Association of Lifestyle Factors with Cognition

As the prevalence of dementia increases globally, the need for both symptomatic and disease-modifying interventions becomes more pressing. The number of people with dementia is expected to double every 20 years, and delaying dementia onset by just 5 years may cut associated costs and prevalence by 50%.^1,2 Whereas investigations into pharmacologic therapies are vital, studies of lifestyle activities have revealed modestly effective but promising parallel approaches to mitigate or delay cognitive decline. By some estimates, modifiable risk factors account for 40% of dementias, and they represent targets ripe for intervention.³ Here, we survey the current evidence regarding lifestyle interventions, such as physical, cognitive, and social activity, across the clinical continuum from cognitively unimpaired individuals to those with dementia (Table 1). This selective review emphasizes the literature on all-cause dementia and Alzheimer disease (AD). Physicians can use these studies to guide evidence-based counseling (Table 2).

Physical Activity

Cognitively Normal Individuals

Both self-report questionnaires and objective measurements are useful surrogate measures of the effect of physical activity on cognition in cognitively unimpaired adults. One of the largest such studies to date, the Trøndelag Health Study (HUNT Study), spanned 16 years in Norway. In this observational cohort, moderate-to-vigorous physical activity, as ascertained by self-report using the International Physical Activity Questionnaire, was associated with a 20% reduction in risk of dementia in nearly 30,000 participants aged 30 to 60 years.⁴

Several studies have used accelerometry to measure the effect of gravity on participants’ movements objectively and noninvasively. In one large observational cohort study conducted in the United Kingdom, cognitively normal adults wore accelerometers over an average of 7 years of follow-up (N=78,430; aged 40 to 79 y). These accelerometers were used to catalog daily steps as a quantifiable measure of physical activity. The investigators found that a minimum of approximately 3800 steps per day was associated with a lower incidence of dementia. Higher step intensity, as defined by steps per minute, also was associated with a lower dementia incidence rate. Based on dose-response curves, just under 10,000 steps per day may be optimal for lowering dementia risk, and participants who achieved this threshold had half the risk of developing dementia.⁵

Recent studies have examined the effects of aerobic exercise on the connectivity of brain networks using functional MRI. One representative study examined this relationship while emphasizing the need for recruitment of diverse participants for dementia risk studies. The authors identified increased neuroplasticity in the medial temporal lobe network after a 20-week dance-based aerobic exercise program in 34 Black participants with an average age of 65 years.⁶ Medial temporal lobe network flexibility also was associated with an increase in mnemonic flexibility, as defined by the ability to generalize previous learning to new task demands. However, other risk factors may alter this effect. A study from the same group demonstrated a diminished neuroprotective value of aerobic exercise in a cognitive assessment of rule learning in healthy older Black participants with ABCA7 rs3764650, a known susceptibility locus for AD. The risk conferred by this gene is highest in Black individuals, in whom one copy of this variant doubles the risk of developing AD.⁷ These studies illustrate that additional research is needed to understand the effects of exercise in diverse populations so that interventions can be targeted to and personalized for those most likely to benefit.

People with Subjective Cognitive Decline

Individuals with subjective cognitive decline (SCD) report progressive cognitive impairment but exhibit normal scores on bedside cognitive evaluations or formal neuropsychological testing. SCD is a risk factor for progression to mild cognitive impairment (MCI) and dementia.⁸ A comprehensive review examined 24 high-quality randomized controlled trials (RCTs) and 17 observational studies and recommended that individuals experiencing SCD should engage in at least 150 minutes per week of individually tailored moderate aerobic physical activity, progressive resistance training, and balance exercises.⁹

People with MCI

As cognitive impairment worsens, the evidence for benefits associated with physical activity and cognition becomes more mixed. In a meta-analysis and systematic review of 5 studies (N=2878), a minimum of 1 year of exercise did not reduce the risk of the onset of dementia and MCI.¹⁰ However, a meta-analysis of 9 studies including participants with MCI found a positive effect of exercise on cognition.¹¹

People with AD

A recent systematic review found that exercise provided significant benefits to people with AD in 18 of 52 meta-analyses (35%); exercise was associated with a lower risk of developing AD. In addition, exercise was associated with positive benefits in terms of outcomes such as cognition, physical performance, and functional independence.¹² An advantage of this review is that each of the meta-analyses included in this study evaluated only high-quality RCTs. However, this analysis was limited by the heterogeneity of primary outcomes in the original studies. Another meta-analysis that investigated 13 RCTs (N=869) found 8 studies showing variable cognitive benefits depending on the intervention, whereas the remaining 5 studies showed no significant benefit from physical activity.¹³ More research needs to be conducted to determine the exercise program that would yield maximum benefit.

