Sending Sepsis Patients Home May Not Harm Them

New research finds no difference in 30-day mortality among sepsis patients treated as inpatients versus outpatients.

Researchers at Intermountain Healthcare are challenging the conventional wisdom for sepsis treatment in emergency departments.

"Based on our data, we are trying to document something that was previously unrecognized in the literature—a significant fraction of patients with clinical sepsis are not admitted to the hospital after presenting to the ED," says Ithan Peltan, MD, MSc, an attending physician at Intermountain and a leader of the research effort.

Peltan and his research team studied 8,239 adult ED sepsis patients at two tertiary hospitals and two community hospitals in Utah. The researchers found that 1,607 of the patients—19.5% of the total—were discharged rather than admitted to the hospital.

"The conventional wisdom assumes that all sepsis patients coming to the emergency department are being admitted, but our data shows some are being discharged. … We need a reconceptualization of who these patients are and how our care guidelines are being formulated," Peltan says.

Peltan's team presented the research last month at an American Thoracic Society conference in San Diego. Although the findings are preliminary, the researchers found that it is probably appropriate for some sepsis patients to be discharged from an ED into outpatient care.

"There was no significant difference in 30-day mortality for discharged versus admitted sepsis patients," the researchers wrote in the abstract they presented in San Diego.

Discharge safety uncertain

The researchers have shown that ED physicians are sending some patients home, and the next step is to characterize which sepsis patients are appropriate for ED discharge, Peltan says.

"We are not at the point where we can recommend routine discharge of any sepsis patients for outpatient management in the community."

He says there likely are several factors that determine whether a sepsis patient in the ED is a good candidate for discharge:

• Patients who are not gravely ill and are not in need of intensive care intervention
   
• Patients who are not at high risk of deterioration
   
• Patients who can get the care they need as outpatients such as compliance with prescribed medications
   
• Patients who can set and attend follow-up visits

 A major element of safely discharging sepsis patients from EDs is developing a risk stratification methodology for sepsis similar to risk tools created for pneumonia, Peltan says. "We need that kind of risk-stratification tool for sepsis."

Risk stratification will help ED physicians sort out the best care path for sepsis patients, he says. "Who are the patients who need to be admitted? Who are the patients we might miss but need to be admitted? Which patients can be managed as outpatients?"

Physician decision-making varies

Peltan's team found significant variation in ED physician decision-making on whether to admit or discharge sepsis patients.

"We looked at physician-level behaviors and found some physicians did not discharge any of their sepsis patients and some physicians discharged nearly 40% of their sepsis patients," he says.

The decision-making variation is a valuable data point, Peltan says.

"Somewhere in the middle, there probably is a happy medium within that range of variation."

The final version of the Peltan team's research is slated for publication in 2019.