Simple System Predicts Mortality Of Older Americans With 81 Percent Accuracy

The index, which weighs different mortality risk factors according to a simple point system, is potentially useful to health care providers, policymakers, and researchers, say the study authors.

The information can be obtained using a 12-question form that "could be completed in a few minutes by a patient or medical office receptionist," according to lead author Sei J. Lee, MD, a geriatric specialist at SFVAMC.

"There's a real need for this kind of prognostic index, for several reasons," says Lee, who is also a research fellow in the Division of Geriatrics at the University of California, San Francisco.

For patients and caregivers, predicting near-term likelihood of death is useful when making decisions about medical tests and clinical care, he says. "For example, is it worth it to order a Pap smear or colonoscopy for a particular patient? Those sorts of screening interventions generally don't help patients until five to eight years after they are given. Doctors need to get a sense of who will survive long enough to benefit."

The study appears in the February 15, 2006 issue of The Journal of the American Medical Association (JAMA).

According to the study authors, policymakers can also use such data when comparing the quality of care between different health care organizations, such as hospitals, and insurance plans. "Accurate risk-adjustment levels the playing field by accounting for differences in health status" of different organizations' patient populations," the paper states.

Finally, prognostic information is helpful for researchers conducting observational studies of patients, notes Lee. "You can use the data to adjust for differences between two groups," he says. "If one group is healthier, this index can capture how much healthier they are. This can help researchers isolate the effect of a treatment from the baseline differences between the two groups."

To create the index, the researchers looked at data collected between 1998 and 2002 from 19,710 community-dwelling adults aged 50 and older who participated in the nationwide Health and Retirement Survey (HRS), a longitudinal study of health, retirement, and aging sponsored by the National Institute on Aging. Participants in the HRS were chosen as a representative sample of all adults in the contiguous United States older than 50 years.

The researchers classified participants according to three broad classes of variables: demographics -- specifically, gender and age; illnesses, such as cancer, diabetes, heart disease, and hypertension; and ability to perform activities of daily living, such as washing, dressing, shopping, and managing finances. They then noted who had died by December 31, 2002 and analyzed to what extent the different variables had predicted mortality.

A patient who scores zero points on the index has a predicted four-year mortality of less than one percent. A patient with a score of more than 14 points has a 65 percent chance of dying within four years.

"The fact that we account for different kinds of risk factors, functional as well as disease-related, allows the scale to be accurate over a very wide range of ages, as well as in all kinds of different people," says Lee. "It gives you the flavor of the relative importance of each risk factor. For example, being unable to walk several blocks is as many points off as having heart failure."

Ideally, says Lee, "I see the index being used as part of a standard intake form in the doctor's office, when the doctor sees the patient for the first time."

Lee cautions that there are many other prognostic indexes, only a few of which have achieved widespread use. As the study notes, however, many are limited to specific populations, focus on single types of risk such as illness or function, or require laboratory testing. Unlike those indexes, Lee says, "this index has the advantage of being applicable to everyone who is seen in a clinic who is older than 50. There aren't many indexes that are as widely applicable." Co-authors of the study are Karla Lindquist, MS, of SFVAMC and UCSF; Mark R. Segal, PhD, of UCSF; and Kenneth E. Covinksy, MD, MPH, of SFVAMC and UCSF.

The research was supported by a grant from the National Institute on Aging that was administered by the Northern California Institute for Research and Education, and funds from the Department of Veterans Affairs and the Agency for Health Care Research and Quality.

UCSF is a leading university that consistently defines health care worldwide by conducting advanced biomedical research, educating graduate students in the life sciences, and providing complex patient care.

The mission of NCIRE is to improve the health and well-being of veterans and the general public by supporting a world-class biomedical research program conducted by the UCSF faculty at SFVAMC.