Trends and disparities in risk factors of ischemic stroke among young and middle-aged adults (20–54 years) globally from 1990 to 2021: A population-based study

Abstract

This study aimed to analyze the global disease burden, temporal trends of ischemic stroke (IS) in adults aged 20 to 54 years (1990–2021), and temporal disparities in attributable risk factors. Using Global Burden of Disease 2021 data, we extracted estimates with 95% uncertainty intervals for IS prevalence, incidence, mortality, and disability-adjusted life years (DALYs) among adults aged 20 to 54 years. Trends were assessed globally and locally, stratified by sex and age subgroup. Age-period-cohort models estimated age, period, and cohort effects on prevalence and incidence, and population-attributable fractions were used to evaluate temporal changes in risk factor contributions. In 2021, adults aged 20 to 54 years had 16,301,644 prevalent IS cases and 1,324,743 incident cases worldwide, with rates of 432.46 and 35.14 per 100,000 population, respectively. IS caused 105,859 deaths and 6,967,488 DALYs, corresponding to mortality and DALY rates of 2.81 and 184.84 per 100,000 population. From 1990 to 2021, prevalence and incidence increased, whereas mortality and DALYs declined. Prevalence was higher among females, while incidence, mortality, and DALYs were higher among males. Marked heterogeneity was observed across socio-demographic index regions, Global Burden of Disease, Injuries, and Risk Factors Study regions, and countries. Greater burdens were concentrated in high–middle socio-demographic index regions, Eastern Europe, Central Asia, and parts of North Africa and the Middle East, while East Asia experienced the largest increase in incidence. Joinpoint regression showed fluctuating but overall increasing trends in prevalence and incidence, alongside sustained declines in mortality and DALY rates after the early 2000s. Age-period-cohort analysis indicated that the risks of IS prevalence and incidence increased with age, particularly after 35 years. Period effects peaked in 2017 to 2021, whereas cohort effects declined progressively from the 1942 to 1946 to the 1997 to 2001 birth cohorts. High low-density lipoprotein cholesterol, high systolic blood pressure, particulate matter pollution, smoking, and high body mass index were the leading contributors to IS-related DALYs. From 1990 to 2021, the population-attributable fractions of particulate matter pollution, smoking, and high-sodium diet decreased, whereas those of metabolic risks increased, particularly high systolic blood pressure and high body mass index. IS imposes a heavy global burden on young and middle-aged adults with an upward trend driven by metabolic, environmental, and lifestyle factors. Early screening and targeted risk management are critical.

1. Introduction

Stroke remains one of the leading causes of death and disability worldwide. According to the World Stroke Organization Global Stroke Fact Sheet 2025, stroke is the 2nd leading cause of death and the 3rd leading cause of death and disability combined among noncommunicable diseases globally.^[1,2] Although stroke temporarily declined in the global rankings of leading causes of death and disability due to the impact of COVID-19, the absolute number of individuals who have died from stroke or are living with stroke-related disability has increased substantially over the past 30 years.^[3] From 1990 to 2021, the absolute burden of stroke increased substantially, with a 70.0% increase in incident strokes, a 44.0% increase in stroke-related deaths, an 86.0% increase in prevalent stroke cases, and a 32.0% increase in disability-adjusted life years (DALYs). The global economic burden of stroke is also considerable, with the estimated global cost exceeding US$890 billion, accounting for approximately 0.66% of global gross domestic product.^[1] The World Stroke Organization-Lancet Neurology Commission estimates^[4] that between 2020 and 2050, the global burden of stroke, including DALYs, mortality, and economic costs, is expected to nearly double, placing an enormous strain on healthcare systems and economies worldwide. Among all incident strokes, ischemic stroke (IS) accounts for the largest proportion, with an estimated 7800,000 cases (95% uncertainty interval [UI]: 6,700,000–8,900,000), representing 65.3% (95% UI: 62.4–67.7) of total stroke cases. It is therefore considered the most common stroke subtype.^.[2]

Although IS is one of the most common conditions among older adults, the number of new cases in this population continues to rise in parallel with global population aging.^[5] However, increasing evidence indicates that IS is not confined to older adults. Stroke in younger populations has become an increasingly important public health issue, and the incidence of IS in young adults has shown an upward trend in several regions.^[6] Due to changes in lifestyle, increased work-related and psychological stress, and genetic susceptibility, IS has shown a concerning trend toward younger onset in recent years. Approximately 10% to 15% of all strokes occur in adults aged 18 to 50 years, and younger patients often show more heterogeneous etiologies and risk profiles than older patients.^[6] The rising incidence among young and middle-aged adults is particularly alarming and warrants urgent public health attention.^[7,8]

The risk profile of IS among young and middle-aged adults is complex and includes both traditional vascular risk factors and age-specific etiologies. Traditional risk factors, such as hypertension, dyslipidemia, diabetes mellitus, obesity, smoking, alcohol consumption, unhealthy diet, and physical inactivity, are increasingly common in younger adults with IS.^[3] In addition, work-related stress, sleep disturbance, and environmental exposures, including particulate matter pollution, may further contribute to cerebrovascular risk in this population.^[2,3] Compared with older adults, young patients may also have more diverse and less common causes, including cervical artery dissection, patent foramen ovale, hypercoagulable states, autoimmune or inflammatory vasculopathies, migraine, pregnancy or puerperium-related hypercoagulability, oral contraceptive use, illicit drug use, and genetic susceptibility.^[9] Traditional metabolic and behavioral risk factors may promote atherosclerosis, endothelial dysfunction, vascular inflammation, and small-vessel disease, whereas young-adult-specific etiologies such as cervical artery dissection, patent foramen ovale, and hypercoagulable states may lead to IS through arterial injury, embolism, or prothrombotic pathways.^[6,9]

As the backbone of both society and family, young and middle-aged adults often bear substantial physical, psychological, and economic pressures. Stroke in this population can severely impair their ability to work and function normally, causing significant distress to families. Moreover, it increases the societal burden by necessitating greater investments in stroke prevention, treatment, and long-term care.^[10] Young stroke patients often require a longer recovery period and are at increased risk of developing a range of complications, which can severely compromise their quality of life.^[11] Because affected individuals are often in their most economically productive years, IS in this age group may result in prolonged work absence, unemployment, reduced income, productivity losses, long-term rehabilitation needs, and increased dependence on family or social care. A societal-perspective study in Australia estimated that stroke among adults aged 18 to 64 years generated an economic burden of AUD 2.0 billion over 5 years and AUD 3.4 billion over 30 years, indicating substantial long-term economic and health losses among younger stroke survivors.^[12] Evidence from China also suggests a heavy economic burden of stroke; a cost-of-illness study estimated that, in 2018, direct costs reached ¥247.8 billion and indirect costs reached ¥704.4 billion.^[13]

Although several Global Burden of Disease, Injuries, and Risk Factors Study (GBD)-based studies have evaluated the global burden of stroke or IS in younger populations, important questions remain insufficiently addressed. Previous studies have provided valuable evidence on the burden and risk factors of IS among young adults aged 15 to 49 years using GBD 2019 data,^[14] and more recent GBD 2021 analyses have assessed the burden and attributable risk factors of overall stroke and its major subtypes among young adults aged 15 to 49 years.^[15] However, evidence specifically focusing on the long-term disease burden of IS among young and middle-aged adults aged 20 to 54 remains limited. Notably, IS in this age group is often characterized by high disability rates, a significant risk of recurrence, and substantial losses in economic productivity. Moreover, existing evidence has not fully addressed the combined assessment of global, regional, and national disparities, segmented temporal trends, age-period-cohort effects, and temporal changes in attributable risk factors in this specific population. Therefore, a comprehensive assessment of the disease burden in this population, along with early and proactive strategies for prevention and management, is critically important. Using data from the GBD 2021, provided by the Institute for Health Metrics and Evaluation,^[16] this study characterized the temporal trends in the disease burden of IS among young and middle-aged adults aged 20 to 54 years from 1990 to 2021. We further examined temporal trends by sex, age subgroup, socio-demographic index (SDI) region, GBD region, and country; assessed age, period, and cohort effects; and evaluated regional and temporal variations in attributable risk factors. The aim was to provide population-specific evidence to support early prevention, targeted risk management, and health policy development for IS in young and middle-aged adults.

