"Data driven" is a relative term. The ICC and Canadian Criteria are a good start in that experienced clinicians and researchers came up with the definitions based on their experience. But clinical experience is just the first step; for a definition to be accurate and feasible, it must be tested on multiple and large groups of patients and this is what I think the CDC might be trying to do. I have a lot of reservations about the CDC but I do respect both Natelson and Kogelnik for their work. Here is something on levels of evidence that is very technical but might be of interest. Randomized controlled trials are at the top and expert consensus based on cases are at the bottom. For diagnosis, it's a little different -- you need a derivation cohort from which you come up with your definitions and then a validation cohort. http://www.cebm.net/index.aspx?o=1025 Supposedly, Dr. Wanda Jones says Dr. Nancy Lee is "data driven." I hope that is true since much of what the federal gov't has done with ME/ CFS has not been data driven.