Much ado is being made about the addition of Total Knee Arthroplasty (TKA) to the Program for Evaluating Payment Patterns Electronic Report or, as we all know it, the PEPPER. I would add my voice to the “ado,” not because I think this indicates coming scrutiny, but instead, my concern is that this may create unnecessary work for hospitals.
So, let us briefly look at TKAs and how the PEPPER analysis provides limited assistance to hospital’s compliance departments.
Understanding the TKA Calculation in the PEPPER
First, the facts. PEPPER metrics are calculated to determine if a hospital is an “outlier” by comparing its billing patterns to other hospitals in a national, Medicare Administrative Contractor (MAC) jurisdiction, and state comparison group or cohort. The term “outlier” in PEPPER is when a hospital’s metric or “target area percent” is in the top 20% when comparing it with its respective cohort.
The PEPPER evaluates how hospitals are statusing TKAs by looking at inpatient total knees divided by all total knees:
Numerator = count of discharges with at least one of the ICD 10-PCS knee replacement procedure codes
Denominator = the numerator + Outpatient claims with CPT® code 27447
What are we to glean from this metric?
For hospitals that are at or over the 80th percentile (high outliers), the PEPPER User’s Guide states that:
“This could indicate that there are unnecessary admissions related to the use of outpatient observation or inappropriate use of admission screening criteria associated with total knee replacement procedures. A sample of medical records for these procedures should be reviewed to determine whether care could have been provided more efficiently on an outpatient basis. Documentation should support the need for an inpatient admission.” [Bold added by author]
If your hospital is under the 20th percentile, the PEPPER User’s Guide states, “Not applicable, as this is an admission-necessity focused target area.”
Why Being Labeled An Outlier Does Not Give Hospitals Valuable Information
Let us focus on a hospital deemed as an outlier because it scored at or above the 80th percentile. This number does not give you valuable information. Why? Well, since we are using the 2-Midnight Rule to establish appropriate admission status, failure to include the length of stay in the analysis limits the conclusions that can be drawn. A tertiary care facility that handles complex patients will have more TKA’s that are longer than 2 midnights and appropriately statused as inpatient while a community hospital will likely have fewer of these cases. Should the tertiary care facility undertake time-wasting audits because the PEPPER is designed to show they may have a problem? Of course not. What if all one-day surgeries are done at a free-standing ambulatory care center, and only cases that will need more than one overnight are done in the hospital as Inpatient? In short, the PEPPER provides us with a percentage of TKAs that are inpatient, but it does not accurately tell us whether that is good or bad because it does not adjust for length of stay.
A New Way Of Evaluating Short Stay Compliance
A more straightforward metric would be better. Hospitals can look at two ratios:
- Ratio 1: the count of discharges with at least one of the ICD 10-PCS knee replacement procedure codes for cases that have a length of stay of less than 2 midnights relative to all less than 2 midnights TKA cases (inpatient and outpatient claims with CPT® code 27447)
- Ratio 2: the count of discharges with at least one of the ICD 10-PCS knee replacement procedure codes for cases that have a length of stay over 2 midnights relative to all over 2 midnights TKA cases (inpatient and outpatient claims with CPT® code 27447)
The first ratio should be lower while the second ratio should be higher. The PEPPER should be amended to include length of stay which would make this metric meaningful to hospitals. Ironically, if you look at it this way when comparing hospitals, there should be almost no variability, as one of the purposes of the 2-Midnight Rule is to bring a consistent application of Medicare Part A and B benefits to the Medicare beneficiary.
Not surprisingly, the PEPPER also errs in ascribing an error to an issue with the underuse of outpatient observation. Observation is not generally used for elective surgeries. The recovery services of a patient who goes home as expected are already covered in the bundled APC and adding observation hours to account for these already covered services is not allowed.
In summary, the PEPPER has good intentions. But it needs to consider length of stay to provide analytic benchmarking that has utility in a post -2-Midnight Rule world. Otherwise, it risks creating crises that do not exist, provokes unnecessary audits, and causes hospitals to spend time chasing outcomes that are non-compliant. In summary…much ado about nothing.
