Temporal trend and attributable risk factors of Alzheimer's disease and other dementias burden in China: Findings from the Global Burden of Disease Study 2021

Abstract

INTRODUCTION

We examined the burden of Alzheimer's disease and other dementias (ADOD) and attributable factors at the national and provincial levels in China.

METHODS

Using the Global Burden of Diseases Study 2021, we estimated incidence, prevalence, mortality rate, disability‐adjusted life years (DALYs), and the ratio of years lived with disability (YLD) to DALYs for ADOD in China. Estimated annual percentage changes (EAPCs) were used to quantify the temporal trends from 1990 to 2021.

RESULTS

In 2021, China experienced the highest ADOD burden among Group of 20 member nations. The EAPCs for age‐standardized rates for incidence, age‐standardized rates for prevalence, and age‐standardized mortality rate were 0.41 (uncertainty intervals [UIs] 0.34–0.49); 0.44 (UI: 0.36–0.52); and –0.19 (UI: ‐0.23 to ‐0.15), respectively. Between 1990 and 2021, the number of people with ADOD increased by 322.18% and DALYs associated with ADOD increased by 272.71%; most of these increases were explained by population aging.

DISCUSSION

Considering the aging Chinese population, targeted strategies to prevent dementia are urgently needed.

Highlights

• China experienced the highest dementia burden among Group of 20 member nations.

• High body mass index, high fasting plasma glucose, and smoking were major risk factors for Alzheimer's disease and other dementias (ADOD) burden.

• Since 1990, the incidence and prevalence of ADOD increased substantially in China.

• The mortality rate related to ADOD decreased consistently.

• Considering the aging Chinese population, targeted strategies are urgently needed.

1. BACKGROUND

Dementia is a growing and challenging worldwide public health issue, and two thirds of affected individuals are living in low‐income and middle‐income countries. ¹ China has the world's largest population of patients with dementia, ² and Alzheimer's disease and other dementias (ADOD) is the fifth leading cause of death in China. ³ From 2000 to 2020, the percentage of the population ≥ 65 years of age doubled from 7% to 14%, positioning China as among the fastest‐aging countries globally, largely due to declining birth rates and increasing life expectancy. ⁴ Recent data from the Seventh National Population Census indicate that 18.7% of the Chinese population is aged ≥ 60 years, and 13.5% is aged ≥ 65 years, totaling 267 million people. ⁵ This number is projected to exceed 400 million by 2034, at which time individuals aged ≥ 65 years would account for 30% of the total population. ⁶ China's aging population faces substantial challenges, including ADOD, which is expected to require urgent governmental intervention. ⁷

The initial report for the Global Burden of Diseases Study 2021 (GBD 2021) concluded that the increase in disability‐adjusted life years (DALYs) was largely due to population aging and growth. ⁸ Several studies stressed that population aging increases the burden of dementia, ⁹ , ¹⁰ , ¹¹ but an updated comprehensive decomposition analysis of temporal trends and attributable risk factors of ADOD burden at the national and subnational levels is needed to guide disease intervention and improve public health. ¹²

Covering the period 1990 to 2021, GBD 2021 integrated literature research, monitoring data and investigative information, inpatient and outpatient medical insurance data, and other information to evaluate the incidence, prevalence, mortality, and DALYs of 371 diseases and injuries by age and sex across 204 countries and territories. ¹³ In the present study, based on data from GBD 2021, we analyzed age‐ and sex‐stratified epidemiological trends of ADOD from 1990 to 2021 in China in terms of number of cases, rate, and age‐standardized rate. We also quantified the ADOD burden attributable to various risk factors at the national and provincial levels in China over the same period.

2. METHODS

2.1. Overview

The GBD 2021 assessed disease burden by cause, age, sex, year, and location, allowing direct comparisons among different populations, time periods, and regions. Details of the study design and methods were described previously based on the GBD 2021 ¹³ and can also be accessed at https://ghdx.healthdata.org/gbd‐2021. The present study uses the GBD 2021 to determine overall disease burden of ADOD and corresponding attributable risk factors in China from 1990 to 2021, on the basis of incidence, mortality rate, prevalence, DALYs, years lived with disability (YLD), and years of life lost (YLL). This study followed the Guidelines for Accurate and Transparent Health Estimates Reporting (GATHER) ¹⁴ and was reviewed and approved by the National Center for Chronic and Non‐Communicable Disease Control and Prevention of the Chinese Center for Disease Control and Prevention. No individual identifiable information was used.

2.2. Definition of ADOD

ADOD was defined based on the 3rd, 4th, or 5th editions of the Diagnostic and Statistical Manual of Mental Disorders or International Classification of Diseases (ICD) case criteria. The diagnosis was accomplished using clinical records, algorithm criteria, National Institute on Aging Alzheimer's disease criteria, 10/66 algorithm criteria, and general practitioner records. The ICD codes for ADOD are as follows: 290, 291.2, 291.8, 294, and 331 for the 9th revision (ICD‐9) and F00, F01, F02, F03, G30, and G31 for the 10th revision (ICD‐10). ¹⁵

2.3. Measures

The cause‐specific mortality rate was estimated using the Cause of Death Ensemble model, and alternative strategies were used to model causes for scenarios for which there were sparse data, substantial changes in reporting over the study period, or unusual epidemiology. ¹³ Mortality rates in municipalities of China were estimated using data from the Chinese Center for Disease Control and Prevention's death reporting system, national disease surveillance points system, and maternal and child health surveillance websites. Prevalence is the number of actual existing cases, and incidence is defined as the number of new cases of dementia. Epidemiological data with known biases were adjusted with correction factors estimated by network meta‐regressions. ⁸ Data on non‐fatal outcomes of ADOD were collected from population‐based surveys and census data, as well as other published and unpublished studies and reports (S1 in supporting information). DALY was calculated as the sum of YLD and YLL and was taken as a summary of the effects of dementia on both the quantity of life (premature mortality) and the quality of life (disability). YLL was calculated as the product of estimated age‐/sex‐/location‐/year‐specific deaths and the standard life expectancy at the age of death for a given cause. ¹³ YLD was calculated with a micro‐simulation process that used estimated age‐/sex‐/location‐/year‐specific prevalence counts of non‐fatal disease sequelae for each cause and disability weights for each sequela as the inputs. ⁸ To estimate the efficiency of the health‐care system in preventing death caused by ADOD, the ratio of YLD and DALYs was calculated. ¹⁶ , ¹⁷

