A diet rich in flavonols is associated with slower cognitive decline

A diet rich in flavonols is associated with slower cognitive decline


  • Participants in a study who had a high dietary intake of flavonols had slower cognitive decline than those who had a lower intake.
  • Higher total flavonol intake was associated with a significantly slower decline in episodic memory, semantic memory, perceptual speed and working memory.
  • Among the flavonols, kaempferol and quercetin were associated with slower cognitive decline, but not myricetin and isorhamnetin.

Flavonoids are polyphenolic compounds found in plants and in large quantities in fruits and vegetables in particular. These compounds are best known for their anti-inflammatory and antioxidant properties. Flavonoids have been associated in several previous studies with slowing age-related cognitive decline and dementia. However, few studies have attempted to identify which flavonoid subclasses and individual molecules are most active in protecting brain health. A recently published American study provides some answers by evaluating the effect of the intake of total flavonols and individual flavonols (kaempferol, quercetin, myricetin, isorhamnetin) on the cognitive performance of the elderly.

The study was conducted among 961 participants from the city of Chicago in the United States, aged 60 to 100, who were part of the Rush Memory and Aging Project cohort, and who were followed for 6.9 years on average. The participants, whose average age was 81 at the start of the study, were mostly female (75%), Caucasian (98%), and had an average of 15 years of schooling. Participants’ diet was assessed using a validated semi-quantitative questionnaire, and dietary flavonol intake was inferred from the collected data. The participants’ cognitive performance was assessed annually with a battery of 19 standardized tests.

A higher dietary intake of total flavonols and individual flavonols was associated with a lower rate of overall cognitive decline and several cognitive domains. A higher intake of total flavonols was associated with a slower decline in episodic memory (memories of personal events), semantic memory (memory of facts and concepts), perceptual speed, and working memory (short-term memory), but had no effect on visuospatial construction ability (understanding and representation of space in 2 and 3 dimensions).

Analysis of individual flavonols indicates that higher intakes of kaempferol and quercetin are associated with slower cognitive decline. In contrast, myricetin and isorhamnetin were not associated with an effect on global cognitive decline. Kale, beans, tea, spinach and broccoli were the foods highest in kaempferol among those consumed in this study. Tomatoes, kale, apples and tea were the foods highest in quercetin in this study.

The mechanisms underlying this favourable association are not yet well understood. The study authors suggest that the anti-inflammatory properties of flavonols may decrease the amplitude or duration of neuroinflammation. In addition, the antioxidant properties of flavonols could reduce or even prevent cell damage caused by oxidative stress, which generates reactive oxygen derivatives (free radicals, oxygenated ions, peroxides).

An earlier study by the same group of researchers reported that green leafy vegetables (spinach, kale, collard greens, lettuce) and certain constituents including kaempferol were associated with slowing overall cognitive decline. The authors concluded that “eating about one serving per day of green leafy vegetables and foods high in phylloquinone, lutein, nitrate, folate, α-tocopherol, and kaempferol may help slow cognitive decline with age.”

The protective role of certain flavonols on cognition has been demonstrated in animal models. Thus, quercetin supplementation improves memory and learning in transgenic mice used as an animal model of Alzheimer’s disease. In another study, kaempferol and myricetin improved memory and learning and reduced oxidative stress in mice used as a model of Alzheimer’s disease.

The prospective design of the American study does not make it possible to establish a causal link between dietary flavonol intake and cognition. Randomized clinical trials would confirm the role of flavonols on cognitive performance and, in the longer term, the prevention of cognitive decline associated with age. This type of study would also make it possible to clarify the dose-response relationship for optimal brain health. In any case, the study also has several strong points: a large number of participants, duration of the study, robust measurement of cognition by the 19 cognitive tests, validated questionnaires. The results were adjusted to minimize residual confounders, since it is possible that a higher dietary intake of flavonols is an indirect effect of a healthier diet. Among the limitations of this study are: self-reported food intake is subject to recall bias; because of their advanced age, participants are at risk of mild cognitive impairment that could cause errors when answering food questionnaires; there remains a possibility of reverse causation (cognitive decline may have altered participants’ eating habits). According to the authors, additional analyses (sensitivity analyses), however, indicate that reverse causation is unlikely.

The results of this study suggest that the consumption of fruits and vegetables (especially green leafy vegetables) in the elderly may not only help them maintain good health in general, but also delay or prevent cognitive decline. However, more studies are needed to confirm and better understand how flavonols slow cognitive and memory decline.

Walking in the forest has positive impacts on the brain

Walking in the forest has positive impacts on the brain

A multitude of studies have shown that the interaction of humans with nature generates several positive effects on health, both physically and psychologically. A recent review of these studies (in French), recently carried out by our team, revealed that these beneficial effects are particularly compelling with regard to the reduction of stress and anxiety that results from an interaction with a natural environment.

