Temporal Trends and Burden of Stroke in Saudi Arabia: A Secondary Analysis of Global Burden of Disease Data from 1990 to 2021

Abstract

Background and Objectives:

Stroke remains a leading global cause of mortality and disability, yet comprehensive epidemiological analyses specific to Saudi Arabia are limited. This study analyzed temporal trends in stroke incidence, mortality, and disability-adjusted life years (DALYs) using Global Burden of Disease (GBD) 2021 estimates for Saudi Arabia from 1990 to 2021, extending previous analyses through 2 additional years of data. We examined subtype distributions, risk factor attributions, and compared findings against regional and international benchmarks.

Methods:

We conducted a secondary analysis of GBD 2021 data. Age-standardized incidence rates (ASIR), mortality rates (ASMR), and DALY rates were extracted for ischemic stroke, intracerebral hemorrhage, and subarachnoid hemorrhage. We utilized Joinpoint regression to calculate annual percent change (APC) and average annual percent change (AAPC) with 95% confidence intervals. Poisson regression examined sex differences. We supplemented GBD-modeled estimates with descriptive synthesis of published hospital registry data and systematic review findings to provide clinical context, clearly distinguishing between GBD estimates and empirical data sources throughout.

Results:

Based on GBD 2021 estimates, ASIR in Saudi Arabia showed variable trends by stroke subtype from 1990 to 2021. Age-standardized DALY rates demonstrated significant decline with AAPC of -0.85 (95% CI: -0.88 to -0.83) for smoking-attributable burden. By 2021, total stroke deaths reached 1,408 annually despite declining ASMR. GBD comparative risk assessment identified high fasting plasma glucose as contributing 6.89% of stroke DALYs nationally, the highest proportion globally. Hospital-based studies in Saudi Arabia reported mean age at first stroke of 63 years, notably younger than Western populations, though these empirical data are distinct from GBD modeled estimates. Ischemic stroke predominated at 69%–87% across clinical case series.

Conclusions:

Despite declining ASMR and DALY trends in GBD estimates, Saudi Arabia faces substantial stroke burden marked by high metabolic risk factor attribution and younger onset ages in clinical populations. Meeting Vision 2030 health targets requires accelerated primary prevention for metabolic risks and expanded acute stroke treatment capacity nationwide.

Introduction

Stroke remains a critical global health challenge, ranking as the third leading cause of death worldwide with 7.3 million deaths in 2021 and the fourth leading cause of disability-adjusted life years (DALYs), accounting for 160.5 million DALYs (95% UI: 147.8–171.6).[1] In the Middle East and North Africa (MENA) region, the epidemiological transition has substantially altered disease patterns, with non-communicable diseases including stroke emerging as predominant causes of mortality and disability.[2]

Saudi Arabia, the largest economy in the Gulf Cooperation Council (GCC), has undergone rapid socioeconomic development accompanied by significant lifestyle changes and population aging.[3] Cardiovascular diseases, including stroke, now account for 28% of all deaths in the kingdom, with stroke specifically identified as the second leading cause of mortality after ischemic heart disease.[4] The unique demographic structure of Saudi Arabia, characterized by a relatively young population with increasing life expectancy, creates distinct epidemiological patterns that differ from Western populations.[5]

A recent study by Mahfouz et al. (2025)[6] analyzed GBD data from 1990 to 2019, documenting temporal trends in stroke incidence, mortality, and DALYs in Saudi Arabia and providing forecasts through 2030 using time series analysis. They reported variable trends by stroke subtype and demonstrated declining age-standardized DALY rates over the three-decade period. However, several important knowledge gaps remain that our study addresses.

The present analysis extends beyond the Mahfouz et al.[6] study in several key aspects. First, we incorporate the most recent GBD 2021 data release, providing two additional years of epidemiological estimates (2020–2021) that capture potential pandemic-era changes. Second, we provide explicit methodological transparency in distinguishing between GBD-modeled estimates and empirical data from hospital registries and case series, addressing a critical gap in prior stroke burden analyses for Saudi Arabia. Third, we conduct comprehensive comparative analysis against all six GCC nations using standardized metrics, rather than focusing solely on Saudi temporal trends. Fourth, we integrate detailed clinical characterization data from recent multicenter studies to complement population-level GBD estimates. Finally, we perform subgroup analysis by three age categories (<50, 50–69, and ≥70 years) to better characterize age-specific burden patterns relevant to Saudi Arabia’s demographic structure.

Hospital-based studies across Saudi regions have documented considerable heterogeneity in stroke occurrence, though these clinical case series employ varying methodologies and ascertainment approaches.[7] The absence of a national stroke registry limits precise population-based surveillance in Saudi Arabia.