RESEARCH IN CONTEXT

1. Systematic review: We conducted a search of online databases, including Medline, Scopus, PsycINFO, ProQuest, Embase, and official websites of the Chinese government, to identify original research articles that report on the disease burden and attributable risk factors, including population aging and growth, for Alzheimer's disease and other dementias (ADOD). Our search terms included “Alzheimer's,” “dementia,” “Global Burden of Disease,” “population aging,” “decomposition,” “incidence,” “mortality,” “disability‐adjusted life years,” and “disease burden,” with no restrictions on publication date or language, up to May 1, 2024. Previous studies highlighted that dementia was emerging as a major health concern both globally and particularly in China. However, despite this growing awareness, there remains a paucity of research that directly compares the temporal trends of dementia in China to those in Group of 20 countries, especially in terms of how and to what degree population aging influences the burden of dementia.

2. Interpretation: We used Global Burden of Diseases Study 2021 data, which incorporated major improvements in terms of expanded data sources, enhanced data accuracy, and improved model estimates, to obtain timely information on the incidence, prevalence, mortality, disability‐adjusted life years (DALYs), years lived with disability, years of life lost, and the main risk factors of ADOD at the national, regional, and provincial levels in China, thus providing a deeper understanding of the disease burden and epidemiological trends of ADOD in the Chinese population. Distinct from previously reported GBD studies, this dementia‐related study quantified disease burden and trends for dementia that are attributable to population aging.

3. Future directions: Despite the decrease in age‐standardized mortality rate, the overall burden of ADOD in China has intensified, as evidenced by rising incidence, prevalence, and DALYs, and this is driven predominantly by population aging. Considering identified modifiable risk factors, including high body mass index, elevated fasting plasma glucose, and smoking habits, it is imperative to adopt geographically tailored public health interventions that promote lifestyle modifications along with the implementation of comprehensive care models and early diagnostic services. This dual approach not only would cater to the specific needs of the aging population in China but also could serve as a model for other regions experiencing similar demographic and health transitions.

2.4. Risk factor assessment

The GBD 2021 used a comparative risk assessment (CRA) method to calculate the proportion of ADOD attributable to a series of risk factors. ¹⁵ Both attributable rate and attributable age‐standardized rate of mortality, DALYs, YLD, and YLL were estimated for high body mass index (BMI), high fasting plasma glucose, and smoking, and the three risk factors combined were estimated according to CRA. Population‐attributable fractions (PAFs) were also calculated using relative risk data, exposure data, and a theoretical minimum level of exposure, which is the level at which the risk of health outcomes is lowest. The attributable burden was calculated by multiplying the cause measure in question by the PAF. Calculations for “all risk factors” represented the proportion of disease burden attributable to all risk factors, as quantified by the total PAF multiplied by the DALY or deaths associated with the ADOD. ¹⁸

2.5. Data analysis

Age‐standardized rates (ASRs; rates per 100,000) for incidence (ASIR), prevalence (ASPR), mortality (ASMR), and DALYs were estimated using a global age structure from 2021, stratified by year, sex, age, and municipality levels. Estimates were compared to the global average and corresponding estimates for Group of 20 (G20) nations, excluding the European Union. Temporal trends in ADOD burden from 1990 to 2021 were examined as the estimated annual percentage change (EAPC) of ASRs, which are commonly used to reflect the variation of ASRs over a specified interval. The point estimates and 95% confidence intervals (CIs) of these indicators were calculated from the mean of 1000 draw values using the GBD 2021 modeling process. We also present 95% uncertainty intervals (UIs) for every metric based on the 2.5th and 97.5th ordered values out of 1000 draws of the posterior distribution. EAPC > 0 and 95% UI > 0 indicated an increasing ASR trend, whereas EAPC < 0 and 95% UI < 0 indicated a decreasing ASR trend. Otherwise, ASRs were regarded as stable over time.

Using a decomposition analysis, we estimated changes in the number of patients between 1990 and 2021 that are attributable to population aging, population growth, exposure to ADOD risk factors, and unexplained attribution. ¹⁸ , ¹⁹ We calculated the contributions of each component based on formulas that consider age‐specific mortality rates and population proportions for different age groups across two time points. This method is not influenced by the order of decomposition or by the choice of reference group, allowing estimation of the impact of population aging on disease burden. Output from this method can be compared across countries and across causes of death to assess the health impact of population aging. All analyses were performed with R software (version 4.3.2).

3. RESULTS

In 2021, an estimated 16.99 million individuals in China were living with ADOD (95% UI: 14.49–19.67 million). The number of new cases was ≈ 2.91 million (95% UI: 2.50–3.35 million), with deaths attributed to these conditions totaling ≈ 0.49 million (95% UI: 0.12–1.33 million). Total DALYs was estimated at 10.07 million (95% UI: 4.95–22.21 million; Table 1). The proportion of individuals aged ≥ 65 years increased annually, with this age group accounting for > 75% of the total number of these measures each year (Table S2 and Figures S1–S3 in supporting information). From 1990 to 2021, the age‐standardized incidence and prevalence of ADOD generally increased, whereas the ASMR and the age‐standardized DALYs each showed steady declines, except for a spike after 2020 (Figure 1). In 2021, for females, all‐age number and age‐adjusted rates of all measures were higher than for males (Figure 1 and Figure S4 in supporting information).

TABLE 1.

All‐age number and age‐standardized rates of all measures for ADOD and percentage changes by sex in China, 1990 and 2021.

