· Lancet (London, England) 2025 · Global epidemiological modeling study · n=55761 data sources (204 countries and territories)

Global burden of 292 causes of death in 204 countries and territories and 660 subnational locations, 1990-2023: a systematic analysis for the Global Burden of Disease Study 2023.

Cited 419 times in the scientific literature.

Level 3 - non-randomized controlled study

Design analogy: large-scale longitudinal epidemiological modeling using population vital statistics and surveillance data

PubMed 41092928 · doi:10.1016/S0140-6736(25)01917-8 · record verified 2026-08-29

What was done

The Global Burden of Disease Study (GBD) 2023 estimated mortality counts, rates, years of life lost (YLLs), probability of dying before age 70 (70q0), and mean age at death for 292 causes across 204 countries and territories and 660 subnational locations from 1990 to 2023. The analysis synthesized 55,761 data sources (vital registration, verbal autopsy, surveys, censuses, surveillance, and cancer registries) using the Cause of Death Ensemble model (CODEm) and generated 95% uncertainty intervals (UIs) from a 250-draw distribution.

What was found

COVID-19 ranked as the leading age-standardised cause of death globally in 2021 but dropped to 20th place by 2023, returning ischaemic heart disease and stroke to the top two positions. Global all-cause mean age at death increased from 46.8 years (95% UI 46.6–47.0) in 1990 to 63.4 years (63.1–63.7) in 2023 (males: 45.4 to 61.2 years; females: 48.5 to 65.9 years). In 2023, all-cause mean age at death was highest in the high-income super-region (females: 80.9 years [80.9–81.0]; males: 74.8 years [74.8–74.9]) and lowest in sub-Saharan Africa (females: 38.0 years [37.5–38.4]; males: 35.6 years [35.2–35.9]). Neonatal disorders remained the leading cause of global YLLs in all years except 2021, despite marked rate reductions. Large declines were documented for diarrhoeal diseases, tuberculosis, stomach cancer, measles, and other vaccine-preventable diseases. The 70q0 metric increased for drug use disorders (in high-income regions and globally for both sexes), conflict/terrorism in females, and diabetes in males.

Why it matters

This study tracks long-term global epidemiological transitions and the resolution of peak COVID-19 pandemic mortality. It reveals persistent geographical disparities in premature mortality and flags rising premature mortality risk from non-communicable diseases and substance use in specific regions.

Limits

The findings rely heavily on statistical modeling and imputation in locations lacking comprehensive, high-quality vital registration systems. Verbal autopsy and surveillance data carry inherent diagnostic misclassification and variability across countries.

Cited by