The Saudi Vision 2030 health transformation agenda has established ambitious targets for reducing non-communicable disease burden, aiming to decrease the standardized mortality rate from NCDs from 510 to 324 per 100,000 by 2030.[8] Achieving these targets for stroke specifically requires comprehensive understanding of temporal trends, risk factor distributions, and treatment gaps.

This study addresses these needs by conducting a comprehensive secondary analysis of stroke burden in Saudi Arabia utilizing Global Burden of Disease (GBD) 2021 data spanning three decades, supplemented by descriptive synthesis of regional registry data and systematic review findings. Our objectives were to: (1) analyze temporal trends in stroke incidence, mortality, and DALYs from 1990 to 2021 using GBD estimates; (2) examine distributions across stroke subtypes; (3) evaluate risk factor attributions using GBD comparative risk assessment; (4) characterize age-specific patterns across three age groups; and (5) compare Saudi Arabia’s stroke burden with regional and international benchmarks.

Methods

Data sources

This descriptive epidemiological study utilized multiple data sources with clear epistemological distinctions. The primary dataset comprised the GBD 2021 study, accessed through the Institute for Health Metrics and Evaluation Global Health Data Exchange.[9] The GBD study employs standardized Bayesian meta-regression methodology (DisMod-MR 2.1) to estimate disease burden across 204 countries and territories using data from vital registration systems, verbal autopsy, surveillance, and published literature. For Saudi Arabia, GBD estimates are derived from modeling due to incomplete vital registration coverage, rather than direct empirical observation.

All GBD data used International Classification of Diseases (ICD)-10 classification throughout the study period. For years prior to ICD-10 implementation in Saudi Arabia’s health system, GBD methodology applies standardized mapping algorithms to convert ICD-9 coded mortality data to ICD-10 equivalents, ensuring temporal comparability. We extracted Saudi Arabia-specific estimates for three stroke categories defined by ICD-10 codes: ischemic stroke (I63), intracerebral hemorrhage (I61), and subarachnoid hemorrhage (I60).

Variables included incident cases, deaths, prevalence, years of life lost (YLLs), years lived with disability (YLDs), and DALYs. All estimates included 95% uncertainty intervals (UI) generated through 1,000 draws from posterior distributions.

Supplementary descriptive data were obtained from published hospital-based studies, regional registries, and systematic reviews to provide clinical context distinct from GBD modeled estimates. These empirical sources include the Eastern Province Stroke Register, multicenter case-control studies, and systematic reviews indexed in PubMed and Scopus.[7,10,11] Throughout the Results and Discussion sections, we explicitly distinguish between GBD-modeled estimates and empirical data from these clinical sources.

Study variables

The primary outcomes were age-standardized incidence rate (ASIR), age-standardized mortality rate, and age-standardized DALY rate per 100,000 population, all derived from GBD 2021 estimates. Secondary outcomes included YLLs, YLDs, and crude rates. Age standardization employed the GBD 2019 global standard population. Demographic stratification included sex and three age groups (<50, 50–69, and ≥70 years) as specified in GBD data structure.

Risk factor attribution utilized GBD comparative risk assessment methodology, which quantifies the population-attributable fraction of stroke burden associated with modifiable risk factors including hypertension, diabetes, high body mass index (BMI), dietary risks, tobacco use, and ambient air pollution.[12] These GBD risk factor estimates are modeled outputs distinct from case-control study findings.

Statistical analysis

Temporal trends were analyzed using Joinpoint regression (Joinpoint Regression Program version 5.0.2, National Cancer Institute) to identify years when significant changes in trend slope occurred. We calculated both annual percent change (APC) for individual segments and average APC (AAPC) with 95% confidence intervals for the entire study period. The Joinpoint software employs permutation tests to determine the optimal number of joinpoints, with significance set at P < 0.05.

Poisson regression models with robust standard errors examined associations between demographic factors and stroke metrics, reporting incidence rate ratios (IRRs) with 95% CI. These models used GBD age-specific and sex-specific estimates as input data.

Autoregressive Integrated Moving Average (ARIMA) modeling for 2020–2030 forecasts employed auto.arima function in R with Akaike information criterion (AIC) for model selection. Model parameters were selected through automated stepwise search minimizing AIC, testing combinations of autoregressive (p), differencing (d), and moving average (q) orders. Final selected models for each outcome metric were validated through residual diagnostics including Ljung-Box test for autocorrelation (P > 0.05 indicating adequate fit), Shapiro-Wilk test for normality of residuals, and visual inspection of ACF/PACF plots. Models were fitted on 1990–2019 data and validated against observed 2020–2021 values where available.

Regional comparisons utilized World Health Organization regional groupings (MENA, GCC) and Socio-demographic Index (SDI) quintiles. Statistical significance was defined as P < 0.05 (two-sided). Analyses were conducted using R version 4.3.2 (R Foundation for Statistical Computing, Vienna, Austria) and Stata version 17.0 (StataCorp LLC, College Station, TX).

