Prevalence and Incidence

The prevalence of all-cause dementia, as well as AD and VaD, dramatically increases with age. An estimated 5.8 million Americans aged 65 and older were living with AD in 2020 [1]. It is estimated that this number will nearly triple to 13.8 million by 2050 [1,Reference Hebert, Weuve, Scherr and Evans2].This increase in the prevalence of AD over the next 50 years will primarily reflect increased life expectancy and the aging of adults who were born during the “baby boom” after World War II (1946–1964), who will make up an increasingly large proportion of our population. The incidence of all-cause dementia, as well as AD and VaD, rises exponentially with age. According to US census population estimates, the annual incidence of AD increases from approximately 53 new cases to 170 and 231 per 1000 people for the age groups 65–74, 75–84, and 85+, respectively. Interestingly, recent declines in dementia incidence within study cohorts have been reported; for instance, the incidence rate of dementia in Europe and North America has declined by 13% per decade over the past 25 years, consistently across 7 cohort studies [Reference Wolters, Chibnik and Waziry3]. The consistency of the finding across multiple epidemiological cohorts in high income countries, while not universal, have been attributed to society-level changes such as increased education and improved healthcare. However, the reasons behind this improvement are not yet clearly understood. Nonetheless, dementia prevalence will increase dramatically over the next few decades due to demographic shifts in both high and low to middle income countries.

Costs

Dementia impacts not only the patients themselves, but also their families and the healthcare system. The direct costs of caring for those with AD and dementia include nursing home or paid healthcare. However, most people with AD live at home, and much of their care is provided informally by family and friends [Reference Rice, Fox and Max4]. The indirect costs of dementia include lost productivity, absenteeism, and worker-replacement costs for these caregivers.

According to the Alzheimer’s Association, the total cost in 2020 for all individuals with Alzheimer’s or other dementias was estimated at US$305 billion and is projected to increase to US$1.2 trillion in 2050 [1]. Costs attributed to nursing home, formal, and informal care represent 75% to 84% of the total attributable cost, including payments for healthcare, long-term care, and hospice for subjects identified as Medicare beneficiaries.

The Potential for Prevention

Given the dramatic increases that are expected in both the prevalence and costs of AD and other dementias, it is critical to identify factors that are associated with increased disease risk and to develop successful prevention and treatment strategies. Because AD and most dementias occur most commonly in very old age, interventions that delay the onset of disease have the potential to significantly reduce prevalence over time. It has been suggested that 10% to 25% reduction in seven modifiable risk factors of AD could prevent as many as 1.1 to 3.0 million AD cases worldwide and 184 000 to 492 000 cases in the US [Reference Barnes and Yaffe5]. The Lancet Commission report on care in 2017 followed up on this work, highlighting that around 35% of dementias is attributable to a combination of nine risk factors: education to a maximum of 11–12 years, midlife hypertension, midlife obesity, hearing loss, late life depression, diabetes, physical inactivity, smoking, and social isolation. These risk factors are modifiable and targetable by intervention [Reference Livingston, Sommerlad and Orgeta6]. In the updated 2020 report, three additional risk factors for dementia were included on account of new evidence: excessive alcohol consumption, traumatic brain injury (TBI), and air pollution. Strategies focusing on several modifiable risk factors for dementia are a promising avenue for primary or secondary prevention.

Sex and Gender

Almost two-thirds of Americans aged 65 and older with AD are women [Reference Hebert, Weuve, Scherr and Evans2]. This larger proportion may be explained by the fact that women live longer, on average, than men and hence experience a longer period of risk [Reference Hebert, Scherr, McCann, Beckett and Evans7,Reference Seshadri, Wolf and Beiser8]. However, it is possible that increased longevity may not entirely explain the higher prevalence in women. Therefore, it remains controversial whether observed gender differences reflect true differences in underlying disease etiology, or if other factors are involved.

