Bundling Payments for Post-Acute Care

The traditional Medicare program pays individual health care providers for the specific services and care they provide to beneficiaries and guarantees that patients have “freedom of choice”[1] to select their Medicare providers.  A current focus of Congress and policymakers is changing Medicare payment policy to pay, instead, for episodes of care for beneficiaries.  One issue in the current discussion is how to pay for care after hospitalization in an acute care hospital, so-called post-acute care (PAC).[2]  “Bundling” post-acute care is an approach that pays a fixed amount for all services provided to a patient after hospitalization for a defined period of time.  Although concern is expressed about “silos” of care and the lack of coordination among PAC providers, a significant motivation is the finding that the largest differences in Medicare costs across the country actually reflect the large differences in the costs of post-acute care.[3]  While advocates support better coordination of care for beneficiaries,[4] we have significant concerns about bundled payments and their potential to result in skimping on care for patients.[5]  Legislation under consideration also raises questions about patients’ rights, conflict of interest, unrecognized Medicaid costs, and observation status.

The Bundling and Coordinating Post-Acute Care Act of 2015

The Bundling and Coordinating Post-Acute Care Act of 2015 (the BACPAC Act of 2015), H.R. 1458, gives a post-acute care coordinator a fixed single amount of reimbursement to pay for post-acute care for 90 days following discharge from an acute-care hospital for a person who had been a hospital inpatient.  The PAC bundle includes most post-acute services, although it explicitly excludes others (physician services, hospice, outpatient hospital services, ambulance, outpatient speech-language pathology, and items and services described in §1861(s)(9)) (“leg, arm, back, and neck braces, and artificial legs, arms, and eyes, including replacements if required because of a change in the patient’s physical condition”)) (§3(b)).

PAC coordinators may be hospitals, PAC providers, insurers, third-party administrators, or combination of hospital and PAC providers (§3(c)).  PAC coordinators are certified by the Secretary and have a PAC agreement with the Secretary (§3(c)).

The Secretary must establish a process for patients to select and change their PAC coordinator, although the hospital can limit patients to the PAC coordinators identified by the hospital.   (Patients must have at least two choices).  (§3(c)(4)(C)(i)).  The bill specifies that if a patient does not choose a PAC coordinator, the Secretary assigns a PAC coordinator (§3(c)(4)(D)).

PAC coordinators must have PAC network agreements “with a sufficient number of PAC providers in a PAC area” to meet “network adequacy requirements as are established by the Secretary” (§2(c)(2)(C)).

The bill requires the Secretary to establish four classifications of beneficiaries “and a methodology for classifying specific PAC bundles within these groups,” using the standardized PAC assessment tool (§3(d)(1)).  It also authorizes adjustments to the four bundles for geographic, risk, and socio-demographic factors (§3(d)(6)).  Socio-demographic factors include dual-eligibility (i.e., eligible for both Medicare and Medicaid)[6] and whether the beneficiary has a willing and able caregiver.  The bill requires a phase-in of PAC bundling, starting first with the highest cost beneficiaries. (§3(e)(4)).

If the cost of PAC care for beneficiaries in the 90-day period is less than the bundled payment, the savings are shared.  PAC coordinators get 55% of the savings; PAC providers get 15%; PAC physician, 15%, hospital, 15% (§3(c)(3)(D)).

Beneficiary Concerns

1. Patient rights.  The bill raises a number of concerns for beneficiaries, particularly about patient rights.

First, must the patient agree to participate in BACPAC?  Is participation in BACPAC mandatory or voluntary for patients?  The bill appears to be silent on this issue, although the provision requiring the Secretary to assign a PAC coordinator to a patient who does not choose one suggests that participation may be mandatory.

Second, will the patient know at the time of selecting the PAC coordinator which PAC providers are in the network and available to the patient?  What standards will the Secretary use to determine network adequacy and provider availability?

Third, the bill does not address the specific rights a patient has after choosing (or being assigned) a PAC coordinator.  Section 2 of the Purposes section suggests that a patient can select his or her preferred providers of PAC services, but is this right limited to choosing one skilled nursing facility (SNF), rather than another, in the network?   Can a patient object to using a particular category of PAC provider?  Suppose, for example, that the patient’s attending physician recommends an inpatient rehabilitation hospital (IRH) and the patient wants to go to an IRH, but the PAC coordinator determines that a SNF is appropriate.  Can the PAC coordinator limit the patient’s choice and overrule the patient and physician?  Can the PAC limit the patient to a choice of home health care provider and deny coverage for care in an IRH or SNF?

These concerns are increased by the bill’s requirement that the Secretary establish four classifications of patient conditions and a methodology to classify the specific PAC bundles within each group (§3(d)(1)).  This section suggests that different PAC coordinators will have different PAC networks, but it does not specify whether the full range of PAC providers must be included for each of the four classifications.  Again, what standards will the Secretary use to determine network adequacy?  Can the Secretary and PAC coordinator limit access to certain PAC services by not including them in the bundle for certain beneficiaries?  Could the Secretary, for example, determine that patients with hip fractures can only get post-acute care in a SNF?  Or would the full range of PAC providers, including IRHs, be required for each classification?  The bill waives current barriers for lower cost care — it waives the 3-day hospital requirement for SNFs (§3(b)(2)(C)(i)) and the homebound requirement and the face-to-face documentation requirement for home health care (§§3(b)(2)(C)(ii), (ii), respectively) – although it also waives the percentage requirements for IRHs and long-term care hospitals (§§3(b)(2)(C)(iv), (v), respectively).

Finally, nothing in the bill appears to require that the Secretary provide any notice and appeal rights to patients. 

2. Conflict of interest.  The PAC coordinator collects the largest portion of shared savings under BACPAC – 55%.  Since the coordinator is responsible for Medicare costs for 90 days, the coordinator could be motivated to identify PAC services that get the patient through the 90-day period (without a new inpatient re-hospitalization) at the lowest cost.  But the lowest cost PAC provider may not provide the patient with the best services for the patient.  (See point 3, below.)
3. Medicaid costs.  The bill looks only at costs to the Medicare program for a 90-day period.  These are not the only relevant costs, however.  The full costs of post-acute care over a longer period than 90 days, including Medicaid costs, need to be considered.

Some studies find that patients who get post-hospital rehabilitation in IRHs not only have better long-term outcomes than patients receiving rehabilitation in SNFs, but also are less likely to need long-term care afterwards, paid for by the Medicaid program.  An analysis of post-acute care for patients receiving hip fracture repair found that IRH patients had shorter lengths of stay than SNF patients and that none of the IRH patients was discharged to a nursing facility, compared to 16% of SNF patients.[7] 

Shifting PAC costs from Medicare to Medicaid does not save federal money.

4. Observation status.[8]  The PAC period ends if a patient is re-hospitalized as an inpatient.  The bill’s exclusion of outpatient hospital services from BACPAC means that a patient who returns to the hospital as an outpatient, including observation status, would continue to be in BACPAC.  This obvious loophole will increase the use of observation status, undermining efforts to reduce re-hospitalization.

Conclusion

Improving coordination of care for Medicare beneficiaries could result in improved quality of care for patients at lower cost to the Medicare program.  But coordination is already mandated.  The Medicare program already requires that hospitals conduct meaningful discharge planning to identify the appropriate PAC setting for inpatients after they are discharged.[9]  Unfortunately, BACPAC adds a financial incentive to skimp on post-acute care.  The bill’s requirement to phase in bundled payments, beginning with the highest cost beneficiaries, also suggests that cost-savings, rather than improved coordination of care for patients, is the key goal of the legislation.

May 2015 – T. Edelman