How many people die by suicide each year? Not 727,000: a systematic review and meta-analysis of suicide underreporting across 71 countries over 122 years.

Background: Suicide underreporting undermines accurate public health assessments and resource allocation for suicide prevention. This study aims at synthesizing evidence on suicide underreporting and to estimate a global underreporting rate. Methods: We conducted a PRISMA-compliant systematic review on suicide underreporting, following a pre-registered protocol. A meta-analytical synthesis was also conducted. Quantitative data from individual studies was extracted to provide an overall global estimate of suicide underreporting (42 studies covering 71 countries out of the initial 770 unique studies, spanning 1900-2021). Most studies used retrospective institutional datasets to estimate underreporting through reclassification of undetermined deaths or comparisons across databases. Demographic and geographic disparities were also examined. Results: The 42 studies selected provