Global, Regional, and Future Predictions of Cardiovascular Disease Burden Among Women of Childbearing Age: A Systematic Analysis of the Global Burden of Disease Study 1990–2021

Introduction

The global total fertility rate declined by more than half from 1950 to 2021,¹ making women of childbearing age (WCBA) health a critical public health priority. Cardiovascular disease (CVD), principally comprising ischemic heart disease (IHD) and stroke, persists as the foremost cause of female mortality worldwide.^2,3 Over the past decade, the overall reduction in the burden of CVD among WCBA has remained largely stagnant.⁴ Although CVD is traditionally considered age-related, emerging evidence demonstrates a trend of earlier onset in Western populations.^5–7 Compared to postmenopausal women, WCBA are increasingly exposed to a complex interplay of risk factors that can be broadly categorized into two domains. First, cardiometabolic risk factors,⁸ including diabetes, dyslipidemia, obesity, and physical inactivity, have risen substantially in younger populations. Second, female-specific and pregnancy-related factors,⁹ such as exogenous hormone use, hypertensive disorders of pregnancy, gestational diabetes, and adverse pregnancy outcomes, are increasingly recognized as early indicators of long-term cardiovascular vulnerability.¹⁰ This dual burden of risk factors highlights the unique pathophysiological and epidemiological profile of CVD in WCBA. Emerging data further underscore that CVD is the leading cause of death in women, especially in WCBA.^11,12 In studies on CVD classification, it is also indicated that the incidence and mortality of WCBA continue to rise, such as RHD,¹³ Stroke,¹¹ and Aortic aneurysms.¹⁴ Moreover, studies show that women are more likely than men to die after myocardial infarction and are at greater risk of developing heart failure within five years following hospital discharge.¹⁵ These observations challenge the long-standing perception of WCBA as a low-risk population and highlight the need for targeted epidemiological assessment. In addition, social determinants of health, such as economic, social, environmental, and psychosocial factors, can also have an impact on the local CVD condition.¹⁶ Expanding research and implementation of both primary and secondary CVD prevention strategies, while addressing socioeconomic disparities and enhancing risk prediction in WCBA, represents a pivotal opportunity to mitigate the growing CVD epidemic.¹⁷

Despite this, current prevention efforts and policy responses have failed to keep pace with the true burden of CVD. In China, existing CVD prevention policies lack specific emphasis on WCBA, contributing to the relative neglect of this population.¹⁸ More broadly, CVD prevention policies remain limited in number and underdeveloped in structure,¹⁹ which not only constrains access to appropriate health-care resources for women with CVD but also limits broader awareness of the importance of cardiovascular health in women. Although dedicated guidelines specifically focused on CVD prevention in women have not been updated since 2011,²⁰ more recent guidelines, such as the 2019 ACC/AHA Primary Prevention Guideline,²¹ have incorporated sex-specific considerations. However, these recommendations remain insufficient to fully address the unique risk profiles and healthcare needs of WCBA. A rigorous evaluation of the global CVD burden and its geospatial distribution among WCBA is therefore essential for developing precision public health interventions. Previous Global Burden of Disease (GBD) studies have primarily focused on the overall population, with limited attention to women of childbearing age as a distinct subgroup. In particular, comprehensive analyses integrating subtype-specific burden, temporal trends, sociodemographic disparities, and future projections for WCBA are lacking. Furthermore, the heterogeneity of CVD subtypes and their divergent trends have not been adequately characterized in this population, limiting the development of targeted prevention strategies.

The GBD 2021 study employs advanced analytical models to systematically quantify global disease and health data, establishing a robust methodological foundation for evaluating CVD burden across 204 countries and territories. This investigation utilizes two principal metrics: incidence and mortality rates, to conduct a comprehensive assessment of CVD burden trends from 1990 to 2021, encompassing 11 major CVD subtypes as defined by GBD 2021. The sociodemographic index (SDI), a composite measure integrating socioeconomic and demographic development indicators, enables quantitative examination of its association with CVD burden. By incorporating Bayesian age-period-cohort (BAPC) model, this study predicts future burden patterns for CVD and its subtypes through 2050. These epidemiological insights will guide therapeutic development priorities, and support data-driven decision-making for health resource distribution and precision public health interventions.

