when you record someone's BMI and it gets thrown in a spreadsheet as patient 123456 has BMI 23.5 the doctor is no longer looking at the patient. perhaps it's good enough, i don't claim to be able to evaluate the methodology of the study (and perhaps using BMI simply results in more skeptical results, by leaving healthy people out of the low risk group, which I'm fine with - not sure this is true though). it really was an incidental note on my part, but i stand by it. i am a little surprised at how much we're talking about it, i guess it is bikeshedding, which is understandable, im not qualified to comment with authority on much of the material myself :)