Decreased activity of the amygdala
To better understand the mechanisms involved in this nature-mediated stress reduction, a team of German researchers investigated the potential involvement of the amygdala, a region of the brain that plays a predominant role in the stress response.

In this study, the researchers recruited 63 participants and randomly divided them into two groups: 1) a “city” group, in which the volunteers (31 participants) had to walk for one hour in an urban area (a commercial street in Berlin) and 2) a “nature” group, in which the volunteers (32 participants) also had to walk for an hour, but this time in nature (the Grunewald forest, located southwest of Berlin).

By measuring the activity of the amygdala of all participants using functional magnetic resonance imaging (fMRI), before and after their assigned route, the researchers observed major differences between both groups: while walking in urban areas had no measurable effect, walking in the forest caused a significant decrease (about 50%) in participants’ amygdala activity. Moreover, this effect was observed equivalently when the participants were exposed to images of neutral faces or to faces expressing fear, which are supposed to induce a higher stress response. It therefore seems that simply interacting with nature for a short period of time is enough to positively influence the brain centre involved in stress.

Even if it is at this stage premature to conclude that this decrease in amygdala activity alone is responsible for the soothing properties of nature, these results nevertheless remain very interesting, because they show for the first time that interacting with nature has measurable positive effects on the activity of certain areas of the brain, particularly in an area involved in the stress response.

Walking associated with reduced risk of dementia, cardiovascular disease and cancer

Walking associated with reduced risk of dementia, cardiovascular disease and cancer


  • Participants in a prospective study who walked more had a significantly reduced risk of dementia, premature mortality (all causes), mortality from cardiovascular disease, and cancer incidence.
  • The optimal beneficial effect of walking is obtained at almost 10,000 steps/day for the risks of dementia, premature death and cancer.
  • As little as 3,800 steps/day is associated with a 25% reduced risk of dementia, half the maximum effect (50%) obtained at 9,800 steps/day.
  • Higher intensity (steps/minute) of walking was associated with favourable effects on premature death and the incidence of cardiovascular disease and cancer.

Several studies published to date suggest that increasing the number of daily steps is important to prevent the development of chronic diseases and premature death. We often hear in the news or on social media that 10,000 steps/day is the target to reach to enjoy the maximum health benefits. However, some studies indicate that at 6,000 to 8,000 steps/day, the maximum protective effect is achieved and that taking more steps does not significantly reduce the risk of death from all causes, for example. However, there is relatively little data on the subject, and the studies published to date do not distinguish between the different types of walking (daily life vs. exercise) and have paid little attention to the relative importance of walking intensity or cadence.

A prospective study recently examined the dose-response association between the amount (number of steps/day) and intensity (cadence) of walking and the incidence of dementia. The study, which lasted an average of 6.9 years, was carried out among 78,430 people aged 40 to 79 who were part of the UK Biobank cohort. Participants wore an accelerometer around their dominant wrist 24 hours a day, 7 days a week for at least 3 consecutive days and had no diagnosed cardiovascular disease, cancer or dementia at the start of the study. From the data collected by the accelerometer, it was possible to determine:

  • the total number of steps/day;
  • the number of steps related to activities of daily living (e.g., walking from one room to another), defined as less than 40 steps/minute;
  • the number of steps taken during intentional physical activity, defined as 40 or more steps per minute (e.g., walking for exercise);
  • the maximum 30-minute cadence (i.e., the average number of steps per minute recorded for the 30 minutes, not necessarily consecutive, when the cadence was maximal).

By the end of the study, 6.9 years after its start, 866 participants had developed dementia. The results show a nonlinear association between the number of total steps/day and the risk of dementia. The risk of dementia decreases with the number of total steps/day up to a 51% decrease at 9,826 steps/day. A number of steps/day of 3,826 is associated with a reduction in the risk of dementia by 25%, or half of the maximum effect.

Figure 1. Dose-response association between total daily steps taken and incidence of dementia. Adapted from del Pozo Cruz et al., 2022.

For the number of steps related to daily life, the optimal dose was 3,677 steps/day, with a 42% reduction in the risk of dementia. Regarding the number of purposeful steps, the optimal dose was 6,315 steps, with a decrease in the risk of dementia by 57%. For the maximum cadence over 30 minutes, the optimal dose was 112 steps/minute, with a reduction in the risk of dementia of 62%.

Associations with risk of mortality, cardiovascular disease and cancer
In a second publication by the same researchers and with the same UK Biobank cohort mentioned above, the dose-effect association between the amount and intensity of walking and premature mortality, mortality due to cardiovascular diseases, and cancer incidence was examined. During the seven years of the study, 1,325 participants died from cancer and 664 participants died from cardiovascular disease.

