Understanding dementia risk research: a guide

Home > About Dementia > Understanding dementia risk research: a guide

This is an advanced reading page on how to understand material available on dementia risk reduction for interested readers.

Although there is no sure way for anyone to avoid dementia, knowledge of the risk factors helps to see what can be done to reduce the probability of getting dementia. This page provides a general discussion to empower readers to navigate available reports, articles and discussions around risk factors and dementia prevention, and see what may be relevant for them. The page does not list the risk factors or consolidate recommendations from reports, though examples and excerpts have been included from some reports to illustrate some concepts.  While the page is designed to be understandable by laypersons, it is “advanced” because it is aimed at readers willing to go deeper into concepts to better understand published material, rather than at readers who want a list of risk factors and know what to do.

For specific reports that may be useful, see  Dementia reports, studies, policy news from India and outside and for the risk factors considered most relevant for dementia, see Risk factors for dementia and what we can do of our page on diagnosis. There is also an infographic here.  

In general, online material and discussions on dementia risk factors are available in published reports and studies, and in articles in the media. Each of these is discussed below.

Sections on this page:

Major published reports by authoritative scientific / medical entities.

Periodically (usually every few years), major entities in health and scientific research publish reports on dementia risk reduction, which discuss risk factors, how much they can prevent dementia, using interventions, etc. These include Alzheimer’s Disease International, Lancet Commission on Dementia Prevention, Intervention, and Care and World Health Organization. These reports are typically based on meta-analysis of many published studies and biological understanding of the processes that may explain the risk factor being studied, and assembled to suggest possible interventions, how strong the evidence for each is, the various stakeholders who may take action, and so on. These are dense, meticulously assembled reports that could contain material relevant to individuals concerned about getting dementia, health care workers and policy makers.

The brief discussion below is aimed at making the process of studying such reports simpler by inducting readers into some basic underlying concepts of such reports.

Interventions to reduce dementia risk are aimed at potentially modifiable risk factors. The authors of the reports on dementia risk reduction look at evidence to understand how each specific risk factor may be contributing to the risk and include those that meet some criteria. In addition to listing the risk factors, they may include a percentage.

For example, the Lancet Commission’s 2024 report Opens in new window puts the overall population attributable fraction (PAF) for all 14 of its risk factors at about 45 per cent — in the report’s words, nearly half of dementias “could theoretically be prevented by eliminating these 14 risk factors”. PAF is a population-level measure: the percentage by which the amount of dementia in a population (its prevalence) could fall if a risk factor were removed from that population altogether. The measure is easy to misread; laypersons may assume that their individual risk goes down by 45% if they address the 14 risk factors mentioned in the report.

For an individual, the effectiveness of addressing a risk factor in terms of how it affects their personal dementia probability, depends on many factors specific to that individual and is not something any such report can estimate.  Also, even if all risk factors are addressed, there is no way to be sure someone can avoid dementia, something the Lancet 2024 points out directly: “However, although this risk modification affects the population, it will not guarantee that any individual will avoid dementia.”

When such reports are discussed in media, a common impression conveyed is that individuals have full agency and responsibility to address risk factors; this sometimes even results in people blaming the person for not having done enough to prevent their dementia. The public health aspect is underplayed while simplifying the report, ignoring the importance of genetics, environment, policies and support systems which impact the access and flexibility individuals have while making choices. The Lancet 2024 key messages point out that “Prevention involves both policy changes at national and international governmental levels and individually tailored interventions.”

In addition to the importance of a public health approach, some themes stand out across reports, and are becoming more prominent, such as the life-course approach and accumulation across a lifetime and also the overlap of many risk factors with those of other non-communicable diseases.

To quote Lancet 2024 again: “Risks can be conceptualised as something that the individual can change, but a public health approach recognises the life-course generation of ill health that is associated with socioeconomic deprivation.”

The WHO July 2026 guidelines titled: “Risk reduction of cognitive decline and dementia: WHO guidelines, second edition”, (see page here Opens in new window) in the background section of its executive summary states:

Because there is no widely accessible disease-modifying treatment or cure for dementia, prevention across the life course remains the most effective strategy to reduce future incidence. The guidelines reflect a life-course understanding of dementia risk, recognizing that exposure to risk factors accumulates over time and that certain risk factors may have a greater impact at specific life stages. This perspective also acknowledges that some associations of risk factors with dementia late in life may reflect early disease processes rather than causal effects, highlighting the importance of timing in risk reduction efforts. Several dementia risk factors overlap with those of other noncommunicable diseases, underscoring the need for integrated public health approaches.

More on the life course model as an overall concept in public health (not specific to dementia) can also be seen at Framework to implement a life course approach in practice Opens in new window.

When looking at such comprehensive reports, it is also helpful to note that risk factors that make it to major reports have to pass many challenges. All major reports include explanations on how the data was gathered and evaluated, and the challenges faced doing so – the methodology – and also links to all referred papers. It is worth perusing these to appreciate the basis for inclusion and exclusion of various risk factors. There should be sufficient studies for the meta-analysis to reach the required standard to decide whether a risk factor should be included or whether there is not enough evidence yet, or there is evidence to the contrary.

