Theodore Dalrymple looks at the NHS in the UK:

It turned out, however, that the costs of prevention were decidedly real, while the savings were inclined to be imaginary. This was for more than one reason. The bureaucratic costs of setting and monitoring health-improvement targets—which were often highly arbitrary—were far greater than anticipated, bureaucracies having an inherent tendency to increase in size and spending power. Many doctors started to be paid for procedures that they were already doing for no charge, like taking their patients’ blood pressure. Screening procedures turned out to be highly equivocal in their efficacy. Thus the overall benefit was much less than anticipated. Some of the more common ills that had been targeted, such as strokes and heart attacks, were in marked decline anyway.

Worse, much of the expenditure on the treatment of disease proved intractable. Technology inexorably increased costs; and even if the health of the population improved rapidly, so that 70 was the new 60, 60 the new 50 and so forth, the proportion of old people in the population meant that the proportion of people ill with expensive chronic diseases increased. In the U.S., there were 37 million people over 65 in 2006, just over 12% of the population. That figure is projected to rise to 71 million, or 20%, by 2030.

In my professional lifetime, procedures such as hip replacement have gone from being relatively new-fangled and exotic to being routine, precisely at a time when there are more people than ever who can benefit from them. Osteoarthritis is no doubt hastened by obesity, but no medical means has yet been found for the prevention of that particular condition.

The key to cost control?  Pay more out of pocket.