Cognitive decline and dementia constitute growing global health challenges, with profound implications for individuals, families, healthcare systems, and societies (The Lancet, 2025[114]). Worldwide, the number of people living with dementia is projected to increase substantially over the coming decades, largely driven by population aging and improved survival (GBD 2019 Dementia Forecasting Collaborators, 2022[35]). The World Health Organization estimates that more than 55 million people worldwide live with dementia, with around 10 million new cases each year (World Health Organization, 2019[129]). Alzheimer's disease is the leading cause, responsible for 60-80 % of cases, followed by vascular dementia and other subtypes (Reuben et al., 2024[95]). The global burden of dementia is projected to rise dramatically, reaching approximately 153 million people by 2050 (GBD 2019 Dementia Forecasting Collaborators, 2022[35]). Dementia is a leading cause of disability (Vargese et al., 2023[121]), and no curative treatments currently exist (Fox et al., 2025[30]). Consequently, identifying modifiable risk factors is essential for delaying or preventing disease onset, as growing evidence indicates that a substantial proportion of dementia cases are linked to lifestyle and environmental factors that are modifiable across the life course (Jones et al., 2024[53]; Livingston et al., 2024[68]).

Despite extensive research efforts, disease-modifying therapies remain limited, and current treatments primarily offer symptomatic relief (Fox et al., 2025[30]; Frisoni et al., 2025[31]). There is increasing emphasis on prevention strategies aimed at delaying onset, slowing progression, or reducing overall risk at the population level that has been summarized in the Lancet Commission on Dementia Prevention, Intervention, and Care reports (Livingston et al., 2024[68]; Livingston et al., 2020[69]; Livingston et al., 2017[70]).

A substantial body of epidemiological evidence indicates that dementia is not an inevitable consequence of aging (Dominguez et al., 2021[22]). Rather, it is influenced by a complex interplay of genetic, biological, environmental, and lifestyle factors operating across the life course (World Health Organization, 2019[129]). Recent estimates suggest that up to 45 % of dementia cases may be attributable to potentially modifiable risk factors, highlighting significant opportunities for prevention (Livingston et al., 2024[68]). Modifiable lifestyle behaviors such as social engagement, sleep patterns, smoking, alcohol consumption, physical activity, and diet have emerged as central targets for intervention because of their broad impact on overall health. (Dominguez et al., 2021[22]).

Dietary patterns have received particular attention given their influence on cardiometabolic health, inflammation, oxidative stress, and neurovascular integrity, all of which are implicated in neurodegeneration (Wen et al., 2025[126]). Adherence to healthy dietary patterns such as the Mediterranean (MeDiet) (Fekete et al., 2025[28]; Wang et al., 2025[124]), Dietary Approaches to Stop Hypertension (DASH) (Song et al., 2024[111]), and Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND) (Wang et al., 2025[124]; Chen et al., 2023[13]) diets has been associated with better cognitive outcomes and reduced dementia risk, whereas diets high in ultra-processed foods, saturated fats, and added sugars may be detrimental (Wieckowska-Gacek et al., 2021[127]). Nonetheless, findings across studies are heterogeneous, and uncertainty remains regarding optimal dietary components, critical exposure windows, and the extent to which observed associations reflect causal relationships.

Similarly, evidence linking other lifestyle factors to cognitive outcomes has expanded rapidly. Regular physical activity and sustained cognitive and social engagement are generally associated with reduced risk of cognitive decline, while smoking, excessive alcohol consumption, physical inactivity, poor sleep, social isolation, and loneliness have been linked to increased risk (Dominguez et al., 2021[22]; Livingston et al., 2024[68]; Jones et al., 2024[53]). However, interpretation of these associations is complicated by methodological challenges, including reverse causation, competing risks, residual confounding, and variability in exposure definitions and outcome ascertainment. Moreover, evidence quality varies considerably across lifestyle domains, with some factors supported by robust meta-analytic data and others by more limited or inconsistent findings.

In recent years, the publication of large prospective cohort studies, updated meta-analyses, umbrella reviews, and multidomain intervention trials has substantially advanced the field. These studies have refined estimates of association, explored dose-response relationships, and provided insights into potential biological mechanisms and interactions among lifestyle factors. At the same time, they have highlighted important gaps, including limited evidence from randomized controlled trials (RCTs), underrepresentation of diverse populations, and insufficient consideration of life-course timing and cumulative exposure.