2. Methods

2.1. Data source

The GBD 2021 study used the latest available epidemiological data and refined standardization methods to comprehensively assess health loss associated with 371 diseases, injuries, and disabilities, as well as 88 risk factors across 204 countries and regions globally. This study extracted estimates of the incidence, prevalence, mortality, and DALYs for IS, along with their 95% UI, from GBD 2021.^[17,18] We also used the SDI, which is a composite measure based on per capita lagged income, average years of schooling for individuals aged 15 years and older, and fertility rates for women aged 25 and under. The SDI ranges from 0 to 1 and is divided into 5 quintiles, including low SDI, low–middle SDI, middle SDI, high–middle SDI, and high SDI groups.^[19] Additionally, the Institutional Review Board at the University of Washington waived the requirement for informed consent to access the GBD data.^[2] This study adhered to the Guidelines for Accurate and Transparent Health Estimates Reporting.^[20]GBD 2021 integrates multiple data sources and applies standardized modeling procedures to generate age-, sex-, location-, and year-specific estimates at global, regional, and national levels. These methods account for differences in data sources and potential biases, thereby improving comparability across locations, years, ages, and sexes. In addition, 95% UIs were generated for all estimates to reflect uncertainty propagated through the estimation process.

2.2. Disease criteria

The disease in this study is IS (International Classification of Diseases, 10th Revision codes I63–I63.9, I65–I66.9, I67.2–I67.848, and I69.3–I69.4). This study adopts the World Health Organization definition of stroke: a rapidly developing focal (or global) neurological dysfunction, with clinical signs lasting >24 hours or leading to death, without a clear cause other than a vascular origin.^[21] Ischemic stroke is defined as all atherosclerotic and thromboembolic events (excluding transient ischemic attacks) that result in impaired blood flow to brain tissue, leading to subsequent infarction.^[22] In addition to disease codes and clinical characteristics, diagnostic neuroimaging evidence is also used to support the diagnosis of ischemic stroke.

2.3. Study population, inclusion and exclusion criteria

The study population was defined as individuals aged 20 to 54 years with IS within the GBD 2021 framework. Estimates for the 20- to 54-year age group were directly extracted from the GBD 2021 results tool.

The inclusion criteria were as follows:

1. GBD 2021 estimates for ischemic stroke defined according to the corresponding International Classification of Diseases codes and GBD cause hierarchy;

2. Estimates for individuals aged 20 to 54 years from 1990 to 2021;

3. Estimates available at the global, SDI regional, GBD regional, or national level;

4. Burden indicators including prevalence, incidence, mortality, DALYs, and risk-attributable DALYs.

The exclusion criteria were as follows:

1. Estimates for age groups other than 20 to 54 years;

2. Stroke subtypes other than ischemic stroke, including intracerebral hemorrhage and subarachnoid hemorrhage;

3. Transient ischemic attacks without cerebral infarction;

4. Estimates outside the study period or not mapped to the GBD 2021 ischemic stroke definition.

2.4. Descriptive analysis

To quantify the global, regional, and national burden of IS among young and middle-aged adults aged 20 to 54 years, we conducted a comprehensive assessment of IS-related prevalence, incidence, mortality, and DALYs within this population. Temporal trends from 1990 to 2021 were evaluated through visualized analyses of the absolute number of cases, crude rates, and age-standardized rates for each burden metric. Comparative analyses were performed between 1990 and 2021 at multiple geographic levels, including globally, across the 5 SDI strata, within 21 GBD regions, and across 204 countries and territories. Furthermore, age-stratified analyses within each SDI region were conducted to elucidate the variation in disease burden across age subgroups within the 20 to 54 years range.

2.5. Trend analysis

Exploring the temporal trends of diseases is an essential component of epidemiology and plays a crucial role in the development of more targeted prevention strategies. To better reflect the time-based changes in the disease burden of IS among young and middle-aged adults aged 20 to 54 years, this study conducted trend analyses of IS prevalence, incidence, mortality, and DALYs from both global and regional perspectives.

First, this study used the estimated annual percentage change (EAPC) to quantify the overall trend in the burden of IS among young and middle-aged adults aged 20 to 54 years.^[23,24] They are summary measures widely used to assess the trend in disease burden over a specific time interval. The EAPC is calculated based on a regression model that describes the pattern of age-standardized rates during a particular period.^[25] The formula model is given by:$Y=\alpha +\beta X$ where $Y=\mathrm{l}\mathrm{n}\left(\mathrm{A}\mathrm{S}\mathrm{R}\right)$ and $X=\mathrm{y}\mathrm{e}\mathrm{a}\mathrm{r}$. The EAPC is then calculated using the formula:

$$\begin{array}{c}{\displaystyle }\mathrm{E}\mathrm{A}\mathrm{P}\mathrm{C}=(\mathrm{e}\mathrm{x}\mathrm{p}\left(\beta \right)-1)\times 100\%\end{array}$$

The 95% confidence interval (CI) for the EAPC is also derived from this model.

Subsequently, this study used the joinpoint regression model (JRM) to analyze the temporal trends in the disease burden of IS among individuals aged 20 to 54 years from 1990 to 2021. The model was used to calculate the average annual percent change as well as the APC for each period identified by the joinpoint regression. The fitting of the JRM and the calculation of the average annual percent change and APC were performed using the Joinpoint software (version 4.9.1; National Cancer Institute [NCI], Bethesda).^[26] The JRM is a method that establishes piecewise regression based on the temporal characteristics of disease data distribution, allowing for a more detailed assessment of disease trends across different intervals within a global time frame. This model identifies all possible joinpoints (segmentation points) by using a grid search approach, and the optimal number of joinpoints and corresponding model parameters are selected through a Monte Carlo permutation test.

