Burden of atherosclerotic cardiovascular disease attributable to high body mass index, high fasting plasma glucose or high low density lipoprotein cholesterol in young and middle-aged adults from 1990 to 2050

Abstract

Aims

This study aims to analyze the global burden of atherosclerotic cardiovascular disease (ASCVD) attributable to high low-density lipoprotein cholesterol (LDL-c), high body mass index (BMI), or high fasting plasma glucose (FPG) in young and middle-aged adults (15–49 years) from 1990 to 2021, with projections extending to 2050.

Methods

Leveraging data from the Global Burden of Disease (GBD) 2021 study, we concentrated on ASCVD burden attributable to three risk factors in young and middle-aged adults. We scrutinized crude death rates, number of deaths, and estimated annual percentage changes (EAPCs) across sexes, countries, regions, and socio-demographic index (SDI) categories. To forecast future trends, we employed the Bayesian age-period-cohort (BAPC) model. Locally weighted scatterplot smoothing (LOESS) regression (span = 0.5) was used to visualize associations between crude death rates and SDI levels across 21 GBD regions and 204 countries.

Results

The ASCVD burden attributable to three risk factors in young and middle-aged adults followed broadly similar epidemiological patterns. In 2021, ischemic heart disease (IHD) burden attributable to three risk factors substantially exceeded ischemic stroke (IS) and lower extremity peripheral artery disease (LEPAD), and men exhibited consistently a higher burden than women. From 1990 to 2021, global trends in crude death rates for ASCVD attributable to three risk factors in young and middle-aged adults remained relatively stable. Significant regional and national variation was observed. At regional SDI levels, crude death rates for IHD showed an inverted M-shaped pattern (peaks at SDI 0.45 and 0.70), while IS and LEPAD demonstrated inverted L-shaped patterns (plateauing at SDI 0.70–0.76). Across 204 countries in 2021, crude death rates from ASCVD attributable to three risk factors increased gradually below SDI 0.60, showed a slight U-shape between 0.60–0.75, and declined sharply above 0.75. Projections indicate that ASCVD attributable to three risk factors will increase to varying degrees in young and middle-aged adults by 2050.

Conclusion

The burden of ASCVD attributable to three risk factors in young and middle-aged adults calls for concerted and integrated management strategies. Urgent public health measures targeting these risk factors are needed to mitigate the projected rise in ASCVD burden.

Introduction

Atherosclerotic cardiovascular disease (ASCVD), including ischemic heart disease (IHD), ischemic stroke (IS), and lower extremity peripheral artery disease (LEPAD), was a leading cause of morbidity and mortality worldwide, imposing a significant burden on global health systems and economies [1]. Over the past few decades, while the prevalence of ASCVD in the elderly has declined, the prevalence of ASCVD in young and middle-aged adults has significantly increased, closely linked to several key metabolic risk factors, including high low-density lipoprotein cholesterol (LDL-c), high body mass index (BMI), and high fasting plasma glucose (FPG) [1]. These risk factors have been identified as major contributors to the development and progression of ASCVD through various pathophysiological mechanisms. High LDL-c levels promote the formation of atherosclerotic plaques, which can lead to ischemic events such as myocardial infarction and stroke [2]. High BMI, particularly when associated with central obesity, is linked to increased cardiovascular risk through mechanisms involving inflammation, insulin resistance, and adverse lipid profiles [3]. Similarly, high FPG levels contribute to the risk of ASCVD through mechanisms such as endothelial dysfunction, oxidative stress, and inflammation, which promote atherosclerosis [4]. The interplay among these risk factors further complicates the assessment and management of cardiovascular risk, highlighting the importance of a comprehensive approach to prevention and treatment.

The Global Burden of Disease (GBD) study [5, 6] has been instrumental in understanding the epidemiology and impact of metabolic risk factors on ASCVD in young and middle-aged adults. In this study, we utilize the GBD 2021 data for the first time to provide an updated and comprehensive analysis of the burden of ASCVD attributable to high LDL-c, high BMI, and high FPG in young and middle-aged adults from 1990 to 2021, and to offer predictions up to 2050. While existing GBD 2021 studies [7–10] analyses have largely focused on CVD burden attributable to metabolic risks across all ages, this study uniquely isolates ASCVD in young and middle-aged adults (15–49 years) to provide the first direct comparison of high LDL-c, high BMI, and high FPG as distinct yet interconnected risk factors in this specific population. Examining three decades of historical trends is critical for evaluating the effectiveness of previous public health interventions, identifying successful prevention strategies, and revealing gaps in current policies. Meanwhile, forward-looking predictions are essential for anticipating the future disease burden amid demographic shifts, and evolving risk factor profiles. These analyses collectively address key policy questions: which risk factors require prioritized investment, how healthcare resources should be strategically allocated across regions and nations, and how prevention programs can be optimized to maximize impact and cost-effectiveness in the coming decades. By integrating three decades of historical data with forward-looking prediction, the study aims to guide policymakers and healthcare providers in prioritizing resources and implementing measures to reduce the burden of ASCVD in young and middle-aged adults.

Methods

Data sources

The data utilized in our study were obtained from the GBD 2021 database [11], encompassing details on 371 diseases and injuries, as well as 88 risk factors, across 204 countries, 21 GBD regions, and five socio-demographic index (SDI) regions, stratified by age and gender [5, 6]. Our study covered data from 1990 to 2021, concentrating on the ASCVD, including IHD, IS, and LEPAD, burden attributable to high FPG, high BMI, and high LDL-c in young and middle-aged adults. The Washington University Institutional Review Board granted approval for the informed consent waiver, given that the GBD 2021 study utilized deidentified and aggregated data.

Definition

To estimate the population attributable burden, we established the theoretical minimum risk exposure level (TMREL) for each risk factor. The TMREL represents the optimal, hypothetical exposure level associated with the lowest possible mortality risk across all health outcomes, serving as the counterfactual reference for calculating preventable burden [10]. Regarding the calculation of the attributable burden, the GBD study employs a comparative risk assessment framework to estimate population attributable fractions (PAFs). Specifically, for each risk factor, a TMREL is established, representing the optimal exposure range associated with the lowest population-level mortality risk. The PAF is then calculated by comparing the current population exposure distribution to this TMREL, using integrated exposure–response relationships derived from meta-analyses of epidemiological studies. The number of attributable deaths is obtained by multiplying the PAF by the total death count for a specific disease. It is important to note that this attribution is based on statistical associations at the population level and does not imply direct clinical causality for individual cases. All estimates are presented with 95% uncertainty intervals (UIs) to reflect the variability in data inputs and modelling assumptions [10]. High LDL-c is considered as any concentration above the theoretical minimum risk range of 0.9–1.4 mmol/L [10]. High BMI for adults (20 and older) is defined as BMI greater than 20–23 kg/m². High BMI for children (2–19 years) is defined as being overweight or obese based on International Obesity Task Force standards [10]. High FPG is considered as any concentration above the theoretical minimum risk range of 4.90–5.30 mmol/L [10]. The classification of CVD, including IHD, IS, and LEPAD, is based on the ICD-9 and ICD-10 systems.

The age range of 15–49 years was selected to focus on ASCVD in young and middle-aged adults, based on both clinical and public health considerations. This period captures the critical period from late adolescence through early middle age, when premature events carry disproportionate public health impact due to steeper risk trajectories, greater years of life lost, and substantial socioeconomic burden [12, 13]. This upper threshold (49 years) lies below conventional high-risk cutoffs (e.g., < 55 years for men/ < 65 years for women) [13], enabling focused identification of early-onset cases and their distinct risk factor patterns. This age band also aligns directly with standard GBD analytic groups, facilitating seamless integration with global burden-of-disease estimates and enhancing comparability for policy applications. Data on number of deaths and crude death rates were extracted from the GBD 2021 database for (1) IHD or IS attributable to high LDL-c, (2) IHD, IS, or LEPAD attributable to high BMI, and (3) IHD, IS, or LEPAD attributable to high FPG.

The SDI, a metric reflecting socio-economic status, is derived from the geometric mean of three elements: the total fertility rate for people under 25, the average years of education for those aged 15 and older, and the lag-distributed per capita income [5, 6]. To facilitate comparative analysis, the 21 GBD regions and 204 countries and territories were categorized into five quintile groups according to their SDI levels.

