November/December 2018

Time to Shake Up PEPPER?
By Selena Chavis
For The Record
Vol. 30 No. 10 P. 22

Industry experts believe the report could be modernized to better fit a value-based care environment—and stand a few other tweaks as well.

For more than 15 years, health care organizations have used the Program for Evaluating Payment Patterns Electronic Report (PEPPER) to identify vulnerabilities in billing. Distributed as a Microsoft Excel file, the report provides hospital-specific comparative data to support a facility's compliance efforts by identifying where it is an outlier for various risk areas.

These data can help identify both potential overpayments and potential underpayments, according to Marianne Durling, MHA, RHIA, CDIP, CCS, CPC, CPCO, associate compliance officer for Duke University Health System, and they can provide guidance on areas where a hospital may want to focus auditing and monitoring efforts. "It's important for hospitals to know how performance stacks up against their comparison group so they can be proactive in their work to make sure an organization is doing everything above board," she says.

Ronald Hirsch, MD, FACP, CHCQM, vice president of regulations and education at R1 Physician Advisory Services, points out that PEPPER has remained relatively consistent over time with some changes to measures here and there. In recent months, he notes that e-mails have been generated from PEPPER drawing attention to the fact that some hospitals are not opening their reports. "It seems to me that there are hospitals out there that are not even looking at their PEPPER," he says. "That's interesting. If Medicare is giving you data on how you compare to other hospitals, wouldn't you want to see it?"

Durling suggests that some hospitals are not using the reports because they are becoming less useful in a value-based environment. "When you look at value-based care, the categories that PEPPER compares are not apples to apples," she says. "I think they will need to refigure their comparison groups to get closer to an apple-to-apples comparison. To truly be integrated value-based care, the structure of PEPPER needs to be tweaked."

Hirsch and Durling agree that while some adjustments are needed for PEPPER to stay relevant, the report still offers great value to hospitals if they know how to use it correctly.

Recent Changes
In the fourth quarter of 2017, PEPPER began reporting on emergency department (ED) evaluation and management CPT code 99285, the highest level ED code available. Data from 2016 were used to give hospitals a view of trending, with outlier status assigned from that information.

Hirsch believes the code will cause more confusion than clarity. For instance, he notes there are currently no standardized guidelines for how the code should be used. "How is Medicare expecting hospitals to compare themselves to others when there is no standardized way of doing it?" Hirsch asks.

Because clear-cut guidelines for assigning code 99285 have been lacking, Durling believes Medicare will use the data to make its first foray into establishing guidelines for ED levels. "The industry has seen a shift in recent years to more higher-level ED visits even when the diagnosis and treatment doesn't fall in line in with [Medicare's] mindset for seriously ill or critically ill patients," she says.

In addition, Hirsch says confusion exists because physicians use the same code for ED visits. He explains that 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, including the cognitive effort expended by the provider based on standardized definitions.

Medicare spending per beneficiary, a measure currently reported on the Hospital Compare website, is another key addition to PEPPER. In 2018, the measure encompassed the costs of every inpatient admission a hospital had in 2016 starting three days prior to admission and 30 days afterward unless the patient went to a skilled nursing facility or other facility where the period of measure ends.

"It's a really noble effort to give hospitals data on how they compare to other hospitals as we enter this bundled payment world," Hirsch says, adding that the problem is that the data are from 2016, and hospitals need access to more recent information to stay on top of the changes underway in bundled payment models. Plus, he notes that the report normalizes all the numbers, mixing cost data from hospitals with a large number of high-cost procedures with those with few.

Last year, PEPPER began excluding admissions with the span code 72 from the one- and two-day inpatient reports. The code indicates that a patient spent one or more midnights at the hospital as an outpatient and that the total stay exceeded two midnights. Hirsch says this is an important distinction because a hospital's number of one-day admissions is an important performance metric for auditors.

When span code 72 was introduced, its use was optional. Now, with the changes to PEPPER, Hirsch says that "it's crucial to start using it."

"If [hospitals] put span code 72 on a claim, the admission won't go into PEPPER data," he explains. "It lowers your number of one-day admissions and it lowers your bar, so you don't stand out as much."

Extracting Value
Hirsch notes that PEPPER is "data rich but information poor" and can be difficult to navigate. "It can be overwhelming with all those graphs and data tables, so it's best to start with areas where you are a high outlier and take a closer look at those," he says.

Hirsch suggests that hospitals start with a review of the volume of cases for each measure to identify the most significant opportunities for improvement. For instance, if a measure contains only a handful of admissions, the difference in one case can move a facility from outlier to normal. But if a hospital is an outlier in a specific area that encompasses more than 100 admissions, it's worth looking closer.

