Short-Stay DRGs: Fix or Fiasco?
By Judy Sturgeon, CCS, CCDS
For The Record
Vol. 27 No. 7 P. 24
To compensate for some of the fallout surrounding the two-midnight rule, the CMS may have created even more sleepless nights for hospital revenue cycles.
Has Medicare's two-midnight rule for hospital inpatient admissions fulfilled its intended purpose? And what is a short-stay diagnosis-related group (DRG)?
Medicare has scrutinized one-day admissions for several years, imposing financial penalties on the worst offenders. Hospitals reacted by using observation services more liberally when they were unsure whether a patient would meet strict admission criteria such as Milliman or InterQual.
In turn, patients who had received medical care in a hospital and believed themselves to be inpatients found to their dismay that this "status difference" also meant that they were required to pay more of the cost of care. If the patient subsequently needed interim care in a skilled nursing facility, the time spent in observation status could not be applied toward the required three-day inpatient stay that was necessary for it to be deemed a covered Medicare service.
Trying to anticipate the appropriate medical status created another administrative burden on hospitals. Increased use of observation status added to the financial strain being shouldered by Medicare beneficiaries. The reaction to one-day stay scrutiny created a need for the Centers for Medicare & Medicaid Services (CMS) to find some way to mitigate its costs without penalizing either its patients or its providers, and without increasing the complexity of the process.
The Medicare final rule for inpatient admissions beginning with fiscal year 2014 included what seemed to be the answer: the two-midnight rule. If a patient needs medically necessary services in a hospital for a span of time that includes at least two midnights, then it should be an inpatient stay. If not, it should be an outpatient stay. The time spent in any outpatient hospital status is included in the time considered for the qualifying two midnights.
For example, if a patient was in the emergency department or in observation status at the first midnight and needed to stay to continue medically necessary services for one more midnight, the physician was to order inpatient admission status with documentation of the expected length of stay and its purpose. This should eliminate unnecessary observation status patients and relieve the provider's struggle to meet the complex admission criteria. If the patient must be in the hospital for two or more midnights, the physician can make him or her an inpatient.
In the rare instance that a patient has been admitted but leaves prior to the second midnight—due to transfer, death, sudden and unexpected improvement in medical condition, etc—the documentation must support the reason for a short inpatient stay with a DRG payment. If an error is found, the hospital will be allowed up to one year to correct the status from inpatient to outpatient retrospectively and to rebill for the qualifying outpatient charges.
This seems to be a simple fix, but several Open Door forums and myriad public discussions and published material have generated only more questions about the new rule. A so-called probe and educate period was announced in which hospitals would be monitored and helped to comply with the new process without recovery audit reviews and related penalties. On April 30, the CMS announced further delay of these audits, stating, "CMS believes this extension will allow for continued education and promote further understanding of the policy."
In a Fix?
The famous idiom that "one thing leads to another" applies to many industries, and health care is no exception. When short stays were targeted for denial and penalty, the use of observation services increased. When the two-midnight rule was enacted, the use of observation services was discouraged and became part of inpatient admissions, and hospitals were no longer penalized financially as long as the provider had an expectation of at least a two-midnight stay. Short-stay inpatient admissions increased.
Medicare must remain budget neutral—whenever a policy change is enacted, other adjustments must be made to ensure that the outgo will not exceed the income. The CMS now is faced with significantly higher expenditures because the DRG payment for an inpatient stay can be significantly greater than the payment for the same services provided in an outpatient status.
Facilities may be faced with significantly reduced payment if they are required to downgrade all short-stay admissions (less than one midnight) and rebill as outpatient. If the provider had actually expected a short stay, that would be clinically appropriate. But what about those patients who appeared to be significantly ill and were expected to need longer intervention? Is it appropriate to disregard nationally recognized medical criteria supporting an inpatient admission and substitute an across-the-board policy based solely on the clock and the calendar?
In recognition of this dilemma, the 2015 Inpatient Prospective Payment System (IPPS) proposed rule calls for public comment on the creation of short-stay DRGs to alleviate payment concerns caused by the two-midnight rule.
In the 2016 proposed IPPS rule, the American Hospital Association (AHA) urged the CMS to offer potential payment solutions for hospital stays of less than two midnights. The AHA researched several options, provided supporting methodology for each, and estimated their expected impact. After rejecting a system similar to the transfer DRG policy already in place for selected DRGs, the AHA proposed the following five short-stay inpatient payment models:
• one short-stay DRG for each major diagnostic category (MDC) except newborns;
• one short-stay DRG for all medical DRGs within an MDC, and one short-stay DRG for all surgical DRGs therein;
• targeted DRGs to include one short-stay DRG for the 61 most common DRGs with the most short stays or the most recovery audit contractor denials;
• one short-stay DRG for each of the base DRGs. For example, if a base DRG offers an option without comorbid conditions (CCs) or major CCs (MCCs), with CCs, and with MCCs, an additional short-stay DRG would be created within the base DRG; and
• DRG refinement, in which every possible DRG receives a separately weighted DRG for short stays, essentially doubling the number of DRGs now available.
Day Egusquiza, president of AR Systems and a member of the American College of Physician Advisors' advisory board, alerted her extensive system of health care providers about the fiscal consequences of such a modification to the current inpatient DRG payment system. In her newsletter, she discussed the effects of the AHA's proposal.
"They are proposing creating an entirely new DRG grouping system for short-stay DRGs. If the patient did not have two midnights as an inpatient, the payment would be reduced to the lower short-stay DRG," she wrote. "I would strongly recommend that you all read this very carefully. It is a huge loss for the hospitals who have two-midnight expectations/presumptions and ended up with transfers out, etc, or who have the two-midnight benchmark, which is one midnight outpatient/one midnight inpatient. All would be paid at a lower than full DRG payment."