Social Activity

Cognitively Normal Individuals

Social engagement for dementia prevention has been increasingly studied. A meta-analysis of 31 cohort and 2 case-control studies encompassing 2,370,452 participants revealed that poor social engagement indices—namely, marital status, living situation, social network size, degree of social support, degree of social satisfaction, frequency of social contacts, and frequency of participation in social activities—were associated with increased all-cause dementia risk.¹⁴ A systematic review examined 256 primary studies and 25 reviews to determine the relationship between social engagement and AD in particular. The primary studies analyzed in the review found low social engagement, poor social networks, and loneliness to be associated with an increased risk of AD. Maintaining regular social contacts, engaging in cognitively stimulating activities consistently, and participating in leisure activities were associated with a decreased risk of AD.¹⁵ However, these studies may be limited by publication bias.

People with MCI

One review examined 24 studies of social engagement in individuals with MCI and found that those with MCI may have different levels of social engagement than those experiencing typical cognitive aging. Individuals with MCI may have smaller social networks and may not feel as socially supported, causing reduced participation in social activities. Higher levels of social support were associated with improved physical health, such as reduced somatic symptoms and better sleep quality. Longitudinal studies indicate that increased frequency of participation in social activities may help preserve cognition and reduce risk of dementia.¹⁶

People with AD

The literature on the effects of social activity on AD dementia is both limited and mixed. One RCT examining the effect of social support groups on 142 participants with dementia demonstrated a benefit of social support on rates of depression, family communication, and quality of life.¹⁷ However, another study that examined the effect of a mobile reminiscing therapy app in both individual and social or group settings did not demonstrate a beneficial effect on mood or quality of life in either group compared with the waitlist control arm, but the authors note that the study was not powered to do so.¹⁸ A small study used gardening as an opportunity for social interaction and, using semi-structured interviews, found that participation in a gardening club gave participants with dementia and their caregivers self-reported feelings of autonomy, connectedness, and peer support that improved their overall well-being.¹⁹ The strength of social connections is important as well; a study of 167 participants with AD found that higher levels of closeness with caregivers were associated with slower cognitive decline. This effect was highest in spouse caregivers.²⁰ Because of the limited number of studies addressing this subject, additional research is needed to characterize the effects of social engagement interventions on people with AD.

Cognitive Activity

Cognitively Normal Individuals

In the My Mind Project, a representative high-quality intervention study examining the effects of cognitive training, participants older than 65 with mild to moderate AD dementia (n=95), with MCI (n=97), or without cognitive decline (n=100) were randomized into 2 groups. The control group received general advice on how to improve memory and health status. The intervention for the cognitively unimpaired group emphasized enhancing working memory and learning processes; the interventions for MCI and AD focused on learning strategies of categorization, clustering, attention, and visuospatial processes. The unimpaired group showed a significant increase in forward verbal span. The MCI group showed significant benefits in cognitive outcomes such as backward verbal span, memory of prose, visuospatial short-term memory, learning and memory of word pairs, and selective attentive processes. The AD group showed significant improvement in backward and forward verbal span, word pairing learning, and total score on the Alzheimer Disease Assessment Scale.²¹

People with AD

Several studies have focused on cognitive training to slow the pace of cognitive decline in people with AD dementia. One meta-analysis of 33 studies (N=2000) suggested that cognitive training may improve global cognition as well as performance on specific neuropsychological tests, such as verbal fluency, but these effects last only a few months.²² An important limitation of the analysis is that the quality of the analyzed studies was determined to be moderate at best, and furthermore, the heterogeneity of study methodologies followed in different studies included in this analysis hinders an accurate comparison of intervention outcomes.