	All‐age number in thousands (95% UI)			Age‐standardized rate, per 100,000 (95% UI)
Measures	1990	2021	Change, %	1990	2021	Change, %
Incidence
Total	703.18 (601.51, 808.63)	2914.11 (2504.73, 3350.74)	314.42 (297.97, 33164)	121.11 (105.5, 137.99)	151.47 (131.22, 173.34)	25.07 (21.27, 28.17)
Female	442.53 (381.57, 507.19)	1836.81 (1593.65, 2101.34)	315.07 (299.99, 332.19)	135.42 (118.39, 154.07)	171.81 (150.12, 195.9)	26.87 (23.21, 29.95)
Male	260.65 (222.34, 301.79)	1077.3 (908.45, 1248.19)	313.31 (291.97, 331.63)	100.25 (86.5, 115.16)	126.48 (107.78, 145.62)	26.17 (21.63, 29.56)
Prevalence
Total	4024.54 (3446.4, 4623.09)	16990.83 (14488.49, 19672.74)	322.18 (304.73, 338.68)	703.14 (608.36, 809.51)	900.82 (770.92, 1043.22)	28.11 (24.47, 31.11)
Female	2512.93 (2165.05, 2892.72)	10828.63 (9315.74, 12515.96)	330.92 (312.6, 348.66)	785.19 (681.22, 900.41)	1025.11 (879.04, 1186.81)	30.56 (26.49, 33.89)
Male	1511.6 (1280.69, 1737.52)	6162.2 (5142.29, 7141.8)	307.66 (285.64, 328.03)	574.55 (493.64, 666.55)	731.21 (618.54, 851.63)	27.27 (22.21, 30.75)
Deaths
Total	119.81 (28.35, 322.1)	491.77 (124.97, 1330.18)	310.47 (248.01, 403.84)	31.39 (7.6, 83.63)	30.82 (7.88, 82.43)	−1.8 (−15.92, 18.32)
Female	80.21 (19.18, 212.44)	328.43 (83.71, 862.46)	309.45 (226.31, 428.17)	34.61 (8.32, 90.6)	33.8 (8.6, 87.19)	−2.34 (−20.58, 23.22)
Male	39.6 (9.25, 113.67)	163.34 (40.66, 466.66)	312.52 (222.53, 428.22)	25.12 (6.70.58)	25.9 (6.51, 73.2)	3.08 (−14.8, 26.19)
YLLs
Total	1894.03 (432.46, 5188)	6612.15 (1661.58, 18403.6)	249.11 (193.27, 327.43)	388.69 (93.2, 1021.16)	376.75 (96.26, 1033.14)	−3.07 (−17.42, 17.64)
Female	1207.92 (280.41, 3189.05)	4224.65 (1064.65, 11432.7)	249.75 (174.2, 356.44)	430.1 (103.45, 1125.07)	415 (105.04, 1115.53)	−3.51 (−22.32, 22.94)
Male	686.11 (161.86, 2019.59)	2387.51 (592.42, 7003.09)	247.98 (165.76, 351.04)	316.67 (74.59, 872.59)	319.98 (79.26, 898.94)	1.04 (−18.56, 27.49)
YLDs
Total	808.46 (545.68, 1082.81)	3460.32 (2394.27, 4632.17)	328.02 (308.16, 346.81)	145.78 (99.22, 193.23)	185.63 (127.98, 246.72)	27.34 (23.93, 30.44)
Female	521.77 (354.16, 698.5)	2275.55 (1555.76, 3049.87)	336.12 (316.78, 354.98)	166.61 (114.06, 222.28)	216.38 (147.99, 288.88)	29.87 (25.94, 33.04)
Male	286.69 (194.75, 376.88)	1184.77 (818.13, 1591.26)	313.26 (292.17, 334.86)	113.18 (77.45, 149.5)	143.7 (99.59, 192.13)	26.96 (21.93, 30.56)
DALYs
Total	2702.48 (1239.18, 6085.39)	10072.48 (4947.15, 22219.15)	272.71 (222.22, 330.14)	534.47 (236.2, 1190.6)	562.39 (271.16, 1238.81)	5.22 (−9.08, 21.94)
Female	1729.68 (790.23, 3750.29)	6500.2 (3171.76, 13681.03)	275.8 (210.42, 349.59)	596.71 (265.21, 1288.3)	631.38 (305.95, 1318.24)	5.81 (−12.64, 26.27)
Male	972.8 (434.09, 2307.85)	3572.28 (1694.72, 8148.48)	267.22 (207.14, 336.03)	429.85 (186.23, 998.45)	463.67 (214.26, 1055.84)	7.87 (−7.06, 27.42)

Abbreviations: ADOD, Alzheimer's disease and other dementias; DALYs, disability‐adjusted life years; UI, uncertainty interval; YLDs, years lived with disability; YLLs, years of life lost.

FIGURE 1.

Trend of age‐standardized incidence, prevalence, mortality rates, DALYs, YLL, and YLD for ADOD by sex in China from 1990 to 2021. ADOD, Alzheimer's disease and other dementias; DALYs, disability‐adjusted life years; YLD, years lived with disability; YLL, years of life lost.

The ASIR of ADOD in 2021 in China was 151.4 (95% UI: 131.22–173.34) per 100,000, the ASPR was 900.82 (95% UI: 770.92–1043.22) per 100,000, the ASMR was 30.82 (95% UI: 7.88–82.43) per 100,000, and the age‐standardized DALYs was 562.39 (95% UI: 271.16–1238.81) per 100,000. Each measure of ADOD burden in China was higher than that reported for G20 member countries and higher than the global average (Table 2). For instance, China's ASPR exceeded the global average of 694.01 (95% UI: 602.88–794.08) per 100,000 and the G20 average of 717.89 (95% UI: 622.67–823.38) per 100,000 (Table 2). The age‐standardized DALYs rate and the ASMR in China were twice as high as those of Mexico (296.84 and 14.8 per 100,000, respectively). Similarly, the ASIR and ASPR were twice as high as those in the Republic of India (78.92 and 16.98 per 100,000, respectively; Tables S3–S6 in supporting information). From 1990 to 2021, the ASPR and ASIR of ADOD in China increased by 28.11% (EAPC: 0.44%, 95% CI: 0.36–0.52) and 25.07% (EAPC: 0.41%, 95% CI: 0.34–0.49), representing the highest increase among the G20 countries and greatly exceeding the global average increase. The decrease in the ASMR of ADOD in China since 1990 was 1.8% (EAPC: –0.19%, 95% CI: –0.23 to –0.15). The 5.22% increase in age‐standardized DALY rate was not significant (EAPC: –0.02%, 95% CI: –0.06 to 0.02; Table 2).

TABLE 2.