Methods

Data Source

The data used in this study was sourced from the GBD 2021 study, publicly available through the Institute for Health Metrics and Evaluation (IHME) via its online data repository (https://vizhub.healthdata.org/gbd-results/). The GBD study utilizes rigorously identified and curated data sources to ensure comprehensive and reliable global health assessments. Details for each data source is uniquely identified and cataloged within the GBD 2021 Sources Tool on the Global Health Data Exchange (GHDx, https://ghdx.healthdata.org/gbd-2021/sources), which serves as a centralized repository. The study applied the meta-regression–Bayesian, regularized, trimmed (MR-BRT) model to adjust cause-related data for known biases, while spatiotemporal Gaussian process regression (ST-GPR) was utilized to analyze heterogeneous and incomplete datasets. This structured and transparent approach provides a robust foundation for analyzing and interpreting global health trends.

In GBD 2021, the International Classification of Diseases (ICD) codes of CVD were combined with GBD health loss data, providing burden estimates for 11 major CVD classifications through indicators such as the incidence rate and DALYs (https://ghdx.healthdata.org/record/ihme-data/gbd-2021-cause-icd-code-mappings). CVD subtypes including atrial fibrillation and flutter (AF/AFL), cardiomyopathy and myocarditis, endocarditis, ischemic heart disease (IHD), lower extremity peripheral arterial disease (lower extremity PAD), non-rheumatic valvular heart disease (NRVHD), pulmonary Arterial Hypertension (PAH), rheumatic heart disease (RHD), stroke, aortic aneurysm (AA), hypertensive heart disease (HHD) and other cardiovascular and circulatory diseases. It should be noted that incidence estimates were not available for certain CVD subtypes (aortic aneurysm, hypertensive heart disease, and other cardiovascular and circulatory diseases) in the GBD 2021 dataset. Therefore, analyses for these subtypes were restricted to outcomes with available estimates.

The SDI quantifies population-level progress through three demographically sensitive proxies: the Age-standardised total fertility rate among women under 25, age-weighted educational attainment, and lag-distributed income per capita. The SDI scale ranges from 0 (lowest development) to 1 (highest development). Following the GBD 2021 methodological protocol, we stratified 204 geopolitical entities into 5 development quintiles: Low SDI, Low-middle SDI, Middle SDI, High-middle SDI and High SDI.²²

Population

The target population of this study is women aged 15 to 49, which demographic the WHO defines.²³ We analyzed rate and number data of CVDs in WCBA from 1990 to 2021 across 204 countries, 21 GBD-defined regions (Appendix S1), 5 SDI regions, and the global level.

Statistics

Both incidence and mortality were utilized to quantify the burden of CVDs. This study implemented a stratified epidemiological analysis to evaluate the burden of CVDs among the WCBA across different age groups, geographical regions, and countries. Age-standardised incidence and mortality rates (ASIR, ASMR) were calculated using the following direct standardization formula:²⁴

In this formula, represents the upper age limit, corresponds to age-specific rate, and represents the population count for a particular age group within the standard population of GBD 2021. All rates are expressed per 100,000 population.

Temporal trends in CVD burden were assessed by estimated annual percentage change (EAPC) through a log-linear regression model: , where t is the calendar year, and β₁ represents the logarithmic trend.²⁵ The EAPC value was derived as formula . A trend was classified as increasing or decreasing if the EAPC and its 95% CI were entirely above or below zero, respectively; otherwise, it was considered stable.

To further explore socioeconomic disparities in the CVDs burden across 21 regions and 204 countries, we employed local regression (loess) smoothing. Spearman correlation analysis was applied to quantify the relationship in 2021, reporting correlation coefficients (r) and p-values. Furthermore, we calculated the EAPC of SDI from 1990 to 2021 for each country and assessed its correlation with the EAPC of ASRs. A significance threshold of p < 0.05 was applied for all analyses.

