Of ED patients who met sepsis criteria, most left without a diagnosis of sepsis, according to the results of a recent study.1 “Emergency providers have been saying for years that many patients who trigger sepsis alerts may not actually have sepsis,” says Michael Puskarich, MD, one of the study’s authors and an associate professor in the department of emergency medicine at the University of Minnesota Medical School.

Puskarich and colleagues analyzed 3,121 patients who met sepsis criteria in the first six hours of their ED visit over an eight-year period. “We were surprised that only one-quarter of patients who, for all intents and purposes had sepsis on arrival, were ultimately assigned an ICD code that would make them eligible for the SEP-1 core measure,” Puskarich reports.

Only 25% and 48% met explicit and implicit criteria, respectively, for a discharge diagnosis of sepsis. “Hospitals are devoting significant quality improvement efforts toward a large group of patients to whom the metric will never apply and in whom these interventions have uncertain effects, including possible harm,” Puskarich observes.

During the first few hours of ED care, it can be difficult to differentiate sepsis from other emergency conditions. To be safe, EPs might treat for sepsis even when it is an unlikely diagnosis. “But once patients start down these pathways, it can be difficult to stop,” Puskarich notes.

Some patients could be harmed from unnecessary sepsis interventions. Of 3,121 patients who met sepsis criteria but were not diagnosed with sepsis, 19% to 36% had at least one risk factor for harm from large-volume fluid resuscitation. “This study highlights two competing risks: the risk of undertreating sepsis and the risk overresuscitating a patient with another cause of their symptoms, with undertreatment of sepsis representing the greater risk,” Puskarich says.

The symptoms of sepsis can vary greatly from patient to patient. “This broad manifestation can result in misdiagnosis,” says William Hopkins, JD, a partner at Austin, TX-based Spencer Fane. “Unfortunately, there is no absolute test for the confirmation of sepsis as a diagnosis. Healthcare professionals are going to sometimes get it wrong.”

The provider’s malpractice defense hinges on medical decision-making and test results. “From a risk management standpoint, if the documentation is solid, that judgment, or even mistake in judgment, can be justified,” Hopkins offers.

If the patient is misdiagnosed with sepsis, it means there was a delay in treatment of their actual diagnosis. “This distraction and delay could result in that actual disease or ailment getting worse. That lost time could result in the further compromise or potential death of the patient,” Hopkins cautions.

Lack of specificity in sepsis criteria causes many ED patients to “fall into the sepsis protocol treatment pathway,” says Monika Smith, DO, MBA, chief of the ED at Virtua Our Lady of Lourdes in Camden, NJ. That could result in bad outcomes for patients who receive a different diagnosis. “Many clinicians lack understanding of the differences between actual clinical sepsis and Centers for Medicare & Medicaid Services [CMS]-required documentation for non-sepsis cases meeting sepsis criteria,” Smith says.

This can lead to ED providers treating sepsis criteria, rather than true sepsis. “These defensive medicine practices put providers and hospitals at risk for medical malpractice [and] financial and reputation losses,” Smith warns.

Many ED patients meet the CMS criteria for sepsis, but do not have clinical sepsis. An example is a patient with a history of atrial fibrillation taking anticoagulants who presents with altered mental status. The patient meets systemic inflammatory response syndrome (SIRS) criteria, based on chronically elevated heart rate and respiratory rate, and meets organ dysfunction criteria based on elevated international normalized ratio (INR). 

If the emergency physician (EP) includes infection on the differential diagnosis, then based on CMS criteria the sepsis bundle should be initiated. This includes early antibiotics. “Although life-saving in sepsis, early antibiotics, if misused and as a reflex therapy, can have a detrimental outcome should the case not be severe sepsis and the infectious source is never found,” Smith says.

Patients can experience allergic reactions or anaphylaxis, contract Clostridioides difficile colitis infections, or develop antibiotic resistance. “Over-aggressive fluid resuscitation in patients without septic shock may lead to increased morbidity and mortality,” Smith says.

Providers must give appropriate doses of intravenous fluids while avoiding fluid overload, as excess fluid may cause edema in the lungs, kidneys, brain, and other organs.

If a patient meets sepsis criteria, but does not have clinical sepsis, good documentation is needed. Smith says EPs can document “Patient does not have severe sepsis,” along with supporting statements why organ dysfunction or SIRS criteria may be caused by something else, such as chronic atrial fibrillation or renal dysfunction. “Compartmentalizing patients into sepsis purely because they have met the sepsis criteria can lead to anchoring bias,” Smith says.

If the provider makes an initial diagnosis of sepsis, the admitting team likely will continue on the same path and miss the actual diagnosis. “For example, misdiagnosis of CHF as pneumonia and initiating sepsis may delay diuretic initiation, leading to adverse outcomes [and] increased morbidity and mortality,” Smith cautions.


  1. Litell JM, Guirgis F, Driver B, et al. Most emergency department patients meeting sepsis criteria are not diagnosed with sepsis at discharge. Acad Emerg Med 2021;28:745-752.