· The Lancet. Public health 2025 · multinational epidemiological modeling study · n=23,782 study-location-years

Global, regional, and national burden of suicide, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021.

Level 3 - non-randomized controlled study

Global descriptive epidemiological modeling and surveillance analysis across population cohorts over time

PubMed 39986290 · doi:10.1016/S2468-2667(25)00006-4 · record verified 2026-08-26

What was done

Using data from the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2021, researchers modeled suicide deaths, age-standardized mortality rates, mean age at death, and firearm-specific suicide rates across 204 countries and territories from 1990 to 2021 by age and sex. Data sources comprised 23,782 study-location-years from vital registration systems, verbal autopsies, and mortality surveillance. Point estimates and 95% uncertainty intervals (UIs) were calculated from 1,000-draw distributions.

What was found

In 2021, an estimated 746,000 suicide deaths (95% UI 692,000–800,000) occurred globally (519,000 males, 227,000 females). The global age-standardized mortality rate declined from 14.9 per 100,000 (12.8–15.7) in 1990 to 9.0 per 100,000 (8.3–9.6) in 2021. The highest 2021 regional rates were in eastern Europe (19.2 [17.5–20.8] per 100,000), southern sub-Saharan Africa (16.1 [14.0–18.3] per 100,000), and central sub-Saharan Africa (14.4 [11.0–19.1] per 100,000). Mean age at death from suicide increased between 1990 and 2021 from 43.0 to 47.0 years in males and from 41.9 to 46.9 years in females. Medically attended suicide attempts were consistently higher in females, while firearm suicides predominated in males, led by the USA, Uruguay, and Venezuela.

Why it matters

This study provides comprehensive global benchmarking for suicide mortality trends across three decades, capturing an overall decline in global rates alongside an aging shift in the average demographic of suicide mortality.

Limits

Estimates rely on statistical imputation in regions with sparse, inconsistent, or absent vital registration, using verbal autopsy data that carries high uncertainty. Misclassification and underreporting driven by legal, cultural, and social stigma surrounding suicide remain substantial sources of potential bias.

Cited by