How to make Medicare data mean something

ONC issues challenge to develop better visualization schemes
By Anthony Brino
11:05 AM

Medicare's release this past week of physician reimbursement data may be, as U.S. CTO Todd Park puts it, an "unprecedented" opportunity for transparency. But what will researchers -- and ultimately seniors and taxpayers -- be able to actually learn from it?

[See also: Doc data release sparks dissent]

Quite a bit, the government is hoping.

Now that the Centers for Medicare & Medicaid Services has released the data, the Office of the National Coordinator for Health IT has put out a challenge for the private sector: Visualize the data for beneficiaries and taxpayers.

"These visualization tools should be created to be used by consumers, for the purpose of assisting in active decision-making processes, especially involving potentially expensive decisions such as referrals or whether to move forward with an intervention or procedure," ONC officials wrote in the announcement of its "Data Supporting Decisions Challenge."

ONC is beckoning developers to the challenge with some cash: $20,000 for the best tool, $10,000 for the second-best and $5,000 for third.

That such a challenge exists does suggest that the data on its own -- or even as it’s been made available in databases by some media organizations -- isn’t all that useful for patients.

One effect, as former ONC chief Farzad Mostashari, MD, argued, is that highly-reimbursed ophthalmologists may be getting more questions about the macular degenerations drugs they prescribe.

And the data, showing that Medicare pays an average of $57 per primary care visit -- just 2 percent of all payment -- also points to some fundamental isses contributing to the U.S. healthcare crisis, namely an underemphasis and underappreciation of primary care, as family doctor and deputy editor of the American Family Physician Journal Kenny Lin, MD, argued on Twitter.

But the Medicare payment data is still far from being useful for patients as-is, which is partly why ONC is hoping to crowdsource a solution and attract technologists who can combine it with other data, such as quality measures.

Doing that could help seniors and their caregivers make more informed choices about which surgeons, for example, to seek out, or even help them decide if they are better off going without the procedure.

It could also "be used to create tools providing deep insight into geographic variations in procedure costs, regional specializations in procedures, and many other uses yet to be uncovered," as ONC officials wrote.

Still, claims data, even with both physician and quality data, leaves informational gaps, including from hospitals, data for which CMS is promising to release in the near future.

Then there is the debate over just what should be measured to reflect quality, in hospitals and physician's office, which may be different for different patients.

Regardless, quality and cost transparency should be sought to at least try to make progress on a common goal: helping patients "make health choices that are the most appropriate for their specific, individual needs," as ONC officials put it.