Given the rapid evolution of the evidence base, an updated synthesis is warranted. This narrative review aims to critically summarize current evidence on lifestyle factors and dietary patterns associated with cognitive decline and dementia risk, with a focus on recent high-quality observational studies and evidence syntheses. By integrating recent findings across domains, this review seeks to clarify areas of consensus and uncertainty, inform prevention strategies, and identify priorities for future research and public health action.

Lifestyle factors are central to dementia prevention (Figure 2(Fig. 2)). Evidence suggests that physical activity, healthy diet, smoking cessation, avoiding heavy alcohol consumption, adequate sleep, and social and cognitive engagement are associated with better cognitive outcomes and lower dementia risk, likely through effects on vascular, inflammatory, metabolic, and neuroplastic pathways. Despite limited causal evidence from long-term trials, consistent observational findings support a multidomain lifestyle approach to reducing dementia risk at the population level.

Observational evidence suggests that the MeDiet, the DASH diet, and their hybrid, the MIND diet (components are illustrated in Figure 3(Fig. 3)), may confer neuroprotective benefits (Chen et al., 2023[13]; Seabrook et al., 2025[101]; Seago et al., 2025[102]; Dominguez et al., 2021[22]), though findings are inconsistent across populations with varying baseline cognitive status. For example, one study reported that overall adherence to the DASH diet was not associated with cognitive performance or cognitive change (Daniel et al., 2021[18]), and a more recent trial observed no differences in cognitive trajectories between participants following the MIND diet and those on a control diet, possibly due to the relatively short follow-up period of three years (Barnes et al., 2023[5]). This variability highlights the complex relationship between diet and brain health. As with exercise, assessing diet's protective effects on cognitive decline and dementia is challenging due to methodological and biological factors. Dietary intake is often self-reported and variable over time, leading to measurement error. People consume complex food combinations, making it difficult to isolate the effects of individual nutrients or their blends, while residual confounding from lifestyle, socioeconomic, and health factors may bias results. Long dementia latency and short-term dietary assessments, along with reverse causation from preclinical cognitive changes affecting diet, further complicate interpretation.

In fact, the World Health Organization's 2019 conditional recommendation for the MeDiet for dementia prevention reflects ongoing uncertainty in the evidence (World Health Organization, 2019[129]), which also likely explains why diet was classified as a potential but insufficiently supported risk factor and excluded from the life-course population attributable fraction estimates in the 2024 Lancet Commission on dementia prevention report (Livingston et al., 2024[68]). Nevertheless, literature on this topic continues to grow in recent years. A systematic review and meta-analysis of 16 cohort studies (follow-up 2.2-41 years) found that higher overall diet quality was associated with an 18 % reduced dementia risk (Liu et al., 2020[67]), with similar findings in studies with longer follow-up and for Alzheimer's disease. However, continuous MeDiet scores showed no significant association (Liu et al., 2020[67]). A larger meta-analysis of three cohorts with over 200,000 participants reported that greater adherence to the MIND diet was linked to a 17 % lower dementia risk (Chen et al., 2023[13]). Another review found protective effects of the MeDiet on global cognitive decline in nearly half of studies, with more limited evidence for incident dementia and Alzheimer's disease (Townsend et al., 2023[118]).

With the aim of summarizing the findings of the vast literature on this topic, we have reviewed evidence from recent systematic reviews and meta-analyses (Table 6(Tab. 6); References in Table 6: Antonelli and Donelli, 2022[3]; Chen et al., 2023[13]; Fekete et al., 2025[28]; Fu et al., 2022[32]; Nucci et al., 2024[84]; Wang et al., 2025[124]; Zuliani et al., 2026[139]). As shown, the recent meta-analytic evidence indicates that adherence to healthy dietary patterns is associated with a reduced risk of cognitive decline and dementia, with notable differences across diet types. The MeDiet, for which there is a greater body of literature, shows robust and reproducible associations with lower risks of Alzheimer's disease, MCI, and all-cause dementia, with relative risk reductions generally ranging from 7 % to 31 % (Zuliani et al., 2026[139]; Fekete et al., 2025[28]; Nucci et al., 2024[84]; Fu et al., 2022[32]; Antonelli and Donelli, 2022[3]). These associations are supported by both observational cohorts and umbrella reviews, although evidence from RCTs remains limited and primarily confined to modest improvements in specific cognitive domains.

By contrast, evidence for the DASH diet-currently lacking a meta-analysis-is more inconsistent, with no clear overall association with dementia risk among older adults. The MIND diet, which integrates elements of the Mediterranean and DASH diets, demonstrates the most consistent protective associations in the few meta-analyses available, including a 20-40 % reduction in dementia and Alzheimer's disease risk. Notably, the protective effect of the MeDiet appears stronger in studies with shorter follow-up durations or specific diagnostic criteria, suggesting potential influences of study design, reverse causation, and outcome ascertainment.

