With over 2000 pages in the CMS 2021 IPPS Final Rule, one might think that CMS would have published useful commentary on its quality programs. But it didn’t. The annual update to CMS’ three primary hospital pay for performance programs (Hospital Readmission Reduction Program, Hospital Value Based Purchasing, and the Hospital Acquired Condition Reduction Program), did little more than add one year to the dates from last year’s commentary. The Inpatient Quality Reporting Program was also far from earth shattering, taking dozens of pages to say that electronic Clinical Quality Measure (eCQM) reporting will gradually transition from one required quarter per year to all 4 quarters per year.

However, what CMS did NOT address is worth talking about, so let me break it down for you. In August, CMS released the 3rd COVID-19 Interim Final Rule with Comment (IFC-3401). While largely addressing long term care facility policies, the 3rd IFC spent a lot of time talking about Extraordinary Circumstances Exceptions (ECEs) for the pay for performance programs. While CMS has already committed to not using any claims data from January through June of 2020 due to the COVID-19 pandemic, many areas of the country experienced their surges through the summer of 2020 or later. How can it be fair to exclude a time period that greatly affected part of the country but not a period that later affected different parts of the country?

The IFC went so far as to say that CMS is considering not scoring any hospitals on any of the three programs, if the available data do not allow for meaningful and fair comparisons. CMS stated “If circumstances warrant, we may propose to suspend prospective application of program penalties or payment adjustments through the annual IPPS/LTCH PPS proposed rule.” Silly me for thinking that the 2021 IPPS Final Rule, which came out a month after the 3rd IFC, in the midst of a generational pandemic, might address data quality issues due to COVID-19. The IPPS Final Rule was silent on the issue.

CMS also combines 2-3 years of data, in general, in its P4P programs because annual data usually contain wide variations in hospital performance. In other words, the signal to noise ratio is not very strong, so longer data periods are necessary to achieve a more reliable and stable measurement. CMS has already committed to removing 6 months of data, shortening evaluation periods to 1.5-2.5 years. Shortening it any further would could cause significant fluctuations in hospital performance and undermine confidence in the measures. Hospitals actually would not mind not being scored in these programs, since two of them are penalty only and one of them is revenue neutral. 

Despite the silence on this issue in the IPPS Final Rule, rest assured that these issues are very real and on the minds of CMS. These programs bring over $1B in net savings to the Medicare Trust Funds every year. So there will be more to follow from CMS on this issue.

The other newsworthy omission from the IPPS quality section comes from the Inpatient Quality Reporting program. Currently, there are only two clinically abstracted measures that are required of hospitals – PC-01 Elective Delivery before 39 weeks and Severe-Sepsis and Septic Shock: Management Bundle. The latter measure, formerly known to most as SEP-1 (not to be confused with sepsis-1, the 1991 definition of sepsis), has been required for reporting since 2015. Conspicuously absent, however, is the inclusion of SEP-1 in future pay for performance determinations. If CMS wished to, it could easily incorporate SEP-1 into the Safety Domain of Value Based Purchasing. In fact, up through the FY 2026 VBP program year, you will find no plans to include the former SEP-1 performance metric into Value Based Purchasing.

Not that CMS doesn’t have time to work it in if they wished, but if you’ve followed the sepsis literature, you know that infectious disease experts for years have been heavily criticizing the interventions contained in SEP-1 as policy grasping from the fringes of supporting science, with interventions that are inappropriate for a national quality measure. National policy overstepping the science behind it – sound familiar? No, I’m not talking COVID-19 here, but sepsis. While many providers just assume that bundle based sepsis care saves lives – the randomized controlled trial (RCT) data supporting that still largely comes only from the original, single site, Early Goal Directed Therapy for severe sepsis trial in 2001. Three well designed international RCTs that followed failed to confirm a mortality benefit from bundle based sepsis care. Over the years, providers from the CDC, the NIH, and other academic centers have criticized the SEP-1 measure as not evidence-based, poorly conceived as an all or none measure, and most importantly not associated with improved sepsis outcomes.

Then, the hammer fell in May of this year, when the Infectious Diseases Society of America, the premier ID society in our country, published a position paper laying bare many of the shortcomings of the SEP-1 measure. (https://pubmed.ncbi.nlm.nih.gov/32374861/) Among the many critical commentaries made, the most striking one was that the SEP-1 measure overtreats patients who end up, after a brief period of investigation, to not have an infectious illness at all. This contributes to C. diff infections and antibiotic resistance, which is not even on the radar of SEP-1 because it only includes patients ultimately diagnosed with sepsis. Equally important is that the evidence base, including newer evidence from the state of NY where sepsis bundles are mandatory, does not support that an antibiotics-first-ask-questions-later approach is beneficial in patients with sepsis without septic shock. We learned that lesson with pneumonia in the early 2000s with antibiotics given in waiting rooms to patients with cough. A detailed review of the paper is out of the scope of this article, but I encourage readers to review the sections on fluid resuscitation, lactate measurement, organ dysfunction definitions, abstracting burden and inter-reviewer reliability issues that the article describes.

The article concludes by stating that the authors were planning to meet with the Quality Measurement and Value-Based Incentives Group at CMS that owns the SEP-1 measure. Word on the street is that CMS is listening, and hopefully that will lead to a future iteration of a sepsis quality metric that is actually evidence-based and targets the population who will benefit most. But don’t expect to read that in the 2021 IPPS Final Rule.