Unfortunately I'm a little out of the loop, relying on secondary research only. The last I read, suggested that the risk / benefit tilt point for statins is around 7% (the 20 % figure set by NICE has more to do with health economics rather than clinical benefit)- hence my tongue in cheek remarks about "pouring statins in the water."
As far as I can see, calcium scoring is no more accurate than population scoring, (Framingham/Sheffield Tables etc) ie, it can identify the risk of, say, you dropping dead of a heart attack in the next ten years, what it can't do (and this is the crucial point) is to identify, with any degree of certainty, which pi lots are going to drop at the yolk.
As you state, homo, anything more sensitive / specific could cause more problems than they solve.
Working in primary care, and being a pragmatist, I'd have to try and work backwards. I'd start by asking how many passengers have been killed as a result of unpredicted heart disease in the pilot.
Wouldn't we be better spending the money we use for stress ecg's etc on oily fish? After all, you never saw a Fijian have a heart attack until they discoverd Burger King.