CMS Releases Price Data

Today CMS released landmark cost data for the first time in 3 decades. While imperfect, this data sheds light on physician costs billed to Medicare. Over the next few days I will be posting analysis of this data as it pertains to dermatology. However, before embarking I should mention a flaws with this data.

1. Downstream costs are not indicated. For example, if Dr. Smith orders 4 times as many MRI exams, or admits 3x as many patients to the hospital as Dr. Jones there is no way to figure that out in this data set.
2. There are no quality indicators.
3. Prescription drugs are one of the key drivers of healthcare costs and the prescribing patterns are not available. For example, Dr. Smith may bill Medicare more but may write for cheap generic drugs that make his total cost lower than Dr. Jones.
4. This data only represents ‘straight’ Medicare patients. Medicare Advantage plan holders are not represented in this data set.

With those caveats in mind this data is extremely powerful. For the first time in my career individual physicians can gauge themselves against their peers. We can argue about cost measures with the hope of creating workable, if imperfect, metrics to use going forward.

To those who would argue that “may patients are sicker” or “I see more rare diseases” – I would counter than with this much data, and with a population that is equally aged, and likely equally infirm, those differences will not explain large variations in cost. While a dermatology patient in Minnesota may be less sun damaged than one in Florida, it is reasonable to lump together all providers in a given state since their patient panels are likely similar once a sufficient number of patients has been obtained.

Below is data, dermatology specific and NOT inclusive of midlevel providers like nurse practitioners and physician assistants (who do not have a specialty listed in the CMS database and require manual lookups), regarding cost of care by state.

-Neil A. Shah, MD, FAAD