Differential development and progression of AD in men and women may reflect differences on a biological basis (i.e., sex differences) such as chromosomal, epigenetic, or hormonal differences. For instance, cardiovascular disease, a known risk factor for dementia, develops on average 10 years later in women than in men. Such factors may contribute to the selective survival, as men who live to older ages are more robust and potentially at lower risk of developing dementia [Reference Chêne, Beiser and Au9]. In addition, gender differences reflected in psychosocial and cultural factors such as access to education and occupational roles may also contribute to observed differences in dementia risk between men and women [Reference Mielke, Vemuri and Rocca10]. A recent study investigated whether differences in educational attainment and experiences of general psychological distress mediate the association between female sex and dementia; the authors found that social (dis) advantage predicted general psychological distress, constituting a possible pathway between female sex, education, and dementia [Reference Hasselgren, Ekbrand and Halleröd11].

Race/Ethnicity

Racial and ethnic disparities in dementia incidence are well documented in the US. Notably, Blacks and Hispanics have higher prevalence and incidence of dementia and AD compared to non-Hispanic Whites. One study of a large population from an integrated healthcare system with equal access to healthcare and a 14-year follow-up period found that dementia incidence was highest among Black and American Indian/Alaska Native, lowest among Asian Americans, and intermediate among Latinos, Pacific Islanders, and Whites. These inequalities in dementia incidence were observed among women and men and across all ages [Reference Mayeda, Glymour, Quesenberry and Whitmer12].

The above findings suggest that efforts aimed at reducing such disparities may affect the incidence of dementia in ethnic/racial minorities. For example, findings from another study suggest that persistent racial differences in cognitive level rather than differences in cognitive decline in old age likely underlie the higher risk of incident AD among Blacks [Reference Weuve, Barnes and Mendes de Leon13]. The results from the Health ABC study suggest that differences in the burden of risk factors, especially socioeconomic status, may contribute to the higher rates of dementia seen among Black compared with White older people [Reference Yaffe, Falvey and Harris14].

Evidence is emerging to suggest that racial disparities in cognitive outcomes may also reflect the unequal burden of cardiovascular disease among Blacks [Reference Xiong, Luo and Coble15,Reference Levine, Gross and Briceño16]. A recent study reported that cumulative blood pressure explains some of the disparities in late-life cognitive decline on global cognition and memory in Blacks compared to Whites [Reference Levine, Gross and Briceño16]. These studies suggest that race/ethnic differences may largely reflect racial disparities over the life course in, for example, education, access to resources including healthcare, exposure to discrimination, and exposure to environmental toxins, rather than genetic factors [Reference Caunca, Odden and Glymour17]. Additional research is needed to determine what underlies these health disparities and to develop interventions that target those at highest risk.

Education

There is considerable evidence that older adults with greater education are less likely to develop AD and dementia [Reference Evans, Hebert and Beckett18,Reference Stern, Gurland and Tatemichi19]. One often cited study analyzed writing samples from nuns when they were 22 years old and found that sisters with high idea density and grammatical complexity early in life were less likely to develop AD later in life [Reference Snowdon, Kemper and Mortimer20]. This may reflect, in part, a measurement bias in which older adults with more education or higher intelligence perform better on cognitive tests, making AD and dementia more difficult to detect. However, there is growing evidence that education may be associated with greater cognitive or neuronal reserves, which may protect against or minimize the impact of neurodegenerative disorders [Reference Jones, Manly and Glymour21]. A recent study suggests that education is positively associated with cognitive ability in early life before reaching a plateau in late adolescence, with relatively few further gains from education after age 20 years [Reference Kremen, Beck and Elman22]. This suggests that education increases the overall level of cognitive reserve (CR), creating a buffer against cognitive decline, thus delaying the onset of clinically significant dementia. However, another study found that years of education contributed to CR only among Whites, but not among Blacks and Hispanics, which could partly be explained by disparities in quality of education; therefore, years of education would be an inaccurate indicator of experiences related to CR [Reference Arce Rentería, Vonk and Felix23]. Literacy is another marker of brain reserve and is considered to be a better reflection of educational attainment than years of schooling. According to a study among high-functioning older adults, low literacy was significantly associated with greater incidence of dementia, especially among individuals who were apolipoprotein E4 (APOE ε4) negative [Reference Kaup, Simonsick and Harris24]. The independent effect of illiteracy on dementia risk may reflect a lower level of cognitive function compared to literate participants, closer to diagnostic thresholds for dementia [Reference Arce Rentería, Vonk and Felix23].

Cardiovascular Factors

Traditionally, AD has been viewed as a neurodegenerative disorder distinct from vascular dementia (VaD). However, autopsy studies have suggested that many patients with clinical AD also have evidence of vascular pathology [Reference Snowdon, Greiner and Mortimer25], especially with increasing age. There is also growing evidence that risk factors for cardiovascular disease, such as diabetes mellitus, obesity, hypertension, and hyperlipidemia, are associated with a higher risk of developing AD and other dementias. Research related to each of these risk factors, as well as the effects of combined cardiovascular risk factors, is described in more detail in the following sections.

Diabetes Mellitus

Population-based, longitudinal studies have consistently found that older adults with diabetes experienced approximately a twofold increase in dementia risk [Reference Ahtiluoto, Polvikoski and Peltonen26,Reference Hassing, Johansson and Nilsson27]. When specific dementia subtypes are examined, diabetes is associated more strongly with VaD than AD [Reference Areosa Sastre, Vernooij, González-Colaço Harmand and Martínez28]. Studies also found that older women with impaired fasting glucose, a prediabetic condition, experienced a twofold increase in the risk of significant cognitive decline [Reference Yaffe, Blackwell and Kanaya29]. Findings have also suggested that hypoglycemia, a common symptom of diabetes, is associated with a twofold increased risk of developing dementia [Reference Yaffe30], suggesting caution in being too aggressive with treatment of diabetes in older adults. In a recent systematic review and meta-analysis, including 122 prospective studies, diabetes conferred a 1.25- to 1.91-fold excess risk for cognitive disorders (cognitive impairment and dementia) [Reference Xue, Xu and Ou31]. Interestingly, prediabetes was also associated with a higher risk for dementia. Moreover, the use of antidiabetic medication reduced the risk of dementia by 47%.

Hypertension

Hypertension in midlife, especially if not treated effectively, has consistently been associated with increased risk of dementia and AD in late life [Reference Qiu32]. In the Honolulu-Asia Aging Study, a longitudinal study of approximately 3700 older Japanese American men, the risk of dementia was almost tripled in men who had untreated systolic blood pressure over 160 mmHg during midlife [Reference Launer, Ross and Petrovitch33]. Moreover, pooled analysis has shown that high blood pressure in midlife is associated with a 60% increased risk of AD [Reference Barnes and Yaffe5]. In late life, both very high systolic blood pressure and very low diastolic blood pressure have been associated with increased risk of dementia and AD [Reference Qiu, Winblad and Fratiglioni34]. High blood pressure may lead to dementia by increasing the risk of ischemia and stroke, whereas low blood pressure in late life may lead to dementia by increasing the risk of cerebral hypoperfusion and hypoxia. Variation in blood pressure seems to also play a role in dementia risk. A large blood pressure variation (both a large rise and fall) has been associated with an increased long-term risk of dementia in older adults and lower cognitive performance in midlife [Reference Rouch, Vidal and Hoang35,Reference Yano, Ning and Allen36].

A review of randomized controlled trials (RCTs) reported that antihypertensive drugs, particularly calcium channel blockers and renin–angiotensin system blockers, may be beneficial in preventing cognitive decline and dementia. However, further randomized controlled trials with longer periods of follow-up and cognition as the primary outcome are needed to confirm these findings [Reference Rouch, Cestac and Hanon37]. In the more recent SPRINT-MIND trial, aggressive systolic blood pressure treatment did not result in a significant reduction in the risk of probable dementia; however, it significantly reduced the risk of mild cognitive impairment (MCI). Due to early study termination and fewer than expected cases, the study may have been underpowered for the end point of dementia [Reference Williamson, Pajewski and Auchus38]. Continued follow-up of SPRINT-MIND participants is crucial to evaluate the long-term effects of intensive systolic blood pressure control on the prevention of cognitive decline.

