“Sometimes I think we’re becoming more of a data analytics company than anything else”
Humana Chief Medical Officer Roy Beveridge, M.D
Why are top payers investing in analytics?
In a recent Fierce Healthcare interview, Humana highlighted their long term commitment to analytics. Humana’s focus on Medicare Advantage means that they are increasingly in data partnerships with hospitals to provide insights and support through population health tools. What is fascinating to me, is this type of partnership would never have been possible in a fee-for-service world, and reflects the move to more value-based-care.
Population Health analytics
Population health tools in this space need to work with the heterogeneous data sets payers receive and are used to characterize and support their members. Many groups I have spoken to stratify high risk individuals in these data sets; for example smokers with heart disease are identified and given guidance by payers on quitting and incentives for exercise programs. I especially admire the way value-based providers and payers are working together to allow advice on high risk individuals to be given directly to the clinicians.
As models of care change, risk stratification requires a more complete view of the individual, and relies on data sets that are vast in scope and complexity, and made up of large amounts of unstructured data in various forms. For example, payers I have spoken to have a growing availability of Electronic Health Records (EHR) from providers, in addition to their traditional claims data, OCR patient records and members’ call center notes. With this variety and volume of data, payers are very much in a big data analytics landscape.
Making use of structured and unstructured data with NLP
A key question as I see it is: with a wide range of business applications to support, how are payers to integrate big data sets to drive improved understanding of their members’ wellness and help reduce costs?
The structured data world is well established for payers, but does not provide the depth of detail that are now needed to fully characterize individuals. Insights into important information about member wellness, such as social determinants of health and life style factors, are trapped in unstructured text. This is where Natural Language Processing (NLP) really comes into play – you need it to translate text into discrete data. Existing analytics tools for structured data, such as SAS and data warehouses, are being joined by investments in Hadoop-type technologies to store and analyze these more varied data sets. The challenge, therefore, is to bring together not only structured and unstructured data but the systems that support these data types as well. Find out more about I2E's Extract Transform Load (ETL) solutions here.
Linguamatics Health is overcoming these challenges first hand. By setting up workflows to take files from Hadoop, extract clinical and other insights, and load these insights into a data warehouse, we have enabled these data worlds and systems to be combined. The NLP pipeline for heterogeneous files and data must be tuned to each source type, so algorithms for EHR data are slightly different to those for PDF patient notes. This ensures a higher level of accuracy and application of spelling correction and OCR error detection to the right document types. Ultimately, resulting in a customizable system that integrates data for our individualized payer customers’ needs.
I am excited about the alignment I am seeing of NLP with big data and analytical tools that is making unstructured data actionable in ways never before possible and enables a true 360 degree view of each member.
To find out more about the use of NLP and big data technologies...