A 15-minute mental test shows promise for identifying people 65 years and older who are most likely to die in the next two years, according to a study at the Harvard School of Public Health.

Given on a regular basis, the test reveals deterioration of such cognitive functions as memory and orientation. “Our study shows that even mild declines in such functions are strongly predictive of death,” says Shari Bassuk, a research fellow in the Department of Health and Social Behavior.

Called the Mini-Mental State Examination (MMSE), the test consists of 30 questions. (“Mini” refers to the shortness of the test, not to diminished mental capability.) Those who answer 28-30 questions correctly receive a rating of high normal. Lesser categories include low normal (24-27), mild impairment (18-23), and severe impairment (0-17).

Matching falling scores to death records showed that a drop from high normal to low normal more than doubles death risk in the following two years for those 65 to 80 years old. If an elder slips from high normal to mild impairment, the risk rises to almost three and a half times (3.45) compared with those who remained at high normal.

A drop from low normal to severe impairment jumps mortality risk a startling eight and a half times (8.51). Going from mild to severe impairment raises the probability of death in two years almost six times (5.74), according to the study.

Test questions are deceptively simple. A person is asked to name the season, to remember three simple words – such as penny, table, and apple – for a few minutes, or to reproduce a simple line drawing. But for the past 25 years, the mini-examination has served as a well-accepted standard for identifying cognitive loss.

Bassuk recommends that the Mini-Mental State Examination be given to people 65 and older during regular physical checkups. Elders can be regularly tested for cognitive function just as they are for blood pressure and cholesterol. Those who show worrisome changes could be referred to a neurologist, psychiatrist, or other specialist for diagnosis and treatment.

Deadly change

About 15 percent of the nation’s population, approximately 5,250,000 people 65 and older, suffer from dementia, mostly Alzheimer’s disease. Experts estimate that milder cognitive deterioration affects two to three times as many elders, or 10-15 million people.

Most studies of oldsters have dealt with dementia and have shed no light on lesser forms of dysfunction. To determine whether the latter can be predictive of a shorter life span, Bassuk, biostatistician David Wypij, and epidemiologist Lisa Berkman decided to look at a sample of old people living on their own. They chose a group of 1,997 men and women 65 years and older living in New Haven, Conn., and began giving them the mini-mental test in 1985.

Follow-up interviews were conducted until 1994. Since then, the researchers have kept track of who died and when. “The predictive power of the test is greatest in the two years after changes in cognitive levels are measured,” Bassuk notes.

Importantly, the researchers found that change is more significant than a person’s location on the dysfunction ladder. Those who started out severely impaired, for example, and remained that way had less a chance of dying in two years than a person who slipped from high normal to mild impairment, or from low normal to severe impairment.

“The category of impairment cannot be discounted in predicting death,” Bassuk cautions, “but it’s less critical than changing for the worse.”

It was also found that the most cognitively impaired elders were more likely to be female, nonwhite, poorer, less educated, socially disengaged, physically impaired, and with symptoms of depression. Bassuk did an earlier study concluding that elders who don’t interact much with other people are more likely to decline than those who are attached to family, friends, and neighbors.

“We don’t know if social disengagement and the other factors cause cognitive impairment or if they’re merely associated with it,” Bassuk admits.

As people reach 80 years of age, their cognitive decline is less clearly associated with death, the study team concludes. “Test scores for these people probably are influenced more by external factors such as motivation, sleep loss, problems with hearing, seeing, and movement, and even the hour of testing,” Bassuk points out. “As a result, the predictive value of the testing is strongest for those between 65 and 79 years old.”

Although it detects cognitive decline, the Mini-Mental State Examination can’t determine the cause of the deterioration. To find that would require more extensive physical and mental tests. The MMSE, however, can serve as an early warning.

“For those 80 years and older, low scores might better serve as a predictor for Alzheimer’s Disease,” Bassuk says. “For the younger elderly, such cognitive declines should alert health care providers to the possibility of an underlying illness, and motivate them to explore potential treatments in order to reduce the risk of death.”