For example, Lancet 2024 has a section titled “The challenges of research into dementia prevention and risk reduction” which discusses problems like distinguishing between true causal effect, reverse causation, or the bidirectional links given the long preclinical phase involved, and the methodological difficulties in designing and conducting intervention trials for dementia prevention. For some things not easy to measure, proxies are used. There are also challenges with respect to ensuring diversity, equity, and inclusivity. Also, some potential risk factors cannot be included as there is insufficient evidence.

As readers we must remember that reports need proof and if a risk factor is not included that indicates absence of proof of their importance rather than proof of absence of any such significance. Lancet 2024 even has a section titled “Potential risk factors considered with insufficient evidence to include in the model”, and names, among the factors it considered but could not include, too little sleep, an unhealthy diet, infections, and mental health conditions.

The scientific understanding of risk factors is evolving with time, both in terms of the biological reasons some factors may be important, and also in terms of actual studies that are done. We can see an example of this if we see the various subsequent versions of the Lancet Commission on Dementia Prevention, Intervention, and Care.

  • The inaugural 2017 report of the Lancet Commission Opens in new window proposed a “novel life-course model of risk” to demonstrate that dementia is not an inevitable consequence of aging but is influenced by factors across the lifespan, and organized the 9 risk factors it identified in three distinct stages: Early life (age <18 years), Midlife (age 45–65 years), and Later life (age >65 years).
  • The 2020 update Opens in new window refined the life course organization and strengthened the message that risk mitigation should occur at every stage, and redefined the three stages, placing its 12 risk factors in early life (younger than 45 years), midlife (45–65) and later life (over 65).
  • The 2024 report, which identified 14 risk factors, defined midlife more broadly (18-65 years); the categories used in the report being early life as < 18 years, midlife 18-65 years, and later life as > 65 years. Some factors previously added in the late life phase were recognized as having their most critical impact in midlife and moved accordingly, thus emphasizing that prevention should aim to keep risk factor levels low throughout life. That is, in their words, “Prevention approaches should aim to decrease risk factor levels early (ie, the earlier, the better) and keep them low throughout life (ie, the longer, the better)”.

Reports also give their strength of recommendation of risk factors in some way. For example, in the WHO report, recommended interventions are identified as strong or conditional. The report explains that “a strong recommendation for an intervention indicates that most individuals should receive the intervention and it can be adopted as policy in most circumstances”, while a conditional recommendation for an intervention “indicates that different choices would be appropriate for different individuals, and policy-making would require substantial debate among different stakeholders”. The Lancet report provides the population attributable fraction (PAF) of each risk factor.

All of this — how report authors weigh their inputs, decide what to include or exclude, and grade the strength of their recommendations — also explains something that can otherwise look like a contradiction: two reports can reach different conclusions about the same risk factor, because they are grading different things by different methods. Diet is an example. The 2024 Lancet report placed an unhealthy diet among the factors with insufficient evidence to include in its model, while the WHO guidelines of July 2026 did make a recommendation on it: that a healthy, balanced dietary pattern may be recommended to adults with normal cognition or mild cognitive impairment, to reduce the risk of cognitive decline or dementia. WHO graded that recommendation conditional, with the certainty of the evidence rated moderate. Lancet was deciding whether the evidence justified including the factor in its model at all; WHO, using the GRADE system, was weighing whether the desirable effects of recommending the intervention outweigh the undesirable ones, and how confident it was about that trade-off.

Reading a report with its aims and methods in view — what it set out to do, how it decided what counted — is what makes it possible to judge which of its suggestions apply to you, and how firmly.

[Back to Top]

Individual studies and papers.

Individual research studies and scientific observations / opinion pieces are published as articles in various newsletters and publications. These provide information based on a specific study (they are not consolidations/ meta-studies), hence need to be read with the understanding of that limitation.

Basically, if reading individual papers, we have to keep in mind that a specific study may find a correlation of a factor with dementia, but this may be a “false positive” or a “false negative.” The result may also be because of a flaw in the experiment design or a bias in the sample. Results cannot be considered proven or general enough to be reliable and universally applicable. Also, it takes many years for preliminary research results to result in medicines or detection methods that can be used for humans.

The main perspective to be gained from such articles is getting an idea of the direction the researchers are considering.

[Back to Top]

Media reports related to ongoing Alzheimer’s/ dementia research findings.

Newspapers and magazines often carry reports of apparently sensational studies. They may even publish a contradictory report a few days later. This is very confusing for persons trying to decide on life changes to make. Remember that media reports are often misleading. Headlines, especially, are clickbait – dramatically written to attract readers. The article may be written in a hurry. It may not be a correct summary of the scientific paper it quotes. Also, newspapers often sensationalize the results of a single study, and don’t bother about whether the claims are supported or replicated by other studies.

To evaluate any claim:

Note that LLMs may not be a very reliable way to do research on such things, as we cannot be sure what sources, and which versions, are being used, so check claims against the paper itself.

