Thus, the increased benefit for extremely depressed patients seems attributable to a decrease in responsiveness to placebo, rather than an increase in responsiveness to medication

is this implying that the placebo effect is masking the drug effect in less severly depressed people?

It's certainly a bit wierd, the evidence is in figure 2 of the paper (btw, Public Library of Science (PLoS) journals like this one are all open access anyone that has a pressing need to give themselves a headache can read the actual articles behind the headlines).

I have no special authority to interpret the data but it seems plausible to me that the small improvement found in people with mild symptoms are masked by the placebo effect but for more seriously ill people, the majority of whom would get worse without intervention, the SSRIs are more effective. Or something like that.

RNZ did eventually talk to a chap who phoned in to say how whichever SSRI it was changed his life, and had a few emails too.

I wish they'd report the story properly in the first place instead of falling into the journalistic cliché "anecdotal evidence trumps what we thought study said" Can't wait for the <i>Listener's</i> inevitable front page story...

Isn't that why we buy Macs - everything can be done by either clicking or dragging and dropping?

You can probably do this trick which involves the a single keystroke before you can begin dragging and dropping

This "world" thing -- is there a torrent for that?

I wouldn't bother, HD has much better resolution...