Overall, these findings support adherence to Mediterranean-based dietary patterns-particularly the MIND diet-as promising non-pharmacological strategies for dementia prevention. However, heterogeneity across studies and the relative scarcity of long-term RCTs highlight the need for further high-quality, longitudinal and interventional research to strengthen causal inference and inform dietary recommendations.

Other dietary patterns with some evidence of neuroprotection include plant-based diets, the Nordic diet, and anti-inflammatory diets (Bigras et al., 2025[6]; de Crom et al., 2023[19]; Townsend et al., 2024[117]; Chen et al., 2019[14]). Predominantly plant-based diets have been linked to favorable cognitive outcomes. A recent systematic review and meta-analysis reported that greater adherence to healthful plant-based diets was associated with lower odds of cognitive impairment and a modest reduction in dementia risk, although results varied across studies. Higher-quality plant-based diet scores, reflecting greater intake of healthy plant foods, showed stronger protective associations than overall plant-based indices (Bigras et al., 2025[6]). However, evidence is not uniform, as some population-based studies found no clear association between general plant-based eating patterns and dementia risk, suggesting that the quality of plant foods may be more influential than plant-based adherence alone (de Crom et al., 2023[19]).

The traditional Nordic diet, characterized by root vegetables, whole grains (rye, barley, oats), berries, fish, and rapeseed oil, has been linked to slower cognitive decline and longer dementia free survival in older adults in some Scandinavian cohorts. This pattern shares key features with other healthy diets (e.g., high in plant foods and unsaturated fats), suggesting that beneficial effects on cognition may generalize beyond Mediterranean style eating (Townsend et al., 2024[117]).

Other healthy dietary patterns, such as anti-inflammatory or prudent diets rich in fruits, vegetables, lean proteins, and whole foods, have been linked to better cognitive outcomes in some studies, though they largely overlap with MeDiet and plant-based diets (Chen et al., 2019[14]).

A growing body of epidemiological and mechanistic evidence indicates that lifestyle- and nutrition-related modifiable cardiometabolic and psychosocial conditions (Figure 4(Fig. 4)) significantly contribute to the development and progression of cognitive decline and dementia (Livingston et al., 2024[68]). These associations are robust across populations and represent key targets for primary prevention.

Large-scale RCTs have shown that lifestyle modification programs significantly lower diabetes risk in individuals at high-risk (Lee et al., 2025[61]). Type 2 diabetes is consistently associated with an elevated risk of dementia, particularly vascular dementia and all-cause dementia, likely due to chronic hyperglycemia, insulin resistance, and vascular injury (de la Monte et al., 2018[20]; Cao et al., 2024[11]; Michailidis et al., 2022[78]). Diabetes promotes cerebrovascular disease, oxidative stress, and inflammation, which accelerate neurodegeneration and cognitive impairment (Michailidis et al., 2022[78]; de la Monte et al., 2018[20]). Epidemiological studies show that individuals with diabetes have a notably higher risk of progressing from MCI to dementia compared with non-diabetics (Ding et al., 2024[21]). Diabetes may contribute to dementia through several mechanisms: insulin resistance, where disrupted brain insulin signaling affects amyloid β metabolism; vascular damage, increasing stroke risk and microvascular injury that lowers cerebral perfusion; and chronic inflammation, which can accelerate neuronal loss (de la Monte et al., 2018[20]; Michailidis et al., 2022[78]). However, Mendelian randomization data suggest that diabetes per se may not directly increase Alzheimer's disease risk independent of metabolic intermediates like insulin and cholesterol, highlighting complexity in causal pathways (Wang et al., 2024[123]). A recent meta-analysis of 15 studies (10.1 million participants; 8.8 million with diabetes) showed that diabetes was associated with a 59 % significantly increased risk of dementia. Diabetes duration < 5 years (RR 1.29, 95 % CI 1.20-1.39) and hypoglycemia (RR 1.56, 95 % CI 1.13-2.16) were associated with higher dementia risk, while glycemic control measures showed no significant effect (Wang et al., 2024[123]).

