Occurrence Span Code 72 Goes from Optional to Necessary

The Program for Evaluating Payment Patterns Electronic Report (PEPPER) is a mystery to many. Sent to hospitals as an Excel workbook, I find it to be data-rich but information-poor.

There are 56 pages of explanations, data, and tables. To make it even more frustrating, no one seems to really know who looks at PEPPER data and how it is used.

Do the Recovery Audit Contractors (RACs) use the data to determine audit targets? Does the U.S. Department of Health and Human Services (HHS) Office of Inspector General (OIG) use the information to decide what hospital to choose for a “routine” audit? Despite our inability to know whether the results are used or not, all hospitals prefer not be labeled outliers. And one way to do that is to ensure that you are aware of the periodic changes made to the PEPPER and that your data accurately reflects your performance on the selected measures.

To that end, two recent major changes to the PEPPER, beginning with the fourth quarter, 2016 reports that were released in early March, deserve everyone’s attention. First, the PEPPER will no longer report separate data for zero-day inpatient admissions for medical and surgical diagnosis-related groups (DRGs), instead merging these into the one-day reports.
Although the report will continue to be called “One-Day Stays for Medical (or Surgical) DRGs,” it will actually report zero- and one-day stays. As on prior versions, the report will continue to exclude patients who are coded as discharging against medical advice, transferring to another short-term hospital, or dying.

But as of this report, the PEPPER “one-day stay” reports will exclude patients whose claims include occurrence span code 72, indicating that the patient spent one or more midnights at the hospital as an outpatient and that the total stay exceeded two midnights. When this is the case such patients thereby qualify as a “benchmark admission,” which is eligible for Part A payment.

When the occurrence span code was introduced, the Centers for Medicare & Medicaid Services (CMS) indicated that its use was optional. As many know, the hospital claim form has fields that indicate the “statement covers” dates, along with the date of inpatient admission. So theoretically, the claim would already indicate that the patient received outpatient care prior to inpatient admission. But the “statement covers” period could also include any services that fell into the three-day window for bundling outpatient services into the inpatient claim, such as an outpatient MRI.

Many were surprised when CMS indicated that occurrence span code 72 was an optional code. In fact, some argued that a National Uniform Billing Committee-approved code cannot be designated as optional, and that once approved, it must be used on every applicable claim. But the option to use it did leave hospitals with an interesting dilemma.

By using the code, a hospital can indicate that the admission met the requirements of the two-midnight rule, thereby reducing the risk of the claim being audited. But on the other hand, if a hospital did not use the code, the RACs – which are kept under a limit of the number of charts they may request for audit – would not know that the admission met the benchmark, and therefore would request the chart and be forced to approve it, wasting one of their precious chances for a denial. In fact, I advocated for not using it in order to frustrate the RACs as much as possible.

But the exclusion of charts from the “one-day stay” measures in the PEPPER now makes use of occurrence span code 72, making it highly recommended, if not mandatory. No one wants to be an outlier on a PEPPER measure, and being an outlier on zero- or one-day inpatient admissions is sure to draw the attention of auditors, especially as the RAC program is resuming and the auditors continue to bombard Congress with propaganda in an attempt to get more audit power. Outlier status here may also draw the attention of the OIG, which highlighted the apparent overuse of short inpatient admissions in a report issued in December 2016. In fact, these admissions are also on the 2017 OIG Work Plan.

With death, transfer, and discharged-against-medical-advice already excluded, using occurrence span code 72 now leaves the “one-day stay” category with far fewer admissions, especially among medical admissions. It should be remembered that many surgeries that are inpatient-only, including total joint replacements and carotid endarterectomies, can have a length of stay of one day, with some patients undergoing total joint replacement even going home the day of surgery. These admissions would show up on the PEPPER, but are compliant. I will go no further into the delineation of the other eligible zero- or one-day inpatient admissions, as this is best left for another day, but needless to say, every one of them should be reviewed prior to billing.

Finally, how does occurrence span code 72 actually work? If a patient spends one or more midnights hospitalized as an outpatient and ends up surpassing the two-midnight benchmark for medically necessary care, then being admitted as inpatient, the coding staff indicates in field 35 or 36 on the UB-04 the dates that span the midnight spent as an outpatient.

As a final thought, it should be noted that inclusion of the occurrence code 72 does not make a record immune to audit; auditors may still pull a sample of such admissions to look for evidence of medical necessity to support the second midnight. So it’s not time to let down your two-midnight rule compliance guard just yet.