Global mortality, prevalence and disability-adjusted life years of Alzheimer’s disease and other dementias in adults aged 60 years or older, and the impact of the COVID-19 pandemic: a comprehensive analysis for the global burden of disease 2021

Abstract

Objective

Alzheimer’s disease (AD) and other dementias are major public health concerns with an increasing global impact. The burden of these conditions varies by region, age, and gender, and the COVID-19 pandemic has further exacerbated these disparities, potentially influencing disease prevalence, mortality, and disability burden. This study aimed to assess the global and regional burden and trends of Alzheimer’s disease and other dementias in adults aged 60 years or older from 1990 to 2021, with a particular focus on the impact of the COVID-19 pandemic on mortality, prevalence, and disability-adjusted life years.

Methods

Using Global Burden of Disease (GBD 2021) data, we analyzed age-standardized death rates (ASDR), incidence rates (ASIR), prevalence rates (ASPR), and disability-adjusted life years (DALYs) from 1990 to 2021. Temporal trends were assessed using the Estimated Annual Percentage Change (EAPC). Projections were modeled using Bayesian Age-Period-Cohort (BAPC) techniques. We evaluated excess mortality by comparing actual versus expected deaths during the pandemic. Decomposition analysis examined the contributions of population growth, aging, and epidemiological shifts. We analyzed health inequality to highlight and address disparities in health status and resource access across regions. All plots and tables were created using Joinpoint Regression model (Version 4.8.0.1), StataMP 18, and R statistical packages (Version 4.4.1).

Results

In 2021, global mortality from AD and other dementias among individuals aged 60 and older reached approximately 1,922,970.75 cases (95% CI: 480,348.08 to 5,104,315.95), and the prevalence was 52,560,253.51 cases (95% CI: 41,399,948.84 to 65,633,448.71). High Body Mass Index (BMI) and High Fasting Plasma Glucose (FPG) were prominent risk factors. Projections suggest a near fourfold increase in AD cases by 2050, driven by population growth and aging, with females disproportionately affected. Health inequalities persist, with higher disease burdens in high-SDI regions. The pandemic impacted mortality unevenly, highlighting regional disparities.

Conclusion

Although incidence rates declined from 1990 to 2021, the overall burden of AD and dementias remains substantial and is expected to rise significantly by 2050. The findings underscore the need for targeted interventions addressing risk factors like High FPG, gender disparities, and the socioeconomic effects of COVID-19, particularly in high-SDI countries.

Introduction

Alzheimer's disease (AD) is a progressive, neurodegenerative condition and the predominant cause of dementia [1], characterized by increasing cognitive deterioration, memory deficits, and behavioral anomalies [2, 3]. Other non-Alzheimer’s dementias include vascular dementia (VaD), and Other Neurodegenerative Dementias [4]. AD diagnosis involves comprehensive clinical evaluations, including patient history, cognitive and functional assessments, and corroboration from a close informant [5]. Although emerging therapies are under investigation, treatments are limited and lack targeted approaches.

Recent studies indicate a significant global increase in the burden of Alzheimer's disease and other dementias. In 2016, approximately 43.8 million individuals were diagnosed with dementia worldwide, nearly doubling from 20.2 million in 1990—a 117% increase over 26 years. This surge is mainly attributed to population aging and growth [6]. The ASIR of AD and other dementias increased by 147.95%, and the ASPR increased by 160.84% from 1990 to 2021. A substantial increase in the burden for women was also presented [7]. Future projections suggest that by 2050, the prevalence of dementia is expected to reach about 152.8 million, largely due to ongoing population growth and aging, with significant variations by region [8].

COVID-19, having become a global pandemic, is known to affect various aspects of human health. While it is plausible that the virus could have implications for AD, for instance, SARS-CoV-2 infection has been shown to exacerbate beta-amyloid neurotoxicity, inflammation, and oxidative stress in AD patients. This suggests that the virus can worsen the pathological processes already present in AD, potentially accelerating disease progression [9]. Surprisingly, there has been a limited effort investigating the direct impact of COVID-19 on AD specifically. Liu et al. performed a systematic review and meta-analysis, finding that the mortality rate of individuals with dementia after being infected with COVID-19 was significantly higher than those without dementia (OR: 5.17) [10]. Wang et al. analyzed electronic health records and found that patients with dementia had an increased risk for COVID-19 (AOR: 2.00) and higher 6-month mortality (20.99%) and hospitalization rates (59.26%) compared to those without dementia [11]. However, existing studies are largely hospital-based, limited to single-country analyses, and reliant on relatively small patient cohorts.

Our study utilizes the Global Burden of Disease (GBD) 2021 dataset, providing the most comprehensive, population-level assessment of dementia burden and its association with COVID-19. Unlike previous studies confined to clinical settings, the GBD framework enables us to capture nationally representative data across multiple regions and socio-demographic index (SDI) levels. It allows us to assess how global economic disruptions, healthcare system strains, and demographic shifts influenced dementia burden during the pandemic. Additionally, the GBD study’s systematic approach to mortality estimation—incorporating death certificate data, verbal autopsy, and predictive modeling—ensures a more robust and standardized evaluation of COVID-19’s impact on dementia mortality compared to studies relying solely on electronic health records or hospital reports. This approach allows us to differentiate between pre-existing trends and pandemic-related disruptions, offering a novel perspective on how the COVID-19 crisis accelerated or altered dementia burden trajectories worldwide.

