Hospital Medicine and EHR for data scientists

As a data scientist and physician scientist trainee I have a uninque perspective on EHR medical data. I’ve used it in production as a 3rd year medical student reviewing labs and vitals before rounds, helped generate it as a 4th year medical student putting in placing lab orders and struggled with analyzing working on clinical research projects utilizing CU’s Health Data Compass

What I hope to accomplish in this post is provide the insight and context medical providers have about how and why EHR data is created to help new data scientists working with it make sense of it.

What is this mess

When I first started trying to work with EHR data it was painful and confusing (and arguably still is). It’s worth saying right off the bat that the structure of EHR data is the way it is because it was designed first to simplify billing, then to be used for lab results.