To project future the long-term trends in CVDs among WCBA, we employed a BAPC model, forecasting ASR from 2022 to 2050.²⁶ The model was specified as:

where represents the rate in age group and period , represents the age effect, the period effect, and the cohort effect. The model used 5-year interval groupings, and model fit was evaluated using the Deviance Information Criterion (DIC) and the Watanabe-Akaike Information Criterion (WAIC).

To predict the long-term trend of CVD burden in the WCBA, we adopted the BAPC model. Compared with traditional regressive methods, the BAPC model can simultaneously consider age, period and cohort effects, and combine Bayesian reasoning to improve the stability and reliability of predictions, especially in the case of sparse data or noise.

Decomposition analysis was applied to disentangle the contributions of population aging, population growth, and epidemiological change to temporal variations in the CVDs burden from 1990 to 2021. Population aging reflected changes in age structure, population growth captured changes in total population size, and epidemiological change represented variations in age-specific rates independent of demographic shifts.

All statistical analyses and visualizations were performed in R (version 4.2.2). Specifically, the epitools package was used for calculating ASR and confidence interval (CI).²⁷ The BAPC analysis was conducted using the BAPC and INLA packages,²⁸ and data visualization was performed using ggplot2.

Results

Global and Regional Burden of CVD Among WCBA

In 2021, the ASIR and ASMR of CVDs among WCBA were 228.19 and 21.06 per 100,000 person-years, respectively. Both the EAPC of ASIR and ASMR between 1990 and 2021 showed slightly decline (EAPC: −0.08, 95% CI: −0.11 to −0.05; EAPC: −1.59, 95% CI: −1.64 to −1.54), while subtype-specific analyses revealed increasing incidence trends for certain major conditions, including IHD and RHD. The case number and deaths of CVDs demonstrated a substantial increase from 2,770,546 and 383,988 in 1990 to 4,492,391 and 419,177 in 2021, respectively (Figure 1A and Table 1).

Table 1 Incidence and Mortality of Cardiovascular Disease Among Women of Child-Bearing Age (15–49 Years) in 2023, and Their Estimated Annual Percentage Changes from 1990 to 2021, by Cause and GBD Region

Figure 1 Regional and national burden of CVD among WCBA. (A) The changing trend of the total number and Age-standardised rate of incidence and mortality in 1990–2021 due to all CVD cases. (B) Age-standardised incidence (left) and mortality (right) rates of CVD across GBD regions in 2021, with countries being labeled within their regions. (C) National age-standardised incidence and mortality rates in 2021, and their estimated annual percentage changes from 1990 to 2021 for CVD. Abbreviations: CVD, cardiovascular disease; WCBA, women of childbearing age.

In the GBD regions, Eastern Sub-Saharan Africa recorded the highest ASIR (289.51 per 100,000 person-years) of CVDs in 2021, followed by Central Sub-Saharan Africa and Southern Sub-Saharan Africa (288.85 and 284.5), while Oceania reported the highest ASMR (59.07). However, the most incidence cases in 2021 occurred in South Asia and East Asia (1242996 and 821368). South Asia also had the highest ASMR (148131). According to the EAPC, ASMR showed an overall downward trend among 21 regions. In ASIR, Eastern Europe, Caribbean and Western Sub-Saharan Africa demonstrated regional upward trends, with EAPC of 0.09 (95% CI: −0.01 to 0.19), 0.08 (95% CI: 0.06 to 0.11) and 0.02 (95% CI: 0.01 to 0.02) (Figure 1B and Table 1).

