CMS Rules for Treating Sepsis Not Linked to Mortality Benefit, Study Finds

ORLANDO — Following an exhaustive literature search, researchers were unable to find compelling evidence that mandated sepsis treatment guidelines from the Centers for Medicare & Medicaid Services (CMS) were favorably associated with mortality.

Of 12 observational studies in their systematic review that addressed the relationship between mortality and compliance to the Severe Sepsis and Septic Shock: Management Bundle (SEP-1), five showed a statistically significant benefit while seven did not, reported James Ford, MD, of the University of California San Diego.

On the question of whether guideline implementation was associated with sepsis mortality, one study showed benefit and four studies did not, according to findings detailed in a late-breaking presentation at the annual meeting of the Society of Critical Care Medicine (SCCM) and published in Annals of Internal Medicine.

“The take-home message: No moderate- or high-level evidence to support [that] SEP-1 compliance or implementation is associated with mortality,” said Ford. He suggested that rigid standardization might not benefit patients or the healthcare system.

SEP-1 was implemented by CMS in October 2015 as a pay-for-reporting measure. The bundle of care requires that all patients with possible sepsis or septic shock have lactate and blood cultures drawn, be administered broad-spectrum antibiotics, and be given intravenous fluids — all in the first 3 hours.

But the sepsis measure has been under fire from groups including the Infectious Diseases Society of America, the American College of Emergency Physicians, and the Society of Hospital Medicine for pushing clinicians to treat all patients who present with sepsis-like symptoms as if they have sepsis. They argue that many serious conditions present in a similar fashion, and that the bundle can be harmful to patients who are not infected, those who are infected with viruses and not bacteria, and can lead to overuse of broad-spectrum antibiotics.

Nevertheless, in a final rule from 2023, CMS said that starting in fiscal year 2026, hospitals will be financially incentivized to comply with SEP-1 as part of the Hospital Value-Based Purchasing Program.

At the very least, Ford said, the study findings suggest that CMS should reconsider that decision.

“I agree that this study shows that one size does not fit all,” said SCCM session moderator Tamas Szakmany, MD, PhD, of Aneurin Bevan University Health Board and Cardiff University in Wales.

“We have similar strong recommendations in our hospital systems as well,” Szakmany said, “that are slightly different from the SEP-1, mostly in the amount of fluids we are required to give, but even that is not universal and even that has not been shown consistently to be the best way of practicing medicine.”

Szakmany suggested a randomized trial be conducted to figure out the best way to handle patients with sepsis.

“I don’t know how difficult that would be, but it would especially have to be done if this is a federal mandate,” he said. “There are healthcare systems outside the U.S. that are similar to the U.S., which might make a randomized trial feasible.”

For their report, Ford and colleagues scrutinized 4,403 unique papers on SEP-1 and eventually identified 17 observational studies for inclusion in an intended meta-analysis.

That plan was abandoned given the heterogeneity of the studies in respect to the populations studied, the interventions studied, the mortality outcomes reported, the measures of association reported, the adjustment for confounding, and the type of study and analysis performed.

For example, of the five studies showing a benefit with SEP-1 compliance, two were small, single-center studies. The other three only looked at components of SEP-1 — just the 3-hour bundle, just septic shock patients, or just Medicare patients; the largest of these was a study of nearly 250,000 Medicare recipients, which showed a modest benefit in sepsis mortality. However, another study of similar size, again involving Medicare recipients, found no mortality benefit.

Ford suggested that the overall level of evidence in the studies would fall into the “low” quality of evidence, since all the studies were observational and there were no randomized trials.

Please enable JavaScript to view the

comments powered by Disqus.