Skip to main content
U.S. flag

An official website of the United States government

Dementia prevention: What does the evidence tell us?

Dr. Richard Hodes
Richard J. HODES,
Office of the Director (OD)

According to the latest projections, by 2050, our world will experience the sobering reality of 131 million people with dementia. By then, many of today’s adults will develop Alzheimer’s disease or a related dementia—unless research can change that course. NIA already funds a wide range of trials designed to test interventions that may be effective against dementia. Do today’s results offer promise of a better future? How can we improve these studies to meet the challenge of this global health priority? While research moves forward, what can we tell the public and clinicians today about actions they might take to protect cognitive health?

A literature review—and a second opinion

To answer these questions, the NIA launched a two-part study in 2015, designed to review and comment on the state of the science on age-related cognitive decline, mild cognitive impairment, and Alzheimer’s-type dementia. The NIA asked the Agency for Healthcare Research and Quality (AHRQ) to conduct an evidence review, so that it could take advantage of AHRQ’s rigorous and well-respected Evidence-based Practice Centers (EPC) program. To provide a second opinion on the literature review and to make suggestions about what advice to give the public about cognitive interventions, the NIA engaged the National Academies of Sciences, Engineering, and Medicine (NASEM) to establish a committee of experts that would review AHRQ’s findings. The committee considered the EPC findings, along with current and future research opportunities and literature not included in the evidence review (such as studies of risk-factor associations), after which it offered recommendations on public health messaging and research gaps.

What do we know?

The AHRQ report found that most interventions “showed no evidence of benefit to delay or prevent” cognitive decline, mild cognitive impairment, or dementia. However, some cognitive training—specifically the approach used in the NIA-funded Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) study—improved performance in the cognitive domain trained, but benefits rarely extended to other cognitive domains. The investigators are continuing to follow-up on longer term outcomes. In addition, the review suggested that though most results for physical activity interventions showed no benefit, the pattern of these results suggested that some benefit is possible. These findings applied to adults with normal cognition.

The NASEM committee then took these results in hand—and as instructed by the NIA, considered additional research from observational studies. The NASEM recommendations pointed to “encouraging but inconclusive” evidence for three types of interventions: cognitive training, blood pressure management in people with hypertension, and increased physical activity. Because research on health outcomes tells us that people should be managing their blood pressure and engaging in regular physical activity, the findings are consistent with what many public health experts and communicators already are already promoting.

The bottom line for the public

Indeed, NIA’s national Go4Life outreach campaign very strongly advocates the benefits of exercise for older adults, even at advanced age. We join in all public health messaging aimed at high blood pressure control as well. We also cite research suggesting that keeping the mind active may be beneficial for well-being with age. As the reviewers noted, these are all healthy behaviors we think it important for older adults to engage in anyway.

That said, however, in light of these most recent recommendations, we will continue to be cautious in what we convey to the public about evidence for any direct connection of these and other interventions to prevention of cognitive decline, mild cognitive impairment, or dementia. Evidence is emerging in these directions, but much more needs to be learned. We owe it to the public to tell them what we know—and what we don’t yet know—on such topics of enormous interest, which is why we commissioned these independent reviews.

Messages for improved research, too

One of the most significant outcomes of the project was the identification of studies that did not meet the strict criteria for inclusion in the evidence review. Specifically, 9,308 of the total 9,448 publications screened by the EPC staff could not be used for further analysis. A number of the screening criteria used by the EPC were related to the quality of the study, such as risk of bias and use of inadequate follow-up time. To this end, the NASEM committee, while noting a range of promising research areas that might continue to be explored, also recommended several methodological improvements for the research community to consider, including but not limited to:

  • Identifying higher risk individuals for future studies
  • Increasing participation by underrepresented populations
  • Testing interventions at younger ages (and with longer follow-up periods)
  • Using consistent cognitive outcome measures

You can learn more about the NASEM committee’s detailed research recommendations, by reviewing the report here, or watching a recent video, that features overall thoughts on the evidence and study findings from members of the expert committee.


Submitted by Robert E Becke… on October 25, 2017

The published evidence of Alzheimer's clinical trials also tells us that there are serious problems with human errors invalidating trials, serious evidence of lax preclinical and clinical work in that numerous Phase II studies with demonstrated or claimed (often with suspicious unplanned post hoc analyses cited as 'preplanned') efficacy used as justification for Phase III trials that have failed, a total disregard by some leading Alzheimer experts of the need for non-contradictory preclinical support for proceeding into humans, a plethora of "me-too" commercial efforts supported by academia even after addressing ABeta has shown no clinically significant influence on the progression of AD.

Isn't there an immediate need for major overhauls if "9,308 of the total 9,448 publications screened by the EPC staff could not be used for further analysis (because of )the quality of the study, such as risk of bias and use of inadequate follow-up time?"

Should not the focus in academia be the preclinical identification and characterization of targets and their effects on disease neuropathological progression, to be followed by small N clinical exploratory trials repeating the preclinical animal work in a third or fourth (i.e. human) species?

Throwing drugs against the wall simply has not worked. Let's get back to the bench for new starts.

An official website of the National Institutes of Health