Thursday, June 14, 2007

Doctors' Scarlet Letter: The Big Blue Dot

Today the New York Times reported that there is a wide discrepancy in reimbursement rates of private insurers from hospital to hospital for patients in need of open heart surgery in Pennsylvania. (Medicare rates are relatively fixed and only vary slightly based on geographic location). They based their report on Pennsylvania’s Health Care Cost Containment Council’s report published today.

While the information is interesting and serves as a reasonable overview, it has no “drill down feature” to look a specifics of the data collected. Instead a complicated, obfuscating, and proprietary weighting scheme (see the last page of these technical notes) was used to determine measures such as expected mortality rates. Cardinal Health claims the trademarked Atlas Outcomes™ methodology has been verified, but then they stand to profit from the data, don’t they?

But what was not discussed in the Times report was that physican mortality data was also presented in the Pennylvania report. Simply. Graphically. You see, rather than reporting an actual number or percentage mortality with details to view, various graphics analogous to Consumer Reports methodology were printed representing mortality data: an open circle (better than expected), dotted circle (expected), or Big Blue Dot (higher than expected).

For Pennylvania physicians that performed with in-hospital mortalities higher than their colleagues in 2005, they are branded with this scarlet letter of health care on the report: the Big Blue Dot. It becomes clear that even one of these Blue Dots might spell disaster for future referrals for a physician. But administrators, insurers, and likely patients will praise the simple graphic.

But health care is anything but simple. Just look at the letter one surgeon in Pennsylvania sent in to the Health Care Cost Containment Council in an attempt to explain his higher mortality rate (looks like three of 43 patients put him at the Blue Dot level). These folks were sick! But his explanation, while posted on Pennylvania’s website, did nothing to change his label.

His Big Blue Dot still stands for eternity for the world to view.

So what will happen with this doctor next year? Surely he will hope to avoid the Big Blue Dot again.

And so, this surgeon might hesitate just a bit before offering surgery to complicated patients with multiple medical problems. It’s not because he wants to be mean-spirited. No, it’s just that now he has to manage the risk to his reputation in concert with the risks of the procedure to the patient. Hopefully, by caring for patients who are less sick, his Blue Dot will be revoked next year and his name cleared. And the sick patients? They’ll either be left to fend for themselves medically, or referred to higher volume centers where, perhaps, the risk will be better absorbed.

And if every surgeon does this, there will be no more Blue Dots. Health care will be better, right?

Well, at least the report will look better.

But for the sickest of patients, they may stand to lose.



TBTAM said...

This is SO upsetting. And we have no control over it....

We are currently fighting with an insurer who shall go nameless except to say that their CEO makes QUITE A BIT, becuase they have branded a number of our docs as below par for performance. When we got the report to see the measures, every single one of them was erroneous! They were claiming we had not done things such as Hgb A1 C's in diabetics (there they wer, in the chart), or echoes on the ardiac patients (Again, in the chart) and HIV testing in our pregnant patients (Again, clearly there for all to see, and paid for by said insurer, buy the way). We had to hire a consultant just to clear up the mess...

These insures are working with erroneous data and using it to smear physician's reputations.

Since when did we make insurer's keepers of the quality measures? If we are to publicize quality, it needs to done by an impartial thrid party with no financial interest in the results.

TBTAM said...

Wow - that was a lot of typos up there! Sorry - the number of typos is directly proportional to my anger as I type...

DrWes said...


Your points are exactly the problem. All too often the "chart abstractors" have little or no medical background, have no incentive to find data that might exist regarding a specific test, and silently work alone creating "data" for a database that has no regulation. And yet each of us can be branded as non-compliant to standards of care without recourse.

I have an idea - want to set up a database for the insurers? We could add fields to our database that examine executive compensation vs. quality of data collected and let the consumers decide! Think employers might ante up to support this initiative?