Hyperlipidemia

High cholesterol level is linked to greater risk of atherosclerosis and cerebrovascular disease, as well as cortical infarcts and white matter lesions [Reference Amarenco, Labreuche and Elbaz39]. In particular, studies on the role of total cholesterol at midlife have found that higher levels of cholesterol are associated with dementia, AD, and cognitive impairment [Reference Shepardson40]. On the contrary, the literature on the role of late-life cholesterol on dementia and cognitive health has been inconsistent [Reference Reitz, Tang and Schupf41]. The evidence on the role of other plasma lipids such as high-density lipoprotein (HDL) or low-density lipoprotein (LDL) cholesterol is limited [Reference Amarenco, Labreuche and Elbaz39,Reference Reitz, Tang and Schupf41]. A recent meta-analysis showed a twofold increased dementia risk for participants with high total cholesterol in midlife compared with normal cholesterol [Reference Anstey, Ashby-Mitchell and Peters42]. Late-life measured HDL cholesterol and triglycerides were not associated with any dementia.

Although some observational evidence indicates that statins are protective against dementia, AD, and cognitive impairment [Reference Beydoun, Beason-Held and Kitner-Triolo43,Reference Dufouil, Richard and Fievet44], the literature has been inconsistent. A more recent meta-analysis suggests that the use of statins may reduce the risk of all-type dementia, AD, and MCI, but not of incident VaD [Reference Chu, Tseng and Stubbs45]. Furthermore, some trials have shown no effect, including results from the PROspective Study of Pravastatin in the Elderly at Risk (PROSPER) [Reference Trompet, Vliet and Craen46]. No RCT to date has clearly shown evidence of either positive or negative effect of statins on cognition. The US Food and Drug Administration (FDA) has issued a safety warning regarding potential memory loss associated with statin use based on several reports of worse cognition [Reference Padala, Padala and Potter47,Reference Muldoon, Ryan, Sereika, Flory and Manuck48].

Obesity

Evidence suggests a U-shaped association between obesity and cognitive decline and AD pathology. Studies have shown that the risk of developing dementia may differ according to the timing of obesity, that is, midlife versus later life obesity [Reference Anstey, Cherbuin, Budge and Young49,Reference Whitmer, Sidney, Selby, Johnston and Yaffe50]. In particular, midlife obesity was associated with greater risk of AD, whereas lower body mass index (BMI) and weight loss in later life were associated with greater risk of AD. Compelling results from the Whitehall study showed that obesity at age 50 years, but not at 60 or 70 years, was associated with a higher risk of dementia. Moreover, the BMI trajectories for participants who developed dementia were characterized by a higher BMI several decades before diagnosis, and lower BMI starting 8 years before diagnosis, compared to the BMI trajectories of matched controls [Reference Kivimäki, Luukkonen and Batty51]. Evidence of protective effects of obesity or overweight on dementia risk in older groups may reflect biases, especially weight loss before dementia diagnosis. Epidemiological evidence on the association between obesity and risk of dementia may partly reflect the role of obesity as a marker of vascular and inflammatory damage [Reference Zeki Al Hazzouri, Stone, Haan and Yaffe52].