At the national level, Afghanistan, Solomon Islands and Kiribati had the highest national ASIR in 2021, reaching 350.31, 345.53 and 337.74 per 100,000 person-years, while Marshall Islands, Micronesia and Tuvalu exhibiting the highest national ASMR of 94.85, 82.28, 74.7 per 100,000 person-years. In 2021, India and China reported the highest incidence numbers (930560 and 791415) and deaths (107644 and 43815) in CVDs. Saudi Arabia showed the most pronounced increase in ASIR, with an EAPC of 0.57 (95% CI: 0.48 to 0.67). What’s more, Zimbabwe reported the most pronounced increase in ASMR, with an EAPC of 4.3 (95% CI: 3.06 to 5.55) (Figure 1C and Table S1).

Geographic Variations and Leading Subtypes of CVD

Among all CVD subtypes, IHD recorded the highest global ASIR in 2021, at 67.53 per 100,000 population (95% CI: 67.42–67.64). IHD also showed the most marked increase in incidence, with an EAPC of 0.44% (95% CI: 0.41–0.47). In terms of case numbers, incident cases of IHD rose from 678,172 in 1990 to 1,349,518 in 2021. Followed by IHD, RHD had the second highest ASIR in 2021, at 65.32 per 100,000 (95% CI: 65.21–65.44), with an EAPC of 0.37% (95% CI: 0.32–0.42). The number of RHD incident cases increased from 807,408 in 1990 to 1,248,510 in 2021. IHD also exhibited the highest ASMR in 2021 (8.64 per 100,000 population, 95% CI: 8.60–8.68) with deaths increasing from 130,835 in 1990 to 172,204 in 2021. Stroke followed IHD in ASMR, at 6.91 per 100,000 population (95% CI: 6.87–6.94), accounting for 137,695 deaths in 2021 (Table 1).

There were also substantial geographic variations in CVD subtypes profiles in terms of incidence and mortality. The proportion of ASIR in 2021 showed intuitive differences between African regions and European regions (Figure S1A). Figure S1B summarized the regional rankings of ASIR and ASMR for each CVD subtype. Within most regions, IHD and RHD exhibited the highest ASIRs, while stroke and IHD had the highest ASMRs.

At the national level, considerable heterogeneity in the burden of CVD subtypes was observed across 204 countries and territories in 2021. RHD emerged as the most incident subtype in the major countries, while stroke dominated in mortality. In the remaining countries, IHD ranked first in terms of both incidence and mortality (Figure 2).

Figure 2 Regional burden of CVD subtypes among WCBA in 2021. (A) Region-specific proportion of incident and mortality cases. (B) Rankings of the Age-standardised incidence rates of 9 CVD subtypes and the Age-standardised mortality rates of 12 CVD subtypes, by region. Abbreviations: CVD, cardiovascular disease; WCBA, women of childbearing age.

Global Patterns of CVD and Subtypes Across SDI Levels

Across different SDI quintiles, the low SDI regions consistently recorded the highest ASIR (276.61 per 100,000 population, 95% CI: 275.94 to 277.28), while the low-middle SDI regions recorded the highest ASMR (32.5, per 100,000 population, 95% CI: 32.34 to 32.67) for CVDs in 2021. (Table S1) Figure 3A reflected the persistence of these relationships from 1990 to 2021, with lower-SDI regions representing higher burden. At the national level, the SDI was significantly negatively correlated with ASIR (r = −0.72, 95% CI: −0.78 to −0.65, P < 0.001) and ASMR (r = −0.48, 95% CI: −0.58 to −0.36, P < 0.001) of WCBA’s CVDs in 2021. (Figure 3B) As illustrated in Figure 3C, countries with lower SDI values experienced disproportionately higher incidence and mortality rates of CVDs, highlighting marked absolute and relative health inequalities associated with human development levels.

Figure 3 SDI-related inequalities in burden of CVD among WCBA. (A) The changing trend of the age-standardised incidence and mortality rates from 1990 to 2021, by SDI region. (B) Associations between the Age-standardised incidence and mortality rates with SDI in 2021. Each dot represents a country, with the size of the dot indicating the number of CVD cases in that country. (C) Concentration curve illustrating SDI-related inequalities in Age-standardised incidence and mortality rates in 2021. Abbreviations: SDI, sociodemographic index; CVD, cardiovascular disease; WCBA, women of childbearing age.

