5 Comments »

Are you really claiming that random variation is solely responsible for McAllen’s ‘outlier’ health care cost status? Would you also then argue that the relatively low cost of health care delivered at the Mayo Clinic is solely due to random chance?
I could see an argument for taking caution before assuming that the reasons for higher cost care in McAllen are generalizable to all cities or practices, but to say that there are no reasons at all beyond random chance sounds absurd to me.

I’m claiming that I really don’t know if random variation COULD account for the outlier status. It’s a possibility that hasn’t been addressed in any of the media.

Integrated delivery systems like Mayo are also outliers, but in reading about them, it makes sense why and it’s intellectually feasible to not believe it’s a matter of chance. However, I wouldn’t compare Mayo to McAllen, because that’s not apples to apples.

I admit to being a fan of Nassim Nicholas Taleb’s book “Black Swan.” It turns out there is a lot more randomness around us than our brains are accustomed to dealing with. Better to start off believing events are random and then disprove it. What is there to disprove McAllen isn’t a fluke in an admittedly dysfunctional health care market?

Thanks for your comment. Maybe I am being absurd but we bloggers like the feedback and the challenge of having to defend ourselves. We like to think it makes for better policy too!

While it is possible that randomness plays a part in the concentration of health care costs and services withing the U.S., this observation begs the question. In fact, health care reform proponents are proposing measures that, for better or worse, would take a large part of the randomness out of the health care distribution process. While we at Benemax have serious reservations about the stronger versions of health care reform that have been proposed, we agree that consumers, employers, plans and carriers should intervene in the market to steer health care toward a more efficient delivery and distribution model.

Great point David. Health care reform is not only about reducing costs but reducing variation, i.e., ‘tightening’ the curve. However, I think there will always be a distribtution of performance around a mean and that there will always be outliers. Drawing lessons from them is dangerous unless you can prove there’s something to it. Drawing lessons AND publishing them in the New Yorker AND then having the President pass it around in the White House is not a good way to seer heatlh care toward a more efficient delivery and distribution model Benemax knows it. Who else does?

Hi-

I take exception with the examples you cite to justify a big logical leap in this article. First, the examples you list are based on sampling a small number of discrete events and looking at the distribution in outcomes. Randomness is easily seen in your examples (or even flipping coins).

However, the healthcare costs in question are already averaged over hundreds of thousands, if not millions of events, it is much harder to attribute large differences in the values of such averages to random chance.

I am not really impressed that the physicians in this area claim that they don’t know what they are doing is expensive. I am sure they have been indoctrinated into thinking a certain way about prescribing medications, ordering tests, etc. How many of you would be satisfied taking your child to the pediatrician for an ear infection and leaving without a prescription? Physicians at Mayo are probably aware of data on the causes of such infections and are less likely to prescribe antibiotics. The Mayo brand gives them moral authority to safely say “no antibiotics for you”. Are the docs of McAllen, TX willing to send their patients away without a script? Is it the community norm to write a script every time they see a sick patient? The costs ad up. Such norms are quickly communicated through word of mouth, and I am sure it is hard to run a practice once word gets around that you didn’t give antibiotics to poor little “johnny”.

What would be interesting is to find numbers for the same community to see what the medical malpractice insurance rates are (as an indirect measure of litigation rate in the community) or to directly discover what the litigation rate was. If the litigation rate is very high, the docs may also feel forced to play defensive medicine to protect themselves - there is a real cost to that, too. Again, it may not be the result of conscious decisions.

Your comment

Name (required)

E-Mail (Required, not published )

Website

HTML-Tags:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>