High blood pressure, especially during midlife, is one of the most consistent vascular risk factors linked to later cognitive decline and dementia (Huang and Aronow, 2024[41]; Levine et al., 2020[63]). Elevated systolic blood pressure contributes to small vessel disease, white matter hyperintensities, and microinfarcts, which are neuropathological substrates of both vascular dementia and Alzheimer's disease (Ungvari et al., 2021[120]). Furthermore, blood pressure lowering with antihypertensive therapy, compared to control, was significantly linked to a reduced risk of dementia or cognitive impairment in a meta-analysis of 14 RCTs (Hughes et al., 2020[43]).

Obesity, particularly in midlife, is associated with higher dementia risk (Qu et al., 2020[90]; Emmerzaal et al., 2015[25]). Adiposity contributes to systemic inflammation, dysregulated adipokine profiles, insulin resistance, and cerebrovascular disease, all of which can adversely affect brain structure and function (Patel and Edison, 2024[87]).

High total cholesterol and dyslipidemia in midlife have been linked to increased dementia risk, particularly Alzheimer's disease (Wee et al., 2023[125]; Iwagami et al., 2021[47]). Proposed mechanisms include cholesterol-mediated effects on cell membrane composition and amyloid processing (Gamba et al., 2015[34]), as well as dyslipidemia-related cerebrovascular atherosclerosis, which may reduce cerebral blood flow (Siegler et al., 2025[107]). Although observational data support an association, clinical evidence on the cognitive benefits of lipid-lowering therapies remains mixed and requires further investigation. Meta-analyses reported inconsistent evidence suggesting that elevated LDL cholesterol in midlife-but not in late life-may be associated with an increased risk of cognitive decline, all-cause dementia, and Alzheimer's disease (Anstey et al., 2017[2]; Wood et al., 2014[128]).

Depression is increasingly recognized as both a risk factor and early symptom of cognitive decline. Longitudinal cohort studies have found that depressive symptoms are associated with an elevated risk of subsequent dementia, independent of other health conditions (Stafford et al., 2022[112]). Potential mechanisms include chronic stress and dysregulation of hypothalamic-pituitary-adrenal axis leading to hippocampal damage, inflammatory pathways that overlap with neurodegenerative processes, and reduced cognitive reserve due to behavioral and lifestyle changes associated with depression (Borges de Souza et al., 2025[8]). Depression may interact synergistically with metabolic risk factors to further elevate dementia risk, underscoring the importance of integrated care. Depression is closely intertwined with lifestyle factors. Unhealthy behaviors-such as poor diet quality, physical inactivity, inadequate sleep, smoking, excessive alcohol consumption, and chronic stress-are associated with an increased risk of developing depressive symptoms, whereas healthier patterns, including regular physical activity, balanced nutrition, and good sleep hygiene, appear to be protective. Although depression has important biological and genetic underpinnings, lifestyle plays a substantial modulatory role in its onset, severity, and persistence. Moreover, lifestyle and depression often interact bidirectionally, creating a self-reinforcing cycle in which depressive symptoms lead to deteriorating health behaviors, which in turn further exacerbate depression and impede recovery (Sato et al., 2025[99]). In essence, while not the only cause, lifestyle factors are powerful contributors and modulators of depression, acting as significant risk or protective elements.

Lifestyle factors are central to the prevention of cognitive decline and dementia. Because complex diseases develop through the cumulative interaction of multiple influences over time, effective prevention is likely to require integrated lifestyle strategies, such as those reviewed in this article (Figure 5(Fig. 5)), many of which show evidence of potential neuroprotective effects.

Successive updates of the Lancet Commission on Dementia Prevention, Intervention, and Care have progressively broadened understanding of modifiable, largely lifestyle-related dementia risk factors across the life course. Emerging evidence has expanded the framework beyond individual medical conditions to include biological, sensory, environmental, and lifestyle factors. The number of identified modifiable risks has increased from 9 in the first report (Livingston et al., 2017[70]) to 14 in the most recent update (Livingston et al., 2024[68]), reflecting growing recognition of multiple, interacting pathways to dementia, including vascular disease, sensory impairment, environmental exposures, and health-related behaviors. Correspondingly, the estimated proportion of preventable dementia cases has risen from ~35 % to 40 %, and most recently to 45 % (Livingston et al., 2024[68]; Livingston et al., 2020[69]; Livingston et al., 2017[70]), underscoring the potential impact of comprehensive, lifespan-wide risk reduction strategies.

