Any one of 10 conditions—ranging from congestive heart failure to major stroke to diabetes—suggests that geriatric patients run a higher risk of dying within a year of being hospitalized, according to a Yale geriatrician. The list of conditions, said Sharon Inouye, M.D., associate professor of medicine and geriatrics, is not for application to individual cases. Instead, it should be used to make outcomes analysis uniform and to foster appropriate programs and policies for an aging population.
“It’s important to be able to compare how sick people are across populations, across hospitals and across studies,” said Inouye, senior author of the study published in March in the Journal of the American Geriatrics Society. Lead author Mayur Desai, Ph.D., working with Inouye and her colleagues, wanted to develop a risk assessment tool that would be easy to use without extensive physicals or detailed chart reviews and which would take into account the high burden of illness among the elderly. “We wanted to come up with a system that is based on administrative data that are readily available, identifies high-risk diagnoses and indicates which segments of the population are at a high risk for mortality. We’re hoping this will be useful to people who do research with older patients or develop new systems to care for older patients.”
Inouye and colleagues found that elderly patients with any of 10 conditions were at higher risk of dying within a year of being hospitalized. In descending order of risk, the conditions are congestive heart failure, pneumonia, chronic lung disease, solid tumor cancer that is localized, metastatic cancer, lymphoma/leukemia, major stroke, acute renal failure, chronic renal failure and diabetes with end-stage organ damage. This system is unique in being developed specifically for use with older persons, based on readily available hospital data.
“Given the potential for misuse or misinterpretation we do not advocate use of this index at the bedside for individual patients,” Inouye said. “The index is recommended for mortality prediction in patient groups or populations.”