EHR data better for sepsis surveillance than claims data
Clinical surveillance using electronic health record data provides more objective estimates of sepsis incidence and outcomes than claims data.
That’s the contention of new research, led by investigators at Brigham and Women’s Hospital, which challenges the use of claims data, which traditionally has been used for sepsis surveillance. Results of the retrospective cohort study, published this week in JAMA, suggest that insurance claims using billing data do not accurately reflect sepsis trends over time.
“Sepsis is a devastating disease that causes a lot of morbidity, mortality and cost to the healthcare system,” says Chanu Rhee, MD, a critical care and infectious disease physician at Brigham and Women’s Hospital. “Quantifying its burden and changes over time is really important, particularly in light of the fact that there are numerous prevention and treatment initiatives that are going on in hospitals around the country. There’s an interest now in trying to measure the quality of sepsis care, which is challenging if you don’t have a reliable and objective way of tracking sepsis.”
According to Rhee, tracking sepsis using claims data is problematic because of the complex syndrome lacks a single confirmatory diagnostic test and tends to be under-recognized by clinicians “while coding can be influenced by reimbursement and policy incentives.”
He contends that his team’s research shows that “if you bypass those extra layers of subjectivity in diagnosis and the changing diagnoses in coding practices and go right to the clinical data” it “gives a clearer and more objective picture of what’s happening with sepsis.”
Leveraging EHR data from nearly 3 million patients admitted to 409 U.S. hospitals in 2014, researchers found that sepsis was present in 6 percent of adult hospitalizations—of whom 21 percent died in the hospital or were discharged to hospice—and in 35 percent of all hospitalizations that culminated in death.
“In contrast to claims-based analyses, sepsis incidence rates using clinical data were stable from 2009-2014; in-hospital mortality rates declined, but there was no significant change in the combined outcome of death or discharge to hospice,” states the article.
Rhee and his colleagues believe that the clinical surveillance method for sepsis is sufficiently developed and validated to the point that it should replace claims data and be routinely implemented in hospital operations around the country.
“Healthcare professionals, hospitals, and policy makers have a better understanding of sepsis incidence, outcomes and trends, which, in turn, could inform better treatment and prevention policies,” he concludes.
At the same time, Rhee emphasizes that their clinical data approach is not perfect. “Currently, there is no perfect way to diagnose and identify sepsis with 100 percent accuracy—it’s not possible,” he says. “But I think our study shows that it brings us a little bit closer to the truth.”