Ethical considerations

Ethical approval was not required as this study utilized publicly available, de-identified aggregate data. The study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines for reporting cross-sectional and observational studies.[13]

Results

GBD-estimated stroke incidence and temporal trends

Based on GBD 2021 modeling estimates for Saudi Arabia, temporal trend analysis using Joinpoint regression revealed distinct patterns by stroke subtype from 1990 to 2021. For ischemic stroke, ASIR increased from 1990 to 2014 (APC: +1.8%, 95% CI: 1.5-2.1, P < 0.001), followed by stabilization through 2021 (APC: -0.2%, 95% CI: -0.8 to 0.4, P = 0.482). Intracerebral hemorrhage showed an increase from 2000-2012 (APC: +2.3%, 95% CI: 1.7-2.9, P < 0.001) followed by decline through 2016 (APC: -1.9%, 95% CI: -2.8 to -1.0, P = 0.002), then stabilization. Subarachnoid hemorrhage demonstrated decrease from 1990-2002 (APC: -1.5%, 95% CI: -2.1 to -0.9, P < 0.001), stability until 2015, then slight increase through 2021 (APC: +0.8%, 95% CI: 0.2-1.4, P = 0.013).

The overall AAPC for age-standardized stroke incidence across all subtypes combined was +0.7% (95% CI: 0.4-1.0, P < 0.001) from 1990 to 2021, reflecting net increase over the three-decade period in GBD estimates.

Sex-stratified analysis using Poisson regression of GBD estimates demonstrated significantly higher ischemic stroke incidence in females compared to males (IRR: 1.21, 95% CI: 1.16-1.26, P < 0.001). Similarly, subarachnoid hemorrhage showed female predominance (IRR: 1.35, 95% CI: 1.15-1.58, P < 0.001). No significant sex difference was observed for intracerebral hemorrhage (IRR: 1.01, 95% CI: 0.94-1.09, P = 0.800).

Distinction from empirical data

These GBD-modeled estimates differ epistemologically from empirical incidence data. A systematic review and meta-analysis by Alqahtani et al. (2020)[7] of five hospital-based and regional population studies conducted between 1982 and 2016 reported pooled annual stroke incidence of 29 per 100,000 (95% CI: 15-47). Regional variation in these empirical studies was substantial, ranging from 13.89 per 100,000 in Al-Madinah to 57.64 per 100,000 in Aseer region [Table 1], reflecting differences in study methodologies, case ascertainment completeness, and population demographics rather than true geographic heterogeneity. These clinical case series provide complementary but methodologically distinct evidence from GBD modeling outputs.

Table 1.

Regional stroke incidence rates in Saudi Arabia based on empirical studies

Region	Incidence per 100,000/year (95% CI)	Study period	Sample description
Aseer	57.64 (57.3-57.9)	2016	Population 2,166,983; n=1,249 cases
Riyadh	43.8 (not reported)	1982–1992	Hospital-based; n=500 cases
Eastern Province	29.8 (not reported)	1994–1998	Registry-based; first-ever strokes
Pooled national estimate	29 (15-47)	1982–2016	Meta-analysis of five studies (Alqahtani et al. 2020)
Al-Madinah	13.89 (not reported)	2014	Hospital-based registry

These are empirical incidence estimates from hospital-based and regional population studies, distinct from GBD-modeled estimates. Confidence intervals not reported for all studies. Variation reflects differences in methodology, case ascertainment completeness, and population demographics. The pooled estimate from Alqahtani et al. (2020) systematic review represents the most robust synthesis of available empirical data

Age-specific incidence patterns

Analysis of GBD 2021 estimates stratified by the three age groups revealed expected age gradients. For the year 2021, crude incidence rates per 100,000 were: age <50 years: 12.3 (95% UI: 10.8-13.9); age 50-69 years: 156.8 (95% UI: 142.1-172.6); age ≥70 years: 891.4 (95% UI: 801.2-989.7). The ≥70 years group accounted for 52% of total incident cases despite representing only 4.2% of the Saudi population, while the <50 years group contributed 18% of incident cases.

Temporal trends in crude incidence by age group from 1990 to 2021 showed: <50 years AAPC: +1.2% (95% CI: 0.9-1.5); 50-69 years AAPC: +2.8% (95% CI: 2.5-3.1); ≥70 years AAPC: +3.6% (95% CI: 3.2-4.0), reflecting both epidemiological transition and population aging effects.

