A new analysis suggests that about 3.4 million Americans age 71 and older — one in seven people in that age group — have dementia, and 2.4 million of them have Alzheimer's disease (AD). The study, supported by the National Institutes of Health (NIH), is the latest in a series of analyses attempting to assess the prevalence of dementia and AD, the most common form of dementia. Published online this week in Neuroepidemiology, the study is the first to estimate rates of dementia and AD using a nationally representative sample of older adults across the United States.

Brenda L. Plassman, Ph.D., of Duke University Medical Center, with Kenneth M. Langa, M.D., Ph.D., and David R. Weir, Ph.D., of the University of Michigan, Robert B. Wallace, Ph.D., of the University of Iowa, and others, conducted the analysis as part of the Aging, Demographics and Memory Study (ADAMS). ADAMS is a sub-study of the larger Health and Retirement Study (HRS), the leading resource for data on the combined health and economic circumstances of Americans over age 50. ADAMS and the HRS are sponsored by the National Institute on Aging, a component of NIH, under a cooperative agreement with the University of Michigan.

The study highlights the nationwide reach of dementia, which affects not only those with the disease, but their families and communities as well. "As the population ages during the next few decades, the prevalence of Alzheimer's disease will increase several-fold unless effective interventions are discovered and implemented," said NIA Director Richard J. Hodes, M.D. "These data underscore the urgency of research in this area."

The study included 856 HRS participants age 71 and older from 42 states in 2001-2003. ADAMS interviewers from Duke University Medical School conducted at-home evaluations to gather information about each participant's cognitive and functional status and symptoms, neuropsychiatric symptoms, current medications, medical history and family history of memory problems. Prior neuroimaging and laboratory results were also obtained.

A team of clinicians reviewed the evaluation information and made a preliminary assessment of each person's cognitive status. A consensus panel of other medical experts then used well-accepted diagnostic criteria to determine if the participant had normal cognitive function, cognitive impairment without dementia, or dementia. Such criteria further were used to discern the type of dementia, including AD or vascular dementia, the second most common cause of dementia in older adults.

Based on the experts' classifications, Drs. Plassman and Langa and co-authors estimated the national prevalence and total numbers of people age 71 and older, by age group, with any dementia and with AD or vascular dementia in 2002. According to their calculations, 13.9 percent of Americans age 71 and older have some type of dementia, 9.7 percent of Americans in that age group have AD, and 2.4 percent have vascular dementia. AD accounted for about 70 percent of all dementia cases among people 71 and older.

As in other studies, the ADAMS analysis showed that the prevalence of dementia increases significantly with age. Five percent of people ages 71 to 79, 24.2 percent of people 80 to 89, and 37.4 percent of those 90 years or older were estimated to have some type of dementia. The estimated rate of Alzheimer's also rose greatly with older age — from 2.3 percent of people ages 71 to 79 to 18.1 percent of people 80 to 89 to 29.7 percent of those age 90 and older. The ADAMS investigators found fewer years of education and the presence of at least one APOE e4 allele, a genetic risk factor for AD, to be strong predictors of AD and other dementias.

Richard Suzman, Ph.D., director of NIA's Behavioral and Social Research Program, which jointly directs the HRS, said the ADAMS data will prove particularly valuable not only in assessing the prevalence of dementia, but also its impact. "ADAMS, with its link to the data about the health, economic, and family resources of individuals in the study, will help us to characterize more fully the burden of dementia on individuals, caregivers and the nation's health care system," he says.

The ADAMS report is the latest published study to estimate the prevalence of dementia and AD among older Americans. "These assessments have provided a range of estimates, based on differing methodologies and approaches," explains Dallas Anderson, Ph.D., program director for population studies in NIA's Dementias of Aging Branch. For example, some studies have included lower age ranges than ADAMS or broader characterizations of dementia, or have sampled participants in a specific community as a base for national extrapolations. A study reported in 1998 (Brookmeyer et al., 1998) combined incidence data from four community-based studies, estimating that national Alzheimer's prevalence among individuals age 60 years or older would rise from 2.3 million in 1997 to 8.6 million in 2047. Widely cited estimates based on the prevalence of Alzheimer's disease in a Chicago-based community (Hebert et al., 2003), and an earlier comparable study using data from East Boston (Evans et al., 1990) forecast the number of those age 65 or older with AD to be 5.1 million in 2010.

Despite the varied approaches and findings, however, NIA experts point out, the numbers of people with dementia, and Alzheimer's specifically, will certainly increase until ways to delay the progression or prevent the dementia are found. Advancing age is the most common known risk factor for Alzheimer's disease.

The HRS is an ongoing national survey of 22,000 adults age 51 and older that began in 1992, providing data that helps researchers, policy makers and others understand the life circumstances of older adults and help address the challenges of the nation's rapidly aging population. The ADAMS and HRS data are made publicly available to researchers seeking to conduct studies about the older U.S. population. For further information about the HRS and ADAMS, visit hrsonline.isr.umich.edu or www.nia.nih.gov/ResearchInformation/HRS.htm.

To reach the investigators, contact Diane Swanbrow, Institute for Social Research, University of Michigan, at 734-647-9069 or swanbrow@umich.edu.

NIA leads the federal effort supporting and conducting research on aging and the medical, social and behavioral issues of older people, including AD and age-related cognitive change. For information on dementia and aging, please visit NIA's Alzheimer's Disease Education and Referral Center at www.nia.nih.gov/alzheimers, or call 1-800-438-4380. For more general information on research and aging, go to www.nia.nih.gov. Please visit the Web sites to sign up for e-mail notification of new information and publications about aging and about age-related cognitive change.