Additionally, this study examined the trends in IS burden related to age, period, and cohort effects. First, the study explored the potential 2-way interactions between age, period, and birth cohort effects (Figs. S1 and S2, Supplemental Digital Content 1). Due to the interactions between age, period, and birth cohort, it is challenging to directly interpret the impact of real-time trends on incidence or prevalence. Therefore, this study further explored the temporal trends in IS incidence and prevalence across the 3 dimensions of age, period, and birth cohort using an APC model. The model was based on a Poisson distribution and accounted for linear relationships among age, period, and cohort effects. To mitigate potential estimation bias, the intrinsic estimator method was employed.^[27,28] The basic expression of the model is as follows:

$$\begin{array}{c}{\displaystyle }\mathrm{l}\mathrm{n}\left(R_{\mathrm{a}\mathrm{p}\mathrm{c}}\right)=\mathrm{I}\mathrm{n}(I_{\mathrm{a}\mathrm{p}\mathrm{c}}/N_{\mathrm{a}\mathrm{p}\mathrm{c}})=\mu +{\alpha }_a+{\beta }_p+{\gamma }_c\end{array}$$

In this model, it $\mathrm{I}\mathrm{n}\left(R_{\mathrm{a}\mathrm{p}\mathrm{c}}\right)$ represents the natural logarithm of the incidence rate; $I_{\mathrm{a}\mathrm{p}\mathrm{c}}$ denotes the number of incident cases; $N_{\mathrm{a}\mathrm{p}\mathrm{c}}$ is the total surveyed population; μ is the intercept; ${\alpha }_a$ represents the age effect for the a-th age group; ${\beta }_p$ represents the period effect for the p-th time period; and ${\gamma }_c$ represents the cohort effect for the c-th birth cohort. In the APC model used in this study, the data series was divided into consecutive 5-year intervals from 1990 to 2021. Data from the years 1990 to 1991 were excluded from the analysis, as they did not span a complete 5-year interval. At the same time, age was grouped into 7 categories based on consecutive 5-year intervals (20–54 years), and birth cohorts were divided into 12 groups spanning consecutive 5-year periods, ranging from 1942 to 1997.

2.6. Attributable risk factor analysis

The GBD 2021 study employed a theoretical framework of comparative risk assessment and used a counterfactual analysis approach to estimate the burden attributable to various risk factors. In this study, risk factors were selected based on clinical relevance and previously reported strong associations with IS in the literature.^[3,29] To estimate the attributable burden, GBD 2021 applies a rule-based synthesis of evidence to generate comparable risk estimates across populations and time periods.^[17] Population-attributable fractions were calculated for each risk factor. The sum of population-attributable fractions across all risk factors may exceed 100%, as many of these risk factors exert effects that are partially or fully mediated through 1 or more other risk factors.^[17] This study used DALYs to quantify the attributable burden of risk factors and conducted comparative analyses of these factors across different regions. In addition, temporal variation analyses of risk factors were performed to quantify changes in the risk profile for IS among young and middle-aged adults over time.

3. Results

3.1. Overall trend analysis of IS burden at the global, regional, and national levels

3.1.1. Global trends in the overall burden of IS

In 2021, IS continued to pose a substantial disease burden among young and middle-aged adults (20–54 years) worldwide, with 16,301,644 cases (95% UI: 14,638,249–17,962,503) and an age-specific prevalence rate of 432.46 per 100,000 population (95% UI: 1161.99–1359.92). From 1990 to 2021, the global prevalence of IS among young and middle-aged adults showed a gradual increase, with an EAPC of 0.22 (95% CI: 0.13–0.31). In 2021, there were 1,324,743 incident cases (95% UI: 1,065,145–1,629,298) of IS among young and middle-aged adults (20–54 years) globally, corresponding to an incidence of 35.14 per 100,000 population (95% UI: 28.26–43.22). From 1990 to 2021, IS incidence in this age group demonstrated a persistent upward trend, with an EAPC of 0.22 (95% CI: 0.06–0.37) (Table 1; Fig. S3, Supplemental Digital Content 2). Despite the increasing trends in both prevalence and incidence of IS among young and middle-aged adults from 1990 to 2021, a notable decline was observed in mortality and DALYs. In 2021, IS was estimated to have caused 105,859 deaths (95% UI: 94,891–118,402) in this population, corresponding to a mortality rate of 2.81 per 100,000 population (95% UI: 2.52–3.14). The EAPC in mortality was −0.67 (95% CI: −1.10 to −0.24), indicating a significant downward trend (Table 1). In 2021, the global DALYs for IS in young and middle-aged adults were estimated to be 6,967,488 cases (95% UI: 6,109,513–7,838,350), with a DALYs of 184.84 per 100,000 population (95% UI: 162.08–207.94). The EAPC in DALYs was −0.41 (95% CI: −0.62 to −0.20), reflecting a significant downward trend (Table 1). Globally, from 1990 to 2021, the incidence and prevalence of IS among young and middle-aged adults showed an overall upward trend, while mortality and DALYs exhibited a general downward trend. In terms of prevalence, women had higher rates than men, while for incidence, mortality, and DALYs, men had significantly higher rates than women (Fig. S3, Supplemental Digital Content 2).

Table 1.

Global burden of IS in 2021 and estimated annual percentage change (EAPC) from 1990 to 2021: Absolute numbers, annual rates of prevalence, incidence, mortality, and DALYs among young and middle-aged adults (20–54 years).

