In a recent memo, the Centers for Medicare and Medicaid (CMS) announced that based on the available evidence, using genetics to predict warfarin responsiveness was not reasonable or necessary.   The evidence is mounting that they are wrong, and more importantly, this is a setback for the general field of genome-informed medicine.   They do not cite the recent paper we published with others providing a global algorithm and showing its advantages over clinical algorithm dosing or fixed dosing.  We estimated that nearly 40% of patients can be prospectively identified as requiring high or low doses using genetics.   As genotyping costs drop, it is clear that genetics can help get the dose more accurately.   The CMS did approve an intermediate action, which approves reimbursement in the setting of clinical trials.  If I was going to take warfarin, I would want to know my genetics up front for sure.   I tried to leave a comment to that effect on the CMS site, but evidently Mac browsers (Firefox and Safari) don't work--so I can't!  Double Argh!