Statistical analysis

To evaluate the burden of ASCVD attributable to three risk factors in young and middle-aged adults, we employed crude death rates (per 100,000 populations), number of deaths, and estimated annual percentage change (EAPC), with these metrics analyzed by sex, country, region, and SDI classification. The crude death rates and number of deaths were directly extracted from the GBD database and shown with a 95% UIs. This UI was calculated from 1,000 resampled estimates for each parameter, with the 95% UI being the range from the 2.5th to the 97.5th percentiles of these resampled values. [5, 6]

A regression analysis using the natural logarithm of crude death rate was conducted, structured as y = α + βx + ε, where y signifies the natural logarithm of the crude death rate and x indicates the calendar year [14]. The estimated annual percent change (EAPC) was computed as 100 × (exp(β)—1), with its 95% confidence interval (CI) obtained from this regression model [14]. The trend of the crude death rate was categorized as increasing if both the EAPC and its 95% CI exceeded zero, as decreasing if they were below zero, and as stable if the 95% CI included zero. [14]

Bayesian Age-Period-Cohort (BAPC) model was employed based on a Poisson distribution to forecast the ASCVD burden attributable to three risk factors in young and middle-aged adults from 2022 to 2050. The foundational population data required for the model were obtained from the World Population Prospects standard dataset published by the United Nations Department of Economic and Social Affairs (https://population.un.org/wpp/). Model parameters were estimated using Integrated Nested Laplace Approximations (INLA) to circumvent convergence issues commonly associated with traditional Markov Chain Monte Carlo (MCMC) methods [15]. To ensure the smoothness and stability of the age, period, and cohort effects over time, Gaussian random walk priors were applied. Specifically, based on the observed trends in disease burden, a second-order random walk prior (RW2) capable of accommodating linear trends was selected. Model goodness-of-fit was assessed using the Watanabe-Akaike Information Criterion (WAIC), and posterior predictive checks confirmed that the model adequately captured the characteristics of the observed data. The BAPC model was chosen for its ability to simultaneously disentangle the three-dimensional temporal effects of age, period, and cohort, offering a forecasting structure more aligned with the natural history of the disease compared to traditional time-series extrapolation methods [16]. Furthermore, the Bayesian framework fully propagates parameter uncertainty into the forecast results, generating probabilistic predictions with credible intervals, thereby providing a more comprehensive risk assessment basis for public health decision-making.

Moreover, to visualize the potential nonlinear associations between crude death rates and SDI levels across 21 GBD regions and 204 countries and territories, we employed locally weighted scatterplot smoothing (LOESS) regression. A smoothing span parameter of 0.5 was selected to balance the capture of local trends with overall curve stability. The resulting curves should be interpreted as smoothed summaries of the central tendency within the data; their specific morphological features are descriptive and may vary with different smoothing parameter choices. All statistical analyses were conducted using R (version 4.2.3).

Results

The ASCVD burden attributable to the three risk factors displayed broadly similar epidemiological patterns in young and middle-aged adults across global, sex, regional, national, and SDI levels.

Global and sex-specific trends and burden

From 1990 to 2021, global trends in crude death rates for ASCVD attributable to three risk factors in young and middle-aged adults remained relatively stable. Crude death rates for IHD or IS attributable to high LDL-c in young and middle-aged adults showed modest declines [EAPC: -0.27(95%CI -0.39 to -0.16) and -0.56 (95%CI -0.71 to -0.41)], with women experiencing a more pronounced reduction than men. In contrast, crude death rates for ASCVD attributable to high BMI or high FPG in young and middle-aged adults increased slightly across all disease categories, with the most pronounced rise for IHD attributable to high FPG [EAPC: 1.54 (95%CI 1.46 to 1.63)] (Tables 1, 2 and 3 and Fig. 1A-C). Notably, gender differences were evident in these trends: for high BMI, IHD or LEPAD rose more rapidly in women, whereas IS increased faster in men; for high FPG, IHD or IS grew more quickly in men, while LEPAD showed a steeper rise in women (Tables 1, 2 and 3).In 2021, the attributable burden for IHD in young and middle-aged adults was consistently and substantially higher than that for IS or LEPAD across all three risk factors. The ASCVD burden for number of deaths and crude death rates in young and middle-aged adults was highest for high LDL-c, followed by high BMI or high FPG. Furthermore, the burden of ASCVD attributable to three risk factors in young and middle-aged adults was generally higher among men than women (Tables 1, 2 and 3).

Table 1.

Number of deaths and crude death rates of atherosclerotic cardiovascular disease attributable to high low density lipoprotein cholesterol among young and middle-aged adults in 2021, and estimated annual percentage changes of crude death rates from 1990 to 2021, by global, region, and socio‐demographic index levels

