Understanding Unstructured Data (Part One)

Learn all about unstructured data and how you can organize all of the data into your healthcare analytic tools.

Understanding_Unstructured_Data_Part_One.jpgAccording to analysts from Merrill Lynch, Gartner and IBM, unstructured data accounts for approximately 80 percent of the data in business organizations. Similarly, the Health Story Project estimates that 1.2 billion clinical documents are produced in the U.S. each year with about 60 percent containing valuable patient care information in an unstructured format.

So what exactly is unstructured data? It’s the kind of data that usually has to be captured, read and analyzed by a person instead of a machine and is stored as free text, making it difficult for processing by a computer. In the healthcare industry, unstructured data includes explanations of benefits, audio voice dictations, handwritten and typewritten notes, diagnostic images, e-mail messages and attachments, text messages, medical claims and more. Radiology reports are responsible for a vast amount of unstructured data.

Hospitals and physician practices generate a substantial amount of unstructured data because many of the standard methods for inputting structured data don’t fit into the typical clinical workflow. These methods include written reports from examinations, information entered by checking a box or using note fields, a drop-down menu and physician dictation notes which are typically in free text format. This kind of unstructured data is especially important because it often presents more of a complete background of a patient’s health and includes reasons certain healthcare decisions were made.

Although the implementation of EHR systems often result in a favorable amount of structured data, they also produce a large amount of unstructured data. For example, EHR systems will allow for free-form entry of clinical notes. While systems will allow for parts of the note to be captured in a structured way, typically only the parts of the note which are pre-designated as key elements are captured into a data structure. Capturing and classifying this free-form data can be a  key to enhancing your analytics offerings to customers.

Issues with Unstructured Data

Although unstructured data can include valuable information, it is often difficult to access and analyze, so traditional analytics tools don’t usually work with it. Most software systems don’t have the capability to interpret unstructured data, and providers require high-quality views of it in order to coordinate care or perform coding or reporting. For example, physician notes may include information about specific patient symptoms but may omit others, so other clinicians who access those notes may not be able to make the right treatment decisions.

Next Steps

By being able to analyze unstructured and structured data together, clinicians can provide a more complete picture of a patient’s history, diagnosis, treatment and outcome. In our next blog, we’ll outline some new techniques and technologies being used to help clarify unstructured data.

Learn More

Syntrix Consulting has the capability to help you capture, classify and integrate unstructured data from your EHR system into your analytics tools. Contact us to learn more about bringing all of your data together to help create a complete clinical picture of your patients and clinical outcomes.

For more information on what Epic reporting is and how to have a successful implementation, download our FREE eBook:

New Call-to-action

Similar posts

Stay on the leading edge of Healthcare Analytics

Discover new ways to enhance and optimize your data analytics function using the most advanced tools and industry knowledge available today.