Impacting sepsis rates starts with a commitment by organizational leadership to prioritize the issue, according to one adviser.


Oct. 4—Amid ongoing measurement challenges, new research indicated hospital sepsis rates stabilized in recent years.

The findings should be taken with a grain of salt, according to a hospital sepsis adviser, because they come amid steep measurement challenges, acknowledged questions about accuracy, and a lack of a “true gold standard for [definitions around] sepsis,” as noted by the study authors.

“If you think about 2009-2014, it’s not that long ago, but in many ways it was a long time ago,” Madeleine Biondolillo, vice president of quality and safety at Premier, said in reference to changing guidance related to sepsis and the new ways that hospitals are identifying and tracking cases using electronic health records.

Amid the caveats, the authors found in 2009-14 a “stable” 0.6 percent annual increase in the incidence of sepsis and a 3.3 percent annual decline for in-hospital sepsis mortality when referring to clinical criteria. In comparison, when using claims data, they found a 10.3 percent annual increase in the sepsis rate and a 7 percent decrease in mortality.

The authors of the study, which was published in September in JAMA, found that “clinical criteria were more sensitive in identifying sepsis than claims” data, with comparable predictive capabilities.

The incidence rate appears to be an improvement compared with other sepsis findings in recent years, including a 2013 study in Critical Care Medicine, which found a 13-13.3 percent average annual increase in the incidence of severe sepsis in a nationally representative sample of hospitals in 2004-09. However, that earlier study also found in-hospital mortality decreased during the six-year period by a cumulative 12.1 percent to 35.2 percent, depending on the measurement used.

“Rates are a function of what you’re checking for,” Biondolillo said, referring to the impact of evolving sepsis-detection guidance that moved toward an increased use of laboratory analyses, for instance, and the spread of those approaches as standards of care.

 Overall, the authors estimated a 5.9 percent annual sepsis incidence among hospitalized patients and a 15.6 percent in-hospital mortality rate, which would translate to about 1.7 million adult hospitalizations and 270,000 deaths in 2014. Those rates fell within a broad possible range of sepsis incidence and death as projected by the 2013 study. The incidence rate was larger than the share of inpatient stays—3.6 percent—for which septicemia was given as the reason for hospitalization, as compiled by the Agency for Health Research and Quality (AHRQ).

The group of 350 hospitals with which Biondolillo’s company works to identify and spread best practices in sepsis prevention, detection, and treatment recently reported an 18 percent overall decrease in sepsis mortality in 2008-16.

“That really does prove that if organizations prioritize this issue and come together to really do that collaborative work across a cohort—look at one another’s data, benchmark it, implement improvement strategies, and share those interventions—they can have a huge impact on mortality from sepsis,” Biondolillo said in an interview.

Hospitals also have a big financial incentive to address the condition, which was the costliest among all those treated by hospitals, according to the latest accountingby AHRQ. Sepsis drove 5 percent of all Medicare readmissions and carried $3 billion in costs in 2013.

Keys to Improvement

“It’s not magic; it’s the same type of effort that goes into any quality-improvement initiative,” Biondolillo said.

Impacting sepsis rates starts with a commitment by organizational leadership to prioritize the effort.

“Right from the C-suite there has to be a message to everybody, ‘We care about our sepsis mortality rate and we are going to bring it down,’” Biondolillo said.

Next, the relevant departments need to analyze issues that are evident in the relevant data, find ways to look at themselves critically, and compare their performance to the latest guidelines and known best practices. The examination should identify areas where the organization is not providing reliable care for the condition and the barriers to doing so.

One Premier client hospital that had particular success, Frederick Memorial Hospital in Maryland, reduced observed-to-expected mortality ratios for sepsis from 1.53 in 2012 to 0.65 in 2016, with a few months in which the ratio was as low as 0.10, the hospital noted in a webinar.

“They determined that the most effective way of alleviating those bottlenecks would be to create nurse-driven protocols,” Biondolillo said. “Once they were able to get that buy-in, they were able to do other things because they had built the trust as a team and the leadership was encouraging of this.”

Pre-Hospitalization

The latest sepsis-rate findings came a month after the Centers for Disease Control and Prevention launched a campaign, Get Ahead of Sepsis, to encourage early recognition and timely treatment of the condition, and efforts to prevent infections that could lead to sepsis.

Hospitals have embraced that call in different ways, including seeking to educate patients about the warning signs of sepsis and encouraging them to alert primary care clinicians, emergency medical service (EMS) technicians, or in-hospital staff if the signs occur.

“Involving EMS in the continuum of care is a very important piece of the puzzle,” Biondolillo said.

About 87 percent of sepsis cases were present on admission, according to the JAMA study, and EMS personnel would be in a position to provide early diagnosis and treatment, including hydration.

Premier is developing protocols to train hospitals’ EMS in sepsis diagnosis and treatment, and plans to start implementing the training for member hospitals in early 2018, according to Biondolillo.

Hospitals also need to reach out to primary care practices to help identify patients who are at risk for developing sepsis, to prevent the onset of the condition, she said.

“These patients are known; they are often people who are complex, with lots of comorbid conditions, usually frequent visitors to the emergency department, and in some cases frequently admitted to the hospital,” Biondolillo said.


Rich Daly is a senior writer/editor in HFMA’s Washington, D.C., office. Follow Rich on Twitter: @rdalyhealthcare

Publication Date: Wednesday, October 04, 2017