Image by Getty Images
Image by Getty Images

To home or SNF?

Few tools and competing incentives make discharge decisions challenging.


An elderly patient admitted for pneumonia aspirates while on a feeding tube and subsequently develops delirium and problems with mobility and cognition. Although frail, the patient is judged medically ready to leave and the hospitalist—under pressure to expedite discharge—must make a quick decision about where to refer her: a skilled nursing facility (SNF) or supervised home care.

The patient had been living independently prior to admission and may be a good candidate for home-based care, but her status at the time of discharge makes the choice challenging.

“We really struggle with patients in this gray area where the choice of home or SNF is not clear-cut,” said Robert Burke, MD, MS, a hospitalist and investigator at the Center for Health Equity Research and Promotion at the Corporal Michael J. Crescenz VA Medical Center and University of Pennsylvania in Philadelphia.

“Besides functional impairment, patients may have comorbid psychiatric illness, or issues related to substance abuse, social isolation, or low health literacy. As providers, we have no standard criteria or evidence-based tools for making these decisions,” he added.

More than 40% of older patients receive postacute care after hospitalization and, of those, 90% are discharged to a SNF or supervised home care, according to a study in the May 2019 JAMA Internal Medicine. SNFs were found to be associated with fewer readmissions compared with home care, a plus for both patients and hospitals.

However, increasing use of SNFs is also a major driver of Medicare postacute care costs, now estimated at upwards of $60 billion annually. To contain spending, CMS has embraced bundled payment systems and accountable care organizations (ACOs), in which hospitals and SNFs are jointly accountable for outcomes and costs.

Such programs have already contributed to reduced costs through lower use of inpatient care and fewer referrals to and shorter stays at SNFs. In a study published in the April 2017 JAMA Internal Medicine, spending on postacute SNF care dropped by 9% in 2014 for patients enrolled in ACOs compared with non-ACO patients.

Another study, published in the Jan. 17 New England Journal of Medicine, reported a modest reduction in Medicare spending for hip or knee replacements during the first two years of bundled payments, without any increase in complications.

“ACOs and some bundled payment models have been clear wins in that there have been savings without ostensible harm to patients so far,” said J. Michael McWilliams, MD, PhD, a co-author of that study, lead investigator of the April 2017 JAMA Internal Medicine study, and professor of health care policy and medicine at Harvard Medical School in Boston. “The evidence suggests that we should continue to pursue and refine these models moving forward.”

Decisions on the front lines

Making a sound decision about where to discharge a patient is complicated by the lack of standard protocols and insufficient time before discharge for careful evaluation.

As a result, SNFs are often viewed as a safety net or default option when it is not clear whether patients would be better off in a SNF or at home, according to a study in the November 2017 Journal of the American Geriatrics Society, in which researchers interviewed 25 clinicians at three hospitals. Hospitalists reported having little knowledge about the inner workings of SNFs, no formalized criteria for making selections, and no follow-up data on outcomes related to their decisions.

Deciding where to refer patients can be more cut and dried after surgical procedures like a joint replacement, noted Joshua Liao, MD, MSc, FACP, associate medical director of payment strategy at the University of Washington in Seattle.

“However, patients with chronic conditions, such as chronic obstructive pulmonary disease or heart failure, can require more time to recover and regain function. It can be harder to identify clear end points,” he said.

Once referred to SNF care, patients and families are typically given a list of facilities sorted by ZIP code and left to select a facility on their own. Many ask for guidance from their care team, but hospital staff can be hesitant to make recommendations due to uncertainty about the legal implications, according to a study published in the August 2017 Health Affairs.

Many hospitals operate under the assumption that Social Security laws guaranteeing patients freedom of choice prevent offering any information beyond names and addresses of facilities, the study found. Clinicians interviewed for this article said they typically do not volunteer any advice or refer patients to the quality data available on Medicare's Nursing Home Compare website.

However, nothing in the statutes precludes hospitals from helping patients make informed choices, and Medicare's interpretive guidelines encourage directing patients to publicly available quality data, the study authors noted.

Nonetheless, many clinicians and discharge planners—some echoing instructions from top executives—believe that providing any additional information could compromise patients' freedom to choose. As a result, most patients end up selecting a facility based on proximity rather than quality, said Dr. Burke.

“Many patients have no idea where to go, so they choose a place close to their home, and within a few hours they're in an unfamiliar place and don't know what to expect,” he said. “It can be a very upsetting and disorienting experience.”

Promising interventions

Some hospitals are trying to improve decisions and efficiencies around postacute care by strengthening partnerships with SNFs—an increasingly essential strategy in a regulatory climate of joint accountability. For example, Bradley Flansbaum, DO, MPH, FACP, associate director of the department of hospital medicine for Geisinger Health System in Danville, Pa., has launched a SNF orientation program for hospitalists.

