With implementation of the patient-driven payment model (PDPM) around the corner, there is much being written about the virtues of basing payment on clinical characteristics and the reduced administrative burden of fewer assessments. While few would argue with both of these benefits, the fly in the ointment is what will happen to provision of rehabilitation services and rehabilitation outcomes without incentives to provide rehabilitation.

The assumption is that providers will provide the amount of therapy that a resident requires, and the classification into physical therapy, occupational therapy, or speech-language pathology groups will ensure that payment is adequate. As simple as this sounds, there are challenges with this assumption.
First, several studies in recent years have demonstrated that the more therapy services a skilled nursing facility (SNF) resident receives, the better the rehabilitation outcomes.

For example, a recent study showed that each additional hour of therapy received per week was associated with a 3.1 percentage point increase in the likelihood a SNF resident was discharged home.

A recent Medicare Payment Advisory Commission (MedPAC) report showed that for every hour of physical therapy per resident day at the facility level there was a 6.9 percentage point increase in the facility rate of mobility improvement and a 16.7 percentage point increase in the rate of community discharge.

The dilemma is that, other than “more is better,” we do not have data on how much therapy is necessary to improve outcomes for different types of residents. Moreover, the development of PDPM did not take outcomes into consideration when determining therapy groups or payment.

This is compounded by the Centers for Medicare & Medicaid Services (CMS) dropping the community discharge outcome measure from public reporting. Furthermore, the functional outcome measures recommended by MedPAC have not been adopted by CMS. Thus, the only measure of rehabilitation outcomes is the Minimum Data Set Quality Measure representing the “percentage of short-stay residents who made improvements in function,” which is notoriously insensitive to outcomes of rehabilitation.

Some evidence exists that just as with nurse staffing hours, there is a level of therapy provision above which SNF residents receive little additional outcome benefit. This amount of therapy is about 720 minutes, equivalent to the current ultra-high rehabilitation resource utilization group (RUG) level. Perhaps it is appropriate and beneficial that, increasingly, residents are receiving this amount of therapy in SNFs. Under PDPM, we will see how many residents still receive this or similar amounts.

If more therapy and a more robust rehabilitation program produces better outcomes for elders, are we setting up SNF residents for worse outcomes under PDPM? Possibly. That is not to say we should go back to payments based on therapy minutes. However, we should invest in rigorous measurement and reporting of rehabilitation outcomes under PDPM, so both providers and residents can optimize care, and CMS can adjust payments accordingly.

As stated by T. Franklin Williams, one of the founding fathers of geriatrics, “The aim of rehabilitation, to restore an individual to his/her former functional and environmental status, or, alternatively, to maintain or maximize remaining function, should be at the heart of all care of aging persons.”