As shown in Table S1 and Figure 4A, the global burden of CVD subtypes varied considerably across SDI regions. Both the combined ASIR and ASMR exhibited a negative correlation with SDI levels. Across SDI settings, the four most common CVD subtypes accounted for approximately 83.4% to 92.7% of total incidence and mortality. In terms of ASIR, IHD ranked first in middle and high-middle SDI regions and remained highly common across other regions. RHD, while ranking first in low and low-middle SDI regions, declined consistently with rising SDI, falling to eighth place in high SDI regions. In contrast, lower extremity PAD was the leading subtype in high SDI regions. Regarding ASMR, IHD and stroke were the top two causes of death across all SDI regions (Figure 4B and C).

Figure 4 SDI-specific burden of CVD subtypes among WCBA in 2021. (A) SDI-specific Age-standardised rates of incidence and mortality. (B) SDI-specific proportion and Age-standardised rates of incidence and mortality for the 4 most common CVD subtypes. (C) Rankings of the Age-standardised incidence rates of 9 CVD subtypes and the Age-standardised mortality rates of 12 CVD subtypes, by SDI region. Abbreviations: CVD, cardiovascular disease; WCBA, women of childbearing age.

In the correlation analysis, the ASIR of endocarditis, lower extremity PAD, and NRVHD showed a positive correlation with the SDI. In contrast, the ASIR of cardiomyopathy and myocarditis, IHD, PAH, RHD, and stroke demonstrated a negative correlation with the SDI. Regarding mortality, the ASMR of endocarditis, IHD, NRVHD, RHD, stroke, HHD, and other cardiovascular and circulatory diseases were all inversely correlated with the SDI (Figures S2–3).

Age-Specific Patterns of Overall and Subtype-Specific CVD Burden

In 2021, age-specific patterns of CVD burden among WCBA revealed that the number and rate of mortality increased progressively with age, peaking in the 45–49-year age group. In contrast, the incidence of CVDs demonstrated a U-shaped trend across age groups, with both the number and rate declining initially and then rising in older age groups (Figure S4).

Subtype profiles of CVDs varied substantially across 5-year age intervals. In 2021, RHD and IHD were the most common CVDs among WCBA. Meanwhile, stroke and IHD were the leading causes of CVDs mortality. The proportion of the global burden of RHD decreased with advancing age, whereas that of IHD increased (Figure 5A). In terms of incidence, RHD was highest in the 15–34 age groups, while IHD was most common among those aged 35–49 years. Throughout the WCBA age range, RHD accounted for the largest share of CVDs mortality, followed by stroke. Deaths from both IHD and stroke rose consistently with age (Figure 5B).

Figure 5 Age-specific burden of CVD subtypes among WCBA in 2021. (A) Global age-specific proportion of incident cases and mortalities by CVD subtypes. (B) Global age-specific counts of incident cases and mortalities for the 4 most common CVDs. Abbreviations: CVD, cardiovascular disease; WCBA, women of childbearing age.

Bayesian Age-Period-Cohort Model Prediction

Based on the GBD data for CVD from 1990 to 2021, we applied the BAPC model to predict trends in the ASIR and ASMR of CVD among WCBA through 2050. The predictions suggested a continuous increase in incidence from 2021 onward, with ASIR expected to reach its historical peaks of 279.12per 100,000 population by 2050, while ASMR will continue to decline and reach its lowest point of 17.12 per 100,000 people in 2025 (Figure S5 and Table S2).

