"In the population we studied, we could not identify a point in time where people experience a sharp decline. Functional loss is a steady process that takes years," said the study's lead author, Kenneth Covinsky, MD, MPH, staff physician at the SFVAMC and UCSF assistant professor of medicine.
Medicare hospice benefits cover only six months of end-of-life care, Covinsky said. "The hospice model doesn't work for this population," he said.
The study appears in the April issue of the Journal of the American Geriatric Society.
In the study, Covinsky and his colleagues used data collected from 917 patients during the last two years of life. The patients had been enrolled in one of 12 nationwide demonstration sites of the Program of All-Inclusive Care for the Elderly (PACE). The program cares for frail older people who meet criteria for nursing home placement, with the goal of keeping the patient at home.
The researchers looked at four measures of function, including the degree of incontinence and the degree of dependence on others in bathing, eating and mobility. Bathing and eating were chosen because, as people become more dependent, their ability to bathe alone is usually the first area in which they need assistance, while eating is the last. Using these data allowed researchers to characterize the trajectory the frail elderly follow in the 24 months preceding death.
The program's nurses assessed functional status and cognitive status upon enrollment in PACE and every three months there after. They used strict definitions for the degree of functional status in the four areas listed above. They used the 10–item Short Portable Mental Status Questionnaire (SPMSQ) to assess cognitive function.
Researchers analyzing the data found that functional decline was more likely among the 64 percent of patients who showed cognitive impairment two years before death. These patients were also more likely to be fully dependent two years before death. Subjects who were cognitively impaired were older at death (85 years vs. 81), more likely to be women (71 percent vs. 61 percent) and less likely to be white (42 percent vs. 54 percent). These patients were also more likely to have Parkinson's disease and less likely to have cancer, diabetes or heart disease than those who were unimpaired.
Researchers found that patients with and without cognitive impairment showed steady decline in functional ability in the two years preceding death. These declines were more apparent in the year before death, but there was only a slight acceleration in decline as death approached. Researchers also found that there was no clear point before death where there was a marked increase in the prevalence of functional impairment.
"More and more people need prolonged years of assistance in the last years of life," Covinsky said. Clinicians who care for older people must recognize frailty as a defining event and refer frail older people to care systems that can provide care for longer than six months. Covinsky said. Such programs, however, are rare and usually are not covered by Medicare. "We need alternatives to the Medicare hospice model that don't require an immediate, bad prognosis," Covinsky said.
Additional authors on the study include Kristine Yaffe, MD, chief of geriatric psychiatry at the SFVAMC and UCSF professor of psychiatry, neurology and epidemiology; Li-Yung Lui, MA, MS, and Laura P. Sands, PhD, both of the SFVAMC division of geriatrics; Catherine Eng, MD, UCSF clinical professor of medicine and Medical Director, On Lok Senior Health Services, San Francisco, California.
Covinsky was supported in part by an independent investigator award from the Agency for Healthcare Research and Quality. Covinsky and Yaffe are Paul Beeson Faculty Scholars in Aging Research.
Journal
Journal of the American Geriatrics Society