The risk of premature mortality decreases with increasing number of total steps/day, up to approximately 10,000 steps/day. For every additional 2,000 steps, the risk of premature death decreases by 8%, 10% and 11%

Figure 2. Dose-response association between total steps walked/day and all-cause mortality (panel A), cardiovascular disease mortality (panel B), and incidence of 13 cancers known to be associated with a low level of physical activity (panel C). Adapted from del Pozo Cruz et al., 2022.

The strengths of this study are the very large cohort size and the use of accelerometry and algorithms that helped distinguish walking steps from other ambulatory activities (a problem in some previous studies). Among the limitations, there is firstly that this type of study, of an observational nature, does not make it possible to establish a causal link. Second, the accelerometry data at the start of the study was collected only once and therefore may not fully reflect participants’ walking habits. However, multiple measurements carried out 4 years after the start of the study indicates that there is little variation between the data obtained during the different measurements. In addition, there remains a risk of reverse causation, i.e. that the disease (dementia, cardiovascular, cancer) could be the cause of a low number of steps, despite certain precautions taken by the researchers to minimize this risk.

The researchers are of the opinion that the right part of the dose-response curves does not reflect a real reduction in the favourable effect of walking after 10,000 steps, but rather reflects the scarcity of walking data or events (diagnoses of dementia, cardiovascular disease, death) for these few great walkers. In addition, some particularly health-conscious participants may have set a target (popularized in the media) of 10,000 steps/day, which could explain why the optimal effect is observed at this amount of walking.

Walking more and walking faster are associated with health and longevity benefits. Although the optimal health effect seems to be achieved around 10,000 steps/day, it is important to note that walking between 5,000 and 8,000 steps/day allows us to greatly benefit from the positive effects associated with walking. Moreover, there is no minimum threshold for the beneficial association between the number of total daily steps walked and mortality and morbidity. For example, as few as 3,800 steps/day are associated with a reduction in the risk of dementia equivalent to half of the maximum effect obtained at approx. 10,000 steps/day. Future recommendations for the prevention of dementia, cardiovascular disease and cancer could take this new evidence into account and encourage the population to walk more and at a faster pace, in order to optimally benefit from the health benefits of walking.

Ultra-processed foods associated with an increased risk of dementia

Ultra-processed foods associated with an increased risk of dementia


  • Among participants in an observational study in the UK, those who ate the most ultra-processed foods had a significantly higher risk of developing dementia.
  • By one estimate, substituting as little as 10% of ultra-processed foods with minimally or unprocessed foods in the daily diet is associated with a 19% decreased risk of dementia.

What is ultra-processed food?
Ultra-processed foods are usually prepared from 5 or more ingredients, often using complex industrial processes. These foods are inexpensive to make, appetizing, high in added sugars, salt and fat, but low in protein, polyunsaturated fat and fibre. Some examples of ultra-processed foods include ready meals, chicken or fish nuggets, sausages, cookies, sweetened fruit yogurts, breakfast cereals, energy bars, soft drinks, sweets. Consumption of these energy-dense but nutritionally poor foods has increased considerably in recent decades, to the point that they now constitute more than 50% of the total dietary intake in some countries, such as the United States (58%).

Ultra-processed foods have been linked to adverse health effects such as cardiovascular disease, diabetes, cancer, depression, and all-cause mortality. Researchers from China and Sweden wondered if there was also an unfavourable association between the consumption of ultra-processed foods and the risk of dementia.

The researchers used data from the UK Biobank, a prospective study of 72,083 participants aged 55 or older who were followed for an average of 10 years. All participants showed no signs of dementia at the start of the study. After 10 years, 518 participants (0.72%) were diagnosed with dementia, including 287 participants (0.4%) who developed Alzheimer’s disease and 119 participants (0.17%) who developed vascular dementia. During the study, participants completed at least two detailed questionnaires about their diet, which allowed the researchers to estimate what percentage of the foods they consumed were ultra-processed. The researchers then divided the participants into four groups (quartiles) based on their level of consumption of ultra-processed foods.

Participants in the first quartile consumed an average of 225 g of ultra-processed foods (9% of daily food intake), while those in the top quartile consumed 814 g (28% of the daily intake). In order, soft drinks, sweets and ultra-processed dairy products were the most consumed ultra-processed foods. One hundred and five of the 18,021 participants in the first group who consumed the least ultra-processed foods developed dementia, compared to 150 of the 18,021 participants in the group (quartile) who consumed the most ultra-processed foods. After adjusting the data for age, gender, family history of dementia and heart disease, and other factors, it was estimated that for every 10% increase in daily intake of ultra-processed foods, participants had a 25% higher risk of dementia. The link was even stronger for vascular dementia (28%) compared to Alzheimer-type dementia (14%).