	Prevalence (95% uncertainty interval)			Incidence (95% uncertainty interval)			Mortality (95% uncertainty interval)			DALYs (95% uncertainty interval)
	Counts (n), 2021	Prevalence (per 100,000 population), 2021	EAPC (95% CI), 1990–2021	Counts (n), (2021)	Incidence (per 100,000 population), 2021	EAPC (95% CI), 1990–2021	Counts (n), (2021)	Mortality (per 100,000 population), 2021	EAPC (95% CI), 1990–2021	Counts (n), (2021)	DALYs (per 100,000 population), 2021	EAPC (95% CI), 1990–2021
Global	16,301,644 (14,638,249–17,962,503)	432.46 (388.33–476.52)	0.22 (0.13–0.31)	1,324,743 (1,065,145–1,629,298)	35.14 (28.26–43.22)	0.22 (0.06–0.37)	105,859 (94,891–118,402)	2.81 (2.52–3.14)	−0.67 (−1.10 to −0.24)	6,967,488 (6,109,513–7,838,350)	184.84 (162.08–207.94)	−0.41 (−0.62 to −0.20)
SDI quintiles
High SDI	2,395,956 (2,166,781–2,621,972)	464.05 (419.67–507.83)	0.04 (−0.04 to 0.13)	164,407 (131,030–203,978)	31.84 (25.38,39.51)	−0.52 (−0.88 to −0.16)	6260 (5602–6975)	1.21 (1.09–1.35)	−1.63 (−2.21 to −1.06)	602,359 (508,720–704,432)	116.67 (98.53–136.44)	−0.79 (−1.02 to −0.55)
High−middle SDI	3,184,086 (2,830,906–3,542,160)	486.87 (432.87–541.62)	0.20 (0.09–0.30)	278,502 (222,340–344,645)	42.59 (34.00–52.70)	0.18 (−0.03 to 0.39)	24,542 (21,582–28,052)	3.75 (3.30–4.29)	−1.40 (−2.04 to −0.75)	1,576,667 (1,378,578–1,790,944)	241.08 (210.79–273.85)	−0.94 (−1.32 to −0.55)
Middle SDI	5,387,073 (4,762,573–6,005,373)	438.15 (387.36–488.44)	0.62 (0.46–0.78)	454,328 (363,205–559,897)	36.95 (29.54–45.54)	0.90 (0.78–1.01)	40,454 (35,734–45,911)	3.29 (2.91–3.73)	−0.10 (−0.51 to 0.32)	2,563,131 (2,237,069–2,880,086)	208.47 (181.95–234.25)	0.08 (−0.11 to 0.28)
Low−middle SDI	3,569,818 (3,230,317–3,917,220)	389.87 (352.79–427.81)	0.26 (0.17–0.36)	291,422 (235,850–361,231)	31.83 (25.76–39.45)	0.29 (0.18–0.40)	25,102 (21,192–30,548)	2.74 (2.31–3.34)	−0.14 (−0.33 to 0.05)	1,580,512 (1,352,026–1,862,639)	172.61 (147.66–203.42)	−0.06 (−0.13 to 0.02)
Low SDI	1,752,085 (1,615,900–1,889,393)	388.39 (358.2–418.82)	−0.24 (−0.35 to −0.13)	135,048 (108,984–167,632)	29.94 (24.16–37.16)	−0.40 (−0.46 to −0.35)	9412 (7520–12,714)	2.09 (1.67–2.82)	−0.73 (−0.98 to −0.49)	639,127 (531,588–786,927)	141.68 (117.84–174.44)	−0.50 (−0.58 to −0.42)
Central Europe, eastern Europe, and central Asia
Central Asia	277,160 (262,617–291,800)	594.32 (563.14–625.72)	−0.06 (−0.25 to 0.13)	25,417 (21,545–30,692)	54.5 (46.20–65.81)	−0.14 (−0.26 to −0.01)	2102 (1820–2433)	4.51 (3.90–5.22)	−1.48 (−2.00 to −0.97)	138,789 (120,977–156,351)	297.61 (259.42–335.27)	−1.13 (−1.42 to −0.84)
Central Europe	231,659 (213,253–249,116)	423.67 (390.01–455.60)	−0.62 (−0.73 to −0.51)	19,904 (16,501–23,910)	36.40 (30.18–43.73)	−1.22 (−1.51 to −0.92)	1585 (1424–1747)	2.90 (2.60–3.20)	−2.53 (−3.28 to −1.77)	105,642 (92,177–117,954)	193.21 (168.58–215.72)	−1.94 (−2.37 to −1.50)
Eastern Europe	628,916 (551,117–710,033)	638.36 (559.39–720.69)	−0.13 (−0.24 to −0.02)	60,949 (48,722–75,942)	61.86 (49.45–77.08)	−0.49 (−0.77 to −0.20)	6206 (5567–6859)	6.30 (5.65–6.96)	−2.06 (−2.91 to −1.21)	371,177 (331,835–410,622)	376.75 (336.82–416.79)	−1.54 (−2.14 to −0.94)
High-income
Australasia	43,780 (40,883–46,747)	300.16 (280.30–320.50)	−0.10 (−0.28 to 0.08)	3231 (2588–3919)	22.15 (17.75–26.87)	−0.48 (−0.83 to −0.13)	41 (37–46)	0.28 (0.26–0.31)	−3.43 (−3.87 to −2.99)	7484 (5868–9065)	51.31 (40.23–62.15)	−1.30 (−1.48 to −1.12)
High-income Asia Pacific	396,438 (351,752–442,530)	471.09 (417.99–525.86)	−0.18 (−0.46 to 0.09)	36,095 (27,924–45,614)	42.89 (33.18–54.20)	−1.01 (−1.70 to −0.31)	579 (531–642)	0.69 (0.63–0.76)	−4.33 (−5.05 to −3.62)	83,739 (65,858–102,209)	99.51 (78.26–121.45)	−1.92 (−2.30 to −1.54)
High-income North America	975,119 (859,413–1,091,583)	580.14 (511.31–649.43)	0.27 (0.12–0.42)	47,005 (36,257–60,212)	27.97 (21.57–35.82)	−0.24 (−0.46 to −0.02)	1223 (1162–1285)	0.73 (0.69–0.76)	−0.83 (−1.27 to −0.39)	178,575 (145,014–213,778)	106.24 (86.28–127.19)	−0.13 (−0.21 to −0.04)
Southern Latin America	109,703 (103,088–117,293)	327.78 (308.01–350.46)	−0.61 (−0.70 to −0.53)	8981 (6872–11,363)	26.83 (20.53–33.95)	−1.38 (−1.71 to −1.06)	191 (177–207)	0.57 (0.53–0.62)	−3.57 (−4.01 to −3.12)	22,698 (18,647–26,756)	67.82 (55.72–79.94)	−2.06 (−2.24 to −1.88)
Western Europe	626,459 (580,382–672,004)	318.74 (295.29–341.91)	−0.41 (−0.46 to −0.36)	49,407 (40,630–59,533)	25.14 (20.67–30.29)	−0.77 (−1.11 to −0.43)	931 (884–977)	0.47 (0.45–0.50)	−3.76 (−4.37 to −3.15)	115,910 (95,130–137,719)	58.97 (48.4–70.07)	−2.04 (−2.29 to −1.79)
Latin America and Caribbean
Andean Latin America	98,801 (92,483–105,313)	303.51 (284.10–323.51)	−0.16 (−0.35 to 0.03)	6779 (5180–8613)	20.83 (15.91–26.46)	−0.51 (−0.65 to −0.38)	316 (248–404)	0.97 (0.76–1.24)	−2.32 (−2.52 to −2.13)	22,291 (18,270–26,618)	68.47 (56.12–81.77)	−1.78 (−1.93 to −1.63)
Caribbean	88,119 (83,201–93,376)	384.16 (362.73–407.09)	0.32 (0.10–0.55)	7274 (5841–8973)	31.71 (25.46–39.12)	0.47 (0.41–0.52)	643 (522–822)	2.80 (2.28–3.58)	0.29 (0.20–0.38)	35,398 (29,571–43,491)	154.32 (128.92–189.61)	0.21 (0.09–0.34)
Central Latin America	399,196 (359,593–440,373)	319.48 (287.79–352.44)	−0.44 (−0.76 to −0.11)	26,409 (19,769–33,864)	21.14 (15.82–27.10)	−1.05 (−1.28 to −0.81)	1355 (1193–1531)	1.08 (0.95–1.23)	−1.87 (−2.15 to −1.60)	92,703 (81,465–104,640)	74.19 (65.20–83.75)	−1.50 (−1.69 to −1.31)
Tropical Latin America	345,628 (300,813–391,575)	296.32 (257.9–335.71)	−0.78 (−0.83 to −0.73)	29,029 (22,397–36,453)	24.89 (19.20–31.25)	−1.89 (−2.16 to −1.62)	2047 (1946–2143)	1.76 (1.67–1.84)	−3.38 (−3.97 to −2.77)	115,572 (106,968–124,583)	99.08 (91.71–106.81)	−2.89 (−3.32 to −2.47)
North Africa and middle east
North Africa and middle east	1,744,273 (1,638,281–1,844,346)	562.17 (528.01–594.42)	−0.04 (−0.15 to 0.07)	131,536 (109,789–158,148)	42.39 (35.38–50.97)	0.09 (0.03–0.16)	16,812 (13,952–20,177)	5.42 (4.50–6.50)	−0.66 (−0.77 to −0.56)	978,646 (827,904–1,140,906)	315.41 (266.83–367.71)	−0.55 (−0.70 to −0.41)
South Asia
South Asia	2,875,582 (2,509,762–3,251,778)	314.35 (274.36–355.47)	0.20 (0.15–0.24)	226,935 (176,523–285,127)	24.81 (19.30–31.17)	0.02 (−0.22 to 0.26)	14,241 (10,784–21,807)	1.56 (1.18–2.38)	−0.60 (−0.81 to −0.39)	1,002,566 (799,281–1,372,196)	109.6 (87.37–150.00)	−0.34 (−0.41 to −0.26)
Southeast Asia, east Asia, and Oceania
East Asia	3,590,431 (3,117,717–4,083,387)	488.03 (423.77–555.03)	0.53 (0.34–0.72)	332,912 (258,685–418,039)	45.25 (35.16–56.82)	2.14 (1.95–2.33)	32,737 (26,128–40,267)	4.45 (3.55–5.47)	0.84 (0.46–1.23)	2,030,477 (1,672,847–2,370,630)	275.99 (227.38–322.23)	0.47 (0.16–0.78)
Oceania	22,619 (21,260–24,053)	358.63 (337.09–381.36)	−0.19 (−0.36 to −0.03)	1483 (1182–1834)	23.51 (18.75–29.08)	−0.14 (−0.17 to −0.10)	112 (82–157)	1.77 (1.29–2.49)	−0.56 (−0.90 to −0.22)	9004 (7243–11,129)	142.76 (114.84–176.46)	−0.43 (−0.75 to −0.11)
Southeast Asia	1,753,610 (1,577,590–1,925,509)	494.7 (445.05–543.19)	0.53 (0.34–0.72)	156,538 (127,269–191,632)	44.16 (35.90–54.06)	0.92 (0.85–0.99)	15,280 (11,318–18,998)	4.31 (3.19–5.36)	0.84 (0.46–1.23)	978,399 (792,429–1,169,945)	276.01 (223.55–330.05)	0.66 (0.51–0.82)
Sub-Saharan Africa
Central sub-Saharan Africa	219,554 (206,629–232,582)	404.01 (380.22–427.98)	−0.29 (−0.55 to −0.03)	18,730 (14,917–23,336)	34.47 (27.45–42.94)	−0.28 (−0.53 to −0.03)	866 (583–1261)	1.59 (1.07–2.32)	−0.80 (−0.89 to −0.70)	65,023 (50,793–82,037)	119.65 (93.47–150.96)	−0.57 (−0.75 to −0.39)
Eastern sub-Saharan Africa	688,711 (636,939–743,929)	401.64 (371.45–433.84)	−0.31 (−0.48 to −0.15)	54,376 (43,254–67,743)	31.71 (25.22–39.51)	−0.51 (−0.65 to −0.37)	2277 (1754–2980)	1.33 (1.02–1.74)	−0.91 (−1.11 to −0.71)	187,328 (153,752–226,201)	109.24 (89.66–131.91)	−0.60 (−0.71 to −0.50)
Southern sub-Saharan Africa	178,242 (158,401–198,195)	453.49 (403.01–504.26)	−0.63 (−1.02 to −0.23)	12,921 (10,162–16,146)	32.87 (25.85–41.08)	−0.80 (−1.03 to −0.56)	789 (684–897)	2.01 (1.74–2.28)	0.10 (−0.36 to 0.56)	57,058 (49,557–64,797)	145.17 (126.09–164.86)	−0.39 (−0.60 to −0.18)
Western sub-Saharan Africa	1,007,648 (926,982–1,087,181)	532.84 (490.18–574.90)	−0.19 (−0.23 to −0.16)	68,831 (56,266–84,869)	36.40 (29.75–44.88)	−0.30 (−0.33 to −0.27)	5526 (4183–7067)	2.92 (2.21–3.74)	−0.44 (−1.47 to 0.60)	369,009 (301,805–445,634)	195.13 (159.59–235.65)	−0.27 (−0.87 to 0.33)

