Saving Billions On End-Of-Life Care: Another Mirage?
“We can no longer afford an overall health care system in which the thought is more is always better, because it’s not.” – Peter Orszag
Could anyone disagree? Not really. Which ought to be the first clue that it’s a meaningless truism. I mean, of course more isn’t always better.
But this hasn’t stopped this truism from becoming one of the most popular refrains in health care reform. Most of the time this is directed at one area: end-of-life care.
The idea is this: Americans spend too much money on heroic, but ultimately futile care at the end of people’s lives. Well-known research out of Dartmouth shows that that huge percentages of health care dollars are spent on end of life care. What’s more, it shows that this kind spending varies by big amounts, depending on which hospital you go to. For example, the federal government spends an average of $85,000 at Johns Hopkins, while spending less than $30,000 at Mayo Clinic. Reformers see the chance to save huge amounts of money by getting rid of these variations. Some say this could save the government a half a trillion dollars over 10 years.
It sounds good.
But here’s something you probably didn’t know: the Dartmouth study only measured spending on patients who died. If the spending saved your life, the researchers didn’t count it.
What the hell?
The New York Times (h/t Kausfiles) reports that some people are trying to take a clearer look at this problem, and conducting research that looks at the living, not just the dead. Those studies show that when you count patients who live, the difference between the most and least expensive hospitals narrows by as much as 44%. Government research that accounts for the fact that sicker patients often end up at more highly skilled hospitals further narrows the gap. Seen in this light, that potential half-trillion dollars evaporates pretty quickly.
This isn’t to suggest that variation doesn’t exist, or that some patients or their families demand care beyond the point where it’s beneficial. But it does show — again — how easy it is to get lured into oversimplifications of the problems in health care.
There are many people in health care who pore over the enormous amounts of data our multi-trillion dollar health care economy generates. They’re looking for patterns, big, systematic problems for which they can devise systematic solutions. But for all their looking, they keep missing the most important lesson of all. Which is that health care is not an assembly line process amenable to one-size fits all solutions. They miss that the answer to the most important question – what is the right way to treat this patient? – is a very unsatisfying: “it depends.” As one of the lead researchers of the Dartmouth study framed it: “Sometimes more medical care is better, but the question is when.”
Getting the right answer to that question isn’t a matter of protocols and financial incentives. Doctors want to get these answers right, regardless of the financial incentives, and so do patients. What is required to do this is a commitment to making sure each and every patient has the time, insight and judgment of their doctor. Yet our system fundamentally undervalues these things, and the reform efforts continue this mindset. If we continue to focus on how much money is spent, rather than whether that money is spent correctly, we will keep making it harder for doctors to get these answers right.
And people will continue to look at data, see puzzling results, and wonder how things ever got that way.
*This blog post was originally published at See First Blog*
Evan–sample size too small–refuted by other Dartmouth data–so says Brownlee in Health Affairs