Location	IHD attributable to high LDL-c			IS attributable to high LDL-c
	Number of deaths (95%UI) in 2021	Crude death rate (95%UI, per 100,000 populations) in 2021	EAPC for crude death rates (95%CI) from 1990 to 2021	Number of deaths (95%UI) in 2021	Crude death rate (95%UI,per 100,000 populations) In 2021	EAPC for crude death rates (95%CI) from 1990 to 2021
Global	323,184 (244,065 to 399,469)	8.18 (6.18 to 10.12)	-0.27 (-0.39 to -0.16)	28,244 (13,783 to 39,071)	0.72 (0.35 to 0.99)	-0.56 (-0.71 to -0.41)
Sex
Female	90,485 (69,966 to 110,202)	4.64 (3.59 to 5.65)	-0.38 (-0.46 to -0.3)	10,817 (5329 to 15,630)	0.56 (0.27 to 0.8)	-1.01 (-1.13 to -0.9)
Male	232,700 (175,706 to 288,639)	11.64 (8.79 to 14.43)	-0.23 (-0.37 to -0.09)	17,428 (8,533 to 24,298)	0.87 (0.43 to 1.22)	-0.25 (-0.44 to -0.06)
GBD regions
Andean Latin America	1,279 (954 to 1,684)	3.66 (2.73 to 4.81)	-1.13 (-1.51 to -0.75)	101 (50 to 151)	0.29 (0.14 to 0.43)	-2.22 (-2.51 to -1.93)
Australasia	427 (342 to 506)	2.96 (2.37 to 3.5)	-2.90 (-3.08 to -2.73)	14 (7 to 18)	0.09 (0.05 to 0.12)	-3.60 (-3.76 to -3.43)
Caribbean	2,039(1,488 to 2,717)	8.52 (6.21 to 11.35)	0.01 (-0.22 to 0.23)	185 (84 to 280)	0.77 (0.35 to 1.17)	0.23 (0.12 to 0.34)
Central Asia	5,243(3,940 to 6,589)	10.75 (8.08 to 13.51)	-1.83 (-2.40 to -1.27)	598 (284 to 836)	1.23 (0.58 to 1.72)	-1.40 (-1.95 to -0.84)
Central Europe	3,890 (3,060 to 4,625)	7.38 (5.81 to 8.78)	-3.66 (-3.90 to -3.42)	418 (215 to 554)	0.79 (0.41 to 1.05)	-2.73 (-2.96 to -2.50)
Central Latin America	9,013(7,012 to 11,108)	6.77 (5.27 to 8.34)	0.58 (0.27 to 0.90)	445 (224 to 611)	0.33 (0.17 to 0.46)	-1.72 (-2.12 to -1.31)
Central Sub-Saharan Africa	2,400 (1,549 to 3,511)	3.68 (2.38 to 5.39)	-0.22 (-0.28 to -0.16)	217(84 to 387)	0.33 (0.13 to 0.59)	-0.74 (-0.81 to -0.66)
East Asia	44,388 (32,001 to 59,082)	6.45 (4.65 to 8.58)	1.11 (0.93 to 1.29)	7,615 (3,529 to 10,835)	1.11 (0.51 to 1.57)	0.71 (0.48 to 0.94)
Eastern Europe	16,326 (12,681 to 19,781)	16.97 (13.18 to 20.56)	-1.92 (-2.74 to -1.10)	1,913 (967 to 2,559)	1.99 (1.01 to 2.66)	-1.47 (-2.18 to -0.76)
Eastern Sub-Saharan Africa	5,986 (4,040 to 8,165)	2.86 (1.93 to 3.9)	-0.06 (-0.16 to 0.03)	602 (256 to 944)	0.29 (0.12 to 0.45)	-0.55 (-0.63 to -0.47)
High-income Asia Pacific	1,868 (1,480 to 2,213)	2.39 (1.89 to 2.83)	-1.20 (-1.39 to -1.00)	154 (79 to 208)	0.2 (0.1 to 0.27)	-3.92 (-4.14 to -3.70)
High-income North America	8,255 (6,364 to 9,826)	4.89 (3.77 to 5.83)	-2.24 (-2.50 to -1.98)	357 (179 to 478)	0.21 (0.11 to 0.28)	-1.56 (-2.08 to -1.03)
North Africa and Middle East	42,900 (32,402 to 54,342)	12.83 (9.69 to 16.25)	-0.55 (-0.64 to -0.46)	5,366(2,624 to 7,672)	1.6 (0.78 to 2.29)	-0.17 (-0.24 to -0.10)
Oceania	1,172 (829 to 1,598)	16.57 (11.72 to 22.54)	0.21 (0.13 to 0.29)	34 (15 to 55)	0.48 (0.21 to 0.78)	-0.54 (-0.65 to -0.43)
South Asia	116,391 (86,219 to 148,024)	11.56 (8.56 to 14.7)	0.38 (0.28 to 0.47)	3,587 (1,529 to 6,306)	0.36 (0.15 to 0.63)	-0.45 (-0.51 to -0.39)
Southeast Asia	38,432(28,565 to 48,952)	10.36 (7.7 to 13.2)	0.69 (0.61 to 0.77)	4,025 (1,856 to 5,895)	1.09 (0.5 to 1.59)	0.88 (0.73 to 1.03)
Southern Latin America	1,164 (929 to 1,380)	3.35 (2.68 to 3.98)	-2.52 (-2.73 to -2.30)	57 (29 to 77)	0.16 (0.08 to 0.22)	-3.51 (-3.67 to -3.35)
Southern Sub-Saharan Africa	1,933 (1,441 to 2,501)	4.48 (3.34 to 5.79)	-0.53 (-0.96 to -0.10)	224 (103 to 331)	0.52 (0.24 to 0.77)	-0.44 (-1.00 to 0.13)
Tropical Latin America	8,205(6,620 to 9,668)	6.85 (5.52 to 8.07)	-1.16 (-1.25 to -1.07)	651 (336 to 855)	0.54 (0.28 to 0.71)	-3.25 (-3.84 to -2.65)
Western Europe	5,091(4,067 to 5,933)	2.70 (2.16 to 3.15)	-3.33 (-3.46 to -3.19)	250 (129 to 325)	0.13 (0.07 to 0.17)	-4.42 (-4.49 to -4.34)
Western Sub-Saharan Africa	6,782 (4,803 to 9,167)	2.96 (2.09 to 4)	-0.11 (-0.27 to 0.05)	1,432 (656 to 2,157)	0.62 (0.29 to 0.94)	-0.12 (-0.25 to 0.02)
SDI levels
Low SDI	29,093 (20,951 to 37,538)	5.36 (3.86 to 6.92)	-0.24 (-0.30 to -0.18)	2,567 (1135 to 4,008)	0.47 (0.21 to 0.74)	-0.20 (-0.24 to -0.16)
Low-middle SDI	103,880 (78,150 to 129,599)	10.22 (7.69 to 12.75)	0.27 (0.20 to 0.33)	6,670 (3249 to 9,989)	0.66 (0.32 to 0.98)	-0.05 (-0.12 to 0.01)
Middle SDI	114,808 (87,513 to 139,986)	9.15 (6.97 to 11.15)	0.66 (0.59 to 0.73)	10,539 (4984 to 14,536)	0.84 (0.40 to 1.16)	0.17 (0.09 to 0.25)
High-middle SDI	51,402 (40,442 to 62,511)	8.16 (6.42 to 9.93)	-1.33 (-1.74 to -0.92)	6,508 (3167 to 8,984)	1.03 (0.50 to 1.43)	-1.11 (-1.49 to -0.73)
High SDI	23,720 (18,413 to 28,991)	4.72 (3.67 to 5.77)	-1.97 (-2.05 to -1.88)	1,936(914 to 2,662)	0.39 (0.18 to 0.53)	-1.47 (-1.63 to -1.32)

CI: Confidence Interval, EAPC: Estimated Annual Percentage Change (measures trend in crude death rates over time, calculated using a generalized linear model with Gaussian distribution and log-link function),GBD: Global Burden of Disease, IHD: Ischemic Heart Disease, IS: Ischemic Stroke, LDL-c: Low-Density Lipoprotein Cholesterol, SDI: Socio-demographic Index (composite index 0-1 based on lag-distributed income per capita, average educational attainment over age 15, and total fertility rate under 25, used to categorize countries into development levels),UI: Uncertainty Interval (95% uncertainty interval around point estimates, accounting for uncertainty from input data, model parameters, and modeling choices)

Table 2.

Number of deaths and crude death rates of atherosclerotic cardiovascular disease attributable to high body mass index among young and middle-aged adults in 2021, and estimated annual percentage changes of crude death rates from 1990 to 2021, by global, region, and socio‐demographic index levels