CI, confidence interval, DALYs = disability-adjusted life years, EAPC = estimated annual percentage change, IS = ischemic stroke, SDI = socio-demographic index, UI = uncertainty interval.

3.1.2. Regional trends in overall burden of IS

The global burden of IS among young and middle-aged adults (20–54 years) exhibits notable variations across regions with different SDI levels. In terms of prevalence, the middle-high SDI region demonstrates the highest rate at 486.87 per 100,000 population (95% UI: 432.87–541.62), whereas the low SDI region shows the lowest prevalence of 388.39 per 100,000 population (95% UI: 358.2–418.82) (Table 1). Temporal trends in prevalence revealed distinct patterns across SDI levels. While the low-SDI region exhibited a declining trend (EAPC: −0.24 [95% CI: −0.35 to −0.13]), all other SDI regions showed an increasing prevalence over time. Similarly, the incidence of IS varied by SDI level, with the middle-high SDI region bearing the highest burden (42.59 per 100,000 population [95% UI: 34.00–52.70]). In contrast, the low-SDI region had the lowest incidence (29.94 per 100,000 population [95% UI: 24.16–37.16]). In contrast to prevalence, the high-SDI region also exhibited a declining incidence over time (EAPC: −0.52 [95% CI: −0.88 to −0.16]) (Table 1). Mortality and DALYs also demonstrated regional disparities. The middle-high SDI region had the highest mortality and DALYs (3.75 per 100,000 [95% UI: 3.30–4.29] and 241.08 per 100,000 [95% UI: 210.79–273.85], respectively). Notably, the high-SDI region recorded the lowest mortality and DALYs (1.21 [95% UI: 1.09–1.35] and 116.67 [95% UI: 98.53–136.44], respectively). Additionally, this region showed the most rapid decline in mortality, with an EAPC of −1.63 (95% CI: −2.21 to −1.06). Furthermore, this study conducted an age-stratified analysis across different SDI regions. The results demonstrated that prevalence, incidence, mortality, and DALYs exhibited exponential increases with age in all SDI groups. Notably, a marked rise in prevalence, incidence, mortality, and DALYs was observed starting from age 30, underscoring the growing trend of IS among younger populations. However, in the 30 to 34 age group, the high-SDI region exhibited significantly lower mortality and DALYs compared to other regions, which may be attributed to more advanced stroke treatment and management strategies in high-SDI settings. Overall, the burden of IS varied significantly across SDI regions, with a clear trend toward younger-onset cases. These disparities likely stem from the combined effects of multiple factors, including regional disparities in healthcare resources, socioenvironmental conditions, lifestyle patterns, and disease management strategies (Fig. 1).

Figure 1.

Age-specific trends in IS disease burden among young and middle-aged adults (20–54 years) across different SDI regions. (A) Prevalence; (B) incidence; (C) mortality; and (D) DALYs. DALYs = disability-adjusted life years, IS = ischemic stroke, SDI = socio-demographic index.

The findings of this study indicate that Eastern Europe has the highest global prevalence of IS among young and middle-aged adults (20–54 years), reaching 638.36 per 100,000 population (95% UI: 559.39–720.69). This was followed by Central Asia, with a prevalence of 594.32 per 100,000 population (95% UI: 563.14–625.72) (Table 1). Interestingly, high-income North America also exhibited a notably high IS prevalence of 580.14 per 100,000 population (95% UI: 511.31–649.43), ranking 3rd highest among all 21 regions analyzed. The high prevalence of IS among young and middle-aged adults in Eastern Europe and Central Asia may be attributed to shared socioeconomic factors, limited health awareness, genetic susceptibility, and environmental influences in these regions. In contrast, the elevated burden observed in high-income North America may be linked to unhealthy dietary and lifestyle habits, as well as disparities in healthcare resource distribution. The analysis further revealed that Eastern Europe and Central Asia remain the regions with the highest incidence rates of IS among young and middle-aged adults, with rates of 61.86 per 100,000 population (95% UI: 49.45–77.08) and 54.50 per 100,000 population (95% UI: 46.20–65.81), respectively. East Asia also exhibited a high incidence rate, ranking 3rd globally at 45.25 per 100,000 population (95% UI: 35.16–56.82). Notably, East Asia showed the most significant increase in incidence over time, with an EAPC of 2.14 (95% CI: 1.95–2.33). In terms of mortality and DALYs associated with IS among young and middle-aged adults, Eastern Europe and Central Asia continue to rank among the highest, occupying the 1st and 3rd positions, respectively. Notably, the North Africa and Middle East region also demonstrated a substantial burden, ranking 2nd among the 21 GBD regions. In this region, the mortality rate was 5.42 per 100,000 population (95% UI: 4.50–6.50), and the DALY rate was 315.41 per 100,000 population (95% UI: 266.83–367.71).

