The New York Times has a great little tool that accesses the massive CMS data dump on provider reimbursement for Medicare Part B. You can look up any Medicare Part B doc that has treated more than 11 patients in 2012 and see that they charged and what they got reimbursed. There are a few caveats to this data set.
- Primary address location is fairly arbitary. I looked up my PCP. He spends 60% of his time at an office a few blocks from my house. He was not there. His “primary” location according to the CMS data set is a location he spends one day a week at. When he first came to the area, he worked at this location 100% of the time, but moved to my neighborhood five or six years ago. His Medicare data profile has not been updated. Unfortunately provider data that is not 100% neccessary for claims payment is splotchy.
- Claims rolling up to a provider’s NPI or Medicare ID. Non-MD/non-D.O. clinicians such as Certified Nurse Practicioners, Physician Assistants, Master and Doctorate level Physical Therapists etc. often will roll their billing up to a doctor’s Medicare billing number. This means we can’t do a simple time management bullshit detection study based solely on “This provider is claiming he is doing 17 Medicare Part B procedures a day. Each of these procedures takes 30 minutes… IMPOSSIBLE”. That type of first level analysis might identify odd situations, but most will be explained by seeing three or four CRNPs/PAs doing most of the work that the doctor than bills for.
- Medicare Advantage is not in this data set. Some regions have lots of Medicare Advantage enrollment. Others don’t Some docs have a lot of Medicare Advantage patients. Others don’t. We can’t generalize too well to the entire Medicare population from the CMS data set.
- No way to determine medical neccessity/particular skill. This is pure counting data, it is not quality data. Counting data is valuable as it can be used to look at odd counts, but there are plenty of good reasons for outliers. For instance, a provider might be particulary good/renowned for putting shoulders back together, so that could be why his shoulder surgery count is so high compared to less aggressive treatment reimbursement codes as he was getting patients referred to him that needed surgery. We can’t tell from this data if the patients were different or the doctor was different in treatment preferences.
This is very valuable data for geeks, but it is caveated and limited. Nicholas Bagley at the Incidental Economist notes that information disclosure is not a particulary effective policy tool in and of itself.
Information disclosure is a common regulatory tool. It’s been studied a lot. And in most settings, it just doesn’t work…. Nor is it clear that employers and insurers will leverage the data in shaping their provider networks or honing their cost-control strategies. An extensive 2000 review of the evidence about publicly available information on provider quality concluded that “[n]either individual consumers nor group purchasers appear to search out, understand, or use the currently available information to any significant extent.”
…. Sure, the data will reveal some outlier physicians with outrageous billing habits. Patients should avoid those doctors. But what about a cardiologist who bills Medicare for stenting an unusually large number of patients? Is that a “bad” doctor with a penchant for inserting medically unnecessary stents? Or a “good” doctor with a thriving practice and a steady hand who inserts stents only where clinically indicated? How would you know?