Chanu Rhee, MD, MPH, is an infectious disease and critical care physician and Medical Director of Infection Control at Brigham and Women’s Hospital. He is also Director of the Center for Sepsis Epidemiology and Prevention Studies (SEPSIS) at the Harvard Pilgrim Health Care Institute and Associate Professor of Population Medicine at Harvard Medical School. Dr. Rhee served on the task force that developed the Betsy Lehman Center’s 2019 report on sepsis care in emergency medicine and is the lead author of a new position paper, “Hospital Strategies to Improve Sepsis Outcomes,” released by the Infectious Diseases Society of America and endorsed by seven other professional societies and the Sepsis Alliance. The paper was published in August in Clinical Infectious Diseases. 

Patient Safety Beat: An expert panel you led recently issued recommendations for improving sepsis outcomes in hospital care. Please describe that project and how it differs from previous recommendations. 

Dr. Rhee: Sepsis remains a major cause of morbidity and mortality, despite substantial efforts over the past decade to improve early recognition and treatment. Those efforts remain essential, but improving outcomes requires a broader approach that looks beyond the first few hours of care and considers what hospitals can do throughout a patient’s course. 

That was the impetus for this project. The goal was to develop practical, hospital-level strategies to help clinicians deliver better sepsis care.  Within that broad framework, we focused particularly on the infection-related aspects of care because getting the diagnosis and management of infection right is fundamental to improving sepsis outcomes. That includes accurately identifying the infection and causative pathogen, optimizing antimicrobial therapy, and preventing additional complications during hospitalization.   

Our recommendations complement existing clinical guidelines, such as those from the Surviving Sepsis Campaign, by focusing on hospital-level systems and processes rather than on the clinical management of individual patients. 

The Task Force drew on both the literature and experiences from our own institutions, asking where gaps exist between current practice and ideal care and which interventions could help close those gaps. We also considered feasibility, cost, and how widely practices are already being implemented. With those considerations in mind, we developed recommendations across several domains: diagnostic testing, antimicrobial management and delivery, surveillance and performance measurement, programmatic infrastructure, infection prevention, and adjunctive therapies.  
 

Patient Safety Beat: Please describe an example of the recommendations outlined in the paper. 

Dr. Rhee: Antibiotic delivery is one example. The Centers for Medicare and Medicaid Services (CMS) sepsis quality measure, known as the SEP-1 bundle, requires administering antibiotics within 3 hours for patients with sepsis.  But when it comes to antibiotic timing, the strongest evidence for urgency is in patients with septic shock, where delays can be particularly consequential.  The panel therefore strongly recommended that hospitals develop workflows to identify patients with septic shock and facilitate rapid antibiotic delivery, ideally within one hour.  

This raises a systems question: How do you actually get the antibiotic to the patient within an hour?  There are multiple steps between recognizing septic shock and starting the infusion, from the clinician recognizing septic shock and ordering the antibiotic, to pharmacy verification and approval, to the nurse obtaining the antibiotic and administering it.  We recommend that hospitals systematically measure how often antibiotics are delivered within one hour of septic shock recognition and use those data to identify delays and drive improvement.  We also recommend tracking the time from the antibiotic order to the start of the infusion, with a goal of less than 30 minutes. That helps hospitals distinguish delays in recognizing and ordering treatment from delays in actually getting the antibiotic to the patient. 

Another example is rapid molecular testing of positive blood cultures. This isn’t really an decision for individual clinicians — it requires a hospital to make the technology available and integrate it into microbiology and clinical workflows. These tests can identify bloodstream pathogens and important resistance markers much faster than conventional methods. We recommend pairing them with active antimicrobial stewardship support to quickly translate the results into decisions to optimize antibiotic therapy. 

We developed approximately 20 recommendations of this kind. They are divided into Tier 1 strategies, which we believe broadly apply to most hospitals, and Tier 2 strategies, which may be beneficial but can be implemented more selectively or phased in depending on local resources and priorities. 
 

Patient Safety Beat: Looking ahead, are there any other emerging strategies that are shaping the future of sepsis care? 

Dr. Rhee: The paper includes a section on emerging and promising strategies. One important area is better early diagnostics. Sepsis can be very challenging to diagnose, as it can resemble a wide range of infectious and noninfectious conditions.  

The Food and Drug Administration recently cleared several new diagnostic tests for sepsis, but we don’t yet have enough evidence to recommend that every hospital implement them. Demonstrating good diagnostic accuracy — the basis for FDA clearance — is an important first step, but it doesn't tell you how best to integrate a test into clinical workflows or, more importantly, whether using it actually improves patient outcomes. Does it help identify sepsis earlier without also driving unnecessary antibiotic use? Are there other unintended consequences?  These new diagnostics are very promising, and we're starting to see the field move beyond retrospective analyses of diagnostic accuracy toward prospective studies of their impact in clinical practice. 
 

Patient Safety Beat: You served on one of the Massachusetts Sepsis Consortium Task Forces that developed recommendations for screening and treatment protocols in 2019. How has sepsis care changed since then? 

Dr. Rhee: The Task Force was working on sepsis just a few years after CMS introduced SEP-1, with its strong focus on early recognition and bundle-based treatment. When SEP-1 was implemented in 2015, there was understandably a great deal of emphasis on improving hospital compliance with the bundle. In recent years, though, there has been growing recognition that bundle compliance alone is not enough. 

National SEP-1 compliance rates have plateaued at around 60 to 70%, and given the complexity and heterogeneity of sepsis, 100% compliance is not necessarily the right goal in every clinical scenario. More importantly, despite strides in sepsis care, mortality remains unacceptably high. We shouldn’t move away from early recognition and treatment, but we need to broaden our approach. 

CDC and CMS recognize this and are working on ways to benchmark hospitals on sepsis outcomes, not just on bundle compliance. That same shift toward a broader approach is happening at the health-system level.  At Mass General Brigham, for example, sepsis has been a major system-wide initiative. Early efforts focused heavily on SEP-1. Now there is a broader push to reduce mortality through a more comprehensive set of strategies spanning early diagnosis and treatment, ICU management, and infection-related aspects of care throughout the patient’s hospitalization. I’m glad to see that shift toward more comprehensive sepsis care.