Location	IHD attributable to high BMI			IS attributable to high BMI			LEPAD attributable to high BMI
	Number of deaths (95%UI) in 2021	Crude death rate (95%UI, per 100,000 populations) in 2021	EAPC for crude death rates (95%CI) from 1990 to 2021	Number of deaths (95%UI) in 2021	Crude death rate (95%UI, per 100,000 populations) in 2021	EAPC for crude death rates (95%CI) from 1990 to 2021	Number of deaths (95%UI) in 2021	Crude death rate (95%UI, per 100,000 populations) in 2021	EAPC for crude death rates (95%CI) from 1990 to 2021
Global	103,255 (43,071 to 160,807)	2.61 (1.09 to 4.07)	1.02 (0.89 to 1.15)	7,742 (1,245 to 14,300)	0.20 (0.03 to 0.36)	1.12 (0.98 to 1.26)	144.79 (33.172 to 337.628)	0.004 (0.001 to 0.009)	1.178 (0.754 to 1.603)
Sex
Female	32,076 (13,080to49,948)	1.65 (0.67 to 2.56)	1.06 (0.96 to 1.16)	3,387 (574to6,186)	0.17 (0.03 to 0.32)	0.69 (0.57 to 0.81)	59.142 (13.937to133.396)	0.003 (0.001 to 0.007)	1.759 (1.403 to 2.116)
Male	71,179 (29,658to111,048)	3.56 (1.48 to 5.55)	1.01 (0.86 to 1.16)	4,355 (682to7,982)	0.22 (0.03 to 0.4)	1.48 (1.31 to 1.65)	85.648 (20.155to206.724)	0.004 (0.001 to 0.01)	0.834 (0.365 to 1.306)
GBD regions
Andean Latin America	597 (249 to 950)	1.71 (0.71 to 2.71)	0.04 (-0.33 to 0.42)	39 (7 to 76)	0.11 (0.02 to 0.22)	-0.96 (-1.24 to -0.69)	0.28 (0.055 to 0.673)	0.001 (0 to 0.002)	7.118 (6.776 to 7.462)
Australasia	187 (79 to 283)	1.3 (0.54 to 1.96)	-1.79 (-2.00 to -1.57)	5 (1 to 9)	0.03 (0.01 to 0.06)	-2.47 (-2.64 to -2.31)	0.509 (0.115 to 1.069)	0.004 (0.001 to 0.007)	0.659 (0.506 to 0.813)
Caribbean	783 (307 to 1,270)	3.27 (1.28 to 5.3)	1.04 (0.83 to 1.26)	59 (9 to 110)	0.25 (0.04 to 0.46)	1.58 (1.48 to 1.69)	2.797 (0.75 to 6.266)	0.012 (0.003 to 0.026)	3.424 (3.169 to 3.680)
Central Asia	2,036 (833 to 3,353)	4.17 (1.71 to 6.88)	-1.11 (-1.71 to -0.50)	194 (32 to 372)	0.4 (0.07 to 0.76)	-0.53 (-1.13 to 0.08)	1.421 (0.268 to 3.972)	0.003 (0.001 to 0.008)	4.146 (3.646 to 4.648)
Central Europe	1,477 (621 to 2,380)	2.8 (1.18 to 4.52)	-3.00 (-3.25 to -2.76)	128 (21 to 2,412)	0.24 (0.04 to 0.46)	-2.05 (-2.30 to -1.81)	12.824 (2.661 to 29.932)	0.024 (0.005 to 0.057)	0.625 (-0.177 to 1.434)
Central Latin America	4,509 (1,976 to 6,961)	3.39 (1.48 to 5.23)	1.73 (1.42 to2.05)	184 (34 to 330)	0.14 (0.03 to 0.25)	-0.58 (-0.97 to -0.18)	5.647 (1.634 to 11.403)	0.004 (0.001 to 0.009)	0.542 (0.177 to 0.907)
Central Sub-Saharan Africa	702 (254 to 1,235)	1.08 (0.39 to 1.89)	2.42 (2.33 to 2.52)	49 (7 to 106)	0.08 (0.01 to 0.16)	2.32 (2.22 to 2.42)	2.449 (0.484 to 7.618)	0.004 (0.001 to 0.012)	4.572 (4.189 to 4.956)
East Asia	13,041 (5,170 to 22,431)	1.89 (0.75 to 3.26)	3.74 (3.55 to 3.93)	1,698 (250 to 3,343)	0.25 (0.04 to 0.49)	4.01 (3.77 to 4.26)	6.969 (1.557 to 18.596)	0.001 (0 to 0.003)	6.241 (5.894 to 6.589)
Eastern Europe	6,011 (2,508 to 9,637)	6.25 (2.61 to 10.01)	-0.80 (-1.66 to 0.06)	587 (92 to 1,054)	0.61 (0.1 to 1.1)	-0.43 (-1.16 to 0.30)	14.773 (2.663 to 40.146)	0.015 (0.003 to 0.042)	0.248 (-0.683 to 1.187)
Eastern Sub-Saharan Africa	1,498 (563 to 2,486)	0.72 (0.27 to 1.19)	1.67 (1.58 to 1.77)	118 (16 to 230)	0.06 (0.01 to 0.11)	1.47 (1.37 to 1.58)	2.307 (0.361 to 7.414)	0.001 (0 to 0.004)	4.822 (4.594 to 5.051)
High-income Asia Pacific	421(168 to 666)	0.54 (0.22 to 0.85)	0.13 (-0.05 to 0.31)	24 (4 to 43)	0.03 (0 to 0.06)	-2.29 (-2.50 to -2.08)	0.552 (0.136 to 1.496)	0.001 (0 to 0.002)	1.715 (1.521 to 1.910)
High-income North America	4,535 (2,056 to 6,637)	2.69 (1.22 to 3.93)	-0.95 (-1.30 to -0.60)	167 (30 to 280)	0.1 (0.02 to 0.17)	-0.23 (-0.69 to 0.24)	34.095 (10.249 to 64.104)	0.02 (0.006 to 0.038)	1.711 (1.096 to 2.329)
North Africa and Middle East	22,351 (9,870 to 34,804)	6.69 (2.95 to 10.41)	0.96 (0.87 to 1.04)	2,385 (439 to 4,339)	0.71 (0.13 to 1.3)	1.25 (1.18 to 1.32)	10.892 (3.03 to 21.876)	0.003 (0.001 to 0.007)	4.017 (3.678 to 4.356)
Oceania	458 (196 to 792)	6.47 (2.77 to 11.19)	0.96 (0.86 to 1.06)	11 (2 to 22)	0.15 (0.02 to 0.31)	0.28 (0.15 to 0.41)	0.285 (0.077 to 0.735)	0.004 (0.001 to 0.01)	2.769 (2.198 to 3.343)
South Asia	26,447 (10,574 to 42,670)	2.63 (1.05 to 4.24)	3.03 (2.88 to 3.19)	617 (84 to 1,341)	0.06 (0.01 to 0.13)	2.60 (2.54 to 2.67)	5.577 (1.257 to 15.772)	0.001 (0 to 0.002)	5.621 (5.545 to 5.697)
Southeast Asia	9,315 (3,607 to 14,831)	2.51 (0.97 to 4)	2.64 (2.47 to 2.81)	719 (99 to 1,365)	0.19 (0.03 to 0.37)	3.27 (2.98 to 3.55)	2.598 (0.544 to 7.501)	0.001 (0 to 0.002)	5.548 (5.416 to 5.681)
Southern Latin America	529 (230 to 814)	1.52 (0.66 to 2.35)	-1.47 (-1.65 to -1.28)	22 (4 to 39)	0.06 (0.01 to 0.11)	-2.39 (-2.52 to -2.26)	0.72 (0.158 to 1.65)	0.002 (0 to 0.005)	1.799 (1.137 to 2.465)
Southern Sub-Saharan Africa	886 (369 to 1,402)	2.05 (0.86 to 3.25)	0.47 (0.04 to 0.90)	90(15 to 161)	0.21 (0.03 to 0.37)	0.64 (0.01 to 1.27)	12.011 (2.104 to 30.911)	0.028 (0.005 to 0.072)	3.773 (3.497 to 4.049)
Tropical Latin America	3,363 (1,440 to 5,218)	2.81 (1.2 to 4.35)	-0.05 (-0.14 to 0.04)	221 (36 to 402)	0.18 (0.03 to 0.34)	-2.03 (-2.62 to -1.43)	15.315 (3.779 to 35.178)	0.013 (0.003 to 0.029)	1.915 (1.515 to 2.317)
Western Europe	1,832 (769 to 2,897)	0.97 (0.41 to 1.54)	-2.39 (-2.56 to -2.21)	72 (11 to 134)	0.04 (0.01 to 0.07)	-3.38 (-3.47 to -3.29)	8.055 (1.595 to 22.566)	0.004 (0.001 to 0.012)	0.921 (0.307 to1.539)
Western Sub-Saharan Africa	2,276 (875 to 3,773)	0.99 (0.38 to 1.65)	1.48 (1.30 to 1.66)	354(52 to 693)	0.15 (0.02 to 0.3)	1.69 (1.52 to 1.86)	4.713 (0.803 to 14.341)	0.002 (0 to 0.006)	3.735 (3.601 to 3.869)
SDI levels
Low SDI	7,257 (2,826 to 11,654)	1.34 (0.52 to 2.15)	1.82 (1.70 to 1.95)	569(80 to 1,115)	0.10 (0.01 to 0.21)	1.99 (1.89 to 2.09)	5.375 (0.973 to 17.807)	0.001 (0 to 0.003)	4.057 (3.816 to 4.299)
Low-middle SDI	29,918 (12,126 to 46,359)	2.94 (1.19 to 4.56)	2.22 (2.12 to 2.32)	1,862 (294 to 3,490)	0.18 (0.03 to 0.34)	1.88 (1.78 to 1.98)	14.129 (3.074 to 36.464)	0.001 (0 to 0.004)	4.860 (4.713 to 5.007)
Middle SDI	36,352 (15,171 to 56,340)	2.90 (1.21 to 4.49)	2.35 (2.28 to 2.43)	2,702 (438 to 4,921)	0.22 (0.03 to 0.39)	2.02 (1.96 to 2.07)	39.771 (9.212 to 94.792)	0.003 (0.001 to 0.008)	3.057 (2.969 to 3.146)
High-middle SDI	18,121 (7,600 to 28,979)	2.88 (1.21 to 4.6)	-0.19 (-0.63 to 0.24)	1,823 (289 to 3,368)	0.29 (0.05 to 0.54)	0.36 (-0.03 to 0.75)	37.599 (7.901 to 92.628)	0.006 (0.001 to 0.015)	0.683 (0.097 to 1.273)
High SDI	11,500 (5,164 to 17,259)	2.29 (1.03 to 3.44)	-0.49 (-0.60 to -0.38)	779 (143 to 1,403)	0.16 (0.03 to 0.28)	0.71 (0.51 to 0.92)	47.598 (13.294 to 97.959)	0.009 (0.003 to 0.02)	1.463 (0.893 to 2.037)

CI: Confidence Interval, EAPC: Estimated Annual Percentage Change (measures trend in crude death rates over time, calculated using a generalized linear model with Gaussian distribution and log-link function),GBD: Global Burden of Disease, IHD: Ischemic Heart Disease, IS: Ischemic Stroke, LEPAD: lower extremity peripheral artery disease, BMI: body mass index, SDI: Socio-demographic Index (composite index 0-1 based on lag-distributed income per capita, average educational attainment over age 15, and total fertility rate under 25, used to categorize countries into development levels),UI: Uncertainty Interval (95% uncertainty interval around point estimates, accounting for uncertainty from input data, model parameters, and modeling choices)

Table 3.

