The problem with this study is that it's comparing self-selected groups: those who drink lots, little, and no coffee/tea. Their conclusions, drawn from comparing coffee addicts to people who don't drink coffee is a little silly. It assumes that everyone's "baseline" is the same, and ignores the well-known fact that people to take caffeine are, like people who smoke cigarettes for nicotine, often self-medicating for a neurobiological deficiency (i.e. in this case probably dopamine). To compare the levels of someone who self-medicates taken off their medication to someone who doesn't need to self-medicate is a complete load of B.S..