3.1.3. National trends in the global burden of IS

The prevalence of IS among young and middle-aged adults (aged 20–54 years) ranges from approximately 192.32 per 100,000 to 876.94 per 100,000. The country-specific distribution of prevalence is shown in Figure 1A and Table S1, Supplemental Digital Content 3. The Republic of Ghana has the highest IS prevalence rate (876.94 per 100,000; 95% UI: 832.62–923.16), while the Republic of Cyprus has the lowest prevalence rate (192.32 per 100,000; 95% UI: 172.99–212.07). These findings highlight the complex and diverse landscape of ischemic stroke epidemiology among young and middle-aged adults across different countries.

At the same time, we analyzed the incidence of IS across different countries. Among all countries, Ukraine (74.67 per 100,000; 95% UI: 59.59–94.71) and Turkmenistan (73.19 per 100,000; 95% UI: 61.30–87.42) had the highest incidence (Fig. 2C, Table S1, Supplemental Digital Content 3). These countries are part of the Eastern Europe and Central Asia regions in the GBD regional classification, which aligns with the previous regional analysis identifying Eastern Europe and Central Asia as the regions with the highest burden of IS prevalence among young and middle-aged adults globally. Overall, the incidence of IS in young and middle-aged adults shows an increasing trend, with the most significant increases observed in Turkmenistan (EAPC: 1.96, 95% CI: 1.80–2.12) and the Republic of Palau (EAPC: 1.57, 95% CI: 1.47–1.67) (Fig. 2D and Table S1, Supplemental Digital Content 3).

Figure 2.

Global disease burden of IS among the 20 to 54 age group by country in 2021, and trends in disease burden changes across nations from 1990 to 2021. (A) The prevalence in 2021; (B) trend of prevalence (EAPC) from 1990 to 2021; (C) the incidence in 2021; (D) trend of incidence (EAPC) from 1990 to 2021; (E) the mortality in 2021; (F) trend of mortality (EAPC) from 1990 to 2021; (G) the DALYs in 2021; and (H) trend of DALYs (EAPC) from 1990 to 2021. ASR = age-standardized rates, DALYs = disability-adjusted life years, EAPC = estimated annual percentage change, IS = ischemic stroke.

Additionally, this study analyzed the mortality and DALYs data for IS among young and middle-aged adults in 2021, and we observed notable consistency among the countries most severely affected by these indicators. The Republic of Bulgaria, the Islamic Republic of Afghanistan, and Turkmenistan performed prominently in both mortality and DALYs. The Republic of Bulgaria had the highest mortality (9.83 per 100,000; 95% UI: 8.02–12.13) and the highest DALYs (522.33 per 100,000; 95% UI: 438.42–620.03) in these categories. The Islamic Republic of Afghanistan ranked 2nd in mortality (9.72 per 100,000; 95% UI: 6.07–15.07) and 3rd in DALYs (504.04 per 100,000; 95% UI: 339.34–752.94). Turkmenistan ranked 3rd in mortality (8.73 per 100,000; 95% UI: 5.56–12.01) and 2nd in DALYs (519.85 per 100,000; 95% UI: 380.93–665.56) (Fig. 2, Table S1, Supplemental Digital Content 3). This overlap indicates that these countries face a high burden of IS among young and middle-aged adults (aged 20–54 years), impacting both mortality and the overall health-related quality of life of their populations. Furthermore, the country with the largest increase in mortality and DALYs due to IS was the Republic of Zimbabwe. From 1990 to 2021, the EAPC in mortality was 3.96 (95% CI: 3.29–4.63), and the EAPC for DALYs was 2.44 (95% CI: 2.11–2.78). For more information on mortality and DALYs, refer to Figure 2 and Table S1, Supplemental Digital Content 3.

3.2. Analyze segmented trends in IS burden using joinpoint regression analysis

To further understand the local trends in the burden of IS among young and middle-aged adults (20–54 years), this study conducted a joinpoint regression analysis of the IS burden, with the results shown in Figure 3. From 1990 to 2021, the number of cases for IS prevalence, incidence, mortality, and DALYs consistently increased. Notably, the most rapid growth occurred after experiencing 5, 4, 5, and 5 joinpoints during the periods of 1996 to 2001, 1998 to 2005, 1997 to 2004, and 1997 to 2004, respectively (Table S2, Supplemental Digital Content 4). Regarding prevalence, the overall trend shows an increase, but the local trends at different joinpoints are inconsistent. Specifically, there was a clear upward trend during the periods of 1995 to 2002, 2011 to 2015, and 2015 to 2021. In contrast, during the periods of 1990 to 1995 and 2002 to 2011, there was a slight decline in prevalence. For incidence, there is also an overall upward trend, though there are fluctuations in different phases. A slight decline in incidence was observed during the periods of 1990 to 1997, 2005 to 2012, and 2019 to 2021. However, the incidence exhibited significant increases in the periods of 1997 to 2005, 2012 to 2015, and 2015 to 2019, with the largest increase occurring in the 2015 to 2019 period. Regarding the trend in mortality, although there were some fluctuations across different periods, the overall trend shows a decline. Specifically, after 2004, there has been a consistent decrease in mortality. The trend in DALYs is similar to that of mortality, also showing a general decline. However, DALYs for IS experienced a significant increase during the 1997 to 2004 period. After 2004, there was a continued downward trend in DALYs (Table S3, Supplemental Digital Content 5). Overall, between 1990 and 2021, the burden of IS among young and middle-aged adults (20–54 years) experienced significant changes.

Figure 3.

Joinpoint regression analysis of IS with 20 to 54 years burden trends globally. (A) Trend of the total number of IS prevalence cases over time; (B) trend of age-standardized prevalence rate of IS over time; (C) trend of the total number of IS incidence cases over time; (D) trend of age-standardized incidence rate of IS over time; (E) trend of the total number of IS-related deaths over time; (F) trend of age-standardized mortality of IS over time; (G) trend of the total number of IS-related DALYs over time; and (H) trend of age-standardized DALY rate of IS over time. APC = annual percentage change, DALYs, disability-adjusted life years, IS = ischemic stroke.

