CMS’ Progress to Cut Medicare Payment Mistakes Stalls as Insufficient Documentation Woes Continue

They say a picture is worth a thousand words, or for the purposes of this blog post, about 600.

The picture in question really is a combination bar chart and line graph that appears on the second page of CMS’ Nov. 15 press release that boasts about how effective the agency has been in reducing improper Medicare fee-for-service payments to providers like hospitals and doctors. You can read the release and see the bar chart/line graph here.   

CMS said its aggressive payment integrity programs have reduced improper payments to providers by nearly $21 billion over the past seven years and that the agency’s improper payment rate hit a “historic low” of 6.26 percent in 2021. CMS defines improper payments as overpayments or underpayments to providers or payments made for which there was insufficient documentation to justify the payment.  

“The continued reduction in Medicare fee-for-service improper payments represents considerable progress toward the Biden-Harris Administration’s goal of protecting CMS programs’ sustainability for future generations,” CMS Administrator Chiquita Brooks-LaSure said in a prepared statement. “We intend to build on this success and take the lessons we’ve learned to ensure a high-level of integrity across all of our programs.”

That’s wonderful news until you scroll down (or turn the page) and look at the bars and lines.  The bars and lines for 2020 and 2021 are the same. That’s your first clue that something’s not quite as it sounds.

Your second clue is the 6.26 percent described as an “historic low.” Unless you’re looking at statistical analyses in a study published in a peer-reviewed journal, you usually don’t see percentages extended to the hundredth of a percent. That’s especially true in press releases where you’re trying to make things as simple as possible for us short attention-spanned journalists. 

So, I downloaded HHS’ 318-page financial report for fiscal 2021 from which CMS extracted the improper payment figures for its press release. There’s where I found the third clue. The report had the same data as the press release, but it didn’t have annual figures going back to 2014. No trend data to construct the bar chart and line graph. Hmmm.   

Time to download HHS’ 326-page financial report for fiscal 2020. And there, on p. 203, is the number I’m looking for. CMS’ improper payment rate for 2020—drum roll please—was 6.27 percent. Yes, it was one hundredth of a percent higher than in 2021. That explains the flat line. Touting it a historic low for 2021 statistically may be accurate, but it’s certainly misleading. 

In 2020, the 6.27 percent translated into about $25.7 billion. In 2021, the 6.26 percent equaled about $25 billion. The difference of a mere $700 million explains why the bars were the same size in the chart.

Also relatively unchanged over that two-year period were the leading reasons why CMS determined that the payments were improper:  

• Insufficient documentation (64 percent in 2021 compared with 63 percent in 2020)
• Medical necessity (13 percent in 2021 compared with 16 percent in 2020)
• Incorrect coding (11 percent in 2021 compared with 11 percent in 2020)
• Other (7 percent in 2021 compared with 5 percent in 2020)
• No documentation (5 percent in 2021 compared with 5 percent in 2020)

What’s the big deal? Why am I being so nit-picky? Isn’t this much to do about nothing? 

You tell me after reading these verbatim headlines in the healthcare trade press: 

• “Consolidation, prior authorization drive improper Medicare payments to historic low”
• “Improper Medicare payments down $20.7B since 2014, CMS data show”
• “Medicare fee-for-service improper payments decline by over $20 billion since 2014”
• “Medicare FFS Improper Payment Rate Hits Low of 6.26%”
• “Medicare FFS Improper Payments Reduced by $20.7B Since 2014”

A more accurate headline might be:

• “CMS’ Progress to Cut Medicare Payment Mistakes Stalls as Insufficient Documentation Woes Continue”

Hey, I think I’ll use that.

Thanks for reading.