Sepsis is a complex complication of bloodstream infections that most patients don’t know about, but most doctors fear.
It kills about 250,000 Americans a year, and one in three people who die in U.S. hospitals have it.
“Sepsis is just as serious as a heart attack,” said Olevia Pitts, the chief medical officer of Research Medical Center.
Last week, for the first time ever, the federal government publicly released scores grading hospitals on how closely they follow guidelines for treating severe sepsis and septic shock.
The scores, available on Medicare’s Hospital Compare website, show a clear pattern in the Kansas City area: the hospitals that fared the best are all run by for-profit national chains.
Four of the highest-scoring hospitals within 30 miles of Kansas City were St. John Hospital in Leavenworth (100 percent), St. Joseph Medical Center in south Kansas City (91 percent), Providence Medical Center in western Wyandotte County (89 percent); and St. Mary’s Medical Center in Blue Springs (88 percent). All are owned by California-based Prime HealthCare Services.
Research Medical Center, which also scored an 88 percent, is owned by HCA Midwest Health, a division of Tennessee-based Hospital Corporation of America.
Representatives of those hospitals say the new ratings are accurate and reflect, in part, the data and resources they can devote to sepsis care because of the size of their parent organizations.
Paryus Patel, the chief medical officer for Prime, said the company has developed its own electronic system to spot possible cases of sepsis, which is notoriously tricky to diagnose early.
Those cases are then flagged by a corporate anti-sepsis team that follows up with local hospital staff to make sure the patients are getting fluids, antibiotics and lab tests at the proper time, even if they’re transferred from one part of the hospital to another.
“That is very unique to Prime,” Patel said. “I’m not sure I’ve heard of any other organizations doing it.”
Prime moved into the Kansas City market in 2013, when it bought St. John Hospital and Providence. It purchased St. Joseph and St. Mary’s two years later.
Jackie DeSouza-Van Blaricum, the CEO of Research Medical Center, said her hospital is also able to leverage data from hospitals throughout the HCA chain to better spot the early signs of sepsis.
That didn’t translate to high scores for all of the HCA Midwest facilities (Centerpoint Medical Center in Independence scored a 46 percent). But DeSouza-Van Blaricum said Research Medical Center has been “laser-focused” on sepsis care because it treats the most complex medical problems in the HCA Midwest chain.
Last year, she said, Research became the first hospital in Missouri to earn a special sepsis designation from the Joint Commission, a national health care accreditation agency.
“We really do consider ourself a sepsis expert,” DeSouza-Van Blaricum said.
Patel said that the new sepsis scores on Hospital Compare are “pretty accurate,” while adding that all health care data have limits.
“It’s the best objective data we have out there,” Patel said, “and the good thing is, it’s apples to apples.”
But, as with other ratings Medicare has posted to Hospital Compare, academic medical centers that serve more low-income patients and people with complex medical needs said it’s not actually apples to apples, precisely because the measurements treat all hospitals alike.
“These numbers unfortunately do not really measure quality of care,” said Jeffrey Hackman, the chief medical information officer for Truman Medical Centers, a teaching hospital for the University of Missouri-Kansas City School of Medicine. “In fact, during the time period of this report, patients with sepsis at TMC actually survived at a higher rate than expected based on predictive models.”
Hospital Compare gave Truman Medical Centers’ Hospital Hill location the metro area’s lowest score, at 25 percent. But Hackman pointed out that Truman’s other location, the Lakewood campus, which is in a more affluent neighborhood, scored twice as high despite operating under essentially the same policies and procedures.
The sepsis scores are based on data from the first three quarters of 2017 and evaluate how often hospitals followed a specific treatment pattern within a few hours of recognizing sepsis.
Hackman said the step-by-step treatment process the hospitals were evaluated against has “been repeatedly revised” and “evidence suggests following the steps may be of no benefit to patients.”
Hackman’s concerns echoed those raised by other physicians in a June article about the new sepsis scores from The Hospitalist, a publication of the Society of Hospital Medicine. Several questioned whether the scores would have any correlation to patient outcomes, and one emergency room doctor called it “check-box medicine.”
Steven Simpson, a critical care doctor at the University of Kansas Hospital, helped write the guidelines the federal government is now using to form the sepsis scores.
He said they’re a good tool for hospitals that had no standard sepsis plan. But he also said they’re a “one-size-fits-all” approach and hospitals like KU, which scored a 29 percent, that have dedicated sepsis specialists are more likely to deviate from them to tailor care to each individual patient.
Simpson and Steve Stites, KU Hospital’s chief medical officer, both said the hospital is measuring itself more on patient outcomes than on whether it followed the federal sepsis treatment protocol, and the sepsis mortality rate at KU is 6.4 percent.
“That’s as good as it gets, anywhere,” Simpson said.
Sean Benson, vice president of Wolters Kluwer Health, said the sepsis scores should be taken with a grain of salt because they’re based on a sample of patients whose Medicare bills were coded for severe sepsis, not the entire patient load.
Because they’re new, Benson said, the federal measurements are also likely to change, as previous ones have.
“My guess is that there is going to be a lot of fine-tuning, adjustments, tweaking going on over the next two quarters,” said Benson, whose company sells an electronic sepsis alert system to hospitals.
Still, Benson said hospitals should take a deep look at their scores. All of them, he said, can probably do better when it comes to sepsis.
“Whether you scored 10 percent or 100 percent, you can’t really state that that definitely shows how you’re performing on your sepsis care,” Benson said.