According to BAPC model, the changing trends of ASIR and ASMR of CVD and subtypes were predicted, excluding those with missing data across the reproductive-age population. RHD, IHD and stroke were identified as the leading causes of CVD incidence burden, with their ASIRs were predicted to be 96.99, 66.35, and 25.45 per 100,000 population, respectively, by 2050. RHD showed the highest predicted increase in ASIR, with a growth rate of 50.17%. Furthermore, although the ASMR of IHD and stroke both declined from 2022 to 2050 with rates of 3.24% and 26.23% respectively, it is predicted that they will remain the main causes of CVD-related mortality by 2050, at 8.35 and 5.09 per 100,000 population, respectively (Figures S6–S7 and Table S2).

Decomposition Analysis

Globally, population growth was the dominant driver of increases in the absolute number of CVD incident cases (77.87%) and deaths (442.15%) among WCBA from 1990 to 2021, whereas population ageing played a greater role in higher SDI settings, with epidemiological improvements partially offsetting mortality increases (Figure 6A, 6B and Table S3). At the regional level, substantial heterogeneity was observed in the drivers of changes. South Asia represented a population growth-dominated pattern, with population expansion accounting for most of the increase in incident cases (79.13%) and deaths (183.29%). In contrast, East Asia exhibited a distinct ageing-driven pattern, in which population ageing was the primary contributor to the increasing burden (132.42%), partially offset by favorable epidemiological changes (171.49%), particularly for mortality (Figure 6C and D).

Figure 6 Decomposition analysis of CVD incidence and mortality among WCBA globally and in 5 SDI regions, 1990–2021. Changes were decomposed into contributions from population growth, population ageing, and epidemiological change. (A) Incidence burden of 5 SDI regions. (B) Mortality burden of 21 GBD regions. (C) Incidence burden of 21 GBD regions. (D) Mortality burden of 21 GBD regions. Abbreviations: CVD, cardiovascular disease; WCBA, women of childbearing age.

Decomposition patterns for major CVD subtypes were broadly consistent with those observed for total CVD, although notable subtype-specific differences were evident (Figure S8). For IHD, one of the dominant contributors, population growth drove increases in low and low-middle SDI regions while ageing predominating in higher SDI settings. In contrast, stroke showed a more pronounced ageing-driven pattern, particularly in East Asia, with substantial epidemiological improvements partially offsetting the mortality increases. A distinct pattern was observed for RHD. Unlike other major CVD subtypes, population ageing contributed negatively to changes in RHD incidence, reflecting a younger age of onset for RHD.

Discussion

This study provides a comprehensive and up-to-date evaluation of the global burden and evolving trends of CVD among WCBA. We confirmed the substantial burden of CVD in this population and identified rising trends in cases and deaths in recent years. Notably, our analysis revealed significant disparities in CVD burden and trends across WCBA subgroups, stratified by age, geographic region, and SDI. While both ASIR and ASMR of CVDs declined from 1990 to 2021, the incidence of certain major subtypes, including IHD and RHD, increased significantly. To be specific: First, the global incidence case and death among WCBA increased, while both ASIR and ASMR decreased from 1990 to 2021. Second, although annual decrease trends of ASIR and ASMR were observed in overall CVDs, several CVD subtypes including AF/AFL, UHD, PAH, RHD increased. Third, among WCBA, the burden of CVDs increased with age groups in 2021, except RHD. Fourth, RHD was most incident among WCBA world-wide, while IHD had the most popular death. Fifth, both ASIR and ASMR showed a consistent downward trend with increasing SDI, while several CVD subtypes especially Lower extremity PAD showed a contrary tendency. Sixth, the Predictions of CVDs among WCBA from 2022 to 2050 showed an increasing ASIR and a decreasing ASMR. At last, decomposition analysis indicated that the improvement in CVD risk was insufficient to completely offset the population expansion and aging of WCBA. These findings update and expand the epidemiological evidence on CVD in WCBA, offering a detailed profile and novel insights.

