The analysis grew out of observations made by corresponding author Joan Teno of Brown University’s Warren Alpert Medical School during her 20 years as a palliative care physician at nursing homes. “One of my frustrations was how we were sending these patients back and forth between the nursing home and the acute care hospital,” Teno told ARF. “My patients would come back with bedsores, additional infections, disruptive behavior from the stress of relocation. They were getting lost in this revolving door.”

Teno collaborated with lead author Pedro Gozalo, a health economist at Brown, and others to take a closer look at this problem. Analyzing federal databases of nursing home and Medicare claims data collected between the years 2000 and 2007, the team identified a cohort of seniors meeting these criteria: 66 years or older with severe cognitive and functional impairment who spent their last 120 days in a nursing home. Patients in the cohort required assistance with all basic functions—eating, dressing, toileting—and had mental deficits comparable to a score of less than 5 on the Mini-Mental State Examination, on which 24 or lower indicates cognitive impairment. “This was a population you would think of as very close to hospice-eligible,” Gozalo said.

Among the 474,829 nursing home residents meeting the study criteria, 90,228 (or 19 percent) experienced at least one burdensome transition in their last 90 days of life. The team considered transitions “burdensome” if they 1) occurred in the last three days of life, 2) involved going from one nursing home to a hospital and then to another nursing home, or 3) happened multiple times. Another key finding was that “bad things happen to these people,” Gozalo said, giving this example: “Patients in the top quintile of burdensome transitions were about twice as likely to have severe bedsores (aka 'stage 4 decubitus ulcers').” Patients in this quintile were also more likely to exhibit other indications of poor end-of-life care relative to patients in the lowest quintile, such as spending time in an intensive care unit or undergoing invasive procedures such as feeding tube insertion.

The percentage of nursing home residents with at least one burdensome transition differed tremendously across U.S. states, from 2.1 percent in Alaska to 37.5 percent in Louisiana. In addition, blacks and Hispanics were more likely than whites to go through a burdensome transition. Patients faced higher risk if they lacked written advance directives or do-not-resuscitate or do-not-hospitalize orders.

A major contribution of the study is that it “highlights the issue and defines the term ‘burdensome transition’ in a way that makes this work feasible and allows further research into causality and intervention,” said Doug Galasko of the University of California, San Diego. “I think this is an important step toward improving what some of us regard as compassionate care.”

As for what prompts burdensome transitions, the authors suggest that current financial incentives under U.S. health programs Medicare and Medicaid play a critical role. This sentiment is echoed in an accompanying commentary by Joseph Ouslander of Florida Atlantic University in Boca Raton, and Robert Berenson of the Urban Institute in Washington, D.C. In some cases, nursing homes have a financial incentive to send low-income residents on Medicaid to the hospital. A three-day hospital stay may qualify these patients for post-acute care covered by Medicare at three to four times the Medicaid rate they would get if treated at the nursing home (see Grabowski, 2007).

Galasko thinks that having a clear diagnosis of the presumed cause of cognitive impairment could figure heavily in how well patients and their families can make plans for end-of-life care and potentially avoid burdensome transitions. In the study, 51 percent of the nursing home residents were diagnosed with dementia (20.9 percent with Alzheimer’s disease, 30.5 with non-AD dementia), and more than 82 percent were judged “severely cognitively impaired” based on test scores. This prompted Galasko to wonder which diseases were present in the remaining 31 percent of the cognitively impaired. “Having a clear diagnosis facilitates discussion between the family and the healthcare team about expectations, and helps set limits to avoid ER trips during the final stages of disease,” Galasko said.

John (Wes) Ashford of Stanford University and the Veterans Affairs Palo Alto Health Care System in California agrees that proper diagnosis and education encourage earlier dialogue and decision-making for dementia patients as they near the end of life. “There is a clear need to educate people—beginning at younger ages, before cognitive problems surface, and going beyond white, educated populations—to make appropriate plans for their late-life healthcare needs,” Ashford wrote in an e-mail to ARF. “The best approach would be to encourage dementia screening, providing education at the time of screening, and giving those with any signs of cognitive difficulties more intense support and guidance in plans and decisions regarding future care.”

Efforts to equip nursing homes to better identify and manage triggering conditions have shown promise at reducing burdensome hospitalizations (see Ouslander et al., 2011 on Interventions to Reduce Acute Care Transfers (INTERACT). Other programs that involve assigning nurse practitioners to oversee care of frail nursing home residents have also saved hospitalizations and money (see Kane et al., 2004; Kane et al., 2003).

“We can improve care and reduce unnecessary complications and expenditures on preventable hospitalizations of nursing home residents,” Ouslander and Berenson wrote. “But it will require a multifaceted approach; commitment of energy and resources; teamwork among healthcare funders, regulators, healthcare professionals, nursing homes, and hospitals; and a true focus on resident-centered care.”—Esther Landhuis.

References:
Gozalo P, Teno JM, Mitchell SL, Skinner J, Bynum J, Tyler D, Mor V. End-of-Life Transitions among Nursing Home Residents with Cognitive Issues. N Engl J Med 2011 Sep 29;365(13):1212-1221. Abstract

Ouslander JG and Berenson RA. Reducing Unnecessary Hospitalizations of Nursing Home Residents. N Eng J Med 2011 Sep 29;365(13):1165-1167. Abstract