The Past and Present: The Prospective Payment System from RUGs to PDPM
In 1998, CMS changed the Medicare reimbursement methodology for SNFs from a cost-based methodology, where SNFs were paid based on what they spent, to the PPS,3 where payments were based on patient characteristics instead of cost. This change was expected to reduce program costs while maintaining or improving patient care. Under the PPS, providers would complete a series of assessments for each patient, called the Minimum Data Set (MDS), and those assessments were used to determine which category (payment level) each patient was assigned to. The categories were called Resource Utilization Groups (RUGs).4 Under the RUGs category system there were effectively only two areas of the patients’ clinical needs assessed: therapy and nursing. Further, the overwhelming emphasis was placed on delivering therapy services to patients. Once a patient received enough therapy, even a very modest level of therapy, nothing else mattered. In other words, once the patient received the minimum level of therapy required to be assigned to a “Therapy RUG,” none of the other clinical criteria captured in the MDS data had any significant influence on how much Medicare would pay for that patient’s care.
Under PDPM the profession will see dramatic changes in how patients are assigned to payment categories. Therapy, although still an essential part of care for most patients, is deemphasized, and the more clinically complex needs of patients (nursing, non-therapy ancillary (NTA) services, diagnosis and other complex services) become the primary drivers for payment.
Adding to the discussion, under the old RUGs model there was only one consideration regarding placement in a therapy category: the volume of therapy delivered: how many minutes of therapy delivered to each patient. The minutes of therapy were captured in the MDS assessment tool (along with over 100 other elements of care). But only the number of therapy minutes determined how a patient would be classified into a Therapy RUG category. This created an incentive to deliver therapy to patients, and the overwhelming majority of providers focused on the delivery of therapy services.
In fact, in recent years, 90 percent of Part A covered SNF days are paid using a Therapy RUG category, and only a small fraction of payment is influenced by non-therapy conditions (i.e., nursing needs).5 The PPS’s overutilization of therapy services under the RUGs model has been roundly criticized by many groups, including CMS,6 the Department of Health and Human Services’ Office of Inspector General7 and the Medicare Payment Advisory Commission (MedPAC).8
In an attempt to address the overemphasis on therapy services, CMS introduced the Resident Classification System, Version 1 (RCS-1) in 2017.9 The RCS-1 proposed rule called for the use of the same MDS assessment tool as the RUG-IV system with some modifications and additions. RCS-1 attempted to align payments with resource use instead of therapy-related financial incentives. To accomplish this, RCS-1 classified patients into separate groups for each of the four case-mix adjusted components: (1) physical therapy/occupational therapy (PT/OT); (2) speech-language pathology (SLP); (3) NTA services;10 and (4) nursing. Each of the four categories has its own case-mix indexes and per diem rates, with the per diem rates for PT/OT and NTA services variable based on changes in a patient’s resource use over a stay. The SLP and nursing component per diem rates would be added to the PT/OT and NTA services component variable per diem rates to arrive at the full per diem. Instead of a consistent rate throughout an assessment period, rates are highest at the beginning of a patient’s stay and decrease over time.
While RCS-1 was never implemented, it did not disappear. In May 2018 it was significantly revised and reintroduced11 as the PDPM.12
The Future: Patient-Driven Payment Model
PDPM considers a much broader range of clinical characteristics, patient diagnoses and overall medical needs when assigning patients to payment components.
Clinical Components Under PDPM
An MDS assessment will be used to identify a classification for each of the five following clinical components:
· Nursing - The identification of medically complex conditions that require more nursing services;
· NTA services;
· PT;
· OT; and
· SLP.
Each of these five components has its own set of categories and payment rates that correspond to those categories. The following shows the number of categories each clinical component will have:
Clinical Component # of Categories
Nursing 25
NTA services 6
PT 16
OT 16
SLP 12
PDPM attempts to limit the emphasis on delivering more therapy minutes by reducing the financial incentives to deliver more minutes of therapy present in the RUGs model. PDPM also aims to provide more accurate reimbursements for medically complex patients through its emphasis on a much broader range of characteristics, including the non-therapy medically complex needs that require more nursing services. PDPM removes therapy minutes as the basis for payment, and instead replaces it with mutually exclusive patient groups based on patient characteristics and additional adjustments. Under PDPM, there are now five unique areas of clinical need assessed. The therapy component, previously driven under PPS only by the number of minutes delivered, is separated into three parts under PDPM, each of which is driven by patient clinical characteristics. The clinical characteristics pointing to the needs for specific types of therapy, not minutes delivered, are assessed in deciding how much to pay for therapy services.
“Non Case-Mix” Components Under PDPM
In addition to the clinical components, there is one rate component — the “non case-mix” component — which pays providers for the non-clinical resources, or “overhead,” needed to deliver care to patients. These include dietary services, housekeeping, laundry, administration, medical records, rent and facility maintenance.
Calculating the Per Diem Rate
The corresponding payment rates for each of the clinical components are summed up and added to the “non-case mix” component to determine the full per diem payment rate for each patient. With five separate case-mix adjusted components that will be combined to arrive at the actual payment rate, there will be far more "categories" (possible combinations) under PDPM than under RUG-IV, where there were 66 categories. Under PDPM there are literally thousands of possible combinations. However, it is anticipated that any given provider will probably see only a few hundred of the possible combinations.
Further, the initial payment amount will decrease over the length of stay, which is different from the RUGs model. For the therapy components (PT, OT and SLP), the rate for each discipline will be reduced by two percent each week after day 21. For NTA services the amount paid in the first three days of the stay will be increased by 300 percent. This 300 percent increase in the first three days for the NTA component reflects the significant additional costs incurred at the beginning of the stay for patients with medically complex conditions. The reductions in payment over time are expected to have some impact on the average length of stay because they reduce the financial benefit (the payments to providers).
MDS Assessments Under PDPM
One of the major changes in the transition from RUGs to PDPM is the reduction in the number and frequency of the MDS assessments. Under RUGs, providers were required to complete a series of MDS assessments at various points in the patient’s stay. For any individual patient there could be five or more separate assessments during the course of the patient’s stay. Under PDPM there is only one assessment required, and it is at the beginning of the stay.
This reduction is not an indication that the MDS assessment becomes less important or less meaningful. The MDS assessment actually becomes more important than ever before. This is because all of the elements of the MDS matter now: They all will contribute to the categories the patients are assigned to, and therefore the amount Medicare will pay the SNF. Under the old system, even though there were a series of MDS assessments, the only component that mattered for the overwhelming majority of patient assessments was the number of therapy minutes delivered, because that ultimately determined the reimbursement rate. Providers will need to place significantly more focus on the MDS assessment under PDPM because the per diem rate is set by an individual patient’s clinical characteristics, and not the minutes of therapy services delivered.
CMS has indicated that it believes providers will save money as a result of the decrease in the number of required MDS assessments, about 183 hours per provider per year, which translates to $12,092.13 That is not necessarily the case. Rather, providers will need to place more focus on the MDS process, and failure to do so can result in potentially costly problems including under/over payments.
Volume of Therapy Services Under PDPM
As noted above, since PDPM deemphasizes the focus on the volume of therapy services delivered to patients, the number of therapy minutes will not determine how much a SNF will be paid by Medicare. However, the number of minutes of therapy will still be reported in the MDS, and CMS will still be monitoring the volume of therapy delivered to patients in order to compare the number of therapy minutes delivered under PDPM to the number of therapy minutes previously delivered under RUG-IV.
PDPM In Practice
MedPAC acknowledges that the transition to PDPM will require “considerable changes” to the SNF profession, but those changes have the potential to be very positive for SNFs, therapy service providers, Medicare, and most importantly, patients. Specifically, because a patient’s per diem will be higher at the beginning of the stay, and less as time passes, patients should have a shorter length of stay because facilities are incentivized to increase the efficiency of rehabilitation and therapy services.14 Only time will tell whether the belief MedPac expresses will be realized.
Moreover, changing the payment incentives from an almost exclusive focus on the number of therapy minutes delivered to a broad range of clinical characteristics will provide additional incentives for SNFs to develop the capabilities and programs necessary to care for more medically complex patients, which means more options for those patients. The addition of NTA services as a component should also increase reimbursement for patients with medically complex conditions and encourage providers to admit more of those types of patients. The same is likely true for the increased focus on nursing services, i.e., the clinical characteristics that point to the need to more nursing time. All parties are interested in whether these changes to nursing and NTA services will be enough to motivate providers to care for more medically complex patients.
Conclusion
Although the shift to PPS 20 years ago resulted in significant unintended consequences, the transition to PDPM has been more methodical. Financially, the transition to PDPM is supposed to be budget-neutral.15 Additionally, CMS intends to continue providing the annual market basket increase (inflating the rates). Whether any individual SNF performs better or worse financially depends on many factors, and success requires preparation and a forward focus on the areas that will help ensure success under PDPM. With proper planning, there is opportunity for all stakeholders to benefit.