Map of life expectancy at birth from Global Education Project.

Friday, August 14, 2009

Credit where it's due department

I complained about them yesterday, but Rutenberg and Calmes have now written the story they should have written in the first place, in which they call the death panel assertions "false" and trace their provenance without fear or favor. They call it a "rumor," which isn't accurate, this isn't spreading by word of mouth or e-mail forwarding, it's being loudly asserted by prominent Republican politicians and media gasbags, but I will let them off the hook for that. I'm not sure I want to let them off the hook for giving a wackjob the last word, which seems to be obligatory in this sort of exposé, but at least we're making progress.

Nonetheless, if we are lucky enough to get some meaningful restructuring of our health care non-system past the screaming mob of liars, we are still going to have to confront the need for practice guidelines, and soon, if we want the reform to succeed. Sufficient unto the day is the evil thereof, and all we can do right now is say no, that isn't in the bill and hope for the best. But if you think this has been hard, it's just going to get harder. So let me try to prepare the ground by making some distinctions clear. I know, I know -- the conservative movement depends on confusing people by conflation among its extensive arsenal of dishonest tactics, and maintaining these distinctions in the face of journalistic corruption and laziness will not be easy. But let me make them anyway.

Number one, more efficient, less costly health care is not synonymous letting people die or go without treatment that might benefit them in some way. Sometimes we run into those difficult moral choices, but that's only a subset of the problem. There are lots of opportunities to get doctors to make choices that save money and benefit patients at the same time.

Here's a simple example, the Ottawa ankle rules. These are simple guidelines that a doctor in the ER can follow to determine if somebody with an ankle injury needs an X-ray. The only reason to take an X-ray is to find out if the person has a bone fracture. Otherwise you've got a sprain and you don't need a cast. X-rays don't just cost money, they're bad for you, creating a small but real additional risk of cancer. Unfortunately, these rules are not followed consistently, even though they are well known and well-validated. There are innumerable similar situations in which we can encourage more conservative approaches to treatment, reduce use of diagnostic procedures, and save patients time, spare them possible pain or risk, save money, and still get optimal results.

Of course, when we save money, that means somebody isn't getting paid, and that person isn't going to like it. So much of the opposition to "rationing" -- which is a good thing, it means allocating resources rationally -- comes from specialty physicians and medical device and drug companies. They're happy to use scare tactics about euthanasia and "bureaucrats" interfering in the sacred doctor-patient relationship if that's going to keep the gravy train running, but let's have no illusions as to what this is all about.

Second, as I wrote yesterday, there is a distinction -- a big one -- between Comparative Effectiveness Research (CER) and Cost Effectiveness Analysis (CEA). The president has encouraged the former, but he has been silent about the latter. CER does not consider cost, it just compares one treatment to another to see which one works better. It's hard to see how anybody can oppose that but they do. See the preceding paragraph. CER in fact does not necessarily result in any cost saving -- in principle, it can just as easily favor the more expensive treatment.

I don't have any empirical data on this, but I suspect that over time, the reality will be that CER does tend to constrain costs, and the reason is that without it, the system has a built-in bias toward excessive spending -- again, because that's how people make money. For example, last week's New England Journal of Medicine published two separate studies finding that a commonly used procedure called vertebroplasty has no beneficial effect. This is when surgeons inject cement into the spine of people with osteoporosis and micro-fractures. Sounds logical, but it turns out that all of the observed effect was a placebo effect. Obviously back surgeons don't like this result, but the fact is that a large proportion of what physicians currently do routinely isn't based on evidence, and we'd probably all be better off if they just did less stuff overall. Less is more.

Only now, after we've already worked through a lot of issues that are going to make drug companies and left kneecap surgeons scream and yell, do we come to actual difficult ethical problems. Cost Effectiveness Analysis, CEA, is where the excrement hits the ventilator. The basic set up for Treatment A vs Treatment B is like this:

Costs more, works worse | Costs more, works better

Costs less, works worse | Costs less, works better

Obviously, the treatment in the upper left-hand corner should not be used. The treatment in the lower right-hand corner, on the other hand, wins. It's the upper-right, lower-left diagonal that presents the problem. (These are equivalent, obviously, just exchanging Treatment A and Treatment B.) Here's where we need to ask how much more spending is worth it.

Honestly folks, nobody is even starting to talk about this problem in this country outside of wonks like my colleagues at the Tufts CEA Registry. You can find lots of info there. I will take this on in my next post.

No comments: