The Apixio Blog

The Path Forward for Healthcare Data is Deeper Insights on Patients and Doctors

A female doctor allows a young patient to try her stethoscope as she gains insight on his condition.

At this point in time, it’s fair to say that healthcare data has not transformed healthcare the way many people thought it would. We have a lot of healthcare data, for sure, but we don’t make effective use of it. This is largely because efforts to leverage healthcare data have been scattered and unfocused. Are we worried about data visualization, or the ability of laypeople to understand data? Are we worried about making sure we give patients more information about costs? Do we want to give them information about their doctor’s social views? Do we need to give employers a broader look at the apps they can offer to their employees? At times, it seems that health IT is all over the place, with no coherent vision for how to fix a broken healthcare system.

The role of cognitive computing

What’s missing here is the understanding that better healthcare doesn’t happen around the edges. It doesn’t happen by addressing symptoms of larger problems in healthcare, by building features instead of platforms, or by solving second-order problems. Instead, we should take a look at the broad challenges in the healthcare system and address them in an organized and targeted way.

Cognitive computing does just that. Many of the problems in healthcare stem from issues of asymmetric–or incomplete–information. Perfect information is a key component of an effective market system because it enables people to make rational decisions. Cognitive computing extracts and analyzes the 80% of healthcare information that’s located in the unstructured portions of medical records, enabling fully informed decisions.

Tackling risk adjustment and quality measurement

What are some of the most important decisions in healthcare, ones that truly require 100% of information? Risk adjustment and quality measurement might seem like unintuitive answers to this question, but they are the right ones.

Risk adjustment enables a better view of patients. Risk adjustment is the process of listing the number and type of chronic conditions a patient has, in order to understand how sick they are relative to the general population. When risk adjusting, a health system or insurer has to go through all the documents they have on a patient and confirm the accuracy of the patient’s  data points. The outcome of this process is a more complete and correct patient record.

Quality measurements enable a better view of doctors. Quality measurement is the process of going through the treatment decisions and outcomes of different doctors and understanding who made the right calls and who didn’t. Just like in risk adjustment, when measuring quality, a health system or insurer has to go through all the documents they have on individual patients, and confirm them for accuracy.

A starting point for more effective care

I think we can all agree that a good starting point for better, more effective healthcare is knowing who we are and who our doctors are. From this point, we can negotiate who should go where and pay how much for what. But it all starts with more information with the fundamentals of who the players are and what we can learn about them. And that only works if we can expand the information we’re working with to include unstructured data, which will yield deeper insights on patients and doctors.