1. the boat has pretty much sailed on this.<p>2. correlation is not causation.<p>3. they've adjusted the numbers based on age instead of breaking out the numbers by age. So, it's very possible age is important in making a distinction. There's no break out of pre-existing conditions among the groups.