3.3. Age-period-cohort analysis on IS (20–54 years) incidence and prevalence

This study conducted an age-period-cohort analysis on the prevalence and incidence of IS among young and middle-aged adults aged 20 to 54 (Fig. 4). After controlling for period and birth cohort effects, the age effect had a significant impact on IS prevalence and incidence risk in this population. Both relative prevalence risk and incidence risk showed a clear upward trend with increasing age, with a marked rise after the age of 35 (Fig. 4, Table S4, Supplemental Digital Content 6). After adjusting for age and birth cohort effects, the period effect also significantly influenced IS prevalence and incidence risk. The period effects for both prevalence and incidence risk exhibited an upward trend, with the 2017 to 2021 group having relative risk values 1.2 and 1.3 times higher, respectively, than those of the 1992 to 1997 group. The highest prevalence and incidence risks were observed in the 2017 to 2021 period (Fig. 4, Table S3, Supplemental Digital Content 5). After adjusting for age and period effects, the birth cohort effect also significantly influenced the prevalence and incidence risk of IS among young and middle-aged adults. The birth cohort effect showed that earlier birth cohorts had a higher risk of incidence and prevalence compared to later cohorts. From the 1942 to 1946 cohort to the 1997 to 2001 cohort, the risks of both prevalence and incidence exhibited a consistent decline (Fig. 4, Table S4, Supplemental Digital Content 6). Notably, the impacts of age, period, and birth cohort effects on incidence and prevalence risk in Figure 4 differed from those in Figures S1 and S2, Supplemental Digital Content 1. This discrepancy may be attributed to the confounding effect of the 3rd factor when analyzing the interaction between any 2 factors in Figures S1 and S2, Supplemental Digital Content 1.

Figure 4.

The effects of age, period, and birth cohort on the relative risk of IS with 20 to 54 years. (A) Prevalence and (B) incidence. IS = ischemic stroke.

3.4. 20 to 54 years IS-related DALYs attributed to risk factors

In 2021, various risk factors had a significant impact on IS-related DALYs among individuals aged 20 to 54 years, and these effects were observed across different strata within this age group. Overall, risk factors such as high low-density lipoprotein (LDL) cholesterol, high systolic blood pressure, particulate matter pollution, smoking, and high body mass index (BMI) had a considerable influence on IS-related DALYs, affecting multiple aspects, including metabolism, environment, and lifestyle. Notably, the predominant risk factors exhibited distinct geographical variations, influenced by regional environmental and sociocultural disparities. In Southeast Asia, East Asia, and Oceania, a diet high in sodium emerged as a critical risk factor. Central Europe, Eastern Europe, and Central Asia required heightened focus on alcohol use, while North Africa and the Middle East exhibited a predominant burden linked to low physical activity. In contrast, South Asia faced a distinct challenge due to low fruit intake (Fig. 5). These regional disparities underscore the necessity for tailored IS prevention strategies that account for location-specific risk factors. Additionally, this study conducted a comparative analysis of attributable risk factors between 1990 and 2021 (Fig. 6). The results revealed that among the 18 risk factors examined, certain risks showed relative improvement over time, including particulate matter pollution (34.53%–28.25%), smoking (25.15%–21.7%), and a diet high in sodium (10.1%–9.8%). However, most other attributable risks demonstrated significant upward trends. Metabolic –factors – particularly high LDL cholesterol, high systolic blood pressure, and high –BMI – emerged as the most substantial contributors to the IS disease burden in young and middle-aged adults. Notably, high systolic blood pressure (45.57%–51.52%) and high BMI (8.28%–13.7%) exhibited the most dramatic increases in attributable risk. Concurrently, other factors including low physical activity, alcohol use, and high temperature also require ongoing attention as significant contributors.

Figure 5.

Impact of different risk factors on DALYs in different regions in 2021. The heatmap shows DALY rates per 100,000 population attributable to selected risk factors at the global, SDI regional, and GBD super-regional levels. DALYs = disability-adjusted life years, GBD = Global Burden of Disease, Injuries, and Risk Factors Study, IS = ischemic stroke, LDL = low-density lipoprotein, SDI = socio-demographic index.

Figure 6.

Attributable risk factors for DALYs in 1990 and 2021. The heatmap and bar plot show population-attributable fractions (%) of DALYs for selected risk factors in 1990 and 2021. DALYs = disability-adjusted life years, LDL = low-density lipoprotein.

4. Discussion

This study conducted a comprehensive analysis of the global, regional, and national burden of IS among young and middle-aged adults aged 20 to 54 years from 1990 to 2021, revealing the complex landscape of temporal changes in the disease burden. The findings highlight that the burden of IS among young and middle-aged adults remains substantial worldwide. In 2021, the number of cases of IS among individuals aged 20 to 54 years globally reached 16,301,644 (95% UI: 14,638,249–17,962,503). Both the number of cases and the prevalence of the disease have shown an upward trend over time. The increasing prevalence and associated burden of IS are attributed to factors such as the younger onset of chronic diseases and insufficient management of comorbidities, which continue to drive the rise in the incidence of IS and its related burden among young and middle-aged adults.^[11,30] These findings are consistent with previous evidence showing that IS is no longer confined to older adults. Putaala reported that a substantial proportion of IS occurs in working-aged individuals and that the incidence of IS at younger ages has increased in many settings since the 1980s.^[31] A recent GBD-based study by He et al also reported increasing numbers of incident cases and deaths among young adults with IS from 1990 to 2021, despite declining age-standardized mortality rates.^[32] Compared with that study, our analysis further focused on adults aged 20 to 54 years and incorporated prevalence, segmented temporal trends, age-period-cohort effects, national disparities, and temporal changes in attributable risk factors.

At the regional level, the burden of IS in terms of incidence is highest in middle-high SDI regions. Mortality and DALYs also show significant regional differences. High SDI regions exhibit the lowest mortality and DALYs, with the fastest decline in mortality occurring in these regions. These findings highlight the persistent socioeconomic disparities in the global burden of IS. The higher incidence and prevalence of IS in middle-high SDI regions are largely attributed to factors such as lifestyle changes and greater investments in early screening. In contrast, high SDI regions benefit from ample healthcare resources, leading to a significant reduction in mortality and DALYs. This reflects ongoing inequalities in the global distribution of disease burden and health resources.^[32,33] Our finding that East Asia experienced the largest increase in incidence is also in line with He et al, who reported that East Asia had the highest EAPC in incidence age-standardized rates among young adults with IS.^[32]

This study also reveals gender differences in the disease burden of IS among young and middle-aged adults aged 20 to 54 years. Men have significantly higher incidence, mortality, and DALYs compared to women. This disparity may be related to lower health management adherence among men, which is reflected in poorer health behaviors such as diet and exercise, higher levels of life and work-related stress, and greater exposure to risk factors such as smoking and alcohol consumption.^[3,34] This interpretation is supported by Putaala, who noted that traditional vascular risk factors tend to accumulate with age, particularly among men.^[31] In the age subgroup analysis, the burden of IS begins to rise significantly from the age of 30, reflecting the trend toward younger onset. This also highlights the cumulative nature of the IS burden over time. Additionally, high SDI regions show lower mortality rates and DALYs in the 30 to 34 age group, which can be attributed to their more advanced stroke treatment and management systems.

When the overall trends were divided into several subperiods, this study found that, while both the incidence and prevalence of IS showed an overall upward trend, mortality and DALYs declined, with noticeable fluctuations across different stages. After 2012, both prevalence and incidence among young and middle-aged adults rose significantly. From 2004 onward, mortality and DALYs showed a sustained decline. The decline in mortality and DALYs after 2004 can be attributed to global prevention and control policies, as well as advancements in treatment technologies. Meanwhile, the increase in incidence and prevalence after 2012 is linked to the younger onset of metabolic risk factors, exacerbation of air pollution, and improvements in diagnostic capabilities.^[35]

Age effects have a significant impact on the prevalence and incidence risk of IS among young and middle-aged adults. Both relative prevalence and incidence risks show a marked upward trend with increasing age, with a significant increase in risk after the age of 35. After controlling for age and birth cohort effects, period effects also showed a notable influence on the prevalence and incidence risk of IS. The period effects for both prevalence and incidence exhibited an upward trend, with the relative risk values for the 2017 to 2021 group being 1.2 and 1.3 times higher, respectively, compared to the 1992 to 1997 group. The highest risks for prevalence and incidence were observed in the 2017 to 2021 period. After controlling for age and period effects, the birth cohort effect also had a significant impact on the prevalence and incidence risk of IS among young and middle-aged adults. The birth cohort effect demonstrated that earlier birth cohorts had higher incidence and prevalence risks than later cohorts, with a steady decrease in risk from the 1942 to 1946 cohort to the 1997 to 2001 cohort. This pattern may reflect improvements in early-life healthcare, sanitation, education, nutrition, and management of underlying conditions across successive generations, while the increasing period effect suggests that recent lifestyle, metabolic, and environmental exposures continue to influence IS risk in this population.

