Q&A: Querying for more accurate pediatric sepsis reporting

Q: What have been the most common queries for pediatric sepsis that providers encounter? In other words, what criteria should coders look out for?

Wendy Arafiles, MD: There are a couple angles. We sometimes send out responses to clinical validation queries when we, as clinicians, document sepsis or do so ambiguously.

For example, for a patient who looks like maybe decompensating on the floor, we might put in our notes “febrile neonate with rule-out sepsis” or “rule-out sepsis evaluation.” In that case, the queries we receive might ask us to validate the diagnosis: “Was it sepsis, or was it not?”

I, personally, don't have a ton of opportunity to add sepsis as a diagnosis. If the patient truly does have sepsis, they’re in the intensive care unit (ICU) in most cases, and our ICU team is obviously very keyed into that diagnosis and is documenting that. Most of the time, if they're in the ICU, there's some sort of organ system dysfunction.

When it comes to clinical criteria, our institution at Phoenix Children's has decided that the diagnosis of sepsis requires some evidence of organ system dysfunction. So, it gets a little tricky when we're looking at the coding for sepsis versus severe sepsis versus septic shock.

We certainly do look for queries for septic shock if we're using vasoactive substances or drips. And we do query for severe sepsis when there's evidence of organ system dysfunction that's not documented, which is fairly rare.

More often, I think we look for a validation of sepsis as the diagnosis because there isn't enough documentation. There may be clinical evidence, but there's not enough documentation for us to support that diagnosis to be included on the list, so we query.

Amy Sanderson, MD: We have received queries regarding our institution’s preferred phrasing. Our institution uses specific nomenclature when it comes to sepsis. We say, “SIRS [systemic inflammatory response syndrome] physiology” or “sepsis physiology” instead of “sepsis” or “septic shock.” Another thing we like to say is “concern for sepsis.”

For example, the provider might say “the patient is admitted to the ICU on three pressors with concern for septic shock,” and the coders might ask “Now, do they actually have septic shock, or are they at risk for septic shock?”

So, we, as clinicians, look for those things because we know what people are trying to say. They're just sometimes not saying it in the right way. So, we look at those clues. But at our institution, problems with institutionally preferred language are particularly prevalent. So, coders should look out for those.

Editor’s note: This article originally appeared in JustCoding. This question was answered by Wendy Arafiles, MD, pediatric hospitalist at Phoenix Children's Hospital in Arizona, and Amy Sanderson, MD, intensivist at pediatric Boston Children’s Hospital and is an assistant professor in anesthesia at Harvard Medical School, during the HCPro webinar, “Defining Pediatric Sepsis: Saving Lives with Early Recognition.”