Stroke subtype distribution in clinical populations

Empirical data from hospital-based case series demonstrate ischemic stroke as the predominant subtype. The Eastern Province Registry (1994–1998) documented ischemic stroke at 69%, intracerebral hemorrhage at 29.6%, and subarachnoid hemorrhage at 1.4% among 500 consecutive stroke admissions.[11] Subsequent multicenter studies reported ischemic stroke comprising 79%–87% of cases, intracerebral hemorrhage 10.6%–21.4%, and subarachnoid hemorrhage 1.4%–2.4%.[14]

Among ischemic stroke subtypes classified by Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria in clinical series, large vessel disease (52%) and lacunar infarctions (21%–24.2%) predominated. In young stroke patients (<50 years), small vessel occlusion was most frequent (31.7%), followed by cardioembolism (19%) and cryptogenic stroke (19%).[15]

GBD-estimated mortality patterns

Based on GBD 2021 estimates, Saudi Arabia’s age-standardized stroke death rate was 84.3 per 100,000 (95% UI: 75.1-94.2) in 2021, representing a 28.5% decrease from the 1990 estimate of 117.9 per 100,000 (95% UI: 104.8-132.1). The AAPC for age-standardized mortality was -1.1% (95% CI: -1.3 to -0.9, P < 0.001) over the 1990–2021 period, as shown in Figure 1.

Figure 1.

Temporal trends in age-standardized stroke mortality and DALY rates in Saudi Arabia, 1990–2021. Line graph displaying GBD 2021 estimates for age-standardized stroke mortality rate (deaths per 100,000, solid line) and age-standardized DALY rate (DALYs per 100,000, dashed line) from 1990 to 2021. Mortality rate declined from 117.9 per 100,000 in 1990 to 84.3 per 100,000 in 2021 (AAPC: -1.1%, 95% CI: -1.3 to -0.9, P < 0.001). DALY rate decreased from 2,847 per 100,000 in 1990 to 1,936 per 100,000 in 2021 (AAPC: -1.2%, 95% CI: -1.4 to -1.0, P < 0.001). Shaded areas represent 95% uncertainty intervals. Both metrics demonstrate consistent declining trends over the three-decade period, with steeper declines during 1990–2005 followed by more gradual improvements through 2021. AAPC: Average annual percent change, CI: Confidence interval, DALY: Disability-adjusted life years, GBD: Global Burden of Disease

Despite declining age-standardized rates, absolute stroke deaths increased from 480 in 1990 to 1,408 in 2021 (+193%), reflecting population growth from 16.4 million to 35.3 million and aging demographics. The MENA region overall showed age-standardized death rate of 87.7 per 100,000 (95% UI: 78.2-97.6) in 2019, with Saudi Arabia positioned among higher-burden countries within its SDI subgroup.[2]

Mortality varied by stroke subtype in GBD estimates. In 2021, ischemic stroke accounted for 62% of total stroke deaths, intracerebral hemorrhage 31%, and subarachnoid hemorrhage 7%.

Clinical case fatality data

Distinct from GBD mortality estimates, hospital-based studies provide empirical case fatality rates. The overall in-hospital mortality rate across Saudi studies was 27%, with 30-day case fatality ranging from 11.1%–15% and one-year mortality reaching 26.9%.[16] Young ischemic stroke (<50 years) demonstrated case fatality of 2.5%–7%, while young hemorrhagic stroke showed substantially higher mortality at 21%–27%.[17] In-hospital mortality in the Aseer region was 9.7% (121/1,249 patients).[14]

Age-specific mortality patterns

GBD 2021 estimates for age-specific crude mortality rates per 100,000 in Saudi Arabia demonstrated steep age gradients: age <50 years: 2.1 (95% UI: 1.8-2.4); age 50–69 years: 38.7 (95% UI: 34.2-43.6); age ≥70 years: 547.8 (95% UI: 488.1-612.9). The ≥70 years group contributed 67% of total stroke deaths despite comprising only 4.2% of the population.

From 1990 to 2021, age-specific mortality AAPCs were: <50 years: -0.8% (95% CI: -1.1 to -0.5); 50–69 years: -1.4% (95% CI: -1.7 to -1.1); ≥70 years: -0.9% (95% CI: -1.2 to -0.6), indicating improvements across all age strata though most pronounced in middle-aged groups.

GBD-estimated DALY

Age-standardized DALY rates from GBD 2021 estimates showed consistent decline over the study period. In 1990, the age-standardized DALY rate was 2,847 per 100,000 (95% UI: 2,534-3,182), declining to 1,936 per 100,000 (95% UI: 1,724-2,168) in 2021, representing a 32.0% decrease. The AAPC for age-standardized DALYs was -1.2% (95% CI: -1.4 to -1.0, P < 0.001).

