WP estimates roughly one million car accident deaths per year (worldwide est for 2007). Heart disease numbers are more like 30M/year, although lots of those people were very old, so it's hard to estimate what's preventable. Compare that with a hypothetical meteor that kills one hundred million people, in a confined geographic area (that's 1/7 of Europe, or 1/5 of N America, or just half a dozen large Asia cities). What're the odds of that meteor? Even the once-per-millenium Tunguska event wasn't large enough to do that much damage. But suppose once-per-millenium meteors WERE deadly enough to kill 100M people, that's still 1/10 the deaths per year. If it's more spending-efficient to prevent the deaths from heart disease, do that. If it's more spending-efficient to prevent the deaths from meteor impact, do that.
On the other hand, extinction events are worth treating with extra care. Which in fact is exactly what meteor watches do, which is why they only look for entry diameters over 100m. Seems reasonable! Extinction events have happened 5 times in the last 540 million years. They're pretty rare. Maybe we should just roll the dice.
Off-topic: That's not what "shrift" means. Maybe you'd like "short order" or just the clearly "in a short time".
My point was rather that you can't compare a catastrophe with disease / accidents simply in deaths-over-time. Catastrophes have dramatic economic consequences as well.
My reply to your core point is that you can so compare (not least because deaths-over-time have dramatic economic consequences too). Maybe we're just gonna disagree about this one. :)
If we have the observatories in place, we should have enough time to come up with a reasonable defense.
I believe if we're wiped out by one of these rocks, it'll be due to our own negligence.