Disease burden of ADOD in 2021, and their percentage changes and EAPC from 1990 to 2021, 19 member countries of the G20 (the 20th member is the European Union) and the world.

	DALY rate			Prevalence			Incidence			Mortality			Ratio of YLDs to DALYs
Countries	ASR, per 100,000, 2021	Percentage change, %, 1990–2021	EAPC, %, 1990–2021	ASR, per 100,000, 2021	Percentage change, %, 1990–2021	EAPC, %, 1990–2021	ASR, per 100,000, 2021	Percentage change, %, 1990–2021	EAPC, %, 1990–2021	ASR, per 100,000, 2021	Percentage change, %, 1990–2021	EAPC, %, 1990–2021	Percent, 2021	Percent change, 1990–2021	EAPC, %, 1990–2021
China	562.39 (271.16 to 1238.81)	5.22 (−9.08 to 21.94)	‐0.02 (−0.06 to 0.02)	900.82 (770.92 to 1043.22)	28.11 (24.47 to 31.11)*	0.44 (0.36 to 0.52)*	151.47 (131.22 to 173.34)	25.07 (21.27 to 28.17)*	0.41 (0.34 to 0.49)*	30.82 (7.88 to 82.43)	‐1.8 (−15.92 to 18.32)	‐0.19 (−0.23 to −0.15)*	33.01	21.02	0.44 (0.38 to 0.5)*
United States of America	509.74 (240.09 to 1068.07)	‐3.6 (−5.49 to −1.94)*	‐0.15 (−0.16 to −0.13)*	773.07 (670.26 to 887.66)	‐4.47 (−6.24 to −3.1)*	‐0.17 (−0.19 to −0.16)*	131.29 (113.93 to 149.6)	‐5.28 (−6.93 to −4.07)*	‐0.19 (−0.2 to −0.17)*	29.17 (7.74 to 73.52)	‐1.58 (−4.27 to 1.64)	‐0.08 (−0.09 to −0.06)*	30.23	−3.73	‐0.12 (−0.13 to −0.11)*
Germany	508.64 (250.83 to 1028.42)	‐0.45 (−5.69 to 5.53)	‐0.05 (−0.08 to −0.02)*	820.5 (713.82 to 930.77)	1.35 (−4.79 to 8.46)	0.06 (0.03 to 0.08)*	142.13 (124.57 to 159.72)	‐2.06 (−7.72 to 4.27)	‐0.1 (−0.13 to −0.07)*	29.25 (7.91 to 69.29)	‐0.84 (−7.54 to 8.31)	‐0.07 (−0.12 to −0.03)*	33.18	0.15	0.04 (0 to 0.08)
Italy	507.4 (243.87 to 1036.59)	‐1.79 (−8.13 to 6.91)	‐0.03 (−0.06 to −0.01)*	759.15 (652.54 to 869.83)	9.55 (4.52 to 15.49)*	0.13 (−0.01 to 0.27)	134.76 (116.74 to 153.65)	15.88 (11.57 to 21.06)*	0.28 (0.09 to 0.46)*	29.75 (8.1 to 72.88)	‐5.56 (−10.26 to 0.57)	‐0.09 (−0.13 to −0.05)*	30.97	13.11	0.21 (0.09 to 0.34)*
Brazil	503.99 (236.06 to 1073.53)	‐1.56 (−4.9 to 1.6)	‐0.06 (−0.07 to −0.05)*	761.45 (661.47 to 869.62)	0.23 (−2.03 to 2.58)	‐0.05 (−0.09 to −0.01)*	127.08 (112.01 to 144.66)	‐1.75 (−3.72 to 0.3)	‐0.11 (−0.15 to −0.08)*	27.29 (7.15 to 69.06)	‐3.08 (−7.13 to 1.7)	‐0.07 (−0.09 to −0.06)*	30.33	1.74	0.01 (−0.03 to 0.04)
Republic of Korea	500.05 (234.92 to 1004.38)	‐13.88 (−22.74 to −2.27)*	‐0.36 (−0.4 to −0.31)*	739.81 (647.7 to 836.31)	‐1.49 (−7.39 to 4.26)	0.01 (−0.04 to 0.06)	124.63 (109.44 to 140.89)	‐2.41 (−8.06 to 3.06)	0.01 (−0.04 to 0.06)	29.02 (7.87 to 69.71)	‐18.69 (−27.21 to −4.49)*	‐0.53 (−0.59 to −0.47)*	30.55	13.7	0.35 (0.28 to 0.41)*
Turkey	491.99 (236.53 to 1031.11)	‐10.79 (−20.51 to −0.8)*	‐0.4 (−0.47 to −0.32)*	819.36 (706.2 to 936.59)	‐6.18 (−9.43 to −2.63)*	‐0.25 (−0.27 to −0.23)*	139.1 (120.57 to 158.16)	‐5.52 (−8.34 to −2.51)*	‐0.22 (−0.24 to −0.2)*	26.16 (6.61 to 69.03)	‐13.04 (−26.74 to 1.83)	‐0.45 (−0.58 to −0.33)*	33.99	4.26	0.12 (0.04 to 0.19)*
Saudi Arabia	456.71 (208.96 to 1019.77)	‐8.85 (−20.67 to 5.1)	‐0.33 (−0.35 to −0.31)*	701.74 (600.97 to 807.91)	‐6.11 (−10.04 to −2.63)*	‐0.2 (−0.21 to −0.19)*	120.44 (104.42 to 138.45)	‐5.42 (−9.2 to −1.91)*	‐0.19 (−0.2 to −0.17)*	25.85 (6.36 to 71.39)	‐9.26 (−25.62 to 11.36)	‐0.36 (−0.39 to −0.33)*	30.35	2.76	0.13 (0.11 to 0.15)*
Japan	456 (221.79 to 920.17)	‐0.99 (−7.76 to 4.54)	0.03 (0.01 to 0.06)*	674.1 (584.78 to 770.24)	3.76 (1.84 to 5.53)*	0.3 (0.25 to 0.36)*	117.23 (102.05 to 133.68)	1.22 (−0.42 to 2.83)	0.2 (0.15 to 0.25)*	26.33 (7.39 to 62.16)	‐3.13 (−10.23 to 5.55)	‐0.07 (−0.1 to −0.05)*	30.97	5.08	0.27 (0.22 to 0.32)*