Number of deaths and crude death rates of atherosclerotic cardiovascular disease attributable to high fasting plasma glucose among young and middle-aged adults in 2021, and estimated annual percentage changes of crude death rates from 1990 to 2021, by global, region, and socio‐demographic index levels

Location	IHD attributable to high FPG			IS attributable to high FPG			LEPAD attributable to high FPG
	Number of deaths (95%UI) in 2021	Crude death rate (95%UI, per 100,000 populations) in 2021	EAPC for crude death rates (95%CI) from 1990 2021	Number of deaths (95%UI) in 2021	Crude death rate (95%UI, per 100,000 populations) in 2021	EAPC for crude death rates (95%CI) from 1990 to 2021	Number of deaths (95%UI) in 2021	Crude death rate (95%UI, per 100,000 populations) in 2021	EAPC for crude death rates (95%CI) from 1990 to 2021
Global	38,297 (32,145 to 45,001)	0.97 (0.81 to 1.14)	1.54 (1.46 to 1.63)	5,046 (3,751 to 6,479)	0.13 (0.10 to 0.16)	0.91 (0.78 to 1.04)	58.341 (46.899 to 74.087)	0.001 (0.001 to 0.002)	1.116 (0.711 to 1.522)
Sex
Female	10,226 (8,233to12,493)	0.52 (0.42 to 0.64)	1.51 (1.44 to 1.59)	1,886 (1,391to2,522)	0.10 (0.07 to 0.13)	0.59 (0.50 to 0.68)	20.015 (15.455to25.445)	0.001 (0.001 to 0.001)	1.960 (1.668 to 2.254)
Male	28,071 (23,377to33,108)	1.40 (1.17 to 1.66)	1.56 (1.44 to 1.68)	3,160 (2,334to4,113)	0.16 (0.12 to 0.21)	1.10 (0.93 to 1.28)	38.326 (30.474to48.986)	0.002 (0.002 to 0.002)	0.754 (0.296 to 1.213)
GBD regions
Andean Latin America	85 (62 to 111)	0.24 (0.18 to 0.32)	0.42 (0.01 to 0.83)	12 (8 to 17)	0.03 (0.02 to 0.05)	-0.89 (-1.21 to -0.57)	0.055 (0.035 to 0.084)	0 (0 to 0)	5.245 (4.879 to 5.612)
Australasia	44 (32 to 61)	0.31 (0.22 to 0.42)	-1.01 (-1.19 to -0.83)	2 (2 to 3)	0.02 (0.01 to 0.02)	-1.76 (-1.90 to -1.61)	0.106 (0.074 to 0.16)	0.001 (0.001 to 0.001)	0.894 (0.689 to 1.099)
Caribbean	231 (178 to 305)	0.96 (0.74 to 1.27)	1.09 (0.87 to 1.31)	31 (22 to 45)	0.13 (0.09 to 0.19)	1.48 (1.34 to 1.62)	1.263 (0.949 to 1.634)	0.005 (0.004 to 0.007)	2.536 (2.321 to 2.752)
Central Asia	458 (366 to 561)	0.94 (0.75 to 1.15)	0.30 (-0.36 to 0.96)	82 (60 to 108)	0.17 (0.12 to 0.22)	0.49 (-0.13 to 1.11)	0.671 (0.497 to 0.92)	0.001 (0.001 to 0.002)	4.057 (3.462 to 4.656)
Central Europe	405 (331 to 486)	0.77 (0.63 to 0.92)	-2.24 (-2.52 to -1.96)	71 (52 to 90)	0.13 (0.10 to 0.17)	-1.46 (-1.74 to -1.18)	4.451 (3.324 to 6.207)	0.008 (0.006 to 0.012)	0.540 (-0.235 to 1.320)
Central Latin America	1,243 (1,006 to 1,499)	0.93 (0.75 to 1.13)	1.72 (1.42 to 2.01)	94 (71 to 122)	0.07 (0.05 to 0.09)	-0.89 (-1.28 to -0.50)	2.315 (1.753 to 2.987)	0.002 (0.001 to 0.002)	0.033 (-0.329 to 0.396)
Central Sub-Saharan Africa	335 (228 to 479)	0.51 (0.35 to 0.73)	0.29 (0.17 to 0.40)	46 (28 to 75)	0.07 (0.04 to 0.11)	-0.51 (-0.65 to -0.37)	1.471 (0.703 to 2.691)	0.002 (0.001 to 0.004)	2.432 (2.095 to 2.771)
East Asia	5,154 (3,760 to 6,858)	0.75 (0.55 to 1.00)	2.95 (2.68 to 3.22)	1,506 (999 to 2,137)	0.22 (0.15 to 0.31)	2.09 (1.79 to 2.39)	4.061 (2.886 to 5.574)	0.001 (0 to 0.001)	3.631 (3.340 to 3.922)
Eastern Europe	1,099 (877 to 1,346)	1.14 (0.91 to 1.4)	-0.39 (-1.34 to 0.56)	212 (157 to 279)	0.22 (0.16 to 0.29)	-0.26 (-1.05 to 0.53)	5.577 (4.198 to 7.326)	0.006 (0.004 to 0.008)	-0.211 (-1.288 to 0.878)
Eastern Sub-Saharan Africa	333 (256 to 433)	0.16 (0.12 to 0.21)	0.32 (0.19 to 0.46)	55 (36 to 80)	0.03 (0.02 to 0.04)	0.14 (0.01 to 0.27)	1.483 (0.791 to 2.979)	0.001 (0 to 0.001)	2.326 (2.115 to 2.538)
High-income Asia Pacific	196 (159 to 242)	0.25 (0.20 to 0.31)	-0.64 (-0.79 to -0.49)	27 (20 to 36)	0.03 (0.03 to 0.05)	-3.46 (-3.70 to -3.21)	0.330 (0.258 to 0.421)	0 (0 to 0.001)	0.378 (0.127 to 0.631)
High-income North America	1,419 (1,170 to 1,719)	0.84 (0.69 to 1.02)	0.41 (-0.01 to 0.83)	88 (67 to 111)	0.05 (0.04 to 0.07)	0.70 (0.30 to 1.11)	11.215 (8.826 to 13.828)	0.007 (0.005 to 0.008)	2.846 (2.246 to 3.449)
North Africa and Middle East	5,795 (4,574 to 7,213)	1.73 (1.37 to 2.16)	1.89 (1.73 to 2.04)	1,149 (800 to 1,538)	0.34 (0.24 to 0.46)	1.81 (1.71 to 1.92)	3.300 (2.275 to 4.669)	0.001 (0.001 to 0.001)	4.277 (3.774 to 4.782)
Oceania	193 (147 to 254)	2.72 (2.08 to 3.59)	0.92 (0.78 to 1.07)	7 (5 to 11)	0.10 (0.07 to 0.16)	-0.12 (-0.29 to 0.05)	0.156 (0.078 to 0.271)	0.002 (0.001 to 0.004)	3.281 (2.757 to 3.808)
South Asia	16,343 (13,111 to 19,940)	1.62 (1.30 to 1.98)	1.75 (1.64 to 1.85)	796 (528 to 1,265)	0.08 (0.05 to 0.13)	0.77 (0.70 to 0.84)	4.491 (2.434 to 8.653)	0 (0 to 0.001)	3.195 (3.085 to 3.304)
Southeast Asia	2,856 (2,297 to 3,454)	0.77 (0.62 to 0.93)	1.55 (1.39 to 1.71)	494 (315 to 700)	0.13 (0.09 to 0.19)	1.71 (1.61 to 1.80)	2.170 (1.496 to 3.199)	0.001 (0 to 0.001)	3.358 (3.074 to 3.642)
Southern Latin America	111 (90 to 135)	0.32 (0.26 to 0.39)	-0.64 (-0.76 to -0.52)	9 (7 to 12)	0.03 (0.02 to 0.03)	-1.93 (-2.04 to -1.82)	0.197 (0.148 to 0.26)	0.001 (0 to 0.001)	1.677 (1.100 to 2.257)
Southern Sub-Saharan Africa	151 (121 to 191)	0.35 (0.28 to 0.44)	0.85 (0.50 to 1.20)	28 (19 to 38)	0.06 (0.04 to 0.09)	1.01 (0.55 to 1.47)	3.683 (2.591 to 4.981)	0.009 (0.006 to 0.012)	2.485 (2.312 to 2.658)
Tropical Latin America	861 (681 to 1,066)	0.72 (0.57 to 0.89)	0.10 (0.03 to 0.17)	112 (84 to 145)	0.09 (0.07 to 0.12)	-2.14 (-2.67 to -1.60)	5.551 (4.253 to 7.12)	0.005 (0.004 to 0.006)	0.874 (0.488 to 1.262)
Western Europe	458 (383 to 542)	0.24 (0.20 to 0.29)	-1.86 (-2.04 to -1.69)	33 (26 to 43)	0.02 (0.01 to 0.02)	-3.15 (-3.23 to -3.07)	3.856 (2.996 to 4.866)	0.002 (0.002 to 0.003)	0.976 (0.423 to 1.533)
Western Sub-Saharan Africa	526 (400 to 689)	0.23 (0.17 to 0.3)	1.11 (0.94 to 1.28)	191 (125 to 286)	0.08 (0.05 to 0.12)	1.02 (0.88 to 1.16)	1.938 (1.007 to 3.836)	0.001 (0 to 0.002)	2.574 (2.456 to2.692)
SDI levels
Low SDI	3,101 (2,495 to 3,761)	0.57 (0.46 to 0.69)	0.67 (0.54 to 0.79)	396 (266 to 608)	0.07 (0.05 to 0.11)	0.64 (0.50 to 0.78)	3.506 (1.771 to 6.411)	0.001 (0 to 0.001)	2.104 (1.913 to 2.295)
Low-middle SDI	12,845 (10,641 to 15,504)	1.26 (1.05 to 1.53)	1.93 (1.83 to 2.04)	1,207 (832 to 1,667)	0.12 (0.08 to 0.16)	1.45 (1.39 to 1.52)	7.295 (5.010 to 10.799)	0.001 (0 to 0.001)	3.563 (3.494 to 3.632)
Middle SDI	13,531 (11,180 to 16,129)	1.08 (0.89 to 1.29)	2.15 (2.09 to 2.22)	1,864(1,374 to 2,443)	0.15 (0.11 to 0.19)	1.33 (1.23 to 1.42)	16.598 (13.108 to 20.954)	0.001 (0.001 to 0.002)	2.021 (1.919 to 2.123)
High-middle SDI	5,163 (4,193 to 6,349)	0.82 (0.67 to 1.01)	0.81 (0.42 to 1.20)	1,118 (809 to 1,492)	0.18 (0.13 to 0.24)	0.64 (0.26 to 1.02)	13.738 (10.667 to 17.429)	0.002 (0.002 to 0.003)	0.266 (-0.420 to 0.956)
High SDI	3,624 (2,982 to 4,329)	0.72 (0.59 to 0.86)	0.54 (0.40 to 0.68)	458 (335 to 600)	0.09 (0.07 to 0.12)	0.68 (0.51 to 0.84)	17.080 (13.763 to 20.657)	0.003 (0.003 to 0.004)	1.987 (1.459 to 2.518)

