This story from a week ago made newly salient thanks to the Connecticut school shooting is, among other things, a great illustration of the principle that you don't want to worry too much about rising health care spending. It turns out that while the murder rate has been falling for 20 years, over the past ten years the number of people getting shot has actually increased. The murder rate continues to fall because if you get shot and then the emergency room staff saves your life, there's no murder. And emergency medicine has gotten much better. A related development in terms of improved battlefield medicine is one reason that many fewer American soldiers died in Iraq and Afghanistan than lost their lives in Vietnam.
Now if you get yourself in the frame of mind where minimizing "costs" (meaning spending aggregates) is the be-all and end-all of health care policy, this looks like a disaster. Guy dies of a gunshot wound on a Thursday night, and all the costs from Friday onward are mortuary costs. Trauma surgeons save the guy's life on Thursday night then you're looking at a potentially expensive recovery process for days, weeks, or months.
Obviously, though, nobody actually thinks it's bad that gunshots are becoming more survivable. But they often say things that seem to logically commit them to that view, treating "health care" as the name of a stationary entity and wishing it would just get cheaper.