Tuesday, July 28, 2009

Don't mention the Germans! 

In Martin Feldstein's column on health care reform today were these two paragraphs about which I'd like to comment:

To support their claim that costs can be radically reduced without adverse effects, the health planners point to the fact that about half of all hospital costs are for patients in the last year of life. I don't find that persuasive. Do doctors really know which of their very ill patients will benefit from expensive care and which will die regardless of the care they receive? In a world of uncertainty, many of us will want to hope that care will help.

I really don't know what to do with this paragraph. My first thought was that this was like that statistics that 1/3 of all accidents happen within a mile of one's home. Well, I think, that's hardly any wonder; you probably drive there more than you drive anywhere else. The press release I linked says it's because somehow we let down our guard, we relax around our homes when we drive, so we're less attentive. How do they know this? They asked drivers whether they relaxed. I don't think survey data is really how one learns this sort of thing.

When do we know it was the last year of life for a patient? It's something you determine ex post. Ex ante we all think we're the one who's going to pull through. We all believe we can beat the odds. This is what the Democrats are proposing: They will keep you from trying to beat the odds. But even if we were willing to allow them to do so, their project presupposes that you know in advance who will die within the next 12 months. And that you trust them with this. Do you? If we line up five top doctors and give them 100 patients with a variety of serious ailments, can they forecast who dies in 12 months? What would be an acceptable margin of error?

We are also often told that patients in Minnesota receive many fewer dollars of care per capita than patients in New York and California without adverse health effects. When I hear that, I wonder whether we should cut back on care, as these experts advocate, move to Minnesota, or wish we had the genetic stock of Minnesotans.
Genetics perhaps matters. Suppose I take this map:
and I overlay it with this map:

Do you see anything there? It isn't obvious. To me it says mostly "be lucky enough to be German, and you might live longer." Certainly up here we do. Heck, we even have the longest living man in the world from Melrose, MN. (That's about a half-hour from my house.) So do German-Americans get the benefit of higher factors on their QALYs?

Of course, mentioning QALY again gives you the answer. The data will likely not include ethnicity or sex. It will mention age, disability, and probably will include whether or not one has "risk factors" like smoking or obesity. If you're lucky enough to be from Minnesota and therefore have the great genes, it won't matter in someone else's decision whether you get treatment.

Labels: ,