Skip Navigation


July 21, 1999

Study Shows Risk of Alzheimer's Disease Drops in the Oldest Old

A study of the prevalence of Alzheimer's disease (AD) in a population of older Americans suggests that the old view of AD--that its incidence and prevalence keep increasing as long as a person ages--is incorrect. Lead author, John Breitner, MD, MPH, professor of Mental Hygiene, Johns Hopkins School of Public Health said, "No matter which genetic group you're in, your risk goes up to a maximum and then it starts to decline, so that there's an increasing likelihood you aren't going to get AD." The study appeared in the July 22, 1999, issue of the journal Neurology.

The researchers studied 5,092 elderly individuals in Cache County, Utah, and found that, in those predisposed to AD, the three normal variants (or alleles) of a gene called apolipoprotein-E (APOE) strongly influenced the timing of the onset of disease. The 4 variant was found to be the most troublesome allele; 25 percent of the population have inherited one 4 allele from either parent, while a much smaller percentage have inherited two, one from each parent.

The study found that the influence of APOE wanes in extreme old age, with 4 conferring a lower risk of AD at ages 85 to 90 than it does at ages 65 to 70. "Till we broke out the 4 allele, it wasn't clear that the risks declined in the oldest old," said Dr. Breitner. "When the population is studied by groups, according to their numbers of 4 alleles, it's clear that all groups show an age of maximum risk and a decline thereafter."

The data also suggested an interaction between female gender and the 4 allele, particularly in those women who inherited a pair of 4's. The Cache County study was the first single study that included sufficient numbers of individuals with two 4's to allow for examination of sex differences in prevalence within the different APOE groups. "Sex is probably a risk factor only in those who have the 4 allele," said Dr. Breitner. "We found no greater occurrence of Alzheimer's disease in females among the 75 percent of the population who lack 4."

The researchers enrolled 5,092 individuals into the study (90 percent of the elderly Cache County population) and attempted to identify and diagnose all prevalent dementia cases. Those with suspected dementia were given a clinical exam and a one-hour battery of neuropsychological tests. Cell samples were collected from 4,932 participants to allow the scientists to tally their numbers of APOE alleles.

The study identified 335 demented individuals, 230 of whom (69 percent) had definite, probable, or possible AD. The 4 allele accounted for 70 percent of the population's risk for AD. Of the 4,932 respondents who consented to genetic analysis, 141 (2.9 percent) had two 4 alleles ( 4/ 4). "APOE appears to be a timing gene," said Dr. Breitner. "For those vulnerable to the disease, there are typical times of onset associated with each gene group. Once you get past that, the odds are good you won't get it. Each genotype has its day."

Thus, the small group of people who had inherited two 4 alleles, one from each parent, usually got AD in their 60s and 70s, with a drop-off after age 80. The group who had a single 4 allele mostly got AD in their 80s, with a drop-off after 89. Individuals without any 4s had a decline in their risk of AD after age 95.

The researchers speculated that the risk effect of 4 might be exaggerated in women who are without the protective effect of estrogen. Because the brain converts testosterone to estradiol, concentrations of estrogen in the brain are probably higher in old men than they are in old women who are not taking hormone replacement therapy.

Support for this study was provided by grants from the National Institutes of Health.

Public Affairs Media Contacts for the Johns Hopkins Bloomberg School of Public Health: Tim Parsons or Kenna Brigham @ 410-955-6878 or