-day standardized mortality ratio (β-coefficient: -.009 [95% CI, -0.018 to -0.0001, P = .048). In this multihospital cohort, we found that robust hospital sepsis programs—as measured by a summary Core Elements score—were associated with guideline-concordant management and lower risk-adjusted mortality in patients hospitalized for sepsis. Furthermore, the Core Elements score was independently associated with management and outcomes of sepsis after adjustment for hospital characteristics. These findings suggest that robust hospital sepsis programs can improve the management and outcomes of sepsis. Study limitations include the dependence on self-report to identify key features of hospital sepsis programs; potential gain or loss of key features over the course of the study that could not be assessed via the survey at a single time point; inability to assess the quality of key program features; and focus on hospitals participating in a statewide collaborative quality initiative that may not generalize to other hospitals. Despite the limitations, these data provide preliminary validation of the CDC Hospital Sepsis Program Core Elements and indicate that hospital sepsis programs have the potential to improve sepsis management and outcomes. More robust evaluation across diverse settings should be encouraged to support national uptake. The authors thank John Blamoun, MD (MyMichigan Health), Paul Bozyk, MD (Corewell Health), Megan Cahil, DO, MBA (Henry Ford Health), Lama Hsaiky, PharmD (Wayne State University; Corewell Health), Maximiliano Tamae Kakazu, MD (Corewell Health), Anurag Malani (St Joseph Mercy Health System), and Stephanie Taylor, MD, MSc (University of Michigan), for feedback on this study, as well as all HMS hospitals for contributing data to the study. H.C.P., P.J.P., and R.B.D. designed the study. H.C.P., E.M., J.H., N.J., S.A.F., and P.J.P. contributed to data collection. M.H. and E.W. analyzed the data. All authors interpreted the data. H.C.P. drafted the manuscript. All authors revised the manuscript for intellectual content and approved it to be submitted for publication. Supplementary material is available at American Journal of ­Respiratory and Critical Care Medicine online. Supplementary material cited in the article can be found at: https://www.mi-hms.org/wp-content/uploads/2025/10/Online-Supplement.docx. Please see the ICMJE disclosure forms, which have been provided as supplementary material. This work was supported by Blue Cross Blue Shield of Michigan (BCBSM) as part of their Value Partnerships program and by the Centers for Disease Control and Prevention (CDC). Although BCBSM and the Hospital Medicine Safety Consortium work collaboratively, the opinions, beliefs, and viewpoints expressed by the authors do not necessarily reflect the opinions, beliefs, and viewpoints of BCBSM or any of its employees. This work was also supported with resources and use of facilities at the Ann Arbor VA Medical Center. This manuscript does not necessarily reflect the position of the Department of Veterans Affairs or the US government. H.C.P., M.H., E.M., J.H., S.A.F., and P.J.P. receive salary support from BCBSM for work on HMS. H.C.P., J.H., and R.B.D. received salary support from CDC for work on the Hospital Sepsis Program Core Elements. H.C.P. received grant funding and salary support, unrelated to this study, from the National Institutes of Health, the Agency for Healthcare Research and Quality, and the Department of Veterans Affairs. H.C.P., N.J., and P.J.P. serve on the advisory board to Sepsis Alliance (unpaid). H.C.P. serves on the Surviving Sepsis Campaign guidelines committee. No artificial intelligence tools were used in writing this manuscript.