The Centers for Medicare & Medicaid Services’ (CMS) current guidelines for treating sepsis should be retired and replaced with more performance-based measures, according to a coalition of medical groups.
The measure, known as the Severe Sepsis and Septic Shock: Management Bundle, or SEP-1, was first implemented by CMS in October 2015. In addition to being several dozen pages long, the measure has another unusual feature — rather than being stewarded by the American Medical Association, CMS, or an infectious diseases association, its steward is Henry Ford Hospital in Detroit.
“The SEP-1 measure requires clinicians to provide a bundle of care to all patients with possible sepsis within 3 hours of recognition,” the groups — which include the Infectious Diseases Society of America (IDSA), the American College of Emergency Physicians, and the Society of Hospital Medicine — said in a statement. “The bundle includes drawing blood cultures, administering broad-spectrum antibiotics, and other measures. However, the SEP-1 measure does not take into account that many serious conditions present in a similar fashion to sepsis.”
“Pushing clinicians to treat all these patients as if they have sepsis leads to overuse of broad-spectrum antibiotics, which can be harmful to patients who are not infected, those who are infected with viruses rather than bacteria, and those who could safely be treated with narrower-spectrum antibiotics,” the statement continued. “Moreover, a series of studies published since the SEP-1 measure went into effect show that in practice SEP-1 has not lowered mortality rates.”
“I don’t think any of us who work in sepsis will deny it’s a big problem, that hospitals should focus on it, and that there’s room for improvement,” Chanu Rhee, MD, MPH, an infectious disease physician at Brigham and Women’s Hospital in Boston and lead author of the recommendations, said in a phone call. “But is SEP-1 the best way to do it, and are we causing unintended consequences by trying to treat everyone with possible sepsis immediately?”
“We think it’s time to move beyond SEP-1 and move to metrics that focus on patient outcomes, so that will encourage hospitals to pay more attention to the full breadth of sepsis care,” said Rhee, who is also an associate professor of population medicine at Harvard Medical School. “There are many things hospitals could do to improve [sepsis care] such as implementing processes to have timely source control, optimizing antibiotic dosing, de-escalation of antibiotics, minimizing sedation for patients who are mechanically ventilated, and preventing hospital-acquired infections. There are a lot of things that go into affecting outcomes for these patients that are getting neglected.”
The recommendations, which were published Oct. 13 in Clinical Infectious Diseases, come just as CMS is increasing its importance in reimbursement. Hospitals have been required to report on their compliance with the sepsis guidelines, but in a final rule published in August, CMS said that starting in fiscal year 2026, hospitals will be measured on how well they comply with the sepsis guidelines as part of the Hospital Value-Based Purchasing Program, and will risk losing money if their compliance isn’t good enough.
This is not the first time the sepsis measure has come under fire. In May 2022, the National Quality Forum rejected an appeal of its endorsement of the SEP-1 measure. The forum, an affiliate of the Joint Commission, defines its role as helping to achieve consensus “on quality measurement and improvement standards and practices that achieve measurable health improvements for all.”
The appeal, which was filed by several of the same groups that authored the new recommendations, came after the NQF had endorsed the measure in 2021 for the third time. The appellants expressed concern about the way the re-endorsement process had been handled.
At a virtual appeals board hearing on April 29, Michael Klompas, MD, MPH, of Harvard Medical School, who spoke for the appellants, focused on the fact that two members of the appeals board were recused from participating due to conflicts of interest. “The apparent conflicts of interest, however, were nothing more than a deep interest in sepsis and having a record of having published technical critiques of SEP-1, including recommendations for possible modifications,” he said. “These are not conflicts in our opinion, but rather bona fides that demonstrate these members’ high-level interest and expertise in sepsis in SEP-1.”
But measure co-developer Sean Townsend, MD, of California Pacific Medical Center, in San Francisco, had a different view. The recused board members “published articles critical of the sufficiency of the evidence supporting [the measure],” he said. “One of the measure evaluation criteria is the sufficiency of the evidence. These opinions conflicted with their duties to serve as impartial judges.” In addition, said Townsend, “both committee members engaged in work with the measure developers. The NQF policy requires that if committee members have done so, that they must be recused.”
CMS Administrator Chiquita Brooks-LaSure was asked about the sepsis issue during a video interview with MedPage Today Medical Editor-in-Chief Jeremy Faust, MD. “Will CMS heed the call of the nation’s leading infectious disease experts and retire SEP-1 and replace it with something better?” Faust said.
“I can’t speak to our future actions, but I do know how much we take what the scientific community says and what the clinical community says seriously,” Brooks-LaSure replied. “We continue to review sepsis and other conditions.”
The agency was more specific in response to an emailed query from MedPage Today. “CMS is reviewing the proposed recommendations in more detail,” a spokesperson said. “However, at this time CMS believes the existing measure is most directly impactful to early evidence-based treatment of sepsis. CMS also notes that some of the recommendations appear to be challenging to implement on a national level because they are looking at interventions that are very dependent upon individual patient needs as opposed to a population of patients, such as optimizing antibiotic dosing.”