Our results demonstrate an increase in the incidence cases of CVD among WCBA from 1990 to 2021, driven primarily by rising cases of IHD, RHD, AF/AFL and PAH. The coexistence of declining overall ASIR and increasing incidence of IHD and RHD may reflect the limitations of aggregated measures, which can obscure heterogeneous trends across subtypes. This rising trajectory appears driven by a combination of demographic aging, increasing population-level exposure to cardiometabolic risk factors, and persistent disparities in healthcare access and quality.^7,29 Notably, modifiable risk factors remain central to the CVD burden in WCBA. It is estimated that approximately 90% of stroke cases in WCBA are attributable to modifiable risks, including hypertension, smoking, diabetes, dyslipidemia, and physical inactivity.³⁰ Meanwhile, there is growing recognition of female-specific and pregnancy-related risk contributors. Pregnancy-related factors such as oral contraceptive use, hypertensive disorders of pregnancy, gestational diabetes, and preterm delivery are increasingly understood not only as transient clinical events but as markers of long-term cardiovascular vulnerability.^31,32 These exposures may unmask underlying predispositions, effectively revealing future CVD trajectories. Despite growing evidence, heterogeneity across studies and limited statistical power underscore the need for further research to elucidate risk-factor pathways and enable targeted prevention strategies.³³ These findings underscore the urgency of implementing sex-sensitive prevention strategies across a woman’s reproductive life course. This mandates the integration of CVD risk screening into standard gynecological and maternal care, with targeted interventions during pre- and inter-pregnancy periods, supported by robust collaboration between cardiology, obstetrics, and primary care.³⁴

Marked regional disparities in the burden of CVD among WCBA underscore the complex interplay between socioeconomic development, healthcare accessibility, and epidemiological transitions.³⁵ Regions such as Sub-Saharan Africa and Oceania continue to be disproportionately affected, with persistently elevated ASIR and ASMR, indicating substantial gaps in preventive and therapeutic cardiovascular care. These findings are consistent with previous evidence showing that low- and middle-income countries disproportionately bear the burden of CVDs due to delayed diagnosis, limited availability of essential medications, and under-resourced health systems.³⁶ Importantly, the increasing incidence of RHD is largely concentrated in low-SDI settings, further highlighting the role of structural inequalities in shaping subtype-specific CVD patterns. Notably, while the overall ASIR has declined in most regions, certain countries still exhibit increasing trends, suggesting that global progress in CVD control is uneven and may be offset by country-specific epidemiological challenges such as conflict, poverty, and rising metabolic risk factors. In addition, the exceptionally severe burden of ASMR in Oceania may reflect the region’s unique profile characterized by a high prevalence of obesity and limited healthcare infrastructure.³⁷ These geographic differences highlight the importance of context-specific strategies in CVD prevention and management. Considering the regional risk factor profile and the capacity of the health care system, tailor-made intervention measures are crucial for achieving fair improvement in cardiovascular health outcomes globally.

Interestingly, our findings indicate that both the incidence and mortality rates of CVD increase with age, aligning with well-established epidemiological evidence.³⁸ Results of decomposition analysis also underscore the significant contribution of population aging to the CVD burden. This age-related pattern is primarily attributed to the cumulative exposure to risk factors over time, progressive deterioration of cardiovascular function, and the chronic low-grade inflammatory state commonly observed in older individuals.^39,40 Among various CVD subtypes, IHD remains the leading cause of death across all age groups, while RHD demonstrates a distinct early-onset pattern and represents one of the few subtypes with increasing incidence, stroke and IHD consistently emerge as the predominant contributors to the CVD burden across all age groups within WCBA, underscoring the importance of age-stratified prevention and management strategies. Given that stroke in early reproductive years is frequently associated with congenital vascular anomalies and inherited coagulation disorders.^41,42 As women progress into mid-to-late reproductive age, the burden of stroke and IHD increases, likely due to reduced physical activity, prolonged exposure to smoking and alcohol, hormonal fluctuations, and a history of adverse pregnancy outcomes.^43–46 Regular screening among high-risk individuals, public education on the harms of smoking and alcohol, promotion of healthy weight, and physical activity are crucial. Furthermore, prenatal risk assessments for older pregnant women may reduce adverse maternal outcomes.⁴⁷ Conversely, the incidence of RHD is highest in younger age groups and decreases sharply with age. This phenomenon may stem from its etiopathogenesis, characterized by an aberrant autoimmune response to group A streptococcal infection in genetically predisposed hosts, a population that has diminished in prevalence over time.⁴⁸ In regions where RHD remains endemic, strengthening primary prevention through timely treatment of streptococcal infections, implementing secondary prevention with long-term penicillin prophylaxis, and improving living conditions are critical strategies to reduce disease burden, particularly in resource-limited settings.⁴⁹