Becky DeGrosky, a product manager at TruCode, has similar concerns about simply defaulting any short stays that fall afoul of the two-midnight criterion for inpatient DRG payment to outpatient status. "Where the issue comes is on the patient—suddenly he has a big out-of-pocket payment and it becomes a big deal," she says. "And I'm kind of on CMS's side because that's Medicare so that's my money and your money, but you have to give the hospital a fair shake. The AHA's comment on the adequacy of the observation payment rate is a critical piece.
"From the hospitals' perspective, they have a daily rate that they charge for an inpatient that covers all of the administrative costs. The fact that they're getting a $1,200 rate for observation if the patient's there for two days but still not for two midnights? That's ridiculous. Medicare needs to look at that and go back to when they were paying an hourly rate for it. The proposed change just encourages hospitals to keep everyone for two days, ie, two midnights."
Keeping patients long enough to meet the clock criteria isn't enough, however, DeGrosky says. "Encounters need to meet medical necessity. That's the minimum basic [standard] of what Medicare says. Wherever the care is rendered, it needs to be what was required; it needs to be necessary," she says.
In terms of coding, there's only so much a coder can affect, DeGrosky says. "As a coder, it's not my job to verify medical necessity. I can only code what's documented," she says. "One more point about the AHA's letter on the list of short-stay DRGs that they propose to create is again from a coding perspective. If you have a surgical DRG, it's driven by an ICD-9 code. But on the outpatient side, we don't report ICD-9 codes for procedures; we report CPT codes."
To require a retrospective change in status from inpatient to outpatient, "you're going to require facilities to dual code with CPT and ICD-9 and PCS codes once we go to ICD-10. That, to me, is crazy," DeGrosky says. "Otherwise, how are you going to have a handle on anything? The codes used for outpatient care can't be used to calculate a DRG, and there is no valid crosswalk. There's no other way to manage it."
Not everyone is stalwartly opposed to payment reviews for short stays. Sandra Routhier, an HIM and coding consultant for HIMpoint, offers some sympathy born of her extensive background in utilization review, coding, and inpatient reimbursement. "I can understand where the AHA is coming from in trying to further refine or make it more fair or objective as to reimbursement for these short stays," she says. "But I still struggle with the need to create something different out of our current DRG reimbursement system for a couple of reasons. First, we already have DRGs that have low relative weights, and those are typically your chest pain patients, your syncopes, and other conditions that have short stays already. That's why they're lower weighted, right? Because they don't consume a lot of hospital resources," she adds.
"Also, inpatient admissions are paid for hospital services by Medicare Part A, which is a different bucket than Part B. If the admission is denied, or if the hospital self-denies, now it gets paid out of Part B benefits," Routhier says.
Medicare is funded from two separate sources, with Part A supported by payroll taxes, taxes on Social Security benefits, and premiums by otherwise noneligible beneficiaries. Part B is financed by a separate trust fund authorized by Congress, premiums paid by enrollees, and interest on trust funds. Each has different patient benefits and costs. Switching payments, charges, and benefits back and forth between Part A and Part B can create significant complications.
Routhier is concerned with the AHA's proposal as it relates to deaths, transfers, and patients who are discharged against medical advice. "Many of those, especially the deaths and transfers, can really be resource-intensive for facilities. To knock the payment down for those cases I think would be wrong. I do oppose that specifically," she says. "When you think about how the MEDPAR data is used, the relative weights of DRGs are adjusted every year based on claims submitted throughout the country. It accounts for how many resources are consumed, and isn't based on if it was used in one day or in two days. So it seems like the methodology today already considers resources fairly."
"The biggest focus point is the two-midnight benchmark," Egusquiza says. "If the patient does not meet the presumption to need two midnights at the first point of contact, usually in the emergency department [ED], then the ED provider, in consultation with the hospitalist or attending physician, should assign the patient to an observation status with an action plan to aggressively manage the patient to safely discharge or convert to an inpatient admission if a second midnight will pass and care is still needed to resolve the condition.
"The AHA plan 'tentatively' eliminates the two-midnight benchmark as a candidate for the full DRG payment, but makes it a candidate for the new complex and lower-paid short-stay DRG. There are also other combinations for when the one-midnight benchmark will come into play with the lower short-stay DRG, but should be paid at the full DRG because they started with the required two-midnight presumption. These include transfers to another acute care facility, unexpected death before the second midnight has been reached, and patients who got well faster and were discharged earlier than expected."
Egusquiza also points out that the AHA indicates it doesn't know how the CMS will address the two-midnight benchmark. As a result, she's concerned "we are creating a major level of complexity while losing significant payments for the majority of inpatients—those under two midnights."
However, Priya Bathija, senior associate director for inpatient reimbursement policy at the AHA, says, "In the inpatient PPS proposed rule that was released in April, CMS indicated that they would address this issue in the outpatient PPS proposed rule it releases this summer."
Your Opinion Counts
Comments regarding the potential restructuring of the inpatient hospital DRG payment methodology and the creation of a separate DRG payment based on one-midnight stays can be shared with the AHA and state hospital associations. These organizations continue to put forth great effort to represent the needs of their members, but they can't truly represent their constituents unless members provide feedback.
— Judy Sturgeon, CCS, CCDS, is the clinical coding/reimbursement compliance manager at Harris County Hospital District in Houston and a contributing editor at For The Record. While her initial education was in medical technology, she has been in hospital coding and compliance for 26 years.