CI: Confidence Interval, EAPC: Estimated Annual Percentage Change (measures trend in crude death rates over time, calculated using a generalized linear model with Gaussian distribution and log-link function),GBD: Global Burden of Disease, IHD: Ischemic Heart Disease, IS: Ischemic Stroke, LEPAD: lower extremity peripheral artery disease, FPG: fasting plasma glucose, SDI: Socio-demographic Index (composite index 0-1 based on lag-distributed income per capita, average educational attainment over age 15, and total fertility rate under 25, used to categorize countries into development levels),UI: Uncertainty Interval (95% uncertainty interval around point estimates, accounting for uncertainty from input data, model parameters, and modeling choices)

Fig. 1.

Trends in the number of deaths and crude death rate for atherosclerotic cardiovascular disease attributable to high low density lipoprotein cholesterol (A), high body mass index (B), or high fasting plasma glucose (C) among young and middle-aged adults from 1990 to 2021. For each panel, bars represent the number of deaths (left y-axis) and lines represent crude death rates (right y-axis). Disease-specific color coding: ischemic heart disease (dark blue), ischemic stroke (orange), and lower extremity peripheral arterial disease (light red)

Regional and national trends and burden

Across 21 GBD regions, from 1990 to 2021, the EAPCs for crude death rates attributable to different risk factors exhibited substantial heterogeneity in young and middle-aged adults. East Asia and Southeast Asia consistently experienced pronounced increases in crude death rates for IHD or IS attributable to nearly all three risk factors in young and middle-aged adults. Conversely, Central Europe, Western Europe, and Australasia achieved substantial declines across nearly all three risk factors. Significant increases in LEPAD burden attributable to high FPG or LDL-c in young and middle-aged adults were observed in Andean Latin America, East Asia, and North Africa and the Middle East, with no decreasing trends across regions (Tables 1, 2 and 3). In 2021, substantial variations were observed in the number of deaths and crude death rates for ASCVD attributable to three risk factors in young and middle-aged adults, as shown in Tables 1, 2 and 3.

Across 204 countries and territories, from 1990 to 2021, the EAPCs for crude death rates attributable to different risk factors also exhibited substantial heterogeneity in young and middle-aged adults. For high LDL-c, the highest EAPC for IHD was in Zimbabwe (4.63, 95% CI: 3.67 to 5.59) and the lowest in Estonia (–6.56, 95% CI:–7.10 to –6.01) (Fig. 2A), while for IS, the highest was in Zimbabwe (4.87, 95% CI:3.80to 5.95) and the lowest in Luxembourg (–8.17, 95% CI: –8.43 to –7.92) (Fig. 2B). For high BMI, the highest EAPC for IHD was in Zimbabwe (6.75, 95% CI:5.62to 7.91) and the lowest in Estonia (–5.95, 95% CI:–6.49 to –5.41) (Fig. 2C); for IS, the highest was in Zimbabwe (7.26, 95% CI: 5.96 to 8.57) and the lowest in Luxembourg (–7.13, 95% CI: –7.40 to –6.86) (Fig. 2D); and for LEPAD, the highest was in Jordan (9.64, 95% CI:8.44 to 10.85) and the lowest in Norway (–4.78, 95% CI:–5.30 to –4.25) (Fig. 2E). For high FPG, the highest EAPC for IHD was in Lesotho (6.18, 95% CI:5.34 to 7.03) and the lowest in Norway (–5.13, 95% CI:–5.30 to –4.95) (Fig. 2F); for IS, the highest was in Libya (6.07, 95% CI:5.70 to 6.44) and the lowest in Luxembourg (–6.41, 95% CI: –6.68 to –6.15) (Fig. 2G); and for LEPAD, the highest was in Georgia (19.90, 95% CI:16.88 to 23.00) and the lowest in Norway (–5.48, 95% CI: –6.00 to –4.95) (Fig. 2H).

Fig. 2.

Global estimated annual percentage changes of crude death rates for atherosclerotic cardiovascular disease attributable to high low density lipoprotein cholesterol (A-B), high body mass index (C-E), or high fasting plasma glucose (F–H) among young and middle-aged adults across 204 countries and territories from 1990 to 2021

Socio-demographic index relationships

Among various SDI regions from 1990 to 2021, crude ASCVD death rates attributable to three risk factors in young and middle-aged adults showed substantial increases or modest declines in low to middle SDI regions and significant decreases or minimal increases in high-middle to high SDI regions. Specifically, crude death rates for IHD or IS attributable to high LDL-c decreased most markedly in high SDI regions but increased notably in middle SDI regions (Table 1). Similarly, IHD attributable to high BMI fell most sharply in high SDI regions and rose most substantially in middle SDI regions. Additionally, all SDI regions exhibited rising trends in IS or LEPAD attributable to high BMI, with the greatest increase in IS seen in middle SDI regions and the largest rise in LEPAD occurring in low-middle SDI regions (Table 2). All SDI regions also experienced increases in crude death rates for ASCVD attributable to high FPG. IHD rose most sharply in middle SDI regions, whereas IS and LEPAD showed larger increases in low-middle SDI regions. Conversely, high SDI regions exhibited the smallest increase in IHD, while high-middle SDI regions experienced more moderate rises in IS and LEPAD (Table 3). In 2021, the highest crude death rates for IHD and IS attributable to high LDL-c in young and middle-aged adults were in low-middle and high-middle SDI regions, respectively, with the lowest rates in high SDI regions (Table 1). IHD attributable to high BMI was highest in low-middle SDI regions, while IS was highest in high-middle SDI regions, and LEPAD attributable to high FPG was highest in high SDI regions (Tables 2 and 3).