Methods

Data acquisition

The GBD 2021 study provides detailed assessments of the burden from 371 diseases across 204 countries and territories, along with 811 subnational regions, utilizing the latest population and epidemiological data. The primary analysis of the data has been previously outlined [12]. Data were sourced from the Global Health Data Exchange (GHDx) website using the GBD results tool as per provided instructions. This enabled the extraction of mortality, prevalence, incidence, and DALYs for individuals over 60 across 204 countries, subsequently classified into 56 GBD regions [13]. Additionally, places were classified into areas according to the Socio-Demographic Index (SDI): High, High-middle, Middle, Low-middle, and Low.

Estimation framework

The GBD 2021 study used advanced statistical methods to estimate the burden of Alzheimer’s disease and other dementias worldwide. To ensure consistency across different regions and age groups, DisMod-MR 2.1 combined available data and filled in gaps where information was missing. In areas with limited data, Spatiotemporal Gaussian Process Regression (ST-GPR) helped predict trends over time and across locations, improving accuracy. To correct for differences in study methods and potential biases, Meta-Regression-Bayesian, Regularized, Trimmed (MR-BRT) adjusted the data before final estimates were made. For estimating deaths, the Cause of Death Ensemble Model (CODEm) used multiple forecasting techniques to provide the most reliable mortality figures. Finally, network meta-analysis ensured that dementia burden estimates accounted for differences in age and sex.

Socio-demographic index

Lag-distributed income, average years of education for individuals over 15, and total fertility rates for women under 25 are combined to create the SDI. The SDI ranges from 0 (lowest socioeconomic status) to 1 (highest socioeconomic status), providing a comprehensive measure of socio-economic development and facilitating health status comparisons across different regions over time.

Statistical analysis

To better conduct comparisons across spatial and temporal limits, we calculated age-standardized rate in people aged over 60 years following this formula: $\frac{\sum_{i=1}^{N}{\alpha }_{i}\times {W}_{i}}{\sum_{i=1}^{N}{W}_{i}}$. To estimate the overall trends, Estimated Annual Percentage Change (EAPC) was calculated using the formula $\text{ln}\left({Rate}_{t}\right)={\beta }_{0}+{\beta }_{1}\times {Year}_{t}$, where $\text{ln}{Rate}_{t}$ is the natural logarithm of the rate at year t, ${\beta }_{0}$ is the intercept, and ${\beta }_{1}$ is the slope of the regression line. EAPC is equal to $\left({\text{e}}^{{\beta }_{1}}-1\right)\times 100$, where e is the natural base of logarithm. A linear regression model was used to determine the 95% Confidence Interval (CI). The Joinpoint Regression model was employed to identify significant (p < 0.05) trend deviations across locations with a Monte Carlo Permutation method [14]. The Age-Period-Cohort (APC) model was utilized to analyze variations in trends across different age groups by combining effects of age and time periods. The general form was $\mathit{ln}\left({R}_{{a}_{P}}\right)=\mu +{\alpha }_{a}+{\beta }_{P}+{\gamma }_{c}$, where ${R}_{ap}$ is the rate for age groups ‘a’ and period ‘p’, $\mu$ is the overall mean, ${\alpha }_{a}$ is the age effect, ${\beta }_{P}$ is the period effect, and ${\gamma }_{C}$ is the cohort effect [15]. Net drifts were derived from the model to demonstrate the annual percentage change specific to each age group. The Bayesian Age-Period-Cohort model (BAPC) automatically generates age-specific and age-standardized projections with Poisson noise added Howson [16]. For the first time, excess mortality due to the COVID-19 pandemic was calculated. It was defined by subtracting the estimated death cases from reported figures.

Decomposition analysis, developed by Das Gupta [17], assesses the impacts of three key factors—aging, population growth, and epidemiological changes—on the overall burden of AD and other dementias. This analysis helps to break down how much each factor contributes to changes in disease prevalence over time. Health inequality analysis was conducted to identify and quantify disparities in health outcomes across 204 countries or regions. The slope index and concentration index were used in this analysis to investigate health inequalities in detail among different subgroups.

Additionally, correlations between the age-standardized rate of AD and other dementias and the SDI were assessed using Spearman correlation analysis. This method is particularly effective for data that exhibit non-linear relationships or do not conform to a normal distribution, providing a more accurate reflection of underlying trends.

All computational processes and data mapping were performed using the Joinpoint Regression program (Version 4.8.0.1), StataMP 18, and R statistical packages (Version 4.4.1).