Indonesia	422 (194.37 to 933.56)	15.59 (−0.41 to 35)	0.41 (0.35 to 0.47)*	662.13 (571.11 to 761.15)	‐2.44 (−3.98 to −0.99)*	‐0.08 (−0.1 to −0.05)*	113.31 (98.38 to 129.52)	‐2.29 (−3.61 to −1.07)*	‐0.07 (−0.09 to −0.05)*	22.97 (5.62 to 63.06)	30.89 (4.09 to 65.48)*	0.77 (0.68 to 0.87)*	31.92	−15.26	‐0.47 (−0.52 to −0.43)*
Canada	417.19 (213 to 829.97)	‐7.45 (−11.05 to −3.63)*	‐0.35 (−0.4 to −0.3)*	792.31 (701.61 to 888.19)	‐10.54 (−14.48 to −6.46)*	‐0.55 (−0.64 to −0.46)*	132.4 (118.02 to 147.29)	‐11.5 (−15.16 to −7.79)*	‐0.55 (−0.61 to −0.48)*	21.15 (5.57 to 54)	‐4.01 (−8.54 to 2.35)	‐0.16 (−0.18 to −0.14)*	38.81	−5.03	‐0.25 (−0.3 to −0.21)*
South Africa	406.65 (185.95 to 901.85)	2.47 (−4.25 to 11.93)	0.07 (0.01 to 0.12)*	609.51 (526.77 to 695.77)	‐5.7 (−7.33 to −4.07)*	‐0.16 (−0.17 to −0.15)*	107.78 (93.63 to 122.59)	‐5.61 (−7 to −4.45)*	‐0.16 (−0.17 to −0.14)*	22.58 (5.4 to 61.8)	6.69 (−2.21 to 19.23)	0.18 (0.1 to 0.26)*	30.28	−8.82	‐0.25 (−0.31 to −0.19)*
Australia	398.66 (187.36 to 824.12)	‐9.19 (−14.67 to −5.21)*	‐0.32 (−0.33 to −0.31)*	589.93 (513.23 to 663.34)	‐16.06 (−21.75 to −11.13)*	‐0.6 (−0.64 to −0.56)*	102.66 (90.27 to 115.36)	‐16.49 (−22.17 to −11.23)*	‐0.62 (−0.67 to −0.58)*	22.89 (5.93 to 57.75)	‐5.41 (−10.49 to 2.06)	‐0.16 (−0.18 to −0.14)*	30.29	−8.92	‐0.35 (−0.39 to −0.31)*
Russian Federation	397.87 (188.83 to 852.89)	‐2.01 (−6.11 to 2.18)	‐0.08 (−0.1 to −0.06)*	661.42 (573.18 to 758.61)	‐1.28 (−2.43 to −0.2)*	‐0.08 (−0.13 to −0.04)*	115.89 (101.23 to 131.98)	‐1.01 (−1.98 to 0.09)	‐0.07 (−0.12 to −0.03)*	21.05 (5.19 to 58.31)	‐2.61 (−8.32 to 4.49)	‐0.09 (−0.1 to −0.08)*	34.27	−0.23	‐0.02 (−0.05 to 0.01)
United Kingdom	390.4 (183.98 to 817.67)	‐4.65 (−6.97 to −2.72)*	‐0.1 (−0.13 to −0.08)*	591.08 (508.13 to 674.77)	‐5.52 (−7 to −4.26)*	‐0.13 (−0.17 to −0.09)*	107.93 (93.37 to 123.89)	‐5.8 (−7.19 to −4.49)*	‐0.17 (−0.2 to −0.13)*	22.61 (5.88 to 57.62)	‐4.56 (−8.21 to −0.25)*	‐0.09 (−0.13 to −0.05)*	31.3	−2.4	‐0.08 (−0.12 to −0.04)*
Argentine	369.59 (176.33 to 772.22)	‐3.36 (−6.69 to 0.43)	‐0.09 (−0.1 to −0.08)*	592.53 (512.44 to 676.59)	‐5.13 (−8.65 to −1.72)*	‐0.18 (−0.2 to −0.16)*	106.73 (92.2 to 122.29)	‐4.87 (−8.09 to −1.41)*	‐0.17 (−0.18 to −0.15)*	20.21 (5.17 to 53.04)	‐3.19 (−7.99 to 3.1)	‐0.06 (−0.07 to −0.04)*	33.34	−2.17	‐0.1 (−0.11 to −0.09)*
France	366.86 (171.08 to 763.86)	‐5.45 (−11.06 to 0.5)	‐0.18 (−0.2 to −0.17)*	525.22 (458.8 to 593.27)	‐5.11 (−9.76 to 0.88)	‐0.2 (−0.23 to −0.16)*	93.63 (82.25 to 105.56)	‐5.8 (−10.17 to 0.02)	‐0.22 (−0.26 to −0.19)*	21.69 (5.61 to 54.3)	‐5.59 (−12.88 to 5.08)	‐0.18 (−0.2 to −0.17)*	29.4	0.35	‐0.01 (−0.04 to 0.02)
India	305.73 (135.42 to 676.67)	16.66 (6.03 to 29.84)*	0.52 (0.48 to 0.55)*	436.13 (376.02 to 501.15)	‐0.89 (−2.34 to 0.33)	‐0.16 (−0.21 to −0.12)*	78.92 (68.29 to 90.58)	‐0.66 (−1.89 to 0.39)	‐0.16 (−0.21 to −0.11)*	16.98 (4.05 to 46.35)	27.19 (12.86 to 49.25)*	0.89 (0.83 to 0.95)*	28.16	−14.09	‐0.63 (−0.68 to −0.58)*
Mexico	296.84 (144.66 to 642.67)	‐4.06 (−9.6 to 0.97)	‐0.11 (−0.12 to −0.1)*	534.2 (460.41 to 612.87)	‐8.73 (−10.31 to −7.36)*	‐0.2 (−0.23 to −0.16)*	97.19 (84.27 to 111.26)	‐8.34 (−9.71 to −7.16)*	‐0.19 (−0.23 to −0.15)*	14.8 (3.49 to 41.39)	‐1.97 (−10.06 to 6.97)	‐0.07 (−0.09 to −0.06)*	36.5	−3.39	‐0.04 (−0.07 to −0.02)*
Global average	450.98 (212.69 to 950.16)	1.18 (−2.83 to 5.15)	‐0.02 (−0.03 to −0.01)*	694.01 (602.88 to 794.08)	3.24 (1.75 to 4.23)*	0 (−0.02 to 0.03)	119.76 (104.96 to 135.89)	2.38 (0.98 to 3.32)*	‐0.02 (−0.04 to 0)	25.16 (6.68 to 64.25)	0.47 (−3.78 to 6.99)	‐0.02 (−0.03 to −0.01)*	31.48	1.44	0.01 (0 to 0.02)

Note: Colored keys indicate rankings for all regions. The dark orange indicates Rank 1 and light gray means Rank 20, with the 95% confidence interval in parentheses. The order of the countries in the table was sorted by DALY ASR from highest to lowest.

