CMS Therapy Payment Model Under Review: What to Expect in 2016

In 2015, the Medicare payment system for therapy services provided by Skilled Nursing Facilities (SNFs) saw intense media and governmental scrutiny. Several news articles and reports were published on the need for reform, many of which highlighted increased billing by SNFs for therapy services at the "ultra-high" (i.e. highest) reimbursement level despite allegedly static beneficiary characteristics.¹

At this point, it appears that the current reimbursement model for therapy will almost certainly change. The only real question is how. Medicare's present payment model establishes a daily reimbursement rate for three different categories: (1) nursing and non-therapy ancillary services; (2) therapy care; and (3) room and board. Since its inception in 1998, the therapy billing model, which classifies patients into different resource utilization groups (RUGs) based on their care needs, has been criticized for encouraging SNFs to provide "unnecessary" rehabilitation therapy services in order to maximize reimbursement rates.²

The reimbursement rate for therapy services is based on the patients' rehabilitation needs and depends, in part, on the RUG to which a patient is assigned. There are five general RUG levels for rehabilitation therapy patients: (1) Rehab Ultra High (RU); (2) Rehab Very High (RV); (3) Rehab High (RH); (4) Rehab Medium (RM); and Rehab Low (RL). The highest daily rate Medicare will pay an SNF is reserved for patients who require "Ultra High" levels of skilled therapy, or a minimum of 720 minutes of therapy per week from at least two therapy disciplines (e.g., physical, speech, and/or occupational), and one therapy discipline must be provided at least five days per week. Under the current payment model, a patient's assigned rehabilitation RUG level depends on the number of skilled therapy minutes and number of therapy disciplines the patient received during a seven-day assessment period, referred to as the "look back period."

Recent criticism of the therapy reimbursement model centers on the steady increase in Medicare reimbursements for ultra-high therapy. According to the Office of Inspector General's September 30, 2015 report, SNFs received an average of $66 per day more than their therapy costs for ultra-high therapy, versus an average margin of just $11 per day when billing for low therapy. The OIG report also challenges an industry-wide claim that the increase in ultra-high billing is mainly the result of greater patient sophistication, demand, and/or acuity. The OIG's data from FYs 2011 to 2013 shows that ultra-high therapy billing rose, even though key patient characteristics, like age and condition, generally remained static. Significantly, the OIG report attributes roughly 80 percent of the $1.1 billion increase in Medicare payments to the overall uptick in ultra-high therapy billing among SNFs.

The Centers for Medicare and Medicaid Services (CMS) acknowledges that therapy reimbursements are currently "based primarily on the amount of therapy provided to a patient, regardless of the specific patient characteristics and care needs." In an effort to address this apparent disconnect between reimbursement and care needs, CMS initiated the SNF Therapy Payment Research project and contracted with a third party (Acumen, LLC) in an effort to "identify potential alternatives to the existing methodology used to pay for services." Phase one of the project involved the contractor's review of past research and policy issues concerning the therapy payment model, as well as options for "improving or replacing the current system." The contractor's conclusions from this phase were published in a report in April 2014. This report identified four potential payment system options:

1. Resident Characteristics Model – uses resident information (e.g., medical, functional or cognitive status) to group residents with similar clinical characteristics and expected cost of care to determine reimbursement
2. Hybrid Model – combines case-mix classification system³ with resource-based pricing adjustment; similar to current therapy payment model's inclusion of resource use into payment calculus
3. Fee Schedule – payment amounts are based on resident's actual therapy use, not expected resource use
4. Competitive Bidding – market-based pricing of therapy services through a competitive bidding process

Of the four options described above, the CMS contractor recommended using the resident characteristics and hybrid model concepts to "inform the specific models that will be developed and tested in the next phase of the project."

Currently, CMS's project is in its second phase, where the contractor is using its above findings to "identify potential models suitable for further analysis." This process is ongoing, and comments/feedback on therapy payment methodology may be submitted any time to SNFTherapyPayments@cms.hhs.gov.

Long term care providers who offer therapy services can expect to see the current scrutiny of therapy billing by CMS to continue in 2016. In conjunction with CMS's exploration of alternative therapy billing models, federal and state governments have stepped up their efforts to detect perceived abuses in therapy billing and have paid particularly close attention to long term care providers who seek what they deem to be "excessive" ultra-high therapy reimbursements. Civil Investigative Demands and other subpoenas requesting company-wide therapy billing records, patient charts and other data are on the rise. In view of the increased attention being paid to therapy reimbursements, it is critical that all long term care providers review their therapy billing records and, if appropriate, conduct an internal audit to verify that the therapy services being provided are necessary, patient-specific and properly documented. Over the past year, Baker Donelson has conducted multiple investigations into therapy billing, supervised internal audits and assisted several clients with responding to state and federal subpoenas.

¹ See, e.g., "The Medicare Payment System for Skilled Nursing Facilities Needs To Be Reevaluated," Department of Health and Human Services, Office of Inspector General Report, Sept. 2015, OEI-02-13-00610; "How Medicare Rewards Copious Nursing-Home Therapy," Wall Street Journal, Aug. 16, 2015; "Nursing Homes Bill for More Therapy Than Patients Need, U.S. Says," The New York Times, Sept. 30, 2015; and "The Need to Reform Medicare's Payments to Skilled Nursing Facilities Is As Strong As Ever," Medicare Payment Advisory Commission and Urban Institute, Joint Report, Jan. 2015.

² Medicare Payment Advisory Commission and Urban Institute, Joint Report, Jan. 2015, at p. 2.

³ Case-mix groups are comprised of residents "expected to receive a similar amount of therapy." These residents could be grouped according to: (1) resource-based options; (2) resident characteristic-based options; and (3) outcomes-based options. In a resource-based payment system, reimbursement would be based on "resource utilization, such as the number of minutes of therapy or the frequency of therapy sessions." The resident-characteristic-based option would group residents based on certain characteristics. The outcomes-based option would "group residents with similar clinical profiles and expected outcomes from therapy."