Results

Global level

The global burden of Alzheimer’s disease (AD) and other dementias has increased substantially over the past three decades, particularly among individuals aged 60 years and older. The number of deaths due to these conditions has nearly tripled, rising from about 650,000 in 1990 to nearly 1.92 million in 2021. This increase reflects both population aging and the growing impact of dementia-related risk factors. However, despite this sharp rise in absolute deaths, the age-standardized death rate (ASDR) has remained relatively stable over time, suggesting that the increase is mainly driven by demographic changes rather than a rising individual risk of death (Table 1). The number of new cases has also surged, with around 9 million people diagnosed with AD and other dementias in 2021, compared to about 3.4 million in 1990—a more than 2.5-fold increase. However, the age-standardized incidence rate (ASIR) has remained relatively unchanged, implying that while more people are developing dementia, this is largely due to population aging rather than a heightened disease risk (Table 2). The total number of people living with AD and other dementias has also seen a dramatic rise, more than doubling from about 19.8 million in 1990 to 52.6 million in 2021. This increase highlights the growing prevalence of dementia worldwide, particularly among aging populations. The age-standardized prevalence rate (ASPR) has increased slightly over time, from 4,977 per 100,000 in 1990 to 5,147 per 100,000 in 2021, underscoring the rising global public health burden (Table 3). In terms of overall disease burden, dementia-related disability and premature death have led to a substantial loss of healthy life years, with the total disability-adjusted life years (DALYs) reaching 34.5 million in 2021. This reflects the increasing toll of AD and other dementias on both individuals and healthcare systems.

Table 1 The number of mortality and Age-standardized Death Rates (ASDR) of Alzheimer’s disease(AD) and other dementias in 1990 and 2021 for both sexes by SDI quintiles and by GBD regions, with Estimated Annual Percentage Change (EAPC) from 1990 to 2021

Table 2 The number of incidence and Age-standardized Incidence Rate (ASIR) of AD and other dementias in 1990 and 2021 by SDI quintiles and by GBD regions, with EAPC from 1990 to 2021

Table 3 The number of prevalence and Age-standardized Prevalence Rate (ASPR) of AD and other dementias in 1990 and 2021 by SDI quintiles and by GBD regions, with EAPC from 1990 to 2021

As for gender disparities in disease burden, women experience a significantly higher burden of AD and other dementias compared to men. In 2021, women had an age-standardized death rate of 219 per 100,000, notably higher than the 162 per 100,000 observed in men. A similar pattern is seen in the incidence rate, with women having an ASIR of 937 per 100,000, compared to 756 per 100,000 for men. The gap persists across all disease burden metrics, with women experiencing higher prevalence and more years of life lost due to dementia. This disparity is likely influenced by longer life expectancy, biological differences, and varying exposure to dementia risk factors (Figs. 1, 2, 3 and 4).

Fig. 1

Age-standardized Incidence Rates (ASIR) and Age-standardized Death Rates for Alzheimer's Disease (AD) and other dementias, across global regions, SDI-stratified regions, and all GBD regions, for both males and females

Fig. 2

Age-standardized Prevalence Rate (ASPR) and Age-standardized DALYs Rate of patients with AD and other dementias in Global, SDI-stratified regions, and all GBD regions for both genders, males and females

Fig. 3

Estimated Annual Percentage Change (EAPC) of ASIR and ASDR across global regions, SDI-stratified regions, and all GBD regions, for both males and females

Fig. 4

EAPC of ASPR and DALYs across global regions, SDI-stratified regions, and all GBD regions, for both males and females

Regional level

For SDI-stratified countries or regions, the highest ASIR, ASDR, ASPR, and DALYs were reported in High-middle SDI areas. For both genders, EAPC in High-middle SDI is 0.22 for ASPR (Table 3), and 0.21 for ASIR (Table 2), indicating a substantial increase in prevalence and incidence of the disease during the study period. Although the increase in ASDR was not statistically significant, the burden of disease is still high due to a general positive EAPC for ASPR, ASIR, and DALYs. The greatest burden was observed in East Asia and Central Sub-Saharan Africa, with the highest ASIR, ASPR, and DALY rates. Specifically, ASIR is 1098.81 per 100,000 while ASPR is 6624.21 per 100,000 in East Asia (Tables 2 and 3). In Central-Sub-Saharen Africa, ASDR and DALYs in 2021 were 274.34 per 100,000 and 4513.15 per 100,000 respectively (Tables 3 and 4).

Table 4 The number and age-standardized DALYs of AD and other dementias in 1990 and 2021 by SDI quintiles and by GBD regions, with EAPC from 1990 to 2021

The greatest EAPC for ASIR and ASPR was observed in East Asia, while enormous EAPC for ASDR and DALYs are spotted in South Asia. This phenomenon indicates that southern and eastern parts of Asia have been experiencing an upward trend of the burden of AD and other dementias. The most negative EAPC values were seen in ASIR and DALYs in Australasia, indicating a significant downward trajectory(Tables 2, 3 and 4). Although ASIR, ASDR, ASPR, and DALYs are still high in most of the regions (Figs. 5, 6, 7 and 8), a general decreased trend could be interpreted from negative EAPC values. Our findings exhibit disparities in the progress of the disease burden in different regions. Countries of Low and Low-middle SDI had witnessed a negative EAPC in ASIR and ASPR. However, ASDR and DALYs exhibited positive EAPC values, indicating an increase in prevalence and DALYs (Figs. 1, 2, 3 and 4). These findings might be due to several factors including underreporting and diagnostic challenges. Socioeconomic and environmental factors might contribute to increasing mortality and DALYs (Figs. 9, 10, 11 and 12).

