Mike learned of the value of data early in his career. While working at a major EMR vendor in 2001, he led a project to help identify patients who were affected by drug recalls. He continued his work in various roles at Allscripts, including reporting, data exchange and systems architecture. From 2006 to 2015, Mike led the technology group at Galen Healthcare Solutions. While the company and his team grew by 50% annually during this time, they became known for excellence, earning awards like Best in KLAS for Technical Services and a Best Place to Work by Modern Healthcare. Mike joined Health Catalyst in 2015 to help with strategic client implementations. He has since joined the product development team to lead Health Catalyst’s text analytics initiative, making information previously locked in text notes available to Health Catalyst’s apps and data architects.
As health systems nationwide advance in their analytics journey, many are ready to expand beyond traditional discrete data sources—“regular data”—and are cautiously curious about the hype around “big data.” In Gartner’s latest report on big data in healthcare, it identified clinical text (physician notes, or radiology and pathology reports) as the most relevant big data source for health systems. In fact, most health systems employ clinical chart abstractors, willing to endure the burden and delay of hiring a team of nurses to manually extract nuggets of information from the free-text content of their medical records because it is invaluable to evaluating medical performance. The work is difficult, time-consuming, costly, and unavoidably retrospective; because of the expert manual effort involved, the questions we are able to ask of free-text data is severely limited to the most essential, often dictated by reporting requirements.
Imagine being able to automate discovery of the left ventricular ejection fraction from the various free-text documents (diagnostic testing reports, cardiology notes), and transform this information into computable data that can be tracked and visualized to monitor progression of heart failure among your patients, enabling analysis of the impact of treatments and exacerbating events, and allowing for the triggering of interventions. Or, imagine we could begin to automate cardiac risk calculation in advance of surgeries by uncovering recent EKG changes, key findings from cardiac stress testing and any recent changes in chest pain symptomatology in addition to the already computable data in blood pressure, ICD-10 coded problem lists, and lab values. The presentation explores the possibilities of text analytics.