One type of recommendation is taking some specific herb or food supplement. Impressive claims are made in a few articles, which are then duplicated across multiple Internet articles and seem all over the Internet. Here are some useful ways to verify claims:

A special note regarding “miracle foods” and supplements:  Many media articles claim that a special herb/ food item has shown some good results in preliminary studies, leaving an impression that adding this to diet is a good idea and harmless, too, – without discussing what exactly the study was, what it proved, and even if considered good for humans, at what dosage and with what cautions, and whether the benefits can only be obtained by using an extract or can also be obtained by taking the herb directly. In some cases, claims rampant in media are based on dramatic anecdotal cases, but no studies have shown any benefits so far. Caution is specially needed when reading such articles, and the links above can help to check the current status by using authoritative sources that consolidate the insight obtained from various studies, and are not swayed by sensationalism.

[Back to Top]

What risk reduction information cannot do for an individual.

There are some overall limitations of any risk reduction work worth remembering. One is that there are many medical conditions that can cause dementia symptoms, and that even if researchers find a way to avoid one medical condition, that does not safeguard an individual against all other conditions that can cause dementia. Take, for example, Alzheimer’s Disease, the most common among the diseases that cause dementia symptoms, and a major focus of research. Even if researchers find how to completely avoid all variations of Alzheimer’s Disease developing in an individual, that individual can still get one or more of the several other dementia-causing diseases.

Another is that any risk factors identified in various reports and recommendations come from multiple studies and as an aggregation across individuals. Individual variations are lost in such work. How a particular action of an individual will benefit that person cannot be predicted.

Given the accumulative nature of risks across the life course, we can see that action to reduce risk can be taken at any time, but also that some accumulation may already have happened. We can only change what comes ahead.

Also, there are a set of unmodifiable risk factors that cannot be altered – like age, genetics, and any impact of biological sex. 

When reading any reports or recommendations on risk reduction or considering how to adapt their recommendations, these are aspects to keep in mind to have a better tuned approach and realistic expectations.

[Back to Top]

Extracting and taking decisions regarding lifestyle choices.

One major challenge is how to integrate reliable information on dementia risk reduction for individual lifestyle choices. The reports give data obtained from studies, and conclusions based on that cannot incorporate the impact of individual variation – how effective each risk factor intervention would be for a specific person, given their genetics, health situation, and many other factors. Selecting which risk factor to address and how is a more individualized decision, and support of a knowledgeable health care professional can be useful.

Reports sometimes say this themselves. For example, a footnote to the WHO guidelines of July 2026 notes that the exact makeup of a healthy diet will vary with individual characteristics, preferences and beliefs, cultural context, locally available foods and dietary customs.

Adopting things based on media reports, especially diet changes and use of some herbs and supplements, needs special caution. Often media articles are quite persuasive and not well grounded in science, depending on a single study and often not giving relevant details even for that. Side-effects, toxicity, interactions with other medicines and supplements, or cautions because of other medical conditions are usually brushed aside in these articles, so evaluation needs to be very careful. The section discussing evaluating media articles above may help get a more usable perspective.

In general, interpret information with care, and before changing diet and lifestyle, please discuss with a doctor who knows your medical history, family history, and current medicines, and also see how much change you can realistically do, as each change is an effort and making changes that do not help (or which can actually harm) means we may miss better and more useful changes.

Also, if we have an increased risk of dementia, we need to be more careful about factors that we can still control. For example, adults cannot change their childhood development. But if they realize that their risk of dementia is higher because of their childhood, they will be more careful about avoiding risk factors now. They will also adopt more protective measures. Research on dementia risk factors is ongoing. Individuals who understand the conclusions can make better choices.

[Back to Top]

Directory of authoritative sites.

Here are some web resources of national and international Alzheimer’s Disease associations and research bodies:

Main sources of important published reports on risk reduction. These are consolidations of research/ meta studies across many published research studies:

  • Alzheimer’s Disease International (ADI) has published reports on this and related topics including reports on nutrition and dementia, dementia and risk reduction etc.  These draw upon research and meta-analysis and include suggested actions.
  • The World Health Organization (WHO) has published reports on evidence-based guidance for a public health response to dementia with recommendations for various stakeholders, including for individuals.
  • The Lancet Commission on Dementia Prevention, Intervention, and Care has been periodically publishing reports consolidating expert opinion on the emerging knowledge of how to prevent and manage dementia.
  • Many national dementia bodies/ international bodies have newsletters you can subscribe to. For example, you can subscribe to the newsletter of alz.org, or follow its blog Opens in new window.
  • One journal is the Alzheimer’s Research & Therapy Opens in new window, an open access journal affiliated with Alzheimer’s Disease International (ADI) that publishes clinical trials, drug development research and epidemiological studies. As it is open access, papers here can be read without a subscription, which is useful when looking for a study quoted in a media report.

[Back to Top]

Previous: Dementia in India: An overview Next: Dementia reports, studies, policy news from India and outside

Dementia Care Notes