Why are only some co-morbidities recorded in SurgicalPerformance but not others?

General Image Close up of male doctor using ipad (ipad image to be photoshopped yet)

 

SurgicalPerformance users sometimes ask why certain co-morbidities are not captured in our databases. For example, hypertension is very common but in SurgicalPerformance we don’t record it.

In SurgicalPerformance we record surgical procedures, clinical outcomes and also confounders of outcomes, such as medical co-morbidities, body mass index, ASA score and age. These confounders are often overlooked but can have a profound impact on the risk of surgical complications.

However, not all medical co-morbidities are associated with poor clinical outcomes.

Mary Charleson, a physician from Cornell University in New York researched what types of medical conditions were associated with patients’ death during their hospital stay.

She identified 19 medical conditions and weighed them according to their severity. More severe co-morbidities and a higher number of co-morbidities resulted in a higher score. Since then a number of publications confirmed that a higher Charleson score relates to a higher risk of death.

Since then the use of the Charleson index has been expanded to predict postoperative, surgical complications and has been validated (including by our group) several times. In one of our papers on laparoscopic versus open endometrial cancer surgery, the Charleson index had an independent and significant impact on the risk of postsurgical complications.

The major breakthrough of the Charleson score is that medical co-morbidities are captured in a structured way by selecting only 19 medical co-morbidities from a very long list. Would we record medical co-morbidities in an unselected way, we would need to work through a very long list every time we record patients details.

While hypertension is common, it is irrelevant in regards to surgical outcomes. In SurgicalPerformance we simply tick the relevant boxes and we are reassured that we capture clinically relevant information.