Buzz Word Bonanza (Part 2)

Nov 25, 2013 | Blog

Boston, MA—November 25, 2013—In my previous blog, I mentioned one of my favorite buzz words in healthcare data analytics, Big Data.  Here is another one that is worthy of comment – Single Source of Truth.

Single Source of Truth – this should be the title of a legal thriller movie.  Its application to healthcare data is the equivalent of trying to reverse the aging process.  Chances are, we will never get there, and spending our time and resources trying to achieve it may make us feel like we have wasted our lives.

Granted, we should do everything we can to clean up the existing data we have to make it useable for analytic purposes, but most healthcare organizations have literally hundreds of various data sources spread around the organization which are used for many different purposes.  Trying to garner all that data and information into one source, standardize it, and manage all of the analytic needs of a large, complex organization is an amazingly difficult task that will likely not yield the results you are looking for, since by the time you have “found the truth”, others in your organization will have already started collecting new data.

Instead, focus on data governance and use plans, which define and identify various data sources, how they are to be used, who should use them, and how to effectively and efficiently access them.  For large, common data sources such as EMR, billing, and supply chain data, combining them into one centralized source for analysis makes sense.

Regarding the concept of a single source of truth, here is an example of why that is a misnomer.  Healthcare provider organizations have tried to determine the cost of specific items or services for years.  Two common approaches for doing this are to implement cost accounting methods or to apply the ratio of cost to charges for items that are billed.  Many organizations use one or both of these approaches, and each will yield a different result for a particular item.

For those who have spent time analyzing healthcare data, you will agree that neither of these approaches provides you with the exact cost.  In summary, neither represents a single source and neither provides the truth, but we find them useful for directional analysis.  Let’s be practical!