I Wonk Because I Care

By Ezra

This is rather weak stuff from CATO's Michael Cannon on the feasibility of universal health care.  Responding to evidence that Canada only has waiting times for elective procedures and France and Germany don't have waiting times at all, Cannon says that 1) waiting times in Canada can still be harmful to mortality; 2) if asked, French and German citizens are unhappy about waiting periods for care; 3) there's some evidence for improved outcomes on very specific diseases in America.

Well, first, he dodged the actual issue.  Ponnuru had written that universal coverage was an impossible goal.  That was flatly untrue, given the accepted definitions of the words "universal" and "coverage."  What Cannon is arguing here is that there are distasteful affects from universal coverage, which are different, though also important, issues.  So, quickly and in order, I grant his point on Canada, though if he thinks it problematic that Canadians have to wait for elective treatment and those periods are occasionally harmful, he must be truly distressed that Americans often simply can't afford necessary treatment and they routinely die.  As for #2, he's using polling data that looks loaded to me.  I have excellent health insurance but am dissatisfied with waiting times.  Why?  Because I don't want to wait at all, ever.  My guess is Germans and French folk feel similarly.  But the question should be how long they wait, not how they feel about it.   And there, Jon Cohn's data lands a clean left hook for the win.  As for Cannon's last point, I've looked over the data on this.  Indeed, I spoke to Gerard Anderson, who conducted the study.  And what it shows is interesting, and I'm going to spend some time talking about it.  Follow below the fold if you feel like some wonkery.

"For prostate cancer," Cannon writes, "you are twice as likely to die of it in Germany as in the U.S. (44 percent vs. 19 percent)...if that’s the case, the appeal of “universal” coverage fades."  I'd noticed him use this statistic before, and I couldn't find it anywhere.  So I asked him (or his assistant, I can't remember) where it came from.  The work of Gerard Anderson, apparently.  So I called Dr. Anderson, who was certain he'd never found any such stat.  It sounded, he said, "ridiculous."  So I went back to Cannon and finally got the original data.  And now it gets complicated.

Anderson's data did indicate that, but he didn't realize it.  The reason is that the finding is a bit more complicated.  Here's the relevant graph:

The question here is whether all cancers are created equal.  Why does Germany have less than half our prostate cancer incidence?  And does it mean that their cancers are more serious, less curable cases?  It's unclear.  But the bottom line is that you're no more likely to die of prostate cancer in Germany than the US, and you're also less likely to die of other things.  Here's the part of the data Cannon leaves out when comparing the health systems:

I'd think our system is looking a bit worse now.  Meanwhile, the question that Cannon, and others of his ideological ilk, need to answer is whether they think the marginal benefits of our system are worth the marginal cost.  We spend around twice what any other country does per capita, and we see very little, if any gain, from the added expense.  Indeed, there's an interesting thought experiment as to how much better their outcomes would be if they pumped up their per capita spending to match ours.  Meanwhile, they cover all of their citizens (the paper Cannon is using defines all the other countries as having, yes, "universal coverage," that state of affairs Ponnuru thinks doesn't exist), while we have a population of 46 million uninsured, and another 15 or so million underinsured.  20,000 Americans die every year because they lack health insurance, and many more perish because they forego care that turns out to be necessary.  That's the sort of quality we're paying for, and I'm not sure Cannon's slightly confusing prostate stats really balance out the ledger.