Zhu · Actas espanolas de psiquiatria 2026 · epidemiological time-series analysis and forecasting study · n=204 countries and territories

Bridging Three Decades: Global Self-Harm Trends From 1990-2021 and Projections to 2040.

Level 3 - non-randomized controlled study

Non-randomized observational epidemiological time-series and modeling study across national populations.

PubMed 42343739 · doi:10.62641/aep.v54i3.2111 · record verified 2026-08-26

What was done

Self-harm data from the Global Burden of Disease (GBD) 2021 study were analyzed across 204 countries and territories from 1990 to 2021. The authors extracted metrics including incidence, prevalence, mortality, years of life lost (YLLs), years lived with disability (YLDs), and disability-adjusted life years (DALYs). Trends were assessed using age-standardized rates and estimated annual percentage change (EAPC). Socioeconomic inequality was measured with the Slope Index of Inequality and Concentration Index. Future burden from 2022 to 2040 was projected using Autoregressive Integrated Moving Average (ARIMA) models.

What was found

From 1990 to 2021, age-standardized self-harm rates decreased globally and across all sociodemographic index (SDI) levels, with the largest declines in high-middle SDI countries and greater reductions among females. DALY inequalities decreased over time but remained higher among females in lower-SDI areas. By 2040, global self-harm deaths are projected to reach 829,853 (95% UI, 262,233-1,397,474) and prevalence is projected to reach 35,863,341 (95% UI, 8,079,108-63,647,574), representing a 131.9% increase from the 2021 baseline of 15,467,153 cases.

Why it matters

Despite decreasing age-standardized rates over the past three decades, total cases and deaths from self-harm are projected to rise substantially by 2040, with persistent health inequities concentrated among females in lower-SDI populations.

Limits

The analysis relies on secondary modeled estimates from the GBD study, which are subject to variations in regional data quality, reporting standards, and stigma-related underreporting of self-harm. The ARIMA model projections have wide uncertainty intervals and cannot account for unforeseen social, economic, or public health disruptions.

Cited by