Fig. 5

Global ASIR for both sexes in 2021

Fig. 6

Global ASDR for both sexes in 2021

Fig. 7

Global ASPR for both sexes in 2021

Fig. 8

Global age-standardized DALYs for both sexes in 2021

Fig. 9

EAPC for ASIR for both sexes from 1990 to 2021

Fig. 10

EAPC for ASDR for both sexes from 1990 to 2021

Fig. 11

EAPC for ASPR for both sexes from 1990 to 2021

Fig. 12

EAPC for age-standardized DALYs for both sexes from 1990 to 2021

To further identify the progression of the burden of AD and other dementias, our joinpoints regression models revealed significant variations in the incidence of the disease in different SDI-stratified regions. On a global scale, ASIR had significantly increased from 2019 to 2021, with an APC of 0.93. Similar trends were also observed in High-middle SDI (APC: 2.02), Low-middle SDI (APC: 1.07), as well as Low SDI (APC: 0.54). In contrast, a substantial decrease in ASIR was observed in High SDI countries, with an APC of −0.38. All of these significant changes were started in 2019 and continued throughout 2020 and 2021 (Figs. 13, 14, 15, 16, 17 and 18).

Fig. 13

Joinpoints regression of Global ASIR for AD and other dementias from 1990 to 2021 in both sexes

Fig. 14

Joinpoints regression of ASIR for AD and other dementias in High SDI countries or regions

Fig. 15

Joinpoints regression of ASIR for AD and other dementias in High-middle SDI countries or regions

Fig. 16

Joinpoints regression of ASIR for AD and other dementias in Middle SDI countries or regions

Fig. 17

Joinpoints regression of ASIR for AD and other dementias in Low-middle SDI countries or regions

Fig. 18

Joinpoints regression of ASIR for AD and other dementias in Low SDI countries or regions

For age-specific distribution in people aged 60 years and over, the greatest burden was exhibited in females aged over 95 years. The most significant disparity was seen in the mortality rate, where substantially higher ASDR was observed. Although ASIR and ASPR were high in people aged 60 to 89 were high, DALYs and mortality rates were relatively low, compared with those aged 90 or older (Figure S1.1, Figure S1.2, Figure S1.3, Figure S1.4). The Age-Period-Cohort study revealed age-specific trends on the burden of disease, where positive and negative annual changes represent an increase or decrease in DALYs. For both genders, positive annual changes of DALYs in people aged 60 to 74 were seen on a global scale, while those aged over 75 years exhibited negative annual changes.

However, although the DALYs rate increased in most of the age groups, a staggering decreased trend was reported in people aged over 95 years. Similar patterns were also reported in High SDI countries, where the DALYs rate increased in 60 to 74 years age groups and decreased in those over 75 years for both genders. In High-middle SDI regions, however, consistence increases in DALYs were observed across all the age groups. In contrast, males over 95 years had seen a decrease in DALYs rate. In Low-middle SDI countries, consistent increases in DALY rates were reported across all age groups. Substantially greater annual change was exhibited in males over 95 years old. Low and Middle SDI regions, on the other hand, demonstrated different patterns. Those aged 60 to 84 witnessed a decrease in DALYs, which was represented by a negative annual change. A substantial increase in DALYs was observed for both genders aged over 85 years old. Global and High SDI countries exhibited a decrease in the burden of disease in the older age groups. Specifically, those aged 60 to 69 generally present with an upward trend of ASIR, ASPR, and DALYs. Additionally, gender disparity was seen in those 75-to-89-year-olds in Middle SDI areas. Notably, increase in ASIR, ASPR, and DALYs for High-middle SDI across all the age groups (Figs. 19 and 20). Although the Prevalence and Incidence rate witnessed a decrease among all the age groups in Low and Low-middle SDI regions, mortality and DALYs saw a staggering increase, indicating poor prognosis for patients with AD and other dementias.

Fig. 19

Age-Period-Cohort study for mortality, DALYs, Prevalence, and Incidence in Global, High SDI, and High-middle SDI countries or regions in females, males, and both sexes

Fig. 20

Age-Period-Cohort study for mortality, DALYs, Prevalence, and Incidence in Middle SDI, Low-middle SDI, and Low SDI countries or regions in females, males, and both sexes

3.3 Risk factors induced Alzheimer’s deaths for SDI-stratified regions

Mortality and DALYs rates of AD and other dementia attributable to High body-mass index (BMI) experienced an increase from 1990 to 2021. High SDI areas consistently showed the highest rates of mortality and DALYs over the course of the year. Followed by High-middle, Middle, Low-middle, and Low SDI regions. These two metrics associated with High FPG have also shown a significant increase. High SDI regions again exhibit the highest death and DALY rates, with a steep increase observed in low-middle and middle SDI regions, indicating a growing burden in these areas. The persistent burden in high SDI regions suggests a sustained impact of obesity on dementia-related mortality, likely influenced by comorbid conditions such as cardiovascular disease, and longer life expectancy that allows for prolonged exposure to high BMI-related risks. Moreover, the impact of High BMI on dementia is compounded by lifestyle factors such as physical inactivity and poor dietary habits, which are common in obese individuals. These factors not only contribute to the development of obesity but also independently increase the risk of cognitive impairment and dementia [18].