According to the same study, it was estimated that substituting 10% of ultra-processed foods with minimally or unprocessed foods (fruits, vegetables, legumes, milk and meat) is associated with a reduced risk of dementia by 19%. More concretely, an increase of 50 g/day of unprocessed food, equivalent to half an apple for example, replacing 50 g/day of ultra-processed food (equivalent to a chocolate bar or a slice of bacon) could reduce the risk of dementia by 3%. Thus, small changes in diet, requiring little effort, could make a big difference in a person’s risk of dementia.

A pro-inflammatory diet increases the risk of dementia

A pro-inflammatory diet increases the risk of dementia


  • In a study on aging and diet conducted in Greece, 1,059 older people reported in detail what they ate for three years.
  • At the end of the study, people with the most inflammatory diet had a 3-fold higher risk of developing dementia compared to those whose diet had a low-inflammatory index.
  • The main anti-inflammatory foods are vegetables, fruits, whole grains, tea, and coffee. The main pro-inflammatory foods that should be avoided or eaten infrequently and in small amounts are red meat, deli meats, refined flours, added sugars, and ultra-processed foods.

Dietary Inflammatory Index
Several studies suggest that the nature of the foods we eat can greatly influence the degree of chronic inflammation and, in turn, the risk of chronic disease, including cardiovascular disease. For example, a pro-inflammatory diet has been associated with an increased risk of cardiovascular disease, with a 40% increase in risk in people with the highest index (see our article on the subject).

Pro-inflammatory diet and risk of dementia
In order to see if there is an association between a diet that promotes systemic inflammation and the risk of developing dementia, 1,059 elderly people residing in Greece were recruited as part of the study Hellenic Longitudinal Investigation of Aging and Diet (HELIAD). Only people without a diagnosis of dementia at the start of the study were included in the cohort. The inflammatory potential of the participants’ diet was estimated using the Dietary Inflammatory Index (DII) based on the known effect of various foods on the blood levels of inflammatory markers . The main pro-inflammatory foods are red meat, deli meats, refined flour, added sugars, and ultra-processed foods. Some of the main anti-inflammatory foods are vegetables, fruits, whole grains, tea, coffee, and red wine.

After a follow-up of 3 years on average, 62 people were diagnosed with dementia. Participants who had the diet with the highest inflammatory index had a three-fold higher risk of developing dementia at the end of the study, compared to those with the least inflammatory index. In addition, there appears to be a dose-response relationship, with an increased risk of dementia that increases by 21% for each unit of the inflammatory index.

Inflammation, interleukin-6, and cognitive decline
The study in Greece is not the first to be conducted on the impact of a pro-inflammatory diet on the incidence of dementia. In a Polish study of 222 postmenopausal women, those with cognitive deficits had significantly higher blood levels of interleukin-6 (IL-6; a marker of inflammation), were less educated, and were less physically active, compared to women with normal cognitive functions. Postmenopausal women who had a pro-inflammatory diet were much more likely to have cognitive impairment compared to those who had an anti-inflammatory diet, even after adjusting for age, height, body mass index, level of education, and levels of physical activity. Each one-point increase in the dietary inflammatory index was associated with a 1.55-fold increase in the risk of cognitive impairment.

In addition to these studies, it is interesting to see that a meta-analysis of 7 prospective studies including 15,828 participants showed that there is an association between the concentration of IL-6 in the blood and the overall cognitive decline in the elderly. Participants who had the most circulating IL-6 had a 42% higher risk of suffering cognitive decline than those with low blood IL-6 levels.

Several studies have suggested that systemic inflammation (i.e., outside the central nervous system) may play a role in neurodegeneration, Alzheimer’s disease, and cognitive decline in older adults. People with Alzheimer’s disease and mild cognitive impairment tend to have high blood levels of markers of inflammation (IL-6, TNF-α, CRP). In addition, a study indicates that people who have elevated levels of markers of inflammation during midlife have an increased risk of cognitive decline in subsequent decades.

Since the studies described above are observational, they do not establish a causal link between inflammatory diet and dementia. They only show that there is an association. Further studies are needed in the future to establish a cause and effect relationship and identify the underlying molecular mechanisms.

Evidence from recent studies should encourage experts to more often recommend diets high in flavonoids that decrease systemic inflammation and are conducive to the maintenance of good cognitive health. Mediterranean-type diets or the hybrid MIND diet (Mediterranean-DASH Intervention for Neurodegenerative Delay) with an abundance of plants are particularly effective in reducing or delaying cognitive decline.