Are older adults with dementia “rehabbed to death”? This is the contention of a perspective published in the New England Journal of Medicine, describing the downward cycle of rehospitalization leading to death that many of these older adults experience. Hospitalization is a particularly significant event for older adults with dementia, who may experience cognitive decline and physical deconditioning during their hospital stay. Thus, patients and families face difficult questions as they approach the end of that hospital stay: how can we meet this older adult’s care needs, and what care setting is most appropriate? Will they recover, rehabilitate, and return to their pre-hospitalization selves? Our findings in two recent papers suggest skilled nursing facility (SNF) care isn’t “futile,” and that for a sizeable minority of older adults with dementia, care delivered in the home following hospital discharge is equivalent to that delivered in a SNF in terms of shorter-term outcomes.

In both studies, we analyzed outcomes for hospitalized adults age 65 and older who were enrolled in fee-for-service Medicare, and received either SNF care or home health care after hospital discharge. We measured hospital readmission and death up to 100 days following hospital discharge. In the first study, we matched more than one million patients with and without dementia based on things like their age, physical function, comorbidities, and characteristics of their inpatient stay (length of stay, whether it included an intensive care unit stay). In the second study, we focused only on patients with a diagnosis of dementia and compared similar groups discharged to a SNF versus home.

In the first study, we found that older adults with a diagnosis of dementia had comparable outcomes to similar older adults without a diagnosis of dementia when they received post-hospital care in SNFs. While a diagnosis of dementia did not seem to differentiate outcomes, the level of cognitive impairment on admission to the SNF was strongly associated with differences in outcomes. In fact, we were surprised to find older adults without a diagnosis of dementia, but who had moderate or severe cognitive impairment, had the worst outcomes of all. These were likely older adults who were so ill they had delirium or other medical problems impairing their cognition. We made a counterintuitive conclusion from this paper: we shouldn’t focus on a diagnosis of dementia when it comes to deciding whether an older adult should receive post-acute care in a SNF. Instead, what is much more relevant is their mental status at the time of hospital discharge—whether or not they have a diagnosis of dementia.  

Second, we found that the outcomes of older adults with a diagnosis of dementia receiving post-hospital care in SNFs were equivalent to those receiving post-hospital care at home delivered by a home health agency. The latter finding is particularly striking given more than 7 in 10 older adults with a diagnosis of dementia are discharged from the hospital to a SNF rather than home. In comparison, similar adults without dementia are equally likely to receive post-hospital care in a SNF or at home with a home health agency. 

These results are important given more than three million Medicare beneficiaries currently carry a diagnosis of dementia, and the prevalence of dementia is expected to increase as the U.S. population ages. Medicare already spends more than $60 billion annually on post-hospital care like SNFs and home health agencies.

Meeting the individual needs of older people living with dementia after hospitalization requires that we provide more support for families caring for these patients at home. We know that unpaid caregivers already supply most of the care to older adults in this country, and that this has a substantial impact on their well-being and finances. In addition to financial support of caregivers, we need to better align medical and non-medical supports. “Skilled” care (medical services) and “home and community-based services” (generally non-medical supports for activities of daily living) are segregated by our current payment mechanisms. Medicare provides short-term “skilled” care for medical needs and Medicaid provides long-term “non-skilled” care to support activities of daily living. Payment models that integrate these supports after hospital discharge are needed. Promising demonstrations to better align these payments and supports are being tested in the Medicare Advantage population, who can now receive non-medical supports through their Medicare Advantage plan. Dual-eligible (Medicare and Medicaid) beneficiaries are participating in the Financial Alignment Initiative, Special Needs Plans, Independence at Home, and the Program for All-Inclusive Care for the Elderly, all of which better integrate these supports but are currently limited in their scope. An integrated care and payment program that addresses the medical and functional needs of patients can help avoid the cycle of being “rehabbed to death,” and provide a better system for meeting these patients’ needs at home.

This post is based on the following studies:

Outcomes of post-acute care in skilled nursing facilities in Medicare beneficiaries with and without a diagnosis of dementia, published on June 25, 2021. Authors include Robert E. Burke, Yao Xu, and Ashley Z. Ritter.

Postacute care outcomes in home health or skilled nursing facilities in patients with a diagnosis of dementia, published on August 12, 2021 in Health Services Research. Authors include Robert E. Burke, Yao Xu, Ashley Z. Ritter, and Rachel M. Werner.


Robert Burke

Robert Burke, MD, MS, FHM

Assistant Professor, Perelman School of Medicine

See More on Improving Care for Older Adults