Regions with low SDI exhibit a gradual although consistent rise over the research period. In contrast to other risk factors, DALYs and mortality attributable to smoking have shown a general decline across all SDI levels, particularly in high and high-middle SDI regions. However, a slight ascending trend is noticeable in high-middle SDI regions from 2020 to 2021, while low SDI regions exhibit a relatively stable trend with minimal decline. For AD attributable to High BMI and High FPG, females had presented with higher mortality and DALYs in males. This disparity is consistence across all the SDI-stratified regions in 2021. On the contrary, Smoking had led to substantially higher mortality and DALYs in males compared with females (Figs. 21, 22, 23 and 24).

Fig. 21

Trends of the number of mortality attributable to High BMI, High FPG, and Smoking in Global, High SDI, High-middle SDI, Middle SDI, Low-middle SDI, and Low SDI countries or regions for both sexes

Fig. 22

Trends of the number of DALYs attributable to High BMI, High FPG, and Smoking in Global, High SDI, High-middle SDI, Middle SDI, Low-middle SDI, and Low SDI countries or regions for both sexes

Fig. 23

Mortality rate of AD and other dementias attributable to High BMI, High FPG, and Smoking in Global, High SDI, High-middle SDI, Middle SDI, Low-middle SDI, and Low SDI countries or regions for both sexes

Fig. 24

DALYs rate of AD and other dementias attributable to High BMI, High FPG, and Smoking in Global, High SDI, High-middle SDI, Middle SDI, Low-middle SDI, and Low SDI countries or regions for both sexes

The number of mortality in both genders caused by different risk factors increased with the passage of years, especially in High FPG-induced AD and other dementias. Among all the elders aged over 60 years, High FPG emerges as a dominant risk factor in regions such as Eastern Europe, Central Asia, the Middle East, and North Africa. In these areas, it accounts for a larger proportion of Alzheimer’s disease deaths and DALYs compared to High BMI and Smoking. In high-income regions like High-income North America and High-income Asia Pacific, High FPG also represents a main factor of the Alzheimer’s disease burden, reflecting the global increase in metabolic disorders. Smoking, while still contributing to Alzheimer’s disease deaths and DALYs, shows a relatively lower proportion compared to High BMI and High FPG in most regions. In regions like Western Europe and High-income North America, the proportion of the burden attributable to smoking is lower, reflecting successful public health measures to reduce smoking prevalence over the years. Our study on Population attributable fraction aligned with previous results, where the contribution of High FPG and High BMI to Alzheimer’s disease has grown over time, underscoring its rising significance as a risk factor for Alzheimer's disease (Figs. 25 and 26), Figure S2, Figure S3).

Fig. 25

Proportion of the number of mortality attributable to High Body Mass Index (BMI), High Fasting Plasma Glucose (FPG), and Smoking across Global, five SDI-stratified regions and all GBD regions for both sexes

Fig. 26

Proportion of the number of DALYs attributable to High BMI, High FPG, and Smoking across Global, five SDI-stratified regions and all GBD regions for both sexes

Future forecasts of the burden of Alzheimer’s disease and other dementias

Globally, the incidence number of AD and other dementias will increase from about 9 million cases in 2021 to over 30 million cases in 2050, which is a nearly four-fold increase (Table S3.2). ASIR had witnessed a staggering increase, from about 879 per 100,000 in 2021 to 1188 per 100,000 in 2050 (Table S3.1). The number of deaths caused by AD and other dementias increased from nearly 2 million in 2021 to 6.6 million mortality in 2050, a three-fold increase. A slightly upward trajectory trend was also observed in ASDR, from approximately 198 per 100,000 to 229 per 100,000 persons over the period (Table S4.1). Detailed projections on different risk factors induced mortality reveal that High FPG and High BMI continue to be leading risk factors for AD (Figs. 27 and 28). Specifically, for High FPG-induced deaths, the number will rise from 287 thousand in 2021 to a staggering 1 million in 2050. ASDR will experience a slight increase, from 29 per 100,000 to about 35 per 100,000 over the period. Although the mortality rate attributable to Smoking experienced a downward trajectory from 1990 to 2021, it will slightly increase from about 6.4 per 100,000 in 2021, to 6.5 per 100,000 in 2050. For ASPR, it will rise from 5 thousand in 2021 to approximately 7 thousand in 2050 for both sexes (Table S5.1). Likewise, elevated FPG remains the primary risk factor for disease burden (Table S6). Similarly, High FPG continues to be the dominant risk factor for the burden of disease. A substantial upward trajectory from over 5 million in 2021 to over 16 million in 2050 was observed, which is another three-fold increase (Table S5.2). Despite the previous decrease trend in Smoking-related AD DALYs was shown, a slight increase in the study period was shown (Figs. 21 and 22). It is also obvious that the gaps between males and females are enlarging, with a faster and more substantial increase in the burden of females (Figure S4.1, Figure S4.2, Figure S4.3, Figure S4.4).

