With election season in full swing, we all have a lot of choices to make about how we would like this country to be run. Of course these decisions are always incredibly complex and nuanced, and seem much too large for any one person to understand. Questions of the economy, foreign policy, and military campaigns are so far out of the scope of our everyday life that it can be difficult to conceive of them. Yet we are obligated to try and make some sense out of them, because we are called on to help create them.
One thing we all know very well, though, is that nothing — no idea, no system, no government — is perfect. Everything breaks down, gets misused, and fails to work the way it should. This is as true of medicine as it is of government, and it of course can be made much worse where government and medicine intersect. So how much inefficiency and error should we tolerate in a system before we throw it out as a bad system? The answer will differ from an economic system to a program of medical treatment, but it’s good to examine the way we look at these questions as we the election approaches.
I spend a lot of time on this blog discussing statistics, and how the presentation of statistics can determine what they say, even to the point of making the same figures say opposite things. Cost-benefit analysis, whether it’s of buying a car or treating a disease, works much the same way. A 30% failure rate, for instance, can make any treatment plan sound terrible and a waste of time — but a 70% success rate is pretty good! The infinite ways to define “failure” and “success” are of course also at play here.
I am always frustrated when my patients do not do well — when they lie to me, or have positive urine drug screens, or are unable to account for all their medications. Certainly, the addiction population is a difficult one and there is no treatment that is 100% effective. I also have many patients who are doing great. They come in on a regular basis, take all the medicines, and are leading productive and useful lives. The question I find myself facing is this: what percentage of people doing well with a particular treatment justifies continuing the treatment? Is there an acceptable “margin of error” — some percentage of the population that can’t be helped by this treatment, but aren’t enough to stop me from performing it? If I were only helping one person out of a hundred, that wouldn’t seem worth it. But how about 50/50? Or 70/30?
I don’t know what the actual number is, of course; no one does. Entire disciplines of medical, social, and economic science are devoted to studying these questions. A similar calculation of help, harm, and ineffectiveness goes into the testing of every new drug and surgical procedure considered for public use.
Of course, with buprenorphine treatment, by perception and the scientific data agrees that patients who do well greatly outnumber those that do not. But it’s the patients who are doing very poorly, who are in pain or still using hard drugs or having family crises, that stick in my mind. On some days it can be hard to remember that the care I give helps anyone at all. It seems like the whole world is nothing but hopeless cases, and it seems then like the cost might outweigh the benefit of any treatment I give. The fact is that the cost of any venture is often more immediate and frightening than the benefit, which tends to be long-lasting and unobtrusive. That’s certainly the political situation we’re in now.
What do you think of cost/benefit, in terms of medicine and politics? Is there a way to measure success and failure, and a way to frame it, that is not misleading? Let us know in the comments!