“We're trying to get hospitalists to see what it's like on the other side and get a feel for what can and can't be done in a SNF,” he said. “Hearing directly from SNF administrators and intake personnel about what happens inside SNFs and what issues concern them is often quite eye-opening.”

The program educates clinicians about issues in postacute care and arranges on-site visits. The idea is to overcome common barriers to successful transitions, including lack of preparation for handoffs and poor communication about the plan and goals of care.

During site visits, hospitalists talk directly with SNF personnel about common problems they encounter, such as incomplete discharge summaries or lack of personal communication from the hospital care team about special concerns, said Dr. Flansbaum. Other key messages have included improving communication about advance care plans and understanding the resource limitations and highly regulated environment of SNFs.

Geisinger is also using a mobility prediction tool built into the electronic health record, and Dr. Flansbaum is working to familiarize hospitalists with it, in order to standardize their assessment of patients' mobility and post-acute care needs.

The health system has adopted the 6 Clicks tool, created at Cleveland Clinic and based on validated patient outcome measures to provide guidance for clinicians on whether a patient can move out of bed or perform basic functions without the help of a physical therapist. The tool gives clinicians a more standardized way of deciding whether patients can be safely discharged home or need more intensive care at a SNF.

Other initiatives focus on improving communication and handoffs between hospitals and SNFs. Knowing more about the type and level of care provided at SNFs helps clinicians make more informed discharge decisions and facilitates smoother transitions, experts said.

For example, the Extension for Community Healthcare Outcomes-Care Transitions project, at Beth Israel Deaconess Medical Center in Boston, currently involves weekly videoconferences between hospital and SNF clinicians to discuss patient transitions. Data from the first two years (2013-2014) of the project were published in the March 2017 Journal of the American Geriatrics Society.

“We learned that the biggest challenge for SNFs, by far, was medication reconciliation,” said Grace Farris, MD, lead investigator, who is now chief of hospital medicine at Mount Sinai West in New York. “Hospitalists gained valuable insight into what happens on the SNF side and helped us uncover problems within our system.”

Health policy implications

Hospitals face varying financial incentives related to the decision of whether to discharge patients to a SNF or home care. Medicare's Hospital Readmissions Reduction Program (HRRP), launched in 2012, penalizes hospitals with 30-day readmission rates exceeding national averages by withholding up to 3% of Medicare payments.

If—as the May 2019 JAMA Internal Medicine study suggests—SNFs are perceived as safer than home care and are associated with fewer readmissions, the HRRP would not motivate reduced use of them. In contrast, ACOs and bundled payment initiatives reward participants for minimizing use of expensive institutional care after discharge.

The new payment models are prompting hospitals to pursue closer working relationships and shared networks with SNFs, said Rachel Werner, MD, PhD, professor of medicine at the University of Pennsylvania Perelman School of Medicine, and lead author of that study. For example, some hospitals are dispatching discharge planners or nurses to manage patients across the transition from hospitals to SNFs.

However, while SNFs are driven to work with hospitals in order to increase and maintain referrals, they can operate under different financial incentives, noted Dr. Liao.

In the most prominent, widely adopted bundled payment models, hospitals are ultimately held accountable for any savings or losses from an episode of care, which are driven in part by SNF length of stay. However, when SNFs are paid per diem by Medicare and not specifically engaged or rewarded for reduced costs (for example, by sharing of financial savings between hospitals and other entities involved in episodes of care), they may not share the financial incentive to reduce patients' postacute stays.

New value-based payment models encourage greater use of home-based care, when appropriate, as a way of lowering overall costs, and research suggests the strategy is working. For example, the April 2017 JAMA Internal Medicine study found that reductions in spending on SNFs grew with longer ACO participation. The decrease was due in part to lower rates of discharge to SNFs as well as shorter lengths of stay.

The finding of fewer readmissions with SNF care compared with home care highlights the importance of moving forward cautiously with new payment models, said Dr. Werner. Pressures to reduce SNF care may result in the unintended consequence of worse outcomes for patients, although that doesn't necessarily mean that more patients should be sent to SNFs, she added.

It's likely that part of the reason for lower readmissions from SNFs versus home care in the study is the higher level of treatment and round-the-clock monitoring available at SNFs, Dr. Werner and colleagues noted in their findings. As a result, SNFs often may be able to handle issues that would prompt hospitalization in home-based patients.

This balance could potentially change if health care payment systems were further revised, experts noted.

“Part of the reason that patients are more likely to be readmitted from home may relate to the standard Medicare benefit that limits patients to one visit a day for a limited period of time,” Dr. Werner said. “A more effective model might be one that offers more intense care at home with more visits and closer monitoring by clinicians.”