Using Crowdsourcing Principles for CDS

One of the emerging challenges for Clinical Decision Support systems is the problem of how to describe the data that is input into the rules engine. For a detailed description of this problem, see Keith Boone's Standards Activity Around Clinical Decision Support.

Of course, with Meaningful Use still in Stage 1, the emphasis is currently on data capture rather than clinical decision support. Healthcare IT has an article on how Google's IP from its failed Wave product may be of help in digitizing hard-to-read physician notes.  Most providers today perform some form of scanning of physical files; the approach proposed in the article essentially involves principles of "crowdsourcing" where a complex translation is farmed out to a large number of people that work on tiny bits of the problem, usually only for a few seconds each.

When the focus shifts towards CDS in Stage 2 of Meaningful Use, a similar crowdsourced approach could be used to create CDS rules databases. Hosted EMR vendors like athenahealth are most likely to be in a position to mine their databases for such an effort.