The primary goal of health wonks and health policy activists is to generate plans and systems that allow us to save money while also providing better health care. That’s what the folks at the Trust for America’s Health are presumably after with their new initiative to encourage policy makers to “move from sick care to health care” by spending more money on the prevention of chronic diseases. Given that about three quarters of U.S. health care spending is on chronic disease, much of which is theoretically preventable, this intuitively seems like a good idea that will save money by making us healthier.

Sadly, that’s not the case. Most economists agree that large scale efforts to spend more on prevention end up costing a lot more in the long run. The problem is that you end up spending a lot more on preventive care for people who were never in need of it just to catch the few who will eventually get the diease.  As the Congressional Budget Office noted during the ObamaCare debate, “for most preventive services, expanded utilization leads to higher, not lower, medical spending overall.” Other studies and experts have come to basically the same conclusion. Indeed, there’s been evidence that prevention programs don’t save money since at least the mid 1980s.

Yet we continue to see proposals to focus on prevention anyway. The appeal is understandable, but with a few narrow exceptions for things like hypertension screening, these proposals are usually misguided.

Much of the appeal, I suspect, stems from the desire to take active measures to save money on health care—to do something. But if we want to save money on health care, then that will probably involve doing less. Perhaps a lot less.

That’s because a large amount of medical care is probably unnecessary.

How much, exactly, varies based on the estimate: Health researchers at Dartmouth Institute for Health Policy and Clinical Practice have suggested that about 30 percent of clinical spending could be avoided without impacting health outcomes. Former Medicare administrator Donald Berwick helped author an estimate suggesting the figure could be as high as 37 percent. But the general consensus is that the number is high.

Much of that unnecessary spending comes from providing treatments that aren’t necessary or don’t work. The fact is, there’s still an awful lot we don’t know about medical care—including treatments we use already. A recent study published in the British Medical Journal reviewed randomized controlled studies for 3,000 medical treatments and found that we don’t know how well about half of them work. It’s a good bet that a fair number of them simply don’t work—or don’t work very well.

The trick, then, is to figure out how to do less. That’s hard, for multiple reasons. One is that people understandably resist the idea that we can probably do without lots of medical care. The other is that we don’t really know what’s worth doing and what’s not—and it can be really hard to tell. This is where liberal health wonks point to government-funded research and cost-cutting advisory panels like the Independent Payment Advisory Board built in ObamaCare. Time-intensive clinical studies and effectiveness research can provide us some information, but in many cases these only tell us what happens on average. That’s useful to know, but it’s not enough, because many individual cases might differ. Which is why I suspect there’s more to be gained by dispersing decisions about health care needs and costs amongst the individuals most affected by those decisions. Health care is enormously complex, and markets driven by individual preferences are usually the best way to manage complex information and preference ecosystems.

The more we do that, the more I suspect we’re likely to find that what happens is that, on the whole, we spend a lot less money, but aren’t noticeably less health. Several years ago, economist Robin Hanson wrote an essay for Cato Unbound pointing out that there’s not much correlation between a society’s overall level of medical spending and its health outcomes. Things like diet, exercise, and education levels are much more predictive. Hanson argued provocatively that we should be able to cut out somewhere in the range of half of our health spending without a significant overall reduction in health outcomes. That rough figure may be high, but the broader idea has merit. If we want to reduce health spending—which is an open question overall but is certainly necessary at the federal and state levels—then what we need to do probably isn’t look for ways to spend money differently, but accept that doing so will actually require us to do and spend less.