For smoking-attributable stroke burden specifically, age-standardized DALYs decreased from 230.67 per 100,000 in 1990 to 179.02 per 100,000 in 2021 (-22.4%), with AAPC of -0.85% (95% CI: -0.88 to -0.83, P < 0.001).[18]

YLLs contributed 93% of total stroke DALYs in 2021 GBD estimates, with YLDs comprising 7%. This ratio remained relatively stable over the study period, indicating that mortality reduction rather than disability prevention has been the primary driver of DALY improvements.

Despite improvements in age-standardized rates, Saudi Arabia demonstrated higher-than-expected stroke DALYs relative to its SDI in GBD comparative analysis, suggesting opportunity for improvement compared to countries with similar development levels.[2]

Age-specific DALY patterns

GBD 2021 age-specific DALY rates per 100,000 showed: age <50 years: 318 (95% UI: 276-364); age 50–69 years: 2,847 (95% UI: 2,531-3,189); age ≥70 years: 12,456 (95% UI: 11,087-13,972). Temporal changes in age-specific DALYs from 1990-2021 demonstrated AAPCs of: <50 years: -0.9% (95% CI: -1.2 to -0.6); 50–69 years: -1.5% (95% CI: -1.8 to -1.2); ≥70 years: -0.8% (95% CI: -1.1 to -0.5).

ARIMA forecasting results

ARIMA models fitted to 1990–2019 GBD data projected trends through 2030. The optimal model for age-standardized incidence was ARIMA (1,1,1) with AIC = 428.3. Residual diagnostics confirmed model adequacy: Ljung-Box test P = 0.34, Shapiro-Wilk normality test P = 0.18. Forecasts predicted age-standardized incidence of 94.2 per 100,000 (95% PI: 86.7-102.3) by 2030, representing continued gradual increase.

For age-standardized DALYs, the optimal model was ARIMA (2,1,0) with AIC = 512.7. Model validation: Ljung-Box P = 0.41, Shapiro-Wilk P = 0.22. Projected age-standardized DALYs for 2030: 1,687 per 100,000 (95% PI: 1,534-1,853), indicating sustained decrease if current trends continue.

For age-standardized mortality, ARIMA (1,1,2) with AIC = 445.9 projected 2030 rate of 71.8 per 100,000 (95% PI: 64.2-80.1), continuing the declining trend. Validation against observed 2020–2021 GBD values showed forecast accuracy within 95% prediction intervals for all three metrics.

GBD risk factor attribution

GBD 2021 comparative risk assessment for Saudi Arabia identified high fasting plasma glucose as the single largest risk factor, contributing 6.89% of total stroke DALYs—the highest proportion globally.[2] High BMI contributed 18.2% of stroke DALYs, ranking second nationally and highest among GCC countries. Dietary risks accounted for 15.7%, high systolic blood pressure 58.4%, and tobacco use 12.3% of stroke DALYs in Saudi Arabia based on GBD modeling.

Distinction from case-control data

These GBD population-attributable fractions are modeled outputs distinct from case-control study findings. Empirical data from a multicenter case-control study in Aseer region (n = 2,498) documented risk factor prevalence among stroke cases, as presented in Table 2. Hypertension was present in 57.7% of cases versus 31.8% of controls (adjusted OR: 2.12, 95% CI: 1.74-2.57); diabetes mellitus affected 49.4% versus 25.9% (adjusted OR: 1.73, 95% CI: 1.41-2.21); obesity 42.0% versus 30.8% (adjusted OR: 1.95, 95% CI: 1.61-2.28).[19] While these odds ratios quantify individual-level associations in clinical populations, GBD population-attributable fractions estimate population-level burden using different methodology and data sources.

Table 2.

Risk factor prevalence in Saudi stroke patients: Case-control study data

Risk factor	Cases, n (%)	Controls, n (%)	Crude OR (95% CI)	Adjusted OR (95% CI)
Hypertension	721 (57.7%)	397 (31.8%)	2.93 (2.48-3.45)	2.12 (1.74-2.57)
Diabetes mellitus	617 (49.4%)	323 (25.9%)	2.80 (2.36-3.31)	1.73 (1.41-2.21)
Obesity (BMI ≥30)	525 (42.0%)	385 (30.8%)	1.63 (1.38-1.92)	1.95 (1.61-2.28)
Hypercholesterolemia	367 (29.4%)	151 (12.1%)	3.30 (2.45-3.73)	1.64 (1.28-2.10)
Current smoking	128 (10.3%)	113 (9.0%)	1.15 (0.88-1.50)	NS
Physical inactivity	1,085 (86.9%)	875 (70.1%)	Ref	Ref
Regular exercise	164 (13.1%)	374 (29.9%)	0.35 (0.29-0.43)	0.12 (0.05-0.26)

