Alleged false claims for ultra-high rehabilitation therapy furnished to SNF residents

All of the SNF CIAs arose, in large part, from allegations that the SNFs provided therapy services that were not medically necessary or reasonable to maximize Medicare reimbursement.

As background, the Centers for Medicare & Medicaid Services (CMS) makes Medicare payments to SNFs on a per diem basis, with the per diem payment rate covering the cost of all covered SNF services furnished to the Medicare beneficiary (subject to limited exceptions).^[5] For each Medicare beneficiary, the per diem payment rate is equal to a federal base rate subject to specified adjustments, including, in relevant part, an adjustment based on the SNF’s case mix (i.e., differences in patient types).^[6]

From October 1, 2010, to September 30, 2019, this case mix adjustment was determined in accordance with a Resource Utilization Groups (RUGs), Version IV (RUG-IV) model.^[7] Under the RUG-IV model, Medicare beneficiaries were assigned to a therapy resource utilization group based on (i) the total number of therapy minutes—for physical therapy, occupational therapy, and speech therapy—provided by the SNF to the patient during an assessment period and (ii) the patient’s “activities of daily living” (ADL) score for bed mobility, transfer, eating, and toilet use. This therapy RUG assignment, in turn, typically would affect the per diem payment amount received by the SNF for that beneficiary. Under the RUG-IV model, the most intensive (and hence highest reimbursing) therapy RUG category was ultra-high rehabilitation. Within this ultra-high rehabilitation therapy RUG category, one of three RUG codes could be assigned, which varied based on the patient’s ADL score, and which could affect the reimbursement to the facility. All three of the ultra-high rehabilitation therapy RUGs required that the SNF provide to the beneficiary 720 minutes of therapy in at least two therapy disciplines (i.e., physical therapy, occupational therapy, or speech therapy) during an assessment period, as reported on the patient’s Minimum Data Set (MDS).^[8] The MDS is a standardized care assessment tool used by SNFs to assess the patient’s condition as it relates to issues such as hearing, speech, and vision; cognitive patterns; mood; behavior; functional abilities and goals; bladder and bowels; active diagnosis; health conditions; swallowing and nutritional status; skin conditions; medications; and special treatments, procedures, and programs.^[9]

For all six of the newly imposed SNF CIAs, as well as the amended SNF CIA, the alleged conduct took place prior to October 1, 2019, when this RUG-IV model was still in place. The SNFs were alleged (directly or through their corporate parents) to have implemented policies and engaged in practices designed to maximize the number of SNF residents billed at the ultra-high rehabilitation therapy RUG level. The alleged result was that the SNFs furnished therapy that was neither reasonable nor necessary for the patient in order to meet this billing goal. Although the specific policies varied, they generally involved alleged company-wide quotas, budgets, or targets that put pressure on SNFs, and therapists in particular, to classify residents at the ultra-high rehabilitation therapy RUG level without regard to the resident’s actual conditions, diagnoses, or needs. Of particular note:

• At least four of the SNF CIAs involved assertions that the SNFs were required by their corporate parents to assign some portion of current or newly admitted residents to the ultra-high rehabilitation therapy RUG level, regardless of the resident’s individual needs.^[10] For example, one corporate owner allegedly mandated that its SNF facilities assign 70% of newly admitted residents to the ultra-high rehabilitation therapy RUG level; two others allegedly required all newly admitted residents to be assigned to this level.^[11]

• Three of the SNF CIAs involved allegations that the SNFs’ corporate owners set unrealistic targets and quotas for therapists. One company, for instance, allegedly required therapists to provide billable services 90% of the time they were at the SNF.^[12] The relator argued that such a mandate left little time for therapists to conduct nonbillable services, such as patient notes, and caused therapists to bill group therapy as individual sessions to increase their productive time. Another company was alleged to have corporate policies that tied therapist productivity to the amount of time the therapist spent on Medicare billable work, which allegedly caused therapists to report unbillable time for things such as their initial patient evaluations and time spent providing unskilled services as skilled therapy time.^[13] A third company allegedly pressured therapists to increase the amount of therapy provided to patients by setting a goal that 100% of Medicare Part A patients would be prescribed rehabilitation therapy.^[14]

• Some of the corporate entities also allegedly assigned SNF budgets based on a presumption that a majority of the SNF’s Medicare Part A patients would be billed at the ultra-high rehabilitation therapy RUG level, putting pressure on the SNF to achieve such targets.^[15]

• In at least one instance, the corporation’s officers and the SNF’s administrative staff instructed nursing staff to reassess patients in order to increase their ADL scores, thus increasing their resulting RUG level and overall billing. Nurses who did not comply were threatened with termination.^[16] Two of the SNF CIAs also involved allegations that nonclinical corporate employees overrode and ignored the recommendations of the SNF therapists and other clinical staff, including by delaying the resident’s discharge to extend the length of stay, thereby increasing reimbursement.^[17]

Some of the SNF CIAs also included allegations of “thresholding” (i.e., claims that the SNF provided just enough therapy minutes for the Medicare beneficiary to qualify for the higher RUG level). CMS noted this trend as early as 2014, when it published a finding that ‘‘the percentage of claims-matched MDS assessments in the range of 720 minutes to 739 minutes, which is just enough to surpass the 720 minute threshold for RU groups, has increased from 5 percent in FY 2005 to 33 percent in FY 2013.”^[18] The SNF CIAs noted in this article demonstrate the continued trend of thresholding since that time. For example, one SNF CIA involved assertions that a SNF’s “Rehabilitation Service Managers” set or adjusted the length of patients’ daily therapy sessions based upon the number of minutes needed to reach the targeted RUG level, despite the fact that these individuals were unfamiliar with the patients’ therapy needs.^[19] Similarly, two other SNF CIAs allegedly monitored the amount of therapy provided to residents billed at the ultra-high rehabilitation therapy RUG to ensure that therapy beyond the required 720 minutes was not provided.^[20] For instance, one company allegedly circulated “overage” reports reflecting the amount of therapy provided above a particular RUG threshold, with corresponding emails explaining that the SNF is “not being paid to provide excess [therapy] minutes.”^[21]

