AHA Scientific Sessions 2026 Abstract · Nov 2, 2026
Socioeconomic Gradients in Stroke Mortality Among Women and Men Globally: Evidence from the Global Burden of Disease Study 1980-2021
By Jaiswal A, Jaiswal V, Chowdhury F et al.
Background: Stroke remains a leading cause of death worldwide, yet the degree to which global mortality trends diverge across levels of socioeconomic development has not been comprehensively quantified over a multi-decade timeframe. Objective: This study examined whether the socioeconomic gradient in stroke mortality has widened between 1980 and 2021 across 204 countries and territories. Methods: This population-based observational study included age-standardized mortality rates (AAMR) obtained from the Global Burden of Diseases 2023 study between 1980 and 2021. The Sociodemographic Index (SDI) was used as an exposure measure. Generalized Additive Mixed Models (GAMMs) with gamma distribution errors, thin-plate regression splines, random intercepts at the location level, and autocorrelation correction for temporal AR1 were used to calculate relative risk rate ratios and absolute excess deaths per 100,000 individuals, based on the 80th SDI percentile. Risk factor decomposition was undertaken for behavioral, metabolic, and environment/occupation groups between 1990 and 2021. A Bayesian hierarchical sensitivity analysis taking into account the uncertainties of the Global Burden of Diseases study was done to validate our main findings. Results: At the 20th SDI percentile (SDI = 0.25), stroke mortality rate ratios relative to the 80th percentile reference were 1.56 (95% CI: 1.44-1.69) in males and 1.84 (95% CI: 1.70-2.00) in females in 1980, representing an 18.2% excess female mortality burden. By 2000, the 20th percentile ratios were 1.56 (95% CI: 1.44-1.70) in males and 1.85 (95% CI: 1.71-2.01) in females (female excess: 18.2%). In 2021, low-SDI populations (20th percentile; SDI = 0.49) showed ratios of 1.49 (95% CI: 1.38-1.61) in males and 1.72 (95% CI: 1.59-1.86) in females (female excess: 15.2%). Contour heatmaps confirmed progressive concentration of the global stroke burden into the lowest SDI percentiles over four decades. Risk factor decomposition demonstrated steep socioeconomic gradients for behavioral, metabolic, and particularly environmental/occupational risks. Stroke-specific rate ratios substantially exceeded all-cause mortality comparators, confirming disease specificity. Bayesian sensitivity analyses corroborated all primary findings after adjusting for registry unreliability.ConclusionsPeople from low SDI suffer from a substantially higher stroke mortality burden compared to high-SDI populations, with females being disproportionately affected, having excess risk greater than males by more than 15%. The stroke mortality inequality gradient appears to be disease specific and in part attributed to the unmitigated deadliness of environmental factors.