In 2021, IHD and stroke remained the two leading contributors to CVD burden among WCBA across all SDI regions. RHD demonstrated a strikingly higher incidence in low-SDI regions, with a steep decline as SDI increased, whereas lower extremity PAD exhibited the highest incidence in high-SDI regions. Notably, the incidence and mortality rates of IHD and stroke showed no pronounced heterogeneity across SDI strata. This pattern likely reflects the counterbalancing effects between advanced medical technology and robust health systems. Although advances in medical interventions have substantially improved the diagnosis and treatment of IHD and stroke,^50,51 lifestyle modifications associated with higher SDI regions—such as sedentary behavior, unhealthy diet, and psychosocial stress—may contribute to elevated risks and poorer prognoses.^52,53 Conversely, in low-SDI regions, despite relatively lower levels of some lifestyle-related risk factors,⁵² limited access to health education and essential medical resources contributes to persistently high mortality rates.⁵⁴ Furthermore, pregnancy-related complications including obesity, diabetes, and pulmonary diseases, serve as major risk factors for IHD throughout the reproductive period, posing serious threats to maternal health and pregnancy outcomes.⁵⁵ The high incidence of lower-extremity PAD in higher SDI regions also reflects an epidemiological transition toward chronic noncommunicable diseases driven by long-term exposure to metabolic risk factors such as diabetes and dyslipidemia.⁵⁶ Local governments and healthcare authorities should emphasize the early and active management of metabolic disorders among WCBA, with strict glycemic control during pregnancy to prevent the progression to overt CVD.⁵⁷ As a preventable disease, the persistently high incidence of RHD in lower-SDI regions underscores the consequences of inadequate primary healthcare infrastructure. According to the WHO, children and adolescents living in overcrowded and unsanitary environments remain the most vulnerable population for RHD.⁵⁸ Therefore, policy interventions should prioritize strengthening primary healthcare systems, integrating penicillin prophylaxis into maternal and child health programs, and promoting simplified echocardiographic screening to enable early diagnosis and effective secondary prevention.⁵⁹

This study has several limitations. First, although GBD 2021 utilized extensive data sources and advanced modeling techniques, the accuracy of CVD burden estimates remains constrained by the quality and availability of input data, particularly in low-resource settings. Second, the classification of CVD subtypes, while aligned with ICD standards, may suffer from clinical overlap, potentially leading to misclassification. Third, the use of EAPC assumes a log-linear trend over time and may not adequately capture potential non-linear or segmented changes in disease burden. Although visual inspection of temporal trends supports the consistency of our findings, future studies could apply more flexible approaches to better characterize complex temporal dynamics. Lastly, the ecological nature of GBD data and lack of individual-level information among WCBA limit the ability to explore specific risk factors or causal inferences.

Conclusion

This study provides a comprehensive assessment of the global burden of CVD among WCBA from 1990 to 2021 using GBD 2021 data. Stroke and IHD remain the leading contributors to both incidence and mortality, while RHD exhibits an early-onset pattern associated with low socioeconomic development. Although ASRs have declined, the substantial increase in absolute case numbers highlights a growing public health challenge. These findings underscore the need for concrete public health actions, including prioritizing resource allocation to low-SDI regions, strengthening primary healthcare systems for early detection and management, and integrating cardiovascular risk screening into routine reproductive and maternal healthcare. In addition, age-specific and region-specific prevention strategies should be implemented to effectively reduce the preventable burden of CVD in this vulnerable population.