Multiple Cardiovascular Risk Factors

Several studies have evaluated the effects of multiple or composite cardiovascular risk factors on dementia risk. Among members of a large health-maintenance organization, participants who had diabetes, hypertension, high cholesterol, or were smokers at midlife were more likely to develop dementia later in life, and the effects of each factor were approximately additive [Reference Whitmer, Sidney, Selby, Johnston and Yaffe50]. Similarly, the “metabolic syndrome,” which is a clustering of cardiovascular disorders, has been associated with increased risk of dementia, cognitive impairment, and cognitive decline, especially in subjects with high levels of inflammation [Reference Yaffe53]. Studies using composites of vascular risk scores showed that ideal cardiovascular health defined by guidelines from the American Heart Association – including the avoidance of overweight or obesity, a healthful diet, nonsmoking, physical activity, total cholesterol, blood pressure, and fasting glucose at goal levels in young adulthood and its maintenance to middle age – was associated with better cognition in midlife [Reference Reis, Loria and Launer54]. Moreover, a recent study showed that ideal cardiovascular health in midlife was associated with lower risk of dementia in late life [Reference Sabia, Fayosse and Dumurgier55]. Studies using the Framingham Risk Score showed that higher risk profiles were associated with conversion from MCI to dementia and accelerated cognitive decline [Reference Viticchi, Falsetti and Buratti56,Reference Yaffe, Bahorik and Hoang57]. Lastly, a large, pooled analysis of cardiovascular risk factors across the life course, especially in early adulthood, were associated with greater late-life cognitive decline [Reference Yaffe, Vittinghoff and Hoang58]. These findings suggest that targeting cardiovascular health in early adulthood could convey the largest benefit for late-life cognition.

Cognitive Activity

There is growing evidence that cognitive activity may help older adults build or maintain a cognitive reserve that may delay the onset of overt dementia symptoms [Reference Stern59]. Several observational studies found that older adults who engage in mentally stimulating activities, such as reading or playing games, are at lower risk of cognitive decline and dementia [Reference Treiber, Carlson and Corcoran60]. Studies with measures of cognitively stimulating activities in mid-life, such as travel, social outings, playing music, art, physical activity, reading, and speaking a second language, found that engaging in such activities was associated with maintaining cognition independent of education, occupation, late-life activities, and current structural brain health [Reference Chan, Shafto and Kievit61] and showed less deposition of beta-amyloid (Aβ) [Reference Landau, Marks and Mormino62]. However, a recent study showed that the relationship between cognitive activity and cognitive health in older age could be interpreted differently. The frequency of leisure activities (e.g., gardening, socializing with friends and family, museum visits, arts and crafts) at study entry, 5 years, and 10 years later in a group of more than 8000 British civil servants aged 45–69 years was investigated in relation to dementia risk [Reference Sommerlad, Sabia and Livingston63]. The authors found that more frequent leisure activity was associated with lower risk of dementia, but only in the short term. Moreover, participants who had a decline in leisure activity over the study period had a higher dementia risk than participants who had continuously lower engagement. These findings suggest that the reduction of or withdrawal from cognitively stimulating activity may reflect ongoing neuropathology and represent an early symptom of dementia.

Physical Activity

Several lines of evidence suggest that physical activity may protect against cognitive decline and dementia in older adults. Multiple prospective observational studies found that older adults who exercise are less likely to experience cognitive decline and dementia [Reference Abbott64–Reference Yaffe, Barnes, Nevitt, Lui and Covinsky66]. A systematic review of prospective and intervention studies in non-demented older adults found that physical activity was inversely associated with risk of AD in most studies [Reference Stephen, Hongisto, Solomon and Lönnroos67]. However, findings from a prospective cohort with a mean follow-up time of 27 years found no evidence of a neuroprotective effect of physical activity [Reference Sabia, Dugravot and Dartigues68]. The authors speculate that findings showing a lower risk of dementia in physically active people may be attributable to reverse causation due to a decline in physical activity levels in the preclinical phase of dementia.

Nevertheless, a meta-analysis of RCTs found that sedentary older adults who were randomized to exercise interventions experienced short-term improvements in cognitive function relative to controls [Reference Colcombe and Kramer69]. In an RCT among older adults with cognitive problems but not yet dementia, physical activity provided modest improvement in cognition [Reference Lautenschlager, Cox and Flicker70]. However, findings from the LIFE RCT showed that among sedentary older adults with no cognitive impairment, a 24-month moderate-intensity physical activity program compared with a health education program did not result in improvements in global or domain-specific cognitive function [Reference Sink, Espeland and Castro71]. Current evidence does not allow translation to practical recommendations concerning the type, frequency, intensity, or duration of physical activity that may be protective against AD or other dementias.