These SNF CIAs also included allegations that the SNFs (directly or through their corporate owners):

• Recorded therapy minutes as individual therapy when concurrent or group therapy was provided to increase the amount of therapy minutes attributable to each resident for billing purposes;^[22]

• Provided unskilled services performed by skilled therapists in order to bill as skilled therapy;^[23]

• Billed for services that were never provided to ensure patients remained at the ultra-high rehabilitation therapy RUG level;^[24]

• Included unbillable time spent on initial evaluations as part of the reported therapy time;^[25]

• Arbitrarily shifted the number of minutes between different types of therapy disciplines to ensure that the ultra-high rehabilitation therapy requirements were met;^[26]

• Adjusted patient assessments to increase the amount of therapy minutes prescribed solely to increase RUG levels;^[27] and

• Inflated resident ADL scores.^[28]

The allegations underlying these SNF CIAs appear to be representative of longstanding concerns raised by the OIG and the Medicare Payment Advisory Commission, an independent congressional agency, regarding how the RUG-IV model created financial incentives for SNFs to bill for more therapy services than medically necessary.^[29] To “better align” the SNF payment system with “resource use,” and to eliminate the “therapy provision-related financial incentives inherent” in the RUG-IV model, CMS replaced the RUG-IV model with a revised case-mix methodology, referred to as the Patient-Driven Payment Model (PDPM), effective October 1, 2019.^[30] Under the RUG-IV model, a SNF reduced everything about a patient into a single case-mix based upon the volume of care needed—the RUG level. The PDPM model is designed to be more flexible to an individual patient’s needs by classifying the patient within each of five different case-mix components—physical therapy, occupational therapy, speech-language pathology, nursing, and nontherapy ancillary services.^[31] Each case-mix component considers different criteria about the patient’s various characteristics in order to make the necessary classification. For example, a patient’s classification under the physical therapy and occupational therapy case-mixes each consider the patient’s clinical category (i.e., diagnosis) and functional score (as reported in Section GG of the MDS), while the patient’s classification under the speech-language pathology case-mix considers whether the patient has any acute neurological conditions, cognitive impairments, swallowing disorders, or speech-language pathology–related comorbidities. The various case-mix classifications are then used to predict the overall therapy costs associated with the individual patient’s medical need. In sharp contrast to the RUG-IV model, none of the case-mix components under the PDPM model consider the number of therapy minutes provided.

While the RUG-IV model is no longer in place, the SNF CIAs discussed above still provide valuable compliance insights, which are discussed in more detail below.

Alleged false claims of substandard services

One of the SNF CIAs also involved the alleged provision of substandard or worthless skilled nursing services. Specifically, the allegations involved a failure to (i) appropriately administer medications to some residents, (ii) follow appropriate pressure ulcer protocols, and (iii) abide by protocols to avoid falls.^[32] In addition, the SNFs allegedly had insufficient staffing to meet the residents’ needs. According to one relator, the corporate entity that owned and managed the SNFs implemented a company-wide staffing policy that restricted staffing at all SNFs to an average of 1.93 hours allotted per day per patient/resident, regardless of the residents’ needs and despite a congressional report suggesting such per day per patient/resident levels were “known to cause widespread and significant care deprivation.”^[33]

Once again, the allegations underlying this SNF CIA appear to be representative of broader government concerns regarding quality of care in the nursing home industry. On March 3, 2020, more than a year before the SNF CIA was imposed, DOJ announced a National Nursing Home Initiative to enhance civil and criminal enforcement against nursing homes that provide “grossly substandard care” to their residents.^[34]

Alleged false claims for medically unnecessary hospice services

Similar to some of the alleged conduct discussed above relating to ultra-high rehabilitation therapy, the sole Hospice CIA arose from allegations that the hospice provider (i) billed for patients who were ineligible to receive hospice services under Medicare requirements, and (ii) billed Medicare, Florida Medicaid, and TRICARE for higher levels of care that were not medically necessary.^[35]

The hospice benefit under Medicare (which is a per diem payment) covers a range of services for “terminally ill” patients, defined as individuals who have received a medical prognosis that their life expectancy is six months or less.^[36] The hospice provider allegedly billed Medicare for at least 48 consecutive months of hospice care, during which time the patients were not eligible for the Medicare hospice benefit because, for at least a portion of the time, they were not terminally ill.^[37]

Additionally, the hospice allegedly provided care that was not medically necessary. Medicare, Medicaid, and TRICARE reimburse for four different levels of hospice care: routine home care, continuous home care, inpatient respite care, and general inpatient care (GIP). GIP is intended for patients who require symptom or pain management that cannot be managed in other settings. As the most intensive kind of hospice care, it is intended to be short term and is reimbursed at the highest level. According to the federal government and relator, the hospice billed at least 15 consecutive days of GIP hospice care in circumstances where patients did not require this level of care for at least a portion of the time, resulting in services that were not medically necessary or reasonable.^[38]