Across 21 regions from 1990 to 2021, the LOESS-smoothed curves (span = 0.5) suggested a complex nonlinear association between regional SDI levels and the crude death rates for all three risk factors in young and middle-aged adults. The curves for IHD attributable to three risk factors were characterized by two relative peaks in crude death rates at SDI levels near 0.45 and 0.70, giving the overall curve an approximate inverted M-shaped appearance. In contrast, the curves for IS and LEPAD attributable to three risk factors were characterized by a sharp initial rise followed by a sustained plateau at higher SDI levels, broadly resembling an inverted L-shape, with the transition plateauing around an SDI of 0.70–0.76. Eastern Europe and Central Europe consistently bore higher-than-expected burdens, whereas Australasia and Western Europe demonstrated significantly lower-than-expected burdens (Fig. 3A-H). Across 204 countries in 2021, the LOESS-smoothed curves (span = 0.5) indicated that the relationship between national SDI levels and crude death rates for ASCVD attributable to three risk factors in young and middle-aged adults was also nonlinear. Under our smoothing parameter, crude death rates tended to increase gradually at SDI levels below 0.60. Within the approximate SDI range of 0.60 to 0.75, the curve exhibited a concave-upward segment, giving the impression of a slight U-shape in this interval. Beyond an SDI of 0.75, crude death rates showed a sharp declining trend (Fig. 4A-H).

Fig. 3.

Atherosclerotic cardiovascular disease attributable to high low density lipoprotein cholesterol (A-B), high body mass index (C-E), or high fasting plasma glucose (F–H) among young and middle-aged adults across 21GBD regions by socio-demographic index levels,1990–2021.For each region, points from left to right depict estimates from each year from 1990 to 2021. The observed patterns represent descriptive summaries of the data rather than deterministic mathematical functions. Smoothed lines represent locally weighted scatterplot smoothing (LOESS) regression with span = 0.5

Fig. 4.

Atherosclerotic cardiovascular disease attributable to high low density lipoprotein cholesterol (A-B), high body mass index (C-E), or high fasting plasma glucose (F–H) among young and middle-aged adults across 204 countries by socio-demographic index levels, 2021. The observed patterns represent descriptive summaries of the data rather than deterministic mathematical functions. Smoothed lines represent locally weighted scatterplot smoothing (LOESS) regression with span = 0.5

Predictive analysis to 2050

Projections indicate varying increases in both number of deaths and crude death rates for ASCVD attributable to three risk factors in young and middle-aged adults by 2050 (Fig. 5). Specifically, number of deaths for IHD attributable to high LDL-c are forecast to increase from 325,795 in 2022 to 515,777 by 2050, with crude death rates climbing from 11.60 to 15.36 per 100,000 populations; number of deaths for IS attributable to high LDL-c are expected to grow from 28,342 to 36,491, while crude death rates will first decrease from 1.01 per 100,000 populations in 2022 to 0.92 per 100,000 populations by 2038 before rising to 1.07 per 100,000 populations by 2050 (Fig. 5A). For high BMI, number of deaths for IHD are projected to rise from 104,835 to 299,488, with crude death rates increasing from 3.03 per 100,000 populations in 2022 to 7.26 per 100,000 populations by 2050; number of deaths for IS are expected to increase from 7,915 to 17,304, with crude death rates rising from 0.23 per 100,000 population in 2022 to 0.46 per 100,000 populations by 2050; and number of deaths for LEPAD are predicted to grow from 147 to 1067, accompanied by crude death rates increase from 0.006 to 0.037 per 100,000 populations (Fig. 5B). Regarding high FPG, number of deaths for IHD are anticipated to surge from 39,188 to 59,279, with crude death rates escalating from 1.38 to 1.72 per 100,000 populations; number of deaths for IS are forecast to rise from 5,125 to 7,928, with crude death rates increasing from 0.18 per 100,000 populations in 2022 to 0.23 per 100,000 populations by 2050; and number of deaths for LEPAD are expected to increase from 58 to 376, with crude death rates climbing from 0.002 to 0.013 per 100,000 populations (Fig. 5C).

Fig. 5.

Trends in the number of deaths and crude death rates for atherosclerotic cardiovascular disease attributable to high low density lipoprotein cholesterol (A), high body mass index (B), or high fasting plasma glucose (C) among young and middle-aged adults from 2022 to 2050. For each panel, bars represent the number of deaths (left y-axis) and lines represent crude death rates (right y-axis). Disease-specific color coding: ischemic heart disease (dark blue), ischemic stroke (orange), and lower extremity peripheral arterial disease (light red)

Discussion

This study presents a comprehensive and forward-looking analysis of the global burden of ASCVD attributable to three pivotal metabolic risk factors in young and middle-aged adults over three decades from 1990 to 2021, with projections extending to 2050. The burden attributable to these risk factors, while distinct in their pathophysiological pathways, demonstrate broadly similar epidemiological patterns and trajectories, underscoring their collective and often synergistic role in driving cardiovascular mortality. At regional SDI levels, crude death rates for IHD showed an inverted M-shaped pattern (peaks at SDI 0.45 and 0.70), while IS and LEPAD demonstrated inverted L-shaped patterns (plateauing at SDI 0.70–0.76). Across 204 countries in 2021, crude death rates increased gradually below SDI 0.60, showed a slight U-shape between 0.60–0.75, and declined sharply above 0.75. Projections indicate that ASCVD attributable to three risk factors will increase to varying degrees in young and middle-aged adults by 2050.

The broadly similar burden patterns observed for these risk factors are a central finding of this study. This homogeneity suggests that these risk factors are not isolated drivers but are deeply intertwined, frequently clustering within individuals and populations due to shared underlying societal and lifestyle determinants. While our analysis quantifies the individual contributions of high BMI, FPG, or LDL-c to ASCVD burden, we acknowledge that these metabolic risk factors are not independent in their pathophysiological effects. The clustering of these factors through insulin resistance creates complex interactions that the current GBD methodology does not fully capture. The sum of individual attributable fractions likely overestimates the true combined metabolic burden, as these factors act in concert rather than additively. Therefore, our estimates should be interpreted as the upper bound of potential prevention impact, recognizing that risk factor overlap may lead to some degree of overestimation when viewed in isolation. Future research employing mediation analysis or causal inference methods could better elucidate the synergistic versus independent contributions of these interconnected metabolic pathways. The globalization of unhealthy diets [17] (processed, convenience, high-energy, and fast foods), widespread physical inactivity [18], and increasing urbanization [19] are common drivers fueling the rise of dyslipidemia, obesity, and hyperglycemia simultaneously. Pathophysiologically, these risks act in concert: high BMI is frequently associated with insulin resistance, which in turn can exacerbate atherogenic dyslipidemia, including high LDL-c [20]. This synergy creates a potent mix that accelerates the development and progression of atherosclerosis in young adults. Our results show that IHD constitutes the overwhelming majority of attributable deaths for all three risk factors, followed by IS and then LEPAD. This hierarchy aligns with the existing studies [21] indicating that, in ASCVD, IHD tend to have higher mortality rates than IS or LEPAD. The notably lower recorded burden for LEPAD may also be partly attributable to its often asymptomatic nature in early stages and significant challenges in diagnosis [22] compared to the more acute and easily identified events of IHD and IS.

The global trends in crude death rates for ASCVD attributable to the three risk factors in young and middle-aged adults have remained relatively stable. Specifically, the diverging trends for high LDL-c compared to high BMI or FPG highlight progress in the widespread adoption and effectiveness of lipid-lowering therapies, such as statins, over recent decades [23]. Conversely, the rising death rate associated with high BMI or FPG points to the escalating global epidemics of obesity and diabetes [24], which are outpacing current prevention and control efforts. The observed sex-specific differences in crude death rate trajectories from 1990 to 2021 suggest potentially variations in biological risk, healthcare access, or behavioral factors between men and women [25–27]. Furthermore, the sustained higher burden experienced by men in 2021 necessitates targeted public health strategies. Therefore, a focused response is essential: public health efforts must strengthen the prevention and control of obesity and diabetes to reverse their rising mortality impact, while consolidating the progress achieved in cholesterol management through sustained lipid-lowering strategies. Simultaneously, gender-specific interventions should be developed and implemented to address the persistently higher ASCVD burden among young and middle-aged men.