Alcohol

A meta-analysis of epidemiological studies found that alcohol use was associated with decreased risk of any dementia [Reference Anstey, Mack and Cherbuin72]. Several studies have found that light to moderate drinkers have a 30%–50% reduction in the risk of cognitive decline, AD, and VaD compared with non-drinkers [Reference Mukamal73–Reference Stampfer, Kang, Chen, Cherry and Grodstein75]. Potential pathways depend on alcohol type, but may include lowering lipid levels and antioxidant effects [Reference Collins, Neafsey and Wang76]. There is some evidence of a U-shaped relationship, in which high alcohol consumption does not appear to be protective. Recent findings from the Whitehall study show that the risk of dementia over 23 years was increased in people who abstained from alcohol in midlife or consumed > 14 units/week, compared to those who consumed 1–14 units/week [Reference Sabia, Fayosse and Dumurgier77]. Moreover, a recent study found that alcohol use disorder among older female US veterans was associated with a more than threefold increase of dementia [Reference Bahorik, Bobrow, Hoang and Yaffe78]. In several countries, current guidelines for harmful alcohol consumption are much higher than 14 units/week. The above findings suggest that a downward revision of such guidelines would be beneficial for cognitive health in old age.

Smoking

Population-based studies have found that the incidence of AD is approximately doubled in older adults who are current smokers [Reference Anstey, von Sanden, Salim and O’Kearney79]. Furthermore, a smoking cessation trial showed that older adults who were successful quitters experienced less cognitive decline than unsuccessful quitters [Reference Almeida, Garrido and Alfonso80]. Recent findings showed a dose-dependent relationship between cumulative smoking in early to mid-life and worse cognitive function in mid-life [Reference Bahorik, Sidney and Kramer-Feldman81]. These findings emphasize the cognitive health risk of moderate and heavy smoking and the potential benefits of quitting. The effects of smoking on cognitive function may involve several pathways. Smoking may increase oxidative stress and inflammation, which could be important pathophysiological mechanisms in developing dementia [Reference Durazzo, Mattsson and Weiner82]. Moreover, smoking is also a critical risk factor for cerebrovascular diseases, such as stroke, hypertension, and diabetes, suggesting that smoking exposure may affect cognitive function via vascular pathways [Reference Durazzo, Mattsson and Weiner82].

Dietary Factors

Growing evidence suggests that specific nutrients or healthy diets are protective against cognitive decline and dementia. Laboratory studies have suggested that oxidative stress may contribute to the pathogenesis of AD [Reference Behl83], leading to the hypothesis that high dietary intake of antioxidants might lower disease risk. Several observational studies have found an inverse association between dietary intake of vitamins E and C and cognitive decline and risk of AD [Reference Engelhart84,Reference Maxwell, Hicks, Hogan, Basran and Ebly85]. On the contrary, a Cochrane review of dietary supplement RCTs (including vitamins, calcium, zinc, antioxidants) reported no evidence of a beneficial effect for cognitive function or dementia risk in middle aged or older people [Reference Rutjes, Denton and Di Nisio86].

In the past 5 years, there has been a shift toward investigating the effects of whole diets or dietary patterns rather than specific nutrients or supplements. Examples include the Mediterranean diet (high intake of vegetables, legumes, fruits, nuts, cereals, and olive oil; low intake of saturated lipids and meat) or the similar Nordic diet (high intake of berries, vegetables, legumes, potatoes, rye; low intake of red meat), which might reduce cognitive decline and dementia [Reference Pistollato, Iglesias and Ruiz87,Reference Chen, Maguire, Brodaty and O’Leary88]. One prospective cohort study with three dietary assessments over 25 years, reported a protective effect of the healthy dietary pattern in dementia risk, only in participants with cardiovascular disease, suggesting that diet might influence dementia risk by reducing the risk associated with cardiovascular risk factors [Reference Akbaraly, Singh-Manoux and Dugravot89]. Moreover, greater adherence to healthy dietary patterns during young adulthood was associated with better midlife cognitive performance [Reference McEvoy, Hoang and Sidney90]. Better understanding of the underlying mechanisms and effectiveness is needed to develop comprehensive and practical dietary recommendations against cognitive impairment among older adults.

Other Factors