Diet Modification

The staples of the Mediterranean diet include fruits and vegetables, whole grains, seafood, nuts, legumes, and olive oil. This diet has been associated with prevention of cognitive decline and dementia.²³ Thirteen meta-analyses of observational studies and 16 meta-analyses of RCTs identified a positive association between adherence to the Mediterranean diet and health outcomes; greater adherence to the Mediterranean diet was associated with a reduced incidence of neurodegenerative diseases, among other cardiovascular diseases and risk factors. A systematic review of cohort studies showed that participants who achieved high adherence to the diet had a 33% reduced risk of MCI or AD.²⁴

The Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND) diet combines the Mediterranean diet and the Dietary Approaches to Stop Hypertension (DASH) diet, which was designed for blood pressure control. All 3 diets emphasize increased intake of natural plant-based foods and limited intake of animal-based foods and foods with high saturated fat content, but the MIND diet specifically emphasizes intake of berries and leafy green vegetables and deemphasizes consumption of other fruits (4 or 5 servings daily in the DASH diet and 3 servings daily in the Mediterranean diet), dairy (2 or 3 servings daily in the DASH diet and 2 cups a day in the Mediterranean diet), potatoes (unrestricted in the DASH and Mediterranean diets), or more than 1 serving of fish per week (6 or fewer servings daily in the DASH diet and 3 servings per week in the Mediterranean diet). One study comparing the MIND diet with the Mediterranean and DASH diets (N=923) in participants aged 58 to 98 years found that, whereas high adherence to the MIND or Mediterranean diet decreased AD risk similarly, more modest adherence to the MIND diet resulted in higher rates of AD reduction compared with the other 2 diets. There was a 53% reduction in the rate of AD for participants who adhered best to the diet and a 35% reduction for those in the middle group of adherence compared with the least adherent group. A limitation of this study is that high adherence may reflect unknown confounders; in addition, because it was an observational study, cause and effect cannot be assumed.²⁵ The MIND diet also is positively associated with a slower decline in global cognition as well as protective effects on specific cognitive domains, including episodic memory, working memory, semantic memory, and perceptual speed.²⁶

Elevated Body Mass Index

In a systematic review and meta-analysis of body mass index (BMI, as a measure of total adiposity) in midlife and the association with dementia in late life, the authors included 19 studies (N=589,649) with follow-up for up to 42 years. They found that midlife (age 35 to 65 years) obesity (BMI ≥ 30), but not being overweight (25 < BMI < 30), was associated with dementia in late life. The risk of late-life dementia in individuals who met BMI criteria for obesity during midlife was 33% higher compared with those with a BMI-defined healthy weight. Excess body weight in midlife may have vascular and neurodegenerative effects underpinning dementia by affecting metabolic pathways.²⁷

Multimodal Interventions

The Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) RCT examined a multimodal lifestyle intervention for cognitively normal Finnish adults aged 60 to 77 years who met criteria for being at risk for dementia using the Cardiovascular Risk Factors, Aging and Dementia (CAIDE) score. The intervention group consisted of 631 participants who received diet, exercise, and cognitive interventions as well as vascular risk monitoring. The control group consisted of 629 participants who received general health advice. The trial demonstrated a significant intervention effect for global cognition, executive functioning, and processing speed. There were no significant differences associated with the memory domain.²⁸

Another RCT studied multidomain lifestyle interventions among at-risk older adults recruited from community centers for dementia established across South Korea. The researchers recruited 32 participants into 1 of 3 arms (offering an intensive lifestyle modification program, the intensive lifestyle modification program plus additional maintenance, or personalized advice on lifestyle modifications to prevent dementia [active control group]). The intensive program group received a 4-week group-based intervention that focused on physical activity, vascular risk factors, dietary habits, cognitive activities, and social engagement. The intensive program with additional maintenance group participated in an extra 20-week maintenance program to consolidate their modified habits. The modified Australian National University–Alzheimer’s Disease Risk Index (ANU-ADRI) score, which assesses individual risk profiles for dementia using odds ratios of risk factors to derive a risk score, was used as the primary outcome. Individual risk profiles for dementia were lower in the intensive intervention with maintenance arm compared with the other 2 arms. In addition, the intensive intervention with maintenance group showed greater improvement in executive function compared with the intensive intervention only group and the control group. There were no significant differences among groups in global cognitive function, but the study was limited by its single-blinded design.²⁹

Early multimodal interventions in cognitively unimpaired individuals may be crucial. The Multidomain Alzheimer Prevention Trial—a large RCT of 1680 participants with subjective memory concerns—did not demonstrate a difference between intervention and placebo groups.³⁰

Summary

Lifestyle interventions ranging from physical activity regimens to changes in weight and diet have demonstrated modest beneficial effects across a spectrum of cognitive and functional stages. Although the data are promising, they are limited to elucidating short-term effects of interventions. More high-quality trials with long-term follow-up are needed to clarify the effects of lifestyle interventions on the prevention or amelioration of cognitive decline, including the types and amounts of interventions that are most effective for individuals across the clinical continuum.