Fig. 27

Future forecasts of the global number of mortality attributable to High BMI, High FPG, and Smoking for both sexes from 1990 to 2050

Fig. 28

Future forecasts of the global mortality rate attributable to High BMI, High FPG, and Smoking for both sexes from 1990 to 2050

Health inequality analysis

The study on the inequality of healthcare revealed disparities in 204 countries with different SDI levels. A positive link exists between the SDI and the crude DALYs rate in both 1990 and 2021, with nations exhibiting higher SDI facing increased burdens of Alzheimer's disease and other dementias (Fig. 29). In 2021, it is estimated that the DALYs rate for the highest SDI level is 1512 digits greater than the lowest, while this differentiation in 1990 is 1242 (Table S9.1). The concentration index for 1990 and 2021 also shows that High SDI countries have a comparatively high disease burden. The concentration index slightly increased from 0.14 (95% CI: 0.13 to 0.16) in 1990 to 0.15 (95%CI: 0.13 to 0.17) in 2021, indicating a slight upward trend of the inequality of health (Table S9.1, Table S9.2).

Fig. 29

Health inequality analysis for 204 countries or regions in 1990 and 2021 for both sexes, including Slope Index and Concentration index

The impact of the COVID-19 pandemic on Alzheimer’s disease and other dementias

In the wake of the global pandemic, the differentiation in the number of deaths varies across different countries, regions, and age groups. The graphs present the excess mortality number or rate in different groups of people in 2020 and 2021. On a global scale, the reported number of mortality was over 5 thousand less than the expected value for both sexes in 2020 (Table S11.1). This trend is consistent in other SDI-stratified regions and both genders. However, disparities occurred in 2021, where global reported deaths were a substantial 15 thousand greater than the projected mortality number(Table S11.1). Middle- and High-Middle SDI nations also reported seeing this. In contrast, High SDI, Low SDI, and Low-middle SDI countries exhibited negative excess mortality in 202 1(Table S11.1). Detailed analysis of age-specific excess mortality rate demonstrated no significant changes in people aged 60 to 79. In 2020, the reported global mortality rate in males was less than the expected value, but the actual death toll was greater in females (Table S10.2, Table S10.3). In High-middle and Low SDI countries, males and females aged 80 to 89 years old exhibited a negative excess mortality rate, while those aged over 90 years old witnessed substantially greater reported deaths than the projection (Table S10.2, Table S10.3). Notably, actual mortality in males aged over 95 years in 2020 was nearly 100 units lower than the expected value (Table S10.3). No significant differences were seen in females of the same age group in Low and High-middle SDI regions in 2020 (Table S10.2). When it comes to 2021, a positive excess mortality rate in all age groups of both genders was exhibited. In High SDI countries, males aged over 90 years exhibited substantially lower reported deaths than expected values (Table S10.3). In contrast, females in the same age group in High SDI areas had a higher actual mortality than the projected one (Table S10.2). High-middle and Middle SDI regions demonstrated positive excess mortality for males and females (Table S10.2, Table S10.3). Additionally, the number increases with age. Both males and females in Middle SDI regions demonstrated consistence positive excess mortality across all the age groups (Fig. 30). Gender diparities in the disease burden,

Fig. 30

Age-specific excess mortality for both sexes, females, and males in Global, High SDI, High-middle SDI, Middle SDI, Low-middle SDI, and Low SDI countries or regions

Decomposition analysis

To better understand current factors affecting the burden of disease, the decomposition analysis on aging, population, and epidemiological changes focused on the period from 2019 to 2021. On a global scale, population growth contributes the most to the overall increase in burden. It results in 444,375.11 (68.32%) incidence cases (Table S13.1) and 29,035.20 (80.99%) death toll (Table S12.1). Females presented with a staggering higher overall increase in the burden of disease than males. Specifically, the overall difference in incidence for females is 414,735.94 and 235,700.07 for males (Table S13.3, Table S13.5), while the overall increase in DALYs for females is 1,524,593.13 and 769,722.13 for males. Similarly, population growth has played a major role in the growing burden of disease. In Low-middle SDI, and Low SDI countries, it contributes to 3102.02 (181.1%) and 2750.07 (159.34%) deaths in the overall burden (Table S12.1). This result is consistent across other SDI-stratified countries (Table S12.1). Notably, Aging has led to a decreased burden of AD and other dementias in Low and Low-middle SDI areas. −1637.97 (−21.68%) and −938.41 (−54.37%) deaths respectively (Table S12.1). Decreased in incidence, prevalence, and DALYs were also presented in the same regions (Table S13.1, Table S14.1, Table S15.1). From 2019 to 2021, epidemiological shifts adversely impacted the disease burden in High SDI countries. Approximately −23,357.11(−22.8%) new cases of AD and −157,622.04 (−27.03%) for the number of prevalence (Table S13.1, Table S14.1). It also had a substantial negative impact on DALYs and mortality in both genders, which resulted in −38,534.95 (−7.55%) DALYs and −1170.33 (−3.26%) death cases (Table S12.1, Table S15.1). Gender disparities were also presented in our decomposition analysis. The effects of aging, population growth, and epidemiological change on females were greater than that in males (Fig. 31).