Abbreviations: ADOD, Alzheimer's disease and other dementias; ASR, age‐standardized rate; DALY, disability‐adjusted life year; EAPC, Estimated Annual Percentage Changes; G20, Group of 20; YLDs, years lived with disability.

^*Indicates statistically significant.

The spatial and geographical distribution of dementia burden in 2021 across China's 33 municipalities was consistent with the proportion of the population over the age of 65 years (Figure 2). Hebei, Henan, and Jilin exhibited the highest ADOD burden, with Hebei recording the highest age‐standardized DALYs (659.51 per 100,000; 95% UI: 308.05–1461.68); this rate was 1.37 times higher than Jiangxi's, which was the lowest (480.68 per 100,000; 95% UI: 230.83–1057.39). Hebei also had the highest ASMR (37.97 per 100,000; 95% UI: 9.73–103.65). Henan had the highest ASPR, at 975.64 per 100,000 (95% UI: 838.77–1134.73), which was 1.27 times that in Jiangxi, which was the lowest at 769.15 per 100,000 (95% UI: 647.07–899.62). The highest ASIRs occurred in Yunnan, Jilin, and Henan, with values of 158.83 (95% UI: 138.02–181.81), 158.43 (95% UI: 137.69–181.42), and 158.05 (95% UI: 136.26–180.68) per 100,000, respectively (Table S7 and Figure S5 in supporting information). From 1990 to 2021, Xinjiang and Qinghai experienced the highest changes in DALYs and mortality rates, whereas Xinjiang and Shanghai showed the most pronounced shifts in incidence and prevalence (Table S8 in supporting information).

FIGURE 2.

Crude incidence, prevalence, mortality rates, and DALYs of ADOD in China in 2021 and a map of Chinese population density. A, Crude incidence. B, Crude prevalence. C, Crude mortality rate. D, Crude DALYs. E, Population density. F, The proportion of people aged ≥ 65 years in 2021. The darker the color on the map, the higher the burden index value of the province. ADOD, Alzheimer's disease and other dementias; DALYs, disability‐adjusted life years.

From 1990 to 2021, the ratio of YLD to DALYs increased from 27.28 to 33.01 in China (Table 1). By 2013, in China, this ratio had surpassed the global average and, thereafter, consistently remained above the highest observed global ratios and the G20 average (Figure 3A). Higher YLD‐to‐DALY ratios were observed in females, in the 80‐ to 89‐year‐old age group, and in Western China (Figure 3B and Figures S6,S7 in supporting information). Moreover, Shanghai and Liaoning had higher YLD‐to‐DALY ratios, which were accompanied by increased DALY in these regions (Figure 3C).

FIGURE 3.

Trends and distributions for the ratio of YLD to DALYs for ADOD in China, the G20, and the world from 1990 to 2021. A, Ratio of YLD to DALY in China, the G20, and the world from 1990 to 2021. B, Distribution of the ratio of YLD to DALYs in China in 2021. C, DALYs and ratio of YLD to DALY for ADOD in different provinces of China in 2021. The abscissa is the ratio of YLD to DALYs; the ordinate is the DALYs per 100,000 population for the indicated provinces; and the dotted lines are taken from the third quartile of the data. The plot is divided into nine groups as indicated. ADOD, Alzheimer's disease and other dementias; DALYs, disability‐adjusted life years; G20, Group of 20; YLD, years lived with disability.

From 1990 to 2021, high fasting plasma glucose levels were the leading risk factor for increased dementia burden in all municipalities (Table S9–S12 in supporting information). In 2021, the leading risk factor for ADOD in China was high fasting plasma glucose for both sexes, especially in individuals aged > 80 years (Figure S8 in supporting information). Considering all ADOD life factors, the attributable DALYs was 2.10 million (95% CI: 0.47–5.20) and there were 0.09 million (95% CI: 0.01–0.30) attributable deaths. Standardized for age, in 2021, the attributable DALYs was 113.41 (95% CI: 24.94–288.96) and there were 5.89 (95% CI: 0.71–18.73) attributable deaths per 100,000 (Table S13 in supporting information). Furthermore, the risk of ADOD associated with smoking increased progressively with age in males, peaking at 80 years (Table S14 in supporting information). The highest attributable DALYs and mortality values across all risk factors were observed in Shandong, Henan, Jiangsu, and Sichuan.

Between 1990 and 2021, DALY associated with dementia increased by 272.71%, from 2.70 to 10.07 million, driven by population aging (218.79% increase), population growth (42.37% increase), and an increase in the age‐specific rate (11.55%; Figure 4). Deaths attributed to dementia increased by 310.47%, from 0.12 million to 0.49 million, with the following contributing factors: population aging (270.59%), population growth (46.17%), and a decrease in age‐specific prevalence (6.29%; Figure 4). The number of people with dementia rose by 322.18%, from 4.02 million to 16.99 million, driven by population aging (218.95%), population growth (45.30%), and an increase in the age‐specific rate (57.93%; Figure S9 in supporting information). From 1990 to 2021, population aging was the primary driver of these increases, accounting for 67.96% of the rise in prevalence, 80.23% of the increase in DALYs, and 87.16% of the increase in deaths, corresponding to 8.81 million prevalent cases, 0.32 million deaths, and 5.91 million DALYs (Figure 4 and Figures S9–S11 in supporting information).

FIGURE 4.

