Rehabilitation at End of Life Study: Look Beyond the Surface

Members | October 10, 2018 | by Aaron Tripp

A recent study published in the Journal of the American Medical Directors Association has received some press coverage due to the authors’ claim that the “study reveals a growing trend of potentially unnecessary - and harmful - high intensity rehabilitation services for residents of nursing homes.” However, there is more to the story than some of the headlines reveal.

The study only reports on MDS data without any comparison group, and by design does not try to explain the reason for the growing trend. It uses long-stay nursing home residents in New York State from 2012 through 2016 as the study population. Results focus on the provision of therapy based on RUG-IV categorization during end of life and make distinctions on the amount of therapy based on facility characteristics such as proprietary status, rural/urban location, and staffing levels among others.

It should be noted that the study does offer support for the non-profit difference. Findings suggest that for-profit facilities and those with higher staffing levels of therapists have a greater proclivity to maximize therapy use. Residents of for-profit facilities have significantly higher risks for all intensity of therapy provision. About 14% of long-stay decedent residents had some therapy in the last month of life and of those, almost 16% had more than 500 minutes prior to death.

Several limitations are acknowledged by the authors, maybe most importantly that because the study was based on administrative data, they cannot determine the extent to which rehabilitative therapy provided may have been beneficial or harmful to the residents receiving the services.

Additionally the story may be incomplete because it does not include data related to resident election into hospice and palliative care provision at the end of life.

Finally, with the move to the patient-driven payment model (PDPM), as included in the 2019 skilled nursing facility final rule, CMS indicates they “believe it is important to remove, to the extent possible, service-based metrics from the SNF PPS [skilled nursing facility prospective payment system] and derive payment from verifiable resident characteristics”. This change has the potential to mitigate the therapy incentives that the study reports on. However, time will tell what the implementation effects of the PDPM system are.