The government dropped a gigantic dataset: details on nearly every single procedure performed by a U.S. doctor on a Medicare patient.
The release was greeted with some serious gnashing of teeth, at least as far as doctors were concerned. The American Medical Association, which has always been staunchly opposed to the release of this sort of data, made sure it’s objection was — again — on the record. MedPage Today leads their doctor-focused site with a story that cataloged criticism. It’s title? “None of Your Business.” The Wall Street Journal said that the data release “sparked an outcry.”
But there’s one group that met the news with a resounding “meh”: the 12,000-plus tweeting doctors that we track online as part of our MDigitalLife database. A quick scan late yesterday suggested that there were just over 200 tweets on the topic. And there was nary a peep of protest. The closest anyone came to an “outcry” was Chicago-based primary care doc Atul Jain, who tweeted out: “Transparency + muddied data = confusion.”
A closer look at the Medicare release may explain why digital docs saw the data dump as something less than than assault on their way of life. As it turns out, doctors who tweet are very, very different from doctors who bill Medicare for millions. Here’s how we looked at the data.
First, we created a list of doctors who appeared both in the dataset of Medicare providers and our MDigitalLife database of verified doctors in the United States with Twitter handles: 8,000 doctors who both used Twitter and received Medicare payments. We compared that group to the top 14,000 or so providers in the Medicare dataset (every single person who received $500,000 or more from the Medicare). There wasn’t a lot of overlap. Only 230 docs made both the top-tweeter and the top-biller list. And among the real outliers — the top 1,000 recipients of Medicare dollars — only 13 were on Twitter, with a measly median follower count of 112.
What’s more, there was an inverse association between Twitter followers and money received from Medicare. The more physician followers a doctor had, the lower his or her Medicare billing tended to be. And that was true not only in aggregate, but for individual specialties, too. It held for ophthalmologists, the per capita leaders in Medicare payments. It held for cardiologists. It held for dermatologists. And it would have held for oncology, too, if not for a single, highly followed oncologist. (Not surprisingly, there was a strong correlation between real-world interactions — referral relationships around Medicare patients — and increased Medicare spending.)
To be sure, the data isn’t clean enough to claim that there is some magic protective effect that makes doctors on Twitter less likely to file millions in claims with Medicare; there are clearly confounding factors that may be at work. And it bears noting that receiving large Medicare payments isn’t necessarily indicative of anything questionable and that the dataset itself has more than a handful of quirks.
Still, it is not a leap to conclude a doctor who chooses to take a plunge into Twitter is showing a commitment to transparency that can and should be applauded by patients. Clearly, that is a group of physicians that don’t fear sunshine.
One of the biggest questions raised by the enormous data dump is what, exactly, patients can glean from dozens of data points on more than 800,000 doctors. And while we plan to take a closer look, one of the tentative conclusions — judging from both the Medicare data and the reaction to the data — is that consumers would be well-served by doing a quick Twitter search. Seeing their provider pop up on the screen is not a guarantee of straight talk, but is sure is a strong signal.