Evidence supporting multidomain interventions as a non-pharmacological strategy for dementia prevention is rapidly growing (Sakurai et al., 2025[98]). However, a definitive answer to whether dementia can be prevented remains elusive. To date, most studies have relied on changes in cognitive function as surrogate outcomes rather than dementia incidence, which would require longer follow-up periods. As multidomain interventions are implemented more broadly at the population level, it will become possible to assess their impact on community-wide dementia incidence and prevalence. Looking ahead, early detection and timely intervention are expected to become as central to dementia care as they are in cardiovascular disease prevention, where there is greater awareness among both healthcare providers and the public.

The Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) was the first large, long-term RCT to show that a two-year multidomain lifestyle intervention improves cognition and reduces cognitive decline in older adults at increased dementia risk. The intervention included diet, exercise, cognitive training, social activities, and vascular risk management, and led to significant benefits in global cognition, executive function, processing speed, and memory, as well as improvements in BMI, diet adherence, and physical activity (Ngandu et al., 2015[82]; Lehtisalo et al., 2017[62]).

Following the encouraging results of FINGER, 15 other multi-component intervention studies have been conducted, and additional 11 studies are ongoing or planned spanning high-, middle-, and low-income countries across all global regions (Sakurai et al., 2025[98]). While early multidomain dementia prevention trials (2015-2019) showed mixed results, most studies conducted in the 2020s have reported beneficial cognitive effects, with negative findings largely attributable to COVID-19 disruptions. Even when primary cognitive endpoints were not met, secondary and subgroup analyses consistently showed improvements in key dementia risk factors, supporting the potential of multidomain interventions to slow cognitive decline in at-risk older adults.

In 2017 the World-Wide (WW)-FINGERS network was launched seeking to harmonize multidomain dementia prevention trials, to promote international collaboration, data and knowledge sharing, and culturally adapted multidomain lifestyle intervention trials across diverse at-risk populations worldwide (Kivipelto et al., 2020[57]; Sindi et al., 2025[108]). Parallel brain health initiatives are also being implemented in primary care, community, and memory clinic settings. Digital, e-health-based interventions (e-FINGERS) are also under investigation (Loukas et al., 2023[71]; Rosenberg et al., 2024[96]).

The first results of the Protect Brain Health Through Lifestyle Intervention to Reduce Risk (POINTER) study have recently been published. In a single-blind, multicenter RCT of 2,111 US adults aged 60-79 years at risk of cognitive decline, a 2-year structured multidomain lifestyle intervention produced a significantly greater improvement in global cognition than a self-guided intervention (Baker et al., 2025[4]), supporting the effectiveness of higher-intensity nonpharmacological approaches for dementia prevention.

Recently, anti-amyloid monoclonal antibodies such as lecanemab and donanemab have become available for individuals with early Alzheimer's disease. However, these disease-modifying therapies are not indicated for all types of dementia, and many patients may not meet eligibility criteria. Their use also entails intensive safety monitoring, substantial costs, and other treatment burdens (Fox et al., 2025[30]; Frisoni et al., 2025[31]). Consequently, the need for effective dementia prevention strategies is increasingly urgent, given their low cost, favorable safety profile, and broad benefits for overall health and quality of life.

The growing evidence presented indicates that behaviors such as sustained social and cognitive engagement, adequate sleep, smoking cessation, moderation of alcohol intake, regular physical activity, and healthy dietary patterns are generally associated with better cognitive outcomes and lower dementia risk. These lifestyle factors appear to influence multiple biological pathways implicated in neurodegeneration, including vascular health, systemic and neuroinflammation, metabolic regulation, and neuroplasticity, and they often exert cumulative effects across the life course. For example, regular physical activity can enhance cerebral blood flow and promote synaptic plasticity, while healthy dietary patterns such as the Mediterranean or MIND diets may reduce oxidative stress and support neuronal function. Similarly, maintaining social and cognitive engagement has been linked to increased cognitive reserve, potentially delaying the onset of clinical symptoms even in the presence of neuropathology.

Although causal inference is constrained by methodological challenges-such as confounding, reverse causation, and the limited availability of long-term RCTs-the overall consistency of observational evidence underscores the potential benefits of a multidomain approach. Integrating several lifestyle factors simultaneously may yield additive or synergistic effects, enhancing resilience against dementia-related pathophysiology. Importantly, these interventions are not only feasible and low-risk but also adaptable, offering a population-level strategy for dementia prevention. Future research should focus on long-term, pragmatic trials that examine the timing, intensity, and combination of lifestyle interventions, further explore interactions with genetic risk factors such as APOE ε4, and evaluate effects on cognitive trajectories, functional outcomes, and underlying biomarkers. By advancing such knowledge, it may become possible to design personalized, culturally adapted prevention strategies that maximize cognitive health across diverse populations and stages of life.