Fig. 31

Decomposition analysis for Incidence, Mortality, Prevalence, and DALYs for Both sexes, females, and males in Global, High SDI, High-middle SDI, Middle SDI, Low-middle SDI, and Low SDI countries or regions from 2019 to 2021

Discussion

This study highlights significant regional disparities in the burden of AD and other dementias. Countries with a High SDI typically have a greater prevalence of AD and other forms of dementia (Fig. 32, Figure S5.1, Figure S5.2, Figure S5.3). While generally protective against dementia, it can also lead to increased diagnosis rates due to better access to healthcare and diagnostic services [19]. Middle- to high-SDI regions reported the highest ASIR, ASPR, and DALYs, with rapid increases notably seen in East Asia and Central Sub-Saharan Africa [20]. In contrast, Australasia displayed a decline in dementia burden, likely due to a well-developed healthcare system effectively managing risk factors like hypertension, diabetes, and hypercholesterolemia [21].

Fig. 32

Spearman correlation analysis between SDI and ASIR among all GBD regions for both sexes

Further analysis using joinpoint regression reveals a significant upward trend in ASIR globally from 2019 to 2021, particularly in high-middle, middle, low-middle, and low-SDI countries. Conversely, a significant decrease was observed in high-SDI countries, likely due to better pandemic adaptations and robust healthcare responses that mitigated some impacts on dementia incidence [22]. These findings underscore the importance of healthcare infrastructure and targeted strategies in managing regional dementia trends.

Demographic analysis shows that the dementia burden is particularly pronounced among individuals aged 60 and older, with the highest DALY increase observed in those aged 90 and above, reflecting the vulnerability of older age groups to cognitive decline [23]. During the pandemic, the resilience of younger elderly populations (aged 60–79) was evident, with relatively lower excess mortality in this group. However, older adults in low-SDI countries faced challenges in following preventive measures, such as mask-wearing and social distancing, due to limited resources, contributing to higher mortality in 2020 [24]. Improvements in healthcare awareness and resources by 2021 may have helped to mitigate these risks, especially in high-SDI countries that quickly adapted telemedicine to maintain continuous care while minimizing infection risks [25].

Gender disparities were evident, with women bearing a significantly greater burden of AD and other dementias compared to men. This gender difference may be partly attributed to the decline in estrogen levels post-menopause, which reduces neuroprotection, and to the higher prevalence of tau and amyloid proteins in women, both of which are associated with AD pathology [26]. These findings underscore the need for gender-specific preventive strategies, with an emphasis on early interventions for women who may be at higher risk due to hormonal and neuropathological factors. During the COVID-19 pandemic, the impact of AD and other dementias on women was more pronounced, aligning with prior research that highlights their greater vulnerability to these conditions. The pandemic may have exacerbated this burden, as women faced increased socio-economic and psychosocial stressors, such as higher rates of employment loss, greater responsibility for caregiving, and heightened levels of gender-based violence [27]. These socio-economic and psychosocial stressors likely contributed to adverse outcomes for women with AD and other dementias during the pandemic, underscoring the need for targeted interventions to mitigate these compounded vulnerabilities.

Lifestyle factors, particularly high FPG, high BMI, and smoking, were identified as major contributors to dementia risk. High FPG, commonly associated with diabetes, is linked to vascular and metabolic effects that exacerbate cognitive decline. A countrywide cohort study in South Korea revealed that mild hyperglycemia and comorbidities are linked to a heightened risk of developing dementia, underscoring the necessity for early care during prediabetic phases [28]. Current management strategies for high FPG to lower the chance of AD and other dementias focus on optimizing glycemic control and utilizing specific antidiabetic medications with potential neuroprotective effects [1]. However, these devices are frequently scarce in regions with lower SDI like Eastern Europe, Central Asia, Middle East, and North Africa, with only a limited number of people having access to them [29]. High BMI, often accompanied by inflammation and metabolic stress, also poses significant risks, particularly in high-SDI regions where these health issues are more prevalent [30, 31]. Smoking remains another critical risk factor, especially among men, reinforcing the importance of anti-smoking policies as a preventative measure against AD [32, 33].

Targeted public health interventions have shown promise in alleviating dementia symptoms and slowing progression, particularly in low-SDI regions. For example, Cognitive Stimulation Therapy (CST) has been effective in enhancing cognition and quality of life for patients in resource-limited settings [34]. Diagnostic improvements and increased public health awareness in middle and high-SDI countries have led to more accurate dementia detection, underscoring the need for accessible diagnostic tools and early intervention across all SDI levels [35].

