Monday, July 13, 2015

ProPublica and the surgeons of America, together . . . or apart

Well, this is worth looking at:

ProPublica analyzed 2.3 million Medicare operations and identified 67,000 patients who suffered serious complications as a result: infections, uncontrollable bleeding, even death. We report the complication rates of 17,000 surgeons – so patients can make an informed choice.

No doubt the doctors with poorer statistics will say the usual: (1) "The data are wrong" and (2) "My patients are sicker."  Some experts offer their own opinions.

Will the information really help patients make an informed choice? The jury's out on that. It depends on whether primary care doctors use the data to guide their referrals or whether they will continue to rely on friendships and anecdotes--all reinforced by the financial pressures of being in the same ACO as those surgeons.


nonlocal MD said...

Well, as you have said in the past, transparency in these matters creates the incentive to improve. I am sure if I am one of those surgeons I will be looking at how to get my rate closer to one of the better ones, even if I am moaning about the data in the meantime.
Also, it helps draw the public's attention to the fact that no doctor is perfect, and some less than others. As with recent calls for medical errors to be included in cause-of-death statistics along with cancer and heart disease, awareness is critical.

Anonymous said...

It will stay the old boy network. I know several surgeons not listed on here. I saw several hospitals not completely excluded. So this might be a start but its not what was hyped.

I think there are additional concerns. I've heard of several people who are not disruptive seem to have a problem with posting. More than a few who simply talk about patient harm. Propublica also seems to highlight general stuff, they shy away from stories that highlight problems, for the most part. It seems like they are cherry picking who can talk on there and who can't. At least from multiple sources, there are concerns that Propublica is truly what it claims to be.

There were many concerns from what I heard, about the change to the Facebook group from Patient Harm to Patient Safety, on behalf of most all who posted there. I'd have to re find out what the story is, but the concern is that they are not going to be focused on patient *harm* from what I understand. It seems the change went that way due to marketing.

Carole said...

Outstanding!!, two of my most favorite commenters...This is about transparency correct? Okay I am in agreement that people aren't perfect and make mistakes, it's wheither or not they own them or hide them that decides what kind of human beings they are, agree? And Patient Harm vs. Patient Safety we need to know what's happened in order to believe something is going to be done about it. I'm sorry I just don't have much faith these issues that effect us all will ever be resolved. We the public will always be mislead for what they decide for us, is the greater good. Ten or twenty years from now we'll still be talking about this and it will always be agree to disagree. Oh how I pray I'm wrong!

John Tucker, PhD said...

"No doubt the doctors with poorer statistics will say the usual: (1) "The data are wrong" and (2) "My patients are sicker."

And in many cases they will be right. Its not just the "doctors with poor statistics" who are criticizing the database: Its most vocal critics on Twitter are two non-surgeons and a urological surgeon with a "perfect" record.

Its easy to attack the motivations of those criticizing the database, but in the long run their motivations don't matter. The quaility of the database does.

According to PP's own calculations, their 95% confidence interval for the adjusted complication rate of a hip surgeon with 100 surgeries is 2.7%. This is larger than the difference between the global average adjusted complication rate for all surgeons (2.8%) and the rate of the "worst" hip surgeon in CA (4.0%). Considerably larger.

If width of the confidence interval for the complication rate of a typical surgeon is greater than the average inter-surgeon variation in complication rates, most of the "high complication rate" surgeons will be false positives.

I strongly support transparency, but releasing ratings that are essentially random numbers does not support that cause.