There are more than 150,000 visits recorded. This is when the power of large numbers kicks in. The statistical trends are very likely to have captured what the researchers say they do. The peculiarities of a single hospital administrator (or even a few) aren’t going to have an effect on the data.

If you’re arguing that the behavior of director B is somehow characteristic of many of the hospitals in the study, that conclusion is at odds with the overall increase in prescription of opioid painkillers. It doesn’t follow.

Your point was that a hospital that consistently under-prescribed for painkillers and treated a larger number of blacks would skew the data.

But the report is not saying that blacks get fewer prescriptions for painkillers, but for opioid painkillers, even in an environment when overall opioid prescription is higher. I don’t think your point holds.

Again, go and look at the abstract, not just the article.

If you trackback to the original abstract, you’ll see that the authors say that race was the only significant factor. Since this is JAM, I’m assuming that they (and their reviewers) know enough elementary statics to control for factors among the hospitals and so on.

Not always a good assumption. But until an analysis of the actual report and not just a newspaper article about it has been done, I think it needs to be taken with a grain of salt or two.

Really, why is this so hard for people to believe?

For the simple reason that statistics are VERY HARD to do right and JAMA isn’t as good at reviewing studies as some people seem to think. As long as the numbers are correct and reasonable controls are done, JAMA will accept the article. The numbers I made up above are accurate and unless more detailed analysis was done at each individual hospital they might not even be noticed as misleading because they are accurate.

If you trackback to the original abstract, you’ll see that the authors say that race was the only significant factor. Since this is JAM, I’m assuming that they (and their reviewers) know enough elementary statics to control for factors among the hospitals and so on.

Also note that minorities (not just blacks) are being prescribed painkilliers, just not the more effective opioid compounds.

Really, why is this so hard for people to believe?

This one-size-fits-all thinking helps alienate patients and increases overall dissatisfaction with medical care in the US - which may actually be a feature.

Insurance companies make money by trying to avoid paying out. The same is true for medical care (mostly paid for by insurance companies).

Another big issue is that we are not actually customers in a medical care “transaction”. We are consumers, i.e. we get the benefit of the medical care we seek. But because we don’t actually pay for the services, we are not the actual customers - the insurance companies are. “He who pays the piper calls the tune…”

Single Payer: It’s the right thing to do…

I ended up switching away from Kaiser — where I do like the doctors — because they wouldn’t even cover my freakin’ rosacea medicine. They kept trying to make me use MetroGel when it doesn’t work for me and in fact makes my rosacea worse. So I was having to pay out of pocket for Noritate, which was just expensive enough to make a dent in my budget ($100 every four months) but not nearly expensive enough for them to deny (IMO, of course). And that’s something for my face, not something vital.

Sorry, that rant’s been building for a while.