Our study highlights the importance of addressing modifiable risk factors, improving healthcare access, and strengthening health systems to mitigate the growing burden of dementia. The rising dementia burden is driven by a combination of demographic, metabolic, and healthcare-related factors. High FPG and BMI are among the leading risk factors, contributing to vascular damage and neurodegeneration. These risk factors are particularly prevalent in high-SDI and high-middle SDI regions, where sedentary lifestyles, poor diets, and rising diabetes rates have intensified the risk of dementia. However, a potential issue in high-SDI regions is overmedicalization, where there is a tendency to overuse medical interventions, which may not always be beneficial and can lead to unnecessary treatments and increased healthcare costs. This can be particularly problematic in dementia care, where the focus should ideally be on holistic and patient-centered approaches rather than excessive pharmacological interventions [36]. Thus, for High and High-middle SDI regions, the focus should be on addressing lifestyle-related risk factors, as well as proactive outpatient care and care coordination to reduce preventable hospitalizations and improve overall care quality. Implementing annual cognitive assessments for individuals aged over 60, and leveraging AI-based diagnostic tools is also useful.

In Low-middle and Low SDI countries, the high mortality and DALYs associated with AD stem from limited healthcare access, underdiagnosis, and lack of dementia-specific policies. This mainly caused by uneven distribution of healthcare resources and insufficient training for medical personnel [37]. As a result, the focus should be on improving diagnostic capacity and basic dementia care services. Strengthen primary healthcare systems by training community health workers to recognize early dementia symptoms and provide referrals is crucial. Integration of dementia care into infectious disease programs and leverages of existing healthcare infrastructure for diseases like HIV and malaria to introduce dementia screening and care could be helpful. Establishing community dementia support groups to provide psychological support and platforms for information exchange would also be highly beneficial.

The COVID-19 pandemic introduced complex challenges to dementia care. Public health measures aimed at mitigating respiratory infections temporarily decreased dementia-related mortality in 2020, particularly in high-SDI areas with strong healthcare systems [38]. However, by 2021, disparities in remote healthcare access contributed to differing mortality trends across regions, underscoring the importance of robust healthcare infrastructure [24]. Long COVID adds further complications to dementia care, with persistent inflammation, vascular issues, and cognitive effects that may accelerate dementia progression, increasing the burden on older populations and those with pre-existing cognitive impairments [38]. These challenges highlight the need for resilient healthcare systems capable of addressing both traditional dementia risks and emerging pandemic-related effects.

Moving forward, global dementia management will require multi-level collaboration, prioritizing both lifestyle interventions and enhanced access to diagnostic and care resources. By doing so, healthcare systems can better support dementia patients, building a sustainable model for managing complex, chronic conditions in a highly interconnected world. The most significant driver of dementia prevalence is the increasing global population aged 60 years and older. Aging alone accounts for a large proportion of the disease burden, particularly in high-SDI countries where life expectancy is longer.

Middle SDI countries faced consistent positive excess mortality, with reported deaths exceeding expectation. In contrast, high and low SDI regions generally experienced fewer deaths than projected. The pandemic overburdened healthcare systems, leading to delayed or missed dementia diagnoses, particularly in low- and middle-SDI countries, where screening programs were paused due to COVID-19 response priorities. Therefore, integrating long COVID monitoring into dementia care to assess potential neurological consequences in essential in High SDI regions. In Low and Low-middle SDI regions, providing dementia care in primary healthcare systems, particularly in rural areas, is crucial. Moving forward, healthcare systems must adapt by prioritizing dementia care resilience, integrating telemedicine, cognitive rehabilitation, and mental health support, and ensuring targeted prevention strategies that reflect the unique challenges of each SDI level. Without strategic, regionally tailored interventions, the pandemic’s long-term impact on dementia burden could exacerbate health inequalities and overwhelm healthcare systems globally.

This study has several limitations. First, reliance on GBD 2021 data means that any limitations within these datasets affect the study’s accuracy. The focus on individuals aged 60 and older may exclude important insights into dementia burden in younger populations. Additionally, the dataset did not include sex minorities, and the reliance on ICD-10 codes excludes some neurodegenerative conditions, potentially underestimating the total burden of AD. Another critical limitation is that our analysis combines data for Alzheimer’s disease and other dementias without distinction, limiting our ability to assess differences in trends, causes, and burdens between AD and other types of dementia. Finally, GBD 2021 depends on the 2017 literature review, possibly omitting more recent findings [34], it may not reflect the most current understanding of dementia. This is particularly relevant given the rapid pace of research in this field, and future studies should incorporate more recent data when available.

Conclusion

Aligned with previous studies, our study highlights the increasing global burden of Alzheimer’s disease and other dementias, primarily driven by population aging and modifiable metabolic risk factors such as high fasting plasma glucose and high BMI, with significant disparities across socioeconomic regions. While age-standardized incidence and mortality rates have declined, absolute cases and deaths continue to rise, exacerbated by the COVID-19 pandemic, particularly in middle-SDI countries. To mitigate this burden, High SDI and High middle SDI regions should prioritize metabolic health programs, routine cognitive assessments, and improved outpatient dementia care, while Low- and Low-middle SDI regions require strengthened diagnostic capacity, integration of dementia screening into primary healthcare, and community-based support initiatives. Addressing gender disparities through targeted interventions for women and adapting healthcare systems for long COVID’s cognitive effects are also crucial. Without region-specific strategies, dementia will continue to strain healthcare systems, reinforcing the need for comprehensive, multi-sectoral action.