October 13, 2017 – Mining the EMR


Anchor lead: What can we learn from data extracted from electronic medical records? Elizabeth Tracey reports

Sepsis, where someone develops an infection that spreads throughout their body and can be life-threatening, appears to be increasing. Yet a new study published in the Journal of the American Medical Association demonstrates two very different conclusions relative to sepsis incidence based on data from the electronic health record. Kevin Gerold, a critical care medicine expert at Johns Hopkins, explains.

Gerold: There’s two kinds of medical information being entered into the electronic record. One is the objective data, the ekgs, the x-rays, the laboratory studies, then there’s the clinician entered data that is highly subjective.   :15

The objective data found stable rates while the subjective data showed an increase.

Gerold: It sends a signal that we need to be careful about what data we extract from the electronic medical record for what purpose. For epidemiologic purposes we need to use the objective data. For direct patient care we can use that subjective impressionistic data, that’s very effective for patient care.  :18

At Johns Hopkins, I’m Elizabeth Tracey.