Digital Waste or Patient Care?
Across all industries, unstructured content represents 80 percent of data. The mere 20 percent is structured.
Unstructured data is vital in just about any industry, and perhaps in healthcare it represents a significant untapped resource that the industry is still struggling to exploit. Capturing clinical data has been a key objective for decades, funneling the data into downstream workflows to gain more in-depth functionality across a wide range of applications from billing to population management.
Adding more and more structured fields, covering everything from a sprained ankle to the intricacies of open heart surgery, has resulted in providing an incomparable level of detail to cover any minute piece of information that may, or may not, be useful. This has also generated digital waste and administrative cost, and whether the quality of patient care has been improved is debatable.
With the rise of the cloud and seemingly unlimited storage, big data seems particularly alluring to healthcare entities that recognize the challenge of extracting value from unstructured content, to capture currently inaccessible nuances and insight locked away in unstructured and semi-structured fields.
To address this, the approach has been to force fit a square peg into a round hole. Despite herculean efforts, there is just no way that a square peg will fit into a round hole, no matter how hard you try. Although you have to give credit to these vendors for their determination and tenacity. The same applies to big data, regardless of industry, except the stakes aren’t as high.
But it’s not a matter of when healthcare must garner meaning from free-form or semi-structured text, the replace the outdated paradigm, it’s a matter of necessity to derive what industry experts like to refer to as ‘actionable intelligence’, to not only improve operations but to also improve patient care.
The earliest traceability of the term ‘natural language processing’ entering vocabulary occurred in the seventeenth century, yes the seventeenth century. Today, it’s still breaking news to many. Talk about getting lost in the shuffle. By definition, natural language processing at its very base level is the ability to build systems that can understand language. A 360 degree turn from identifying a structured data field, such as a social security number.
The problem remains. The majority of vendors can’t or won’t seem to leave the traditional approach, the square peg into a round hole scenario. Time’s a-wastin’. Content is ballooning out of control, patients are expecting, if not demanding, new levels of care, and the medical professional is hampered by making decisions based on human intelligence, experience, and reasoning because they can’t access the data. All round a no-win situation for everyone. Unfortunately, the consequences can be a matter of life and death.
If you are a medical professional, do you feel access to unstructured content would be advantageous to improve decision making, a higher level of care, and the bottom line? Thoughts on your experience welcome.