Decomposition analysis of DALYs and mortality rate changes for ADOD in China in 2021. A, DALY changes. B, Mortality rate changes. Attributable risk factors of ADOD burden in China: population aging, population growth, high BMI, high fasting plasma glucose, smoking, and unknown risk factors. If the contribution of a risk factor is negative, the histogram of the indicator is to the left of the reference abscissa, which, in this figure, is 0. ADOD, Alzheimer's disease and other dementias; BMI, body mass index; DALYs, disability‐adjusted life years.

From 1990 to 2021, the highest total percentage changes for DALYs and mortality were observed in Chongqing. Compared to other risk factors, impacts from population aging and population growth were more pronounced (Figure 4). In 2021, the provinces of Sichuan, Heilongjiang, and Jilin exhibited the highest proportion of dementia burden attributable to population aging. Sichuan led in aging‐related contributions to DALYs, mortality, and prevalence: Sichuan's age‐standardized DALY rate was 2.94 times higher than that of Tibet, its ASMR was 2.44 times higher than that of the Macao Special Administrative Region, and its ASPR was 3.08 times higher than that of Tibet (Tables S15–S17 and Figures S12–S14 in supporting information). 

From 1990 to 2019, the G20 member countries and the global population experienced an exponential increase in mortality rate, prevalence, and DALYs attributed to population aging. After 2019, however, these metrics markedly declined. In contrast, China has seen a continuous rise in these indicators from 1990 to 2021, with a significant acceleration after 2005. The observed changes in mortality, prevalence, and DALYs can be attributed to factors such as population growth, aging, and shifts in age‐specific rates. Notably, in China, the impact of population aging on the overall changes in ADOD prevalence and associated mortality and DALYs has been disproportionately higher compared to the G20 and global averages (Figures S15–S21 in supporting information). ,  

4. DISCUSSION

Using GBD 2021 data, we estimated the disease burden and regional distribution of ADOD in China from 1990 to 2021. In 2021, China had the highest ASMR, ASIR, ASPR, and age‐standardized DALYs rates of ADOD among all G20 countries. Although China has witnessed a decrease in ASMR over the past 3 decades, it experienced increases in ASIR, ASPR, and age‐standardized DALYs driven primarily by population aging at both national and provincial levels. High BMI, elevated fasting plasma glucose levels, and smoking were identified as risk factors for dementia. Considering the accelerated population aging, the disease burden caused by ADOD in China is becoming an increasingly serious public health problem and requires immediate attention. 

The number of deaths, incidence, prevalence, DALYs, YLD, and YLL increased substantially in all sex and age groups from 1990 to 2021. However, while there was a trend toward declining ASMR and age‐standardized YLL, no such decrease was observed in ASIR and ASPR rates. There was a spike in ASPR and age‐standardized DALYs after 2019; one potential reason is that the Coronavirus Disease 2019 (COVID‐19) pandemic significantly impacted the care and disease burden of individuals with ADOD. ²⁰ , ²¹ , ²² Indeed, GBD 2021 found that the burden of COVID‐19, as measured by DALYs, increased between 2020 and 2021. ⁸ The observed trend toward decreasing age‐standardized mortality and YLL rates can be attributed to improvements in health care, early detection, and better management of ADOD. Enhanced medical care and support systems also may have contributed to reducing mortality rates and extending the lives of individuals with dementia. In contrast, the age‐standardized incidence and prevalence rates did not decrease, which could be due to several factors. First, there has been a significant increase in awareness and diagnostic capabilities for ADOD, which has probably led to more individuals being diagnosed and contributed to increasing incidence and prevalence. Second, as more people live longer, the absolute number of individuals at risk for developing dementia increases. Third, as medical care improves, individuals with dementia live longer, contributing to higher prevalence rates, even as mortality rates decline. Finally, other emerging risk factors or changes in lifestyle and environmental factors may continue to drive the incidence and prevalence of ADOD.

From 1990 to 2021, due to population aging, the ADOD prevalence and associated mortality and DALYs in China increased by 67.96%, 80.23%, and 87.16%, respectively. Possible reasons for the aging population include low fertility, high life expectancy, and family planning policies. ²³ Over the next 30 years, China will experience a wave of population aging that will increase the number of the oldest olds (aged ≥ 85) and empty‐nest elderly as well as increase elderly dependency. ²⁴ Although the Chinese government recently revised its one‐child policy, it is unlikely that the country's population aging trend will be reversed in the near future. Consequently, the significant burden of dementia on the public health system is likely to persist for the foreseeable future.

Japan is recognized as a super‐aging country and reports the highest global incidence of dementia according to GBD 2019 data: by 2045, > 25% of Japanese individuals > 65 years are projected to suffer from dementia. ²⁵ Japan has implemented comprehensive national policies, anchored by the National Framework for Promotion of Dementia Policies, to address the burden of dementia, ²⁶ as well as the “Orange Plan,” focusing on community‐centered support and care. ²⁷ Some countries have proposed forward‐looking, constructive, and practical national strategies for ADOD prevention, including the National Alzheimer's Project Act in the United States, ²⁸ Living Well With Dementia in the United Kingdom, ²⁹ the Nationale Demenzstrategie (National Dementia Strategy) in Germany, ³⁰ and “Together We Aspire” in Canada. ³¹ Although the implementation details vary, these strategies share common goals: ³² enhancing public and professional awareness, supporting research and innovation, improving diagnosis and care, and providing support for caregivers and families. ³³

To address the challenges of aging in China, we call for national strategies to coordinate ADOD care, raise public and professional awareness, and guide research and policies. China's Action Plan for Healthy China 2030 clearly sets out an implementation plan to promote healthy aging. ³⁴ The National Health and Family Planning Commission of the People's Republic of China issued the Work Program for Exploring Special Services for Dementia Prevention and Control ³⁵ and resolved to implement a nationwide program of dementia prevention and care promotion from 2023 to 2025. ³⁶ The objective is to achieve an 80% increase in public awareness of dementia prevention and control as well as an 80% screening rate for cognitive function in the elderly community. ³⁵ Furthermore, the State Council of China issued a guideline to promote the development of national undertakings for the aged and to improve the elderly care service system during the 14th Five‐Year Plan period (2021–2025), ³⁷ and six National Clinical Research Centers for Geriatric Disorders and the National Center for Neurological Disorder were established. These centers have collaborated across disciplines, with a focus on early diagnosis and screening, to increase the number of patients being accurately diagnosed.