"Small volumes of patients to me are less worthy of attention because the data are not showing trends or potential problems," Hirsch explains. "It's just too narrow."

Durling points out that PEPPER proved a valuable resource when she held an HIM director position and worked closely with clinical documentation improvement teams in a small rural hospital.

"Using PEPPER showed us where we were both high and low. If we were a low outlier, it made us reexamine our documentation and coding to make sure we weren't missing things and to make sure our coding was as accurate as it needed to be," she says, adding that if the hospital was a high outlier, it was important to determine the reason to avoid becoming ripe for regulatory audits.

Durling emphasizes that there can be good reasons for being an outlier. "For some, it might be that they are a high outlier in a stroke area because they are a stroke center," she points out.

While use of PEPPER can't directly prevent denials, Hirsch says the report can provide key insights into how to lower exposure to denials and future audits. For example, he recalls one hospital that was a high outlier for cardiovascular procedures. When clinical teams conducted a further investigation into the reasons, they found that doctors were placing cardiac stents on elective patients undergoing angiograms. "They thought it was appropriate because patients were admitted overnight," he notes.

In that case, the hospital corrected the problem by educating physicians on why the order of stents was not part of routine recovery for that patient population.

Alternatively, a hospital with the same issue may find that the problem is due to the transfer of a high population of heart attack patients from other hospitals. The high outlier status in this case is appropriate due to patient population.

At a minimum, having insights into the "whys" of outlier status better positions hospitals to dispute denials and respond to audits, according to Durling. "Once you've done the research and you know you are not an outlier from an error or a problematic coding standpoint … and you do get a denial, you kind of already have some of that supporting documentation to say, 'We are the only stroke center, so we have higher-acuity patients,'" she says. "Hospitals are better prepared to dispute denials, which helps you respond faster and more succinctly."

A Fit for Value-Based Care?
Hirsch says it's difficult to speculate on PEPPER's relevance in a value-based care environment because the Centers for Medicare & Medicaid Services has not yet mastered defining "value." "They are trying lots of things, but for now, the Medicare spending per beneficiary report is a fair representation of a hospital's past performance if they were in a bundled payment world," he says, further pointing out that if a hospital was not in a bundled payment program, it would want the report to depict the institution as a high spender. This way, when they transition to value, there is plenty of room to improve. "If you are already a low-cost provider and are told to improve from your baseline, that's going to be much more difficult."

The lack of PEPPER's timeliness is a problem for value-based care, according to Hirsch, who believes hospitals would rather see data right after the quarter ends even if they are somewhat incomplete. For instance, the September report featured data compiled from January to March.

Durling believes PEPPER's structure needs to be revised to be particularly useful in a value-based climate. Currently, PEPPER compares a hospital's performance with three groups: the nation, the hospital's Medicare Administrative Contractor (MAC) region, and a state comparative. "It really is not as beneficial in those categories," Durling explains. "The way the MACs are distributed can be vastly different in terms of the populations served and services provided."

For instance, comparing a state with a host of small regional hospitals with a state with large academic centers does not provide an apples-to-apples comparison. "Where I am now, we have large academic health centers in close proximity to each other," Durling says. "You know those centers will provide higher-level care at a higher cost than those smaller hospitals in the same area. It's now an equal comparison for everyone."

Other Areas for Improvement
Hirsch believes PEPPER could be improved in multiple aspects. For example, he says that setting the outlier limits at 20% from the top and bottom is too broad. "Any statistician would [laugh] if they saw that," he says. "That means 40% of hospitals are outliers, and only 60% are doing it right. Outlier is usually defined as 2% on each end."

He also points out that some measures, such as hospice care, are off the mark in today's health care climate. "They put it in red," he says. "Philosophically, I think hospitals that spend more on hospice should be in green because we underutilize hospice. When hospitals are in the red, I say congratulations, you're doing great."

Other measures such as debridement and stroke should be taken off altogether, Hirsch adds. With debridement, he notes that hospitals rarely have the volume needed, and the stroke measure high outliers are indicative of a positive. "It's confusing," he says. "It doesn't fit the rest of the theme."

Finally, Hirsch says that through PEPPER, Medicare should provide hospitals with the cases that make up each measure to allow them to take deep dives. "I have never seen a hospital that can reproduce the PEPPER numbers from their own system," he notes.

— Selena Chavis is a Florida-based freelance journalist whose writing appears regularly in various trade and consumer publications, covering everything from corporate and managerial topics to health care and travel.