Developing Story: Two New Issues, Added Confusion Regarding PEPPER

Clarity is badly needed for the quarterly health data report.

Every quarter, the quality representatives at most hospitals receive their Program for Evaluating Payment Patterns Electronic Report (PEPPER), and every quarter the quality representatives struggle to figure out what to do with the data.

And this quarter, TMF, which produces the PEPPER from data sent to them from the Centers for Medicare & Medicaid Services (CMS), has added two measures that will likely create more confusion than clarity.

Part One: ED Facility Visit Charges, or Holding Hospitals Accountable to a Non-Existent Regulation

 The first new issue involves the facility billing of emergency department (ED) evaluation and management (E&M) CPT® code 99285, the second-highest-level code that can be used. It is important to differentiate this from coding by physicians who also use the same code for their visits. Facility fee coding reflects the volume and intensity of resources used by the facility to care for the patient, while professional fee codes are determined based on the complexity and intensity of the provider’s visit, and this includes the cognitive effort expended by the provider based on standardized definitions. Because they are measuring completely different elements, one cannot correlate the two in any way.

In the PEPPER, a hospital will be able to see the percentage of ED visits coded with 99285 in comparison to other hospitals in their jurisdiction, state, and nation, along with the volume of visits. The value of the addition of this measure is limited for several reasons. First, there are no standardized guidelines for assigning a facility E&M code, and the acuity of patients at different hospitals, even within the same community, can vary greatly.

CMS summarized the problems with E&M coding of ED visits in the Federal Register 72 FR 66789, wherein it noted that “until national guidelines are established, hospitals should continue using their own internal guidelines to determine the appropriate reporting of different levels of clinic and emergency department visits. We would not expect individual hospitals to necessarily experience a normal distribution of visit levels across their claims, although we would expect a normal distribution across all hospitals as currently observed and as we would also expect if national guidelines were implemented. We understand that, based on different patterns of care, we could expect that a small community hospital might provide a greater percentage of low-level services than high-level services, while an academic medical center or trauma center might provide a greater percentage of high-level services than low-level services.”

It should also be noted that 99285 is not the highest-level ED facility code; that honor goes to critical care, CPT 99291, but I suspect that code is not used enough to warrant its own PEPPER analysis. In a recent edition of Report on Medicare Compliance, a PEPPER representative is quoted as stating that their analysis demonstrated a wide variation in the use of 99285, so the incentive for upcoding led them to choose this measure.

Of course, as with many PEPPER measures, being an outlier means very little. As astutely noted by CMS, an ED in an area with an abundance of primary care physicians and urgent care centers is likely to have more ED visits with higher acuity, whereas an ED in an area where many patients use the ED as their source of primary care will have fewer acuity visits. Likewise, some hospitals will triage lower-intensity patients to an in-house urgent care area, where billing does not involve the use of ED facility codes.

With no nationally accepted guidelines on determining the proper visit code, as there are with every other service, what is a high outlier to do? The first step would be to ensure that the coders are choosing the right codes based on the hospital’s internal guidelines, perhaps by conducting inter-rater reliability testing.

Second, if a hospital is using internal coding guidelines, it may be worth considering comparing them to externally published guidelines, such as those published by the American College of Emergency Physicians, available at this link.

Low outliers similarly should look at their data and try to determine if their lesser use of high-level codes is due the lower acuity of their patient population or under-coding of their visits. Perhaps with this new attention to ED facility coding, along with UnitedHealthcare’s (UHC’s) decision to automatically downcode some high-level ED visits, as described in my past RACmonitor.com article, the coding community will finally develop national standards.

Part Two-: Medicare Spending per Beneficiary Produces Colorful Graphics, But Limited Value

The other new data provided in the PEPPER is a bit more nuanced than the reporting of ED facility codes. PEPPER will now provide hospitals their data on Medicare spending per beneficiary (MSPB). This is a measure that is now reported by CMS on its Hospital Compare website, as spending per beneficiary encompassing more than just the inpatient admission is the basis of many of the bundled payment programs.

Unfortunately, it appears the data that will be provided to hospitals will be of limited value because of its age. The current PEPPER, which provides data to hospitals on performance on most measures for July through September 2017, provides MSPB data from calendar year 2016. Furthermore, a TMF representative has informed me that this data will be updated yearly, so we will not get 2017 data until the first PEPPER is released in March 2019. And we think that PEPPER data on other measures, which is six months old, is untimely…

Nonetheless, when presented with data, it is worth at least taking a look. To do that, you need to understand the basics of the report. This looks at every Medicare beneficiary admitted as inpatient to a subsection (d) hospital, which eliminates critical access hospitals, cancer hospitals, and others, and looks at all costs beginning three calendar days prior to admission through 30 days post-discharge. It also excludes deaths, patients who transfer, and no-pay inpatient admissions.

But the costs do not stop accruing at 30 days; if a patient begins a home health episode of care or enters a skilled nursing facility (SNF) within 30 days of an admission, the total cost of that episode of care, even if it ends more than 30 days after discharge, will be reported as accrued in that beneficiary’s 30-day spending.

To account for acuity, the spending will be risk-adjusted using hierarchal condition categories (HCCs), along with age and disability status, and there will be adjustments for the influence of outliers. They are also able to back out any funds paid to hospitals as part of medical education or the disproportionate share program, and adjustments will be made to equalize variable wage indices. Double-checking CMS’s math is not for the faint of heart, but those interested can find the formulas on QualityNet here.

Unlike the other PEPPER measures, your performance is only compared to that of the nation, so there is no opportunity to account for regional differences in care or differences based on rural or non-rural location. Since CMS will not release any spending data by patients prior to their arrival or after discharge, it is impossible for hospitals to duplicate this data to achieve real-time process improvement.

When you get the opportunity to review this, you will see three sets of data. The first reflects spending by claim type, including inpatient, outpatient, physician fees, skilled nursing facility (SNF), home health, hospice, and durable medical equipment. The next set is overall spending by time frame, pre-admission, during the admission, and post-discharge. And the third set breaks each time frame down into the claim type, inpatient, outpatient, etc.

What you choose to do with the data is an individual decision. You may see some glaring variations from the norm, such as an inordinate amount of spending on SNF care or post-hospital inpatient care at acute rehabilitation or long-term acute care, but other than that, I don’t see any great value in this data as it is presented beyond allowing TMF to use green, pink, red, and blue graphics on the same page.

And finally, because you expect me to be critical, I will point out that they use the term “carrier” to denote claims paid by the Medicare Administrative Contractor (MAC) for physician fees and other Part B-covered items not included in other categories, such as Part B drugs and vaccines. Now, I cannot blame TMF for this, because the term is used by QualityNet in their information. But as we all learned as children, two wrongs do not make a right. TMF should have corrected QualityNet and called them by their proper name: the MAC.

Perhaps TMF realized the limited value of this data since they added the tab to the end of the PEPPER. But if you do make it that far, this overview should help you realize that your efforts are likely better spent elsewhere.

Program Note

Listen to Dr. Hirsch every Monday on Monitor Mondays, 10-10:30 a.m. EST.

Comment on this article