Yale Researcher Designs More Accurate Measure of Mortality Risk for the Elderly

Functional impairment and laboratory abnormalities, combined with diseases, offer the most accurate prediction of which elderly persons are most likely to die within one year after being hospitalized, a Yale researcher has found.

In a study published this month in the journal Medical Care, principal investigator Sharon Inouye, M.D., professor of medicine and geriatrics at Yale School of Medicine, said her new Burden of Illness Score for Elderly Persons (BISEP) is an alternative to the traditional risk adjustment index. The existing systems are generally limited to a list of diagnoses, but just as important in calculating the risk of mortality, she says, is functional impairment, such as difficulty walking, and physiological impairment, as reflected by laboratory abnormalities.

“There really are no measures that have looked at functional impairment in older people, yet study after study shows functional impairment is an even more important determinant of the outcome than diagnoses,” Inouye said. “If patients can’t walk, can’t feed themselves, can’t bathe when they leave the hospital, they are not going to do well when they go home.”

Her article offers a new way to conceptualize the current method of assessing mortality risk, and all of the new information is readily available in hospital database systems, in discharge information, and in medical records.

Inouye’s study forecast the outcomes for 525 patients age 70 and older at one university hospital. This group was compared with outcomes for 1,246 patients 65 years and older from 27 hospitals. Five risk factors each were selected from diagnoses, laboratory tests and functional status. These factors were high risk diagnoses, levels of albumin and creatinine, dementia and walking impairment. The new approach is timely. Medicare now mandates that the functional status of certain patients be assessed before they leave the hospital.

BISEP demonstrated superior performance when compared with several other risk measures.

Inouye said the index is not intended to predict mortality in individual patients. BISEP is recommended for risk adjustment in populations or patient groups for the purpose of identifying high risk groups, comparing treatment effectiveness between groups, or evaluating clinical performance across health sytems.

Share this with Facebook Share this with Twitter Share this with LinkedIn Share this with Email Print this