Tuesday, February 09, 2010

CMS' New Drug Spending Projections

“Prediction is very difficult, especially if it's about the future.” Niels Bohr

The economists at the Centers for Medicare & Medicaid Services (CMS) just released their 2010 projections for U.S. National Health Expenditures (NHE). Here are two key observations about the prescription drug numbers.
  • The Government Pays More, Starting Now—Public funds continue to crowd out private payers. CMS now projects that private insurance and public funds will each pay 40% of retail drug spending in 2010. By 2019, public funds will be almost half of outpatient retail prescription spending.

  • Drug Spending Will Look Higher Next Year—Using the stated CMS methodology, I estimate the National Health Expenditure (NHE) data series currently understate actual prescription sales by about $18.5 billion (~8%). Therefore, prescription drug's share of overall health care spending is also underestimated by a full percentage point. Expect an upward revision in headline prescription drug spending in next year’s NHE data. Remember, you read it here first!
Make it to the bottom of the post and you will also be rewarded with a health economics knock knock joke.


CMS economists have extended their Source of Payment projections through 2019. Savor the fun for yourself:
Here’s my summary of the updated data. The projected future mix of payers has not changed since last year’s numbers.

CMS projects that public funds will cover 47% of outpatient retail prescription spending in 2019, while private insurance will pay for only 36%. Public funds include Medicare Part D, Medicaid and State Children’s Health Insurance Program (SCHIP). Consumer out-of-pocket funds include the sum of cash-paid prescriptions and consumer co-payments.

The public/private streams will cross this year—CMS projects that private insurance and public funds will each pay 40% of retail drug spending in 2010.

CMS refuses to comment on whether crossing the streams will lead to total protonic reversal, as Dr. Egon Spengler has predicted.


In next year’s NHE data, I expect that drug spending will appear to be a full percentage point more of total U.S. spending—11% versus 10% today. This is an important difference given the political scrutiny over the pharmaceutical industry.

This discrepancy is an artifact of CMS’ methodology. At the bottom on Drug Forecasts: Oops!...They Missed It Again, I point out that CMS’ estimates for total outpatient drug spending are consistently lower than estimates from private sources. Here's why.

The current NHE Prescription Drug series is benchmarked to product line sales from the 2002 Economic Census of Retail Trade (NAICS 44-45). The relevant Product and Service Code is 20161 (“Prescriptions”).

The 2007 Economic Census data are still dribbling out. But based on the initial release, the Census data show total retail prescription drug sales that are close to the IMS/NACDS data but much higher than the current NHE data.

In other words, the NHE data for 2007 are $18.4 billion (-7.5%) below the Census data and $22.4 billion (-9.0%) below the IMS/NACDS data. By comparison, the 2002 Census data and the 2002 NHE data are comparable (difference=1.2%).


In 2005, CMS projected that total outpatient prescription drug spending in 2014 would be $521.6 billion.

In 2010, CMS projects that total outpatient prescription drug spending in 2014 will be $322.1 billion.


Given these numbers, here’s a conversation starter for your next health care economist cocktail party. (Say aloud.)

Knock knock.
Who’s there?
Economist who?
Economist the drug spending forecast but I promise to do better next time.


Don’t miss Joe Paduda’s well-done Health Wonk Review- So, what do we do now?


  1. Adam,

    Checked out your link to the HealthWonk review.

    "Thanks Joe.
    Is there a way to fairly compensate the medical community and retain our collective shirts?"

    Gosh, it seems you may be softening a bit. What's up with that?

    Nice post today.

  2. Okay...I had to read the punch line twice before I got the joke.

    So in 2005 CMS overstated the forecast 9 years out by $200B. It does not sound like their forecasting methodology has changed much between 2005 and 2010. Does that mean we shouldln't read to much into their 2019 projections? How much off the mark were their 2005 projections 4 -5 years out?

  3. Never fear, I'm not going soft. My comment is *above* my name below the Health Wonk post. Some one named Joe Swan wrote the comment about fair compensation.


  4. shouldn't it be "Economist up the drug spending forecast?"

    A minor quibble, though. Surely the funniest knock knock joke I've read about health economics all afternoon.