An interview with Charlie Harp, CEO of Clinical Architecture
We have more data than we know what to do with in healthcare. We have clinical data held in EHRs, results from lab systems, images from imaging platforms, as well as wellness data from personal devices and apps. So how do we sift through it all to find the relevant information? The solution starts with quality data.
I recently sat down with (well, maybe not “with” in the traditional sense – it was a virtual meeting) Charlie Harp, CEO of Clinical Architecture to learn about the importance of data quality in healthcare. Harp had some great insights on data – especially the new responsibility we have in Health IT to ensure the data we use is of high quality.
“To do value-based care well,” stated Harp. “You have to be able to communicate how you’re doing, understand how you’re doing, and then be able to leverage how you can improve how you’re doing,”
For years, data was a byproduct of activities in healthcare. Services were delivered to patients and records were kept of those encounters. In the beginning that data was primarily used for billing purposes, but with the growing sophistication of analytics, that data has newfound value. It can hold the key to improved treatments, operational efficiencies, and better patient engagement.
According to Harp, in order to unlock the true value of your collected data, you must first focus on improving its quality. What is quality data? Harp defined it this way:
These four “Cs” are the guiding principle behind Clinical Architecture’s solutions.
I think most of us understand the importance of the first three Cs – current, correct, and complete. Comprehensible, however, was not something I expected to see. Harp explained it simply – in order to leverage data in any software system, that system must be able to not only ingest that data, but also understand/comprehend that data.
In some data repositories, 60-80% of the most valuable information is stored in the form of unstructured text (aka clinical notes). To make this data comprehensible, it must first be codified, then normalized so that it can be made useful for other systems. If a system can’t understand that data, it can’t use the data.
Data can give us a complete picture of a patient, but it is difficult to know which parts of the picture are relevant to the task/problem at hand. A medical record can tell you that a patient broke their arm when they were 16 years old and that they were put on an antibiotic 5 years ago, but neither are useful to help diagnose why they are in your office with a fever and shortness of breath.
Harp would like to see us get to a place where systems know the right contextual information to use from the mountain of healthcare data.
At Clinical Architecture, Harp and his team are helping organizations understand and leverage data. But perhaps most importantly, they’re doing the heavy lifting organizations need to do to ensure the quality of their data. They recently helped an organization normalize medications and labs in over 100 facilities in just two months – a project that would have taken about 5 years to complete had that organization done it with internal resources. They know data, they know how to normalize it, and they know how to help organizations use it right.
Check out my conversation with Charlie Harp, CEO of Clinical Architecture to learn:
- What the “informonster” is
- How data gives people superpowers
- How to avoid “artificial uncertainty”
- Why we work in “big pixels” and what it means for bridging gaps in understanding
This interview is part of our #HealthIT100in100 initiative.