By Makoto Ishii, MD, PhD

Assistant Professor of Neuroscience and Neurology, Feil Family Brain and Mind Research Institute, Department of Neurology, Weill Cornell Medical College

Dr. Ishii reports no financial relationships relevant to this field of study.

SYNOPSIS: In community-dwelling people from Bronx County, New York, there was a sharp decrease in dementia incidence in those born after mid-1929, which could not be readily explained by changes in the prevalence of cardiovascular diseases, higher education, or increased racial/ethnic diversity.

SOURCE: Derby CA, Katz MJ, Lipton RB, Hall CB. Trends in dementia incidence in a birth cohort analysis of the Einstein Aging Study. JAMA Neurol 2017;74:1345-1351.

With an increasingly aging population and no cure or vaccine for Alzheimer’s disease (AD) or age-related dementias, dementia is poised to become a medical epidemic. Surprisingly, investigators recently have found that for unclear reasons, there may be a decrease in dementia incidence. Identifying the underlying causes for any such decrease would be critical for developing effective dementia prevention strategies. In this study, investigators from the Einstein Aging Study sought to determine whether there was evidence of a decrease in dementia incidence across sequential birth cohorts. As evidence mounts supporting the role of cardiovascular disease as a significant contributor to dementia, Derby et al also sought to determine whether trends in cardiovascular disease could explain any observed dementia trends.

The Einstein Aging Study is an ongoing study that has been recruiting noninstitutionalized individuals from Bronx County, New York, since 1993. Eligibility criteria were ≥ 70 years of age, fluent in English, and no dementia at study entry. Each study participant underwent an annual follow-up assessment that included a clinical neurological examination, comprehensive neuropsychological assessments, medical history, blood pressure, anthropometrics, and psychosocial assessments. Self-report of physician diagnosis was used to determine the prevalence of myocardial infarction, stroke, or diabetes. Global cognitive performance was assessed using the Blessed Information Memory Concentration (BIMC) test, while depression was assessed using the Geriatric Depression Scale. Diagnosis of dementia was based on standardized clinical criteria from DSM-IV and assigned at consensus case conference. There were 1,348 participants in the study after excluding those who died or were unavailable for follow-up before their first annual visit.

Using crude dementia incidence rates as a function of age and dates of birth, the overall trend of increasing dementia incidence with increasing age was seen as expected. When the investigators fit locally weighted scatter plot smoothing functions using generalized additive models, there was a consistent pattern within each age range of decreasing incidence with sequential birth years and an accelerated decrease in those born in the middle to late 1920s. Investigators next identified birth years when there was a significant change in incidence rates by fitting Poisson regression models with change points. After adjusting for age, sex, race, and education level, a significant change point was found for individuals born after July 1929. Investigators then examined whether decreasing rates of cardiovascular disease affected the dementia incidence rates. While the age-specific prevalence of myocardial infarction and stroke decreased in the cohort across sequential birth cohorts, there was an increase in diabetes prevalence in later birth cohorts. However, adjusting for changes in prevalence of cardiovascular diseases did not change the results. Additionally, in the more recent birth cohort, there was a higher proportion of African-American and Hispanic individuals, as well as higher years of education and baseline cognitive status. None of these changes explained the decreased dementia incidence.


Although earlier studies found no significant changes in dementia incidence for cohorts followed before 1990, the major findings from this study are consistent with more recent studies that also found decreased dementia incidence, particularly in the United States and Western Europe. Of note, the analytic approach of this study differs from other recent studies by examining incidence of dementia according to year of birth, as opposed to dementia rates in specified age brackets enrolled during different periods. This should allow for more precise separation of age and cohort effects. Another strength of this study is the use of standardized diagnostic criteria applied to all participants across the study period in a community-based sample. Also, there was a high concordance with the clinical diagnosis and pathological changes found in the subset of subjects who received autopsy, giving confidence to the accuracy of the clinical diagnosis. A significant limitation is the small number of dementia cases for the more recent birth cohorts. This is important as the largest effect in decreased dementia incidence was seen in this cohort. In addition, Derby et al could not distinguish between Alzheimer’s disease and other age-related dementias. Furthermore, while the authors attempted to address whether improved cardiovascular risk factors contributed to the decreased dementia incidence, they used self-reported diagnosis as opposed to more direct measures such as hemoglobin A1c levels, which may lead to errors or bias in the data. The authors also did not address other factors that may be important in more recent birth cohorts, such as improved infection control and treatment, better nutrition, and other societal changes.

Additional studies clearly are needed to replicate these findings and, importantly, to identify any potential factors that contribute to decreasing dementia incidence. Moreover, it is unknown if this trend will continue, as a significant increase in the prevalence of obesity and related cardiovascular diseases in more recent birth cohorts may reverse any gains made. Although this study shows promise that dementia incidence may be decreasing, caution should be exercised before declaring any victory against the still rising tide of a dementia epidemic.