Detecting risk factors is crucial for the prevention of IS, and adjusting modifiable risk factors can help reduce the disease burden. This is also a key strategy in formulating public health policies. Overall, risk factors such as high LDL cholesterol, high systolic blood pressure, particulate matter pollution, smoking, and high BMI have a significant impact on IS-related DALYs among young and middle-aged adults. High systolic blood pressure can damage the endothelium, leading to oxidative stress and arterial sclerosis. High LDL cholesterol contributes to the deposition of pro-inflammatory plaques, increasing the risk of thrombosis. Smoking exacerbates ischemia through vasoconstriction and platelet aggregation. Particulate matter pollution induces inflammation and oxidative stress. These mechanisms work synergistically to cause vascular narrowing or occlusion, ultimately triggering ischemic stroke.^[29,36] Our findings are consistent with previous studies showing that traditional vascular and lifestyle-related risk factors remain important in young IS. Putaala reported that traditional vascular risk factors are more prevalent in young stroke patients than previously recognized and tend to accumulate with age.^[31] Ohya et al further found that smoking, drinking, and obesity were more frequent in young adults with IS than in non-young adults, and that diabetes mellitus and dyslipidemia in young adults older than 40 years were comparable to those in non-young adults.^[37] These findings support our observation that metabolic and lifestyle-related factors, particularly high LDL cholesterol, high systolic blood pressure, smoking, and high BMI, substantially contribute to IS-related DALYs among adults aged 20 to 54 years. In addition to conventional risk factors, IS in younger adults may have more heterogeneous etiologies. Ohya et al reported that embolic sources, including patent foramen ovale and intracardiac thrombus, as well as uncommon causes such as arterial dissection, moyamoya disease, cerebral venous thrombosis, antiphospholipid syndrome, and protein S deficiency, were more prevalent in young adults than in non-young adults.^[37] This suggests that the prevention and management of IS in young and middle-aged adults should combine traditional vascular risk control with attention to age-specific etiologies.

At the same time, due to geographic, environmental, and cultural differences, the risk factors for IS-related DALYs exhibit notable regional characteristics. In Southeast Asia, East Asia, and Oceania, a diet high in sodium, such as that from pickled foods and soy sauce,^[38,39] contributes to elevated blood pressure, significantly increasing the risk of IS through endothelial damage and arterial sclerosis.^[40,41] In Central Europe, Eastern Europe, and Central Asia, cold climates combined with a culture of alcohol consumption lead to alcohol use, which triggers hypertension and arrhythmias, directly exacerbating thrombosis formation.^[42] In North Africa and the Middle East, hot climates limit outdoor activities, and combined with sedentary lifestyles, obesity and metabolic syndrome are promoted, indirectly facilitating vascular changes.^[43] In South Asia, economic constraints and a grain-based diet result in a diet low in fruits, leading to a lack of antioxidants, which exacerbates oxidative stress and dyslipidemia.^[44] Moreover, compared to 1990, by 2021 metabolic risk factors – such as high LDL cholesterol, high systolic blood pressure, and high BMI – had the greatest impact on the disease burden of IS among young and middle-aged adults, with the most pronounced increases in associated risk. In addition, other factors such as low physical activity, alcohol use, and high temperature also warrant ongoing attention. Given the evolving trends in risk factors, there is an urgent need to develop targeted prevention and management strategies for IS in younger populations, particularly those that address lifestyle and dietary habits. These strategies should be culturally and regionally tailored to effectively mitigate the growing burden of IS in this age group. Therefore, the treatment and management of IS must take into account regional dietary habits and lifestyle characteristics to develop appropriate stroke prevention and treatment strategies.

This study, which utilized Global Burden of Disease data to analyze the burden and risk factors of IS among young and middle-aged adults aged 20 to 54 years, provides valuable insights for the early treatment, prevention, and management of IS. However, several limitations should be acknowledged. First, the completeness and timeliness of data sources may vary, potentially affecting the accuracy of the results. Second, the reliance on modeling assumptions and wide uncertainty intervals may reduce the precision of the estimates. Additionally, subnational variations and socioeconomic disparities within countries may not be fully captured, limiting the depth of the analysis. Lastly, as GBD data primarily reveal associations rather than causality, further validation using country-specific and locally sourced data is essential to support context-specific interpretation and policy-making. Furthermore, some young-adult-specific etiologies, such as patent foramen ovale, arterial dissection, and inherited or acquired thrombophilia, are not fully captured in the GBD risk factor framework; therefore, clinical registry studies are still needed to complement population-level GBD estimates.

5. Conclusion

Based on GBD 2021 data from 1990 to 2021, this study demonstrates that ischemic stroke among young and middle-aged adults aged 20 to 54 years remains a substantial and evolving global health burden. Over the past 3 decades, the prevalence and incidence of ischemic stroke increased, whereas mortality and DALY rates declined, suggesting improved survival but a growing number of affected individuals in this population. Marked geographic disparities were observed across SDI regions, GBD regions, and countries. Metabolic risk factors, particularly high systolic blood pressure, high LDL cholesterol, and high BMI, have played an increasingly prominent role in ischemic stroke-related DALYs, alongside behavioral and environmental risks such as smoking, alcohol use, and particulate matter pollution. These findings underscore the need for age-targeted and region-specific public health strategies. Such strategies should include early screening for hypertension, dyslipidemia, obesity, and diabetes; lifestyle interventions aimed at reducing smoking, alcohol use, and unhealthy dietary patterns; and environmental policies to reduce air pollution exposure. Strengthening primary prevention, risk factor management, and equitable access to stroke care is essential to mitigating the growing burden of ischemic stroke among young and middle-aged adults.

Acknowledgments

We acknowledge the exceptional contributions made by the collaborators of the Global Burden of Diseases, Injuries, and Risk Factors Study 2021. We sincerely appreciate the Institute for Health Metrics and Evaluation institution for providing the GBD data.

Author contributions

Conceptualization: Xiaoying Lyu, Liyun He.

Data curation: Xiaoliang Zhao, Yuning Qin, Enshi Lu, Xiaoying Lyu, Liyun He.

Formal analysis: Xiaoliang Zhao, Yuning Qin, Enshi Lu, Mengqi Wang.

Investigation: Xiaoliang Zhao, Yuning Qin, Enshi Lu, Mengqi Wang.

Methodology: Xiaoying Lyu, Liyun He

Validation: Xiaoliang Zhao, Xiaoying Lyu.

Writing – original draft: Xiaoliang Zhao, Yuning Qin.

Writing – review & editing: Mengqi Wang, Yudong Sheng, Liyun He.