Exercise training showed positive outcomes on memory in patients with obesity, irrespective of endurance or the type of resistance training used. ³⁸ Aerobic exercise improves cognition and executive function in older adults with glucose intolerance, ³⁹ and behavioral lifestyle modification through multifactorial approaches is a priority to prevent and delay onset of dementia in patients with diabetes. ⁴⁰ Adherence to a combination of healthy lifestyle factors, including a balanced diet, regular physical activity, and abstinence from harmful habits such as smoking and excessive alcohol consumption, especially at a young age, was associated with a significantly lower risk of dementia in participants with multi‐morbid conditions. ⁴¹ China has the most smokers of any tobacco product, nearly 300.7 million, ⁴² and ≈ 183.5 million current smokers have tobacco dependence. ⁴³ Promoting smoking cessation, particularly for middle‐aged males, may be beneficial in reducing the risk of developing dementia. Furthermore, digital interventions such as web‐based and electronic health interventions for public policy makers warrant further consideration. ⁴⁴

Substantial regional differences existed in the burden of dementia across provinces in China. From 1990 to 2021, Xinjiang and Qinghai had the greatest change in DALYs and mortality, and Xinjiang and Shanghai had the greatest change in incidence and prevalence. The precise reasons underlying these trends are not clear. Environmental risk factors, such as extreme temperature events, ⁴⁵ short‐term exposure to ambient air pollution, ⁴⁶ lithium in the drinking water, ⁴⁷ and seasonal variations in vitamin D levels, ⁴⁸ might provide possible explanations. The provinces of Sichuan, Heilongjiang, and Jilin had the highest proportion of ADOD burden attributable to population aging, and population aging contributed most significantly to DALYs, mortality, and prevalence in Sichuan. These differences may reflect varied cultural factors and social support systems among provinces in terms of elder care. Nearly half of dementia cases could theoretically be prevented by eliminating 14 modifiable risk factors, as stressed by the 2024 update of the Lancet Commission on dementia, ⁴⁹ and modifying these risk factors has even more potential for reducing dementia risk in low‐income and middle‐income countries, like China. Tailoring intervention programs to region‐specific risk factors could prove invaluable, for example smoking cessation in Hunan ⁵⁰ and Zhejiang ⁵¹ provinces, promoting physical exercise in Jiangxi province, ⁵² and improving education for awareness of dementia in Hubei. ⁵³ Further research, including in‐depth epidemiological surveys in targeted areas, are needed to obtain more detailed and specific regional data that may explain regional differences in ADOD burden. A meta‐analysis emphasized standardized reporting practices to better understand the challenges posed by AD in China. ⁵⁴ A national direct reporting system for ADOD is needed; toward this goal, the China Brain Cognitive Health Index may provide an excellent resource for these data within in each province. ⁵⁵

GBD 2021 comprehensively estimates the global burden of and risk factors for ADOD, albeit with certain inherent limitations. ⁸ , ¹³ , ¹⁵ , ⁵⁶ , ⁵⁷ First, GBD 2021 relies on models that use province‐level data and extant literature from China. The lack of data from some regions and the omission of distinctions between urban and rural areas could introduce biases into the estimates. Second, weighted PAFs may provide a more comprehensive estimate of the combined impact of multiple risk factors. ⁵⁸ , ⁵⁹ However, this approach is not standard in GBD studies due to methodological complexities and potential challenges in accurately estimating the joint effects of multiple risk factors. Third, GBD datasets do not distinguish between subtypes of dementia, such as vascular dementia or dementia with Lewy bodies, each of which is associated with differing burdens and risk profiles, and additional factors such as genetic predisposition and modifiable risk factors should be considered in risk profiles for dementia subtypes. ⁴⁹ Fourth, global updates to diagnostic criteria, biomarkers, medical records, and insurance codes for dementia have introduced heterogeneity in data over the past 30 years, which may contribute to discrepancies between our findings and the actual conditions. Finally, this analysis was conducted in the first 2 years after the emergence of COVID‐19, and longer term consequences of the pandemic on cognition or other aspects of nervous system health are not fully quantified. ¹⁵ Therefore, our results should be interpreted with caution, and further research is necessary to refine these findings.

5. CONCLUSIONS

Despite the observed decrease in ASMR, the overall burden of dementia in China has intensified, driven predominantly by population aging. This escalation places China at the forefront of dementia burden among the G20 nations. Considering identified modifiable risk factors for dementia, it is imperative to adopt geographically tailored public health interventions that promote lifestyle modifications alongside the implementation of comprehensive care models and early diagnostic services to meet the specific needs of the aging population in China and serve as a model for other regions experiencing similar demographic and health transitions.

AUTHOR CONTRIBUTIONS

Yi Tang and Maigeng Zhou conceived the research idea, designed the study, and provided overall guidance. Kun Yang directed the study and is responsible for the integrity of the data and the accuracy of the analysis. Xuan Yang performed data collection and management and drafted the manuscript. Peng Yin, Kun Yang, Xuan Yang, Yi Tang, and Maigeng Zhou critically revised the manuscript. All authors read and approved the final manuscript. All authors had full access to all the data in the study and had final responsibility for the decision to submit the manuscript for publication.

CONFLICT OF INTEREST STATEMENT

All authors have no actual or potential competing interests to declare. Author disclosures are available in the supporting information.

CONSENT STATEMENT

All participants provided written informed consent.

ETHICS STATEMENT

All data were analyzed anonymously, so ethical approval was not required.

Supporting information

Supporting Information

Supporting Information

Yang K, Yang X, Yin P, Zhou M, Tang Y. Temporal trend and attributable risk factors of Alzheimer's disease and other dementias burden in China: Findings from the Global Burden of Disease Study 2021. Alzheimer's Dement. 2024;20:7871–7884. 10.1002/alz.14254

DATA AVAILABILITY STATEMENT

GBD data used in this study can be obtained from the Global Health Data Exchange GBD 2021 website (https://ghdx.healthdata.org/gbd‐2021). Authors who work for the Chinese Center for Disease Control and Prevention can provide information on accessing these data.