The substantial regional and national heterogeneity in crude ASCVD death rates attributable to three risk factors in young and middle-aged adults from 1990 to 2021 reflects a complex interplay of differential epidemiological transitions, health system capacities, and policy implementations. East Asia and Southeast Asia’s pronounced increases in IHD and IS mortality attributable to three risk factors likely reflect rapid economic growth driving nutrition transitions toward processed diets [28], reduced physical activity [29], and elevated metabolic risk profiles [24], compounded by healthcare systems historically oriented toward infectious disease management that have struggled to scale up chronic disease prevention [30]. Conversely, Central Europe, Western Europe, and Australasia’s substantial declines likely reflect decades of healthcare system strengthening [31], and evidence-based policy interventions [32].At the national level, Zimbabwe and Lesotho’s extreme crude death rate increases for multiple risk factors are driven by political instability, healthcare fragmentation, and resource constraints in low-income settings [33, 34]. Conversely, the notable progress observed in Estonia and Luxembourg is attributable to the vigorous efforts of their formidable healthcare systems in the prevention, treatment, and rehabilitation of disease [35–38]. When further stratifying these geographic patterns by SDI levels, low to middle SDI regions exhibited the most concerning trends, with crude ASCVD death rates showing substantial increases or only modest declines. LOESS analysis further supported this finding, revealing IHD death rates with dual peaks at SDI ~ 0.45 and 0.70 (inverted M-shape) and IS/LEPAD rates plateauing at SDI 0.70–0.76 (inverted L-shape) across SDI regions from 1990–2021, and a slight U-shape between SDI 0.60–0.75 in 204 countries in 2021. These patterns may reflect rapid urbanization, the "nutrition transition" towards diets high in ultra-processed foods, sugars, and unhealthy fats, and increasingly sedentary occupations and lifestyles [39–41]. Compounding this, these regions and countries often grapple with healthcare systems that are under-resourced, fragmented, and overwhelmingly focused on acute care and infectious diseases [42, 43], leaving little capacity for the long-term management of chronic cardiometabolic conditions. Furthermore, issues of access and affordability to essential medicines and healthy foods create insurmountable barriers for large segments of the population [44, 45]. In stark contrast, high-middle to high SDI regions exhibited the most encouraging trends, with crude ASCVD death rates showing substantial increases or only modest declines. These findings were further substantiated by LOESS analysis. High-income regions and countries have long benefited from robust public health initiatives, improved access to effective treatments, and widespread preventive care [46–49]. These regions and countries have seen declining death rates due to the successful implementation of comprehensive health policies [46]. For example, the widespread use of statins and anti-diabetic drugs has significantly reduced mortality from CVD [47–49]. Additionally, high-income regions and countries have benefited from advanced healthcare infrastructure, well-trained healthcare professionals, and comprehensive health insurance systems that ensure access to care for all segments of the population [46]. To address these disparities, key public health priorities should including strengthening primary care and chronic disease management, particularly in low- to middle-SDI settings; implementing multisectoral policies to promote healthy diets and physical activity; ensuring affordable access to essential cardiovascular medicines within universal health coverage frameworks; enhancing health information systems for targeted surveillance; and fostering global collaboration to share effective prevention strategies.

The predictive analysis indicates a significant increase in the number of deaths and death rates for ASCVD attributable to three risk factors in young and middle-aged adults by 2050, which may be associated with the anticipated continuous increase in these risk factors [50–52] and the projected expansion of the young and middle-aged population. However, the drivers of this increase vary substantially across SDI levels. In low- to middle-SDI regions, rising number of deaths are driven by population growth in young and middle-aged adults [53] combined with intensifying risk exposure [52]. In contrast, high- and high-middle SDI regions face a more concerning pattern. Despite plateauing or declining young and middle-aged adults due to aging and sub-replacement fertility, absolute death counts continue to rise [53], indicating that worsening metabolic risk profiles are outpacing demographic contraction. This alarming projection underscores the urgent need for public health measures to address these risk factors. Effective interventions should focus on promoting healthy lifestyles, including balanced diets, regular physical activity, and obesity control. Additionally, expanding healthcare infrastructure and workforce capacity is crucial to accommodate growing young and middle-aged populations in low- to middle-SDI regions and ensure early detection and management of these risk factors. In high- and high-middle SDI regions, healthcare systems must prioritize targeted metabolic risk reduction programs to offset the impact of deteriorating risk profiles amid demographic contraction. Public health campaigns aimed at raising awareness about the dangers of three risk factors, and encouraging regular health check-ups, can play a pivotal role in reducing the burden of ASCVD in young and middle-aged adults. Moreover, policy initiatives that support the availability and affordability of nutritious foods, as well as the creation of safe and accessible spaces for physical activity, are essential components of a comprehensive strategy. Strengthening health systems to provide comprehensive care for individuals with these risk factors, including access to lipid-lowering therapies, diabetes management, and weight management programs, will also be critical in mitigating the projected rise in ASCVD burden in young and middle-aged adults.

Limitations

This study has several limitations. First, these estimates, derived using the Cause of Death Ensemble model (CODEm), depend on the quality and availability of input data from vital registration systems, verbal autopsies, and other sources, which may vary in availability, diagnostic accuracy, and coding consistency [6–10]. Second, the calculation of attributable burden is sensitive to the TMREL and risk-outcome definitions, which are based on current epidemiological evidence and may evolve; additionally, the analysis assumes risk factors are independent, and PAF-based estimates cannot fully account for confounding, mediation, reverse causation, or complex interactions [6–10]. Third, LEPAD is likely under-ascertained, particularly in asymptomatic stages and in settings without routine specialized testing [22], potentially leading to underestimated prevalence, especially in low- and middle-income countries. Fourth, The “M-shaped” or “L-shaped” patterns was based on descriptive smoothing techniques (LOESS). While these patterns serve as a useful visual and descriptive characterization of the data, their exact morphology should not be overinterpreted as a definitive functional relationship, but rather as a reflection of underlying complex associations. Fifth, GBD 2021 does not provide data disaggregated by race or ethnicity, precluding subgroup analyses to examine disparities across these demographics. We acknowledge that understanding racial and ethnic differences in ASCVD burden would be of significant scientific and public health value, and hope that future iterations of the GBD study or other population-based datasets may enable such investigations. Finally, the BAPC model for prediction assumes that past trends will continue and cannot account for unexpected technological breakthroughs, dramatic policy shifts, or pandemics that could alter the trajectory of ASCVD burden. [10].

Conclusion

These findings underscore the urgent need for integrated, coordinated strategies targeting three risk factors to curb the projected rise in ASCVD burden among young and middle-aged adults by 2050. Public health efforts must prioritize obesity and diabetes prevention while sustaining cholesterol management gains, with tailored interventions addressing persistent sex disparities and the unique challenges faced by low- to middle-SDI regions.

Author contributions

Hua Zhang and Jun Li contributed equally to the conceptualization, methodology, and original draft preparation of this manuscript. Kaixin Qin was instrumental in data analysis and interpretation. Yuan Zhao, Qi Yan, and Wenhao Wu provided valuable contributions in data collection and initial manuscript review. Jin Li oversaw the study, offering critical feedback and approval of the final manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (No.82270915), Special Fund for Science and Technology Innovation Teams of Shanxi Province (No. 202304051001021), Shanxi Funding for High Level Overseas Returns (2022), Shanxi Province Higher Education "Billion Project" Science and Technology Guidance Project (BYJL010) and Basic Research Program of Shanxi (202103021224425).

Data availability

The data resources for the GBD study 2021 were accessible online through the Global Health Data Exchange (GHDx) query tool ( [http://ghdx.healthdata.org/gbd-results-tool](http:/ghdx.healthdata.org/gbd-results-tool)).

Declarations

Ethics approval and consent to participate

Not applicable. This study utilized publicly available, de-identified data from the Global Burden of Disease Study 2021, which is exempt from ethical approval and consent requirements.

Consent for publication

Not applicable. This manuscript does not contain any individual person’s data in any form.

Competing interests

The authors declare no competing interests.