The Unstructured Data Trailer
When the curtain opened on big data analytics, unstructured data entered the VIP line. Unstructured data in medical records especially is often text-heavy (think medical histories) or images, such as X-rays. In a word, unstructured data is unruly. It lacks an actionable field in the underlying database—making it more difficult to utilize traditional data mining techniques that are readily applied to defined data. Adding to the complexity is the sheer volume of information since according to estimates, 80% of the world’s data is unstructured. Meanwhile, only the remaining 20% of structured data is searchable by basic algorithms, (i.e., easily understood by machines).
Spotlight on Unstructured Data in Medical Records
The original medical record residing in a provider’s EMR includes mostly unstructured data. From the perspective of health plans receiving medical record copies that are PDFs, almost all data is unstructured. The result is that—without the types of machine assistance available for structured data—most healthcare payers are utilizing manual, laborious processes to find the information needed during medical record reviews. For example, within many health plans, resources manually index a medical record and scan or read specific sections of the imaged medical record to make a clinical or coding determination.
CAVO Takes the Unstructured Medical Record Data Stage
But what if healthcare payers could streamline these steps and directly search the unstructured data within a medical record, almost immediately locating the necessary information to make a claim decision? With CAVO, health plans can—and immediately—realize productivity increases for their payment integrity and/or utilization review departments with little IT lift.
CAVO intakes medical record documents and images from multiple platforms, creates searchable data, and establishes accessibility from a central, cloud-based repository. Our proprietary OCR technology creates full-text search capabilities, which renders medical record categorizing and/or indexing unnecessary. Plus, thanks to CAVO’s open architecture, health plans can utilize a broad platform of APIs to support their workflows, machine learning initiatives, etc. But we’ll save those details for discussion in a sequel! For now, let’s chat about CAVO – popcorn not included.