Subscribe to eTOC

Nursing Home Placement and Mortality Rates for Those with Significant Cognitive Decline Found to Differ Among Minorities and Whites


Investigators reported that older African-Americans and Latinos had 60-70 percent fewer placements in nursing homes compared to older whites and a 70 percent decreased hazard for mortality compared to older whites.

Older African-Americans and Latinos with significant cognitive decline have a lower likelihood of nursing home placement despite the fact that their rates of cognitive decline are similar to that of whites. And African-Americans and Latinos also live longer compared to older white adults with cognitive decline.

Those were the findings from an eight-year longitudinal study looking at nursing home placement and mortality among adults with significant cognitive decline participating in the US Health and Retirement Survey. The data were discussed at the International Conference on Alzheimer's Disease in Hawaii in July.

Lead investigator Kala M. Mehta, DSc, of the Division of Geriatrics at the University of California-San Francisco, said the lower rate for nursing home placement coupled with the longer survival for minorities indicates the need for attention on the part of clinicians to the likelihood that older African-Americans and Latinos with cognitive impairment are being cared for in the home.

“Our findings support the need for culturally appropriate dementia care, making sure that African-American and Latino patients get good in-home support services and other culturally appropriate resources,” she told Neurology Today.

More surprising is the finding of longer survival for minorities in the study, though both Dr. Mehta and at least one neurologist with expertise in epidemiology who reviewed the report said the finding may stem from any number of social and cultural factors — as well as factors related to nursing home placement — that make survival comparisons difficult to interpret.

“Though we took into account several factors which could have accounted for race/ethnic differences, our study couldn't measure all factors, which may explain the differences in survival,” Dr. Mehta told Neurology Today. “The findings may also reflect differences in genetics, biological or cultural factors, such as how decisions are being made for elders when they are being cared for in their homes, as opposed to when they are being cared for in a nursing home.”

The Health and Retirement Study (HRS) is a longitudinal study of health, retirement and aging sponsored by the National Institute on Aging. The HRS surveys more than 22,000 Americans over the age of 50 every two years.


From 1998 to 2006, 7,513 participants in the survey with a mean age of 73.5 years were followed receiving biennial cognitive testing that consisted of immediate and delayed free recall, orientation, naming, and the serial 7s test. (The latter tests an individual's ability to count backwards by seven from 100.)

Participants self-reported their race or ethnic group — African-American, Latino, and white. Significant cognitive decline was defined as a change in cognitive test scores of 1 standard deviation or more.

Over the eight years, 14 percent of the HRS participants experienced significant cognitive decline. The proportion of persons with significant decline did not vary by race.

However, of those with significant cognitive decline, older African-Americans and Latinos had significantly less placement in nursing homes compared to older whites, and decreased hazard for mortality compared to older whites.

These analyses were adjusted for several potential mediators, such as age, gender, education, total net worth, self-reported medical history of hypertension, heart disease, diabetes and stroke, body mass index, vigorous activity performed three times per week, and cognitive function over time.

Dr. Mehta acknowledged that the mortality findings run counter to expectations. However, she noted that there is a growing literature documenting a “mortality crossover effect,” whereby African-Americans experience higher mortality up to age 80, but after age 80 tend to live longer than whites. However the current analysis of data from the HRS did not examine whether the mortality outcome was related to that effect.


Claudia Kawas, MD, professor of neurology at the University of California-Irvine, who reviewed the study for Neurology Today, emphasized that the mortality findings should not be interpreted to mean that older minorities with cognitive decline generally are at lower risk of mortality than white counterparts. She noted that the researchers controlled for a number of factors — hypertension, diabetes, and body-mass index greater than 30 percent — all of which disproportionately affected both minority groups in the study.

“If you look at diabetes, no matter the age, diabetics have shorter life, and if you look at the characteristics of the cohort there is twice as much as diabetes as the white group,” Dr. Kawas said. “So it doesn't mean that minorities aren't dying prematurely of preventable conditions, but only that when you control for those variables they show a longer survival.”

Dr. Kawas said the nursing home placement rates almost certainly reflect non-medical factors, especially payment ability and insurance status. But those rates may also reflect cultural differences between minorities and whites about how to care for elderly family members.

And the mortality figures showing longer survival for minorities could theoretically reflect an effect of in-home care, though it is not possible to determine that conclusively.

“All of us feel that no matter how good the nursing home is if you have a loving family surrounding you it doesn't get any better than that,” Dr. Kawas said. “These kinds of decisions are very much ethically and culturally driven.”

She added that the mortality findings could also be imbedded in factors related to the protocols surrounding nursing home placement. “When someone is in a nursing home we are much more likely to talk to families about do-not-resuscitate orders and about what conditions under which a patient wouldn't be taken to the emergency room or admitted to the hospital,” she said. “So the care in a nursing home might be extremely good but the decisions about what kind of care to provide are likely to be different than when someone is being care for in a home.”