Big Data

How Cognitive Computing Will Get Healthcare Out of the Dark Ages in 2016

  • author image

Today, healthcare has a data problem. Each year the U.S. produces 1.2 billion clinical-care documents. If all the data in these documents were aggregated and made available, it would lead to groundbreaking insights. This would also enable more accurate and effective patient care. But this is not happening, at all. To use a blunt metaphor, data-wise, healthcare is in the Dark Ages. 

The problem is that nearly 80 percent of the patient clinical data is unstructured and in clinical notes. The data is “unstructured” in the sense that it is written in free-text form in the medical record. Traditionally, computers have not been able to interpret meaning from uncategorized medical text. A cognitive computing platform (think, IBM Watson) could be trained (or taught) to “read” this data, but it’s something few healthcare organizations currently have access to. 

This data is also tough to access because the average medical chart is stored in multiple fragments and formats across various locations and systems. For example, your primary-care physician has a record of you but not the record from your cardiologist or from the ER doctor you saw six months ago for bronchitis. Without a unifying system to access, process and interpret medical record data, much of it is left unused and unanalyzed. 

The fact that 80 percent of patient data is not used for better care should be infuriating to anyone who cares about high-quality healthcare. Imagine having only 20 percent of the information about a car before purchasing it. In this case, you would know what the car looks like but not know the gas mileage, the quality of the audio and display, the way it rides, the warranty terms or even how it is priced. This would be an unacceptable state of affairs in the automobile industry (as Volkswagen now knows well) and should be so in healthcare.

2016: The year that cognitive computing transforms healthcare

In 2016, cognitive computing software will use statistical techniques that can assemble individuals into cohorts. This will enable physicians to answer questions about what treatment works best in the care of specific diseases in specific populations in real-world settings.

Tens of millions of clinical records are generated and updated daily, which provides a treasure trove of knowledge about clinical medicine and healthcare delivery. Cognitive computing software will tap into and analyze healthcare patients’ messy, unstructured data and change the delivery and consumption of healthcare in several important ways. Here are three predictions for how cognitive computing will transform healthcare:

  1. Care providers will have timely access to the complete patient care profile, which will have a dramatic impact on providing user-directed or personalized care. For example, when faced with a 30-year old African-American woman with a new diagnosis of type 2 diabetes and a history of high blood pressure, a physician could draw on the database to use the experience of similar patients with this profile to determine the best initial medication to control her blood sugar.
  2. The insights from unstructured data will enable timely feedback on clinical decisions. How does the application of a general evidence-based guideline work for a person of a specific clinical profile? For example, does watchful waiting for early-stage prostate cancer in an older man result in a favorable outcome?
  3. Physicians will use the widespread adoption of electronic health records (EHRs) and unlocked unstructured data to drive clinical trials that are truly tailored to the individual, rather than randomized. A greater number of treatments could be studied for their efficacy (otherwise known as comparative effectiveness in the medical literature), and a greater body of solid literature could be established for what does and does not work at an individual level. By coupling this knowledge with the relative cost of treatments, we can assemble a value-based rating for each treatment – the amount of incremental quality or outcomes achieved for each dollar spent.

The time for a data science intervention is now

The time is now for a data science revolution in healthcare: the cost of computing is getting cheaper with the cloud, the use of electronic records is strong and new payment models are in place to achieve high outcomes and provide strong value. 

Healthcare and technology are at a crossroads, and 2016 marks the cusp of a data-rich healthcare future. If doctors and healthcare organizations know more about patients, they can make more informed decisions that will supercharge the value of care. 

Dr. Darren Schulte is CEO of Apixio. Apixio’s Iris is a cognitive computing platform designed to bring advanced data insights to healthcare. He has over 11 years’ experience in healthcare analytics and technology. He served as the chief medical officer and president of Apixio prior to becoming CEO in 2014. Earlier he served in executive roles at Alere, Anvita Health and Resolution Health. He co-developed 25 clinical measures to assess ambulatory care quality measurement. Follow him on Twitter. 

 

 

 

 

 

 

 

 

 

 

Post Your Comment




Leave another comment.

In order to post a comment, you must complete the fields indicated above.

Post Your Comment Close

Thank you for your comment.

Thank you for submitting your comment, your opinion is an important part of SandHill.com

Your comment has been submitted for review and will be posted to this article as soon as it is approved.

Back to Article

Topics Related to this Article