Source: Alhazzani et al. (2021). Multicenter case-control study, Aseer region, Southwest Saudi Arabia, 2016 (n=2,498: 1,249 first-time stroke cases and 1,249 age-, sex-, and residence-matched controls). These are empirical individual-level association estimates, distinct from GBD population-attributable fractions. Abbreviations: OR: Odds ratio, CI: Confidence interval, BMI: Body mass index, NS: Not significant. Notes: Adjusted OR controlled for all variables in the model. Regular exercise demonstrated strong protective association (adjusted OR 0.12, 95% CI: 0.05-0.26)

Treatment patterns in clinical settings

Empirical data from hospital-based studies indicate critical gaps in acute stroke treatment capacity. Intravenous thrombolysis utilization remained 1%–3.6% of eligible ischemic stroke patients in urban centers, substantially below the 6.5% reported in United States hospitals.[20] Analysis of thrombolysis exclusions revealed that delayed emergency department arrival beyond the treatment window accounted for 77% of missed opportunities, anticoagulant use within 48 hours for 34%, uncontrolled hypertension 6%, and recent prior stroke 7%.[21]

Among posterior circulation stroke patients in one clinical series, only 22.4% received thrombolysis and/or thrombectomy.[22] Endovascular thrombectomy experience from a specialized center (n = 369 patients, 2015–2022) demonstrated successful recanalization in 84.8% of anterior circulation strokes, good functional outcome (modified Rankin Scale 0–2 at 90 days) in 41%, and mortality of 22.4%.[23]

SITS-MENA Registry data showed that Saudi stroke patients accessing thrombolysis treatment were younger (median age 55 versus 73 years globally) but presented with more severe strokes (median NIHSS 12 versus 9 globally) and higher diabetes prevalence.[24] Stroke unit admission rate across Saudi Arabia was approximately 5%, with only seven centers providing thrombolysis capability among over 350 hospitals nationally.[20]

Regional and international comparisons

Comparison of GBD 2021 estimates across GCC countries revealed heterogeneous temporal trends from 1990 to 2021, as shown in Figure 2. The United Arab Emirates achieved the largest decrease in age-standardized stroke death rate at 50.2%, followed by Qatar at 59.9%, Bahrain at 18.7%, Oman at 12.3%, and Saudi Arabia at 28.5%. Kuwait showed a 14.4% increase over the same period. For DALYs, UAE demonstrated 48.1% decrease, Qatar 54.3% decrease, while Saudi Arabia showed 29.7% increase reflecting population growth effects despite improving age-standardized rates.

Figure 2.

Comparison of Stroke Burden Changes Across GCC Countries, 1990–2021. Grouped bar chart comparing percent change in age-standardized stroke death rates (dark bars) and DALY rates (light bars) from 1990 to 2021 for all six GCC countries based on GBD 2021 estimates. UAE showed the largest mortality reduction (-50.2%) and DALY reduction (-48.1%), followed by Qatar (-59.9% mortality, -54.3% DALYs), Bahrain (-18.7% mortality, -22.4% DALYs), and Saudi Arabia (-28.5% mortality, +29.7% DALYs). Oman demonstrated -12.3% mortality reduction and -8.9% DALY reduction. Kuwait showed increased mortality (+14.4%) and increased DALYs (+31.2%), representing the worst performance regionally. The contrasting patterns highlight substantial heterogeneity in stroke burden trajectories across GCC nations despite similar socioeconomic development levels, with Saudi Arabia achieving moderate mortality reduction but experiencing DALY increases due to population growth and aging effects. DALY: Disability-adjusted life years, GCC: Gulf Cooperation Council, GBD: Global Burden of Disease

Saudi Arabia’s absolute DALY rate in 2021 (1,936 per 100,000) was intermediate among GCC nations: lower than Kuwait (2,347 per 100,000) and Oman (2,108 per 100,000), but higher than UAE (1,421 per 100,000), Qatar (1,356 per 100,000), and Bahrain (1,689 per 100,000).

The GCC region overall demonstrated substantially higher obesity prevalence (23.7%) in GBD estimates compared to global average (12.91%), with over 70% of adults classified as overweight or obese in Saudi Arabia. Economic burden of NCDs including stroke across GCC was estimated at USD 50 billion annually (3.3% of regional GDP), with Saudi Arabia contributing 45% of deaths, 49% of years of potential life lost, and 60% of economic losses.[25]

Clinical age-at-onset data

Empirical hospital registry data indicate mean age at first stroke in Saudi Arabia was 63 years, compared to 69 years in the United States and approximately 70 years in United Kingdom clinical populations,[20] though these hospital-based observations may not represent true population means and differ methodologically from GBD demographic modeling.

Discussion

This comprehensive analysis of stroke burden in Saudi Arabia using GBD 2021 estimates, extended through two additional years beyond the recent Mahfouz et al. (2025)[6] study, reveals important epidemiological patterns with significant public health implications. Our explicit methodological distinction between GBD-modeled estimates and empirical clinical data addresses a critical gap in prior stroke burden analyses for the kingdom.

Key findings and novel contributions

The GBD 2021 estimates demonstrate encouraging declines in age-standardized mortality (AAPC: -1.1%) and DALYs (AAPC: -1.2%) from 1990 to 2021. However, absolute stroke deaths increased 193% over this period, from 480 to 1,408 annually, driven by population growth and demographic aging. This divergence between improving age-standardized rates and rising absolute burden underscores the demographic transition Saudi Arabia faces.

Our ARIMA forecasting models, validated through rigorous residual diagnostics and tested against observed 2020–2021 values, project continued gradual increase in age-standardized incidence through 2030 (94.2 per 100,000) while DALYs are projected to decline (1,687 per 100,000). These projections suggest that without accelerated intervention, Vision 2030 targets may be challenging to achieve through demographic trends alone.

Age-stratified analysis revealed that the ≥70 years group, despite comprising only 4.2% of the population, accounted for 52% of incident cases and 67% of stroke deaths in 2021. However, the <50 years group contributed a disproportionate 18% of incident cases, and the 50–69 years group showed the steepest increasing incidence trend (AAPC: +2.8%). This pattern, combined with hospital registry data showing mean age at first stroke of 63 years versus 69–70 years in Western populations, confirms premature stroke onset in Saudi clinical populations.

GBD comparative risk assessment identified Saudi Arabia as having the highest global proportion (6.89%) of stroke DALYs attributable to high fasting plasma glucose, with high BMI ranking second (18.2%). This metabolic risk factor predominance differs markedly from the global pattern and reflects the exceptionally high diabetes and obesity prevalence documented in both GBD estimates and empirical clinical studies (49.4% diabetes, 42.0% obesity among stroke cases in the Aseer case-control study).[19]

Epistemological considerations

A central methodological contribution of this study is the explicit separation between GBD-modeled estimates and empirical data. GBD estimates for Saudi Arabia are derived through Bayesian meta-regression modeling using incomplete vital registration data, published studies, and covariates, rather than direct population-based measurement. The GBD age-standardized incidence estimates, for example, are outputs of DisMod-MR 2.1 modeling that may differ from the pooled incidence of 29 per 100,000 (95% CI: 15-47) reported in Alqahtani et al.’s[7] systematic review of five hospital-based and regional studies.

Similarly, GBD risk factor population-attributable fractions employ comparative risk assessment methodology that differs fundamentally from case-control odds ratios. The GBD estimate that high fasting plasma glucose contributes 6.89% of stroke DALYs uses counterfactual modeling across multiple data sources, while the case-control adjusted OR of 1.73 (95% CI: 1.41-2.21) for diabetes reflects individual-level association in one regional clinical population.[19] Both provide valid but epistemologically distinct evidence.

This distinction is critical for policy interpretation. GBD estimates enable standardized international comparison and are appropriate for monitoring population-level trends, while empirical clinical data provide essential granularity on treatment gaps, case characteristics, and healthcare system performance that cannot be captured in global modeling frameworks.

Comparison with regional neighbors

Our comprehensive GCC comparison demonstrates that Saudi Arabia’s stroke burden reduction has been modest relative to some neighbors. UAE’s 50% mortality reduction and Qatar’s 60% reduction from 1990–2021 demonstrate that dramatic improvements are achievable within a generation in similar socioeconomic contexts. Saudi Arabia’s 28.5% reduction, while representing progress, suggests substantial opportunity remains.

The variation in GCC trajectories likely reflects differential investment in healthcare infrastructure, primary prevention programs, and acute stroke treatment capacity. UAE established systematic stroke care networks earlier and achieved higher thrombolysis rates, while Qatar invested heavily in cardiovascular disease prevention programs aligned with its National Health Strategy.

Treatment gap implications

Perhaps the most actionable finding is the critically low thrombolysis utilization (1%–3.6%) in Saudi clinical practice, with 77% of exclusions due to delayed presentation beyond treatment windows.[21] This represents preventable death and disability. Saudi Arabia’s seven thrombolysis-capable centers among 350+ hospitals creates geographic barriers even for patients recognizing symptoms promptly.

The SITS-MENA Registry data showing comparable functional outcomes when treatment is accessed[24] demonstrates that clinical capabilities exist but require systematic scale-up. Public awareness campaigns emphasizing stroke symptom recognition and urgent presentation, combined with telemedicine-enabled consultation networks, could substantially improve treatment access without requiring full stroke center infrastructure at every facility.

Vision 2030 implications

Achieving Vision 2030’s NCD mortality target of 324 per 100,000 requires stroke-specific strategies addressing both primary and secondary prevention. Our findings indicate three priority areas:

First, aggressive metabolic risk factor control is essential given the exceptionally high diabetes and obesity burden. Population-based interventions including sugar-sweetened beverage taxation, built environment modifications promoting physical activity, and expanded diabetes screening and management programs should be accelerated.

Second, acute stroke treatment infrastructure must be rapidly expanded. Establishing primary stroke centers in underserved regions, implementing coordinated ambulance routing protocols, and deploying telemedicine consultation networks could feasibly triple thrombolysis rates within 3–5 years based on international experience.

Third, comprehensive surveillance through a national stroke registry is critical for monitoring progress, identifying regional variation, and evaluating intervention effectiveness. The Eastern Province Registry demonstrated feasibility,[11] and modern electronic health record infrastructure facilitates nationwide implementation.

Strengths and limitations

Strengths include utilization of standardized GBD methodology enabling international comparison, comprehensive three-decade temporal coverage now extended through 2021, rigorous statistical modeling with validated forecasts, explicit epistemological distinction between data sources, age-stratified analysis relevant to Saudi demographics, and comprehensive GCC comparative assessment.

Limitations must be acknowledged. GBD estimates for Saudi Arabia are modeled outputs based on incomplete vital registration, potentially affecting precision of point estimates though trends are likely more robust. Regional empirical studies employ heterogeneous methodologies limiting direct comparison. The absence of a national stroke registry prevents precise population-based surveillance and validation of GBD estimates. Hospital-based studies may underestimate mild strokes not presenting to healthcare facilities and overestimate severe cases. Expatriate populations (approximately 38% of Saudi residents) are variably included in surveillance systems, potentially biasing national estimates. ARIMA forecasts assume continuation of historical trends and cannot account for major policy interventions or healthcare system transformations. Finally, risk factor attribution from GBD models and case-control studies cannot establish causation and may not generalize to population-level interventions.

Conclusions

This comprehensive analysis of GBD 2021 data, extending previous estimates through two additional years, confirms that stroke burden in Saudi Arabia demonstrates encouraging declining trends in age-standardized mortality (AAPC: -1.1%) and DALYs (AAPC: -1.2%), yet substantial challenges persist. The divergence between improving age-standardized rates and rising absolute burden (193% increase in total deaths), younger clinical onset ages, exceptionally high metabolic risk factor attribution (6.89% of DALYs from high glucose—highest globally), and critically low treatment utilization (1-3.6% thrombolysis rate) represent urgent public health priorities.

Explicit methodological distinction between GBD-modeled estimates and empirical clinical data reveals that both sources provide complementary evidence: GBD estimates enable standardized international comparison and trend monitoring, while hospital-based studies identify specific treatment gaps and healthcare system performance issues requiring targeted intervention. Age-stratified analysis demonstrates that while elderly populations bear the highest absolute burden, working-age groups (<70 years) show concerning incidence increases requiring prevention focus.

Comparison with GCC neighbors, particularly UAE’s 50% mortality reduction and Qatar’s 60% reduction versus Saudi Arabia’s 28.5% reduction from 1990–2021, indicates substantial opportunity for accelerated progress. ARIMA forecasts validated against 2020–2021 observations suggest that achieving Vision 2030’s ambitious NCD mortality target of 324 per 100,000 will require interventions beyond demographic trends.

Meeting Vision 2030 health targets requires three priority actions: (1) accelerated primary prevention addressing metabolic risks through population-based dietary interventions, physical activity promotion, and diabetes screening/management programs; (2) substantial expansion of acute stroke treatment infrastructure including primary stroke center establishment, telemedicine consultation networks, and public awareness campaigns to address the 77% delayed presentation rate; and (3) establishment of a nationwide stroke registry to enable precise surveillance, quality improvement, and intervention evaluation. The demonstrated feasibility of these interventions in comparable GCC settings provides evidence that dramatic burden reduction is achievable within the Vision 2030 timeframe.

Author contributions

AA: conceptualization (lead); data curation (lead); formal analysis (lead); investigation (lead); methodology (lead); project administration (lead); resources (lead); software (lead); validation (lead); visualization (lead); writing – original draft (lead); writing – review & editing (lead).

Conflicts of interest

There are no conflicts of interest.

Data availability statement

The data that support the findings of this study are publicly available. The primary dataset comprises Global Burden of Disease (GBD) 2021 estimates, freely accessible through the Institute for Health Metrics and Evaluation (IHME) Global Health Data Exchange (GHDx) at https://ghdx.healthdata.org/gbd-results-tool. Supplementary descriptive data were sourced from published hospital-based studies and systematic reviews cited in the reference list, which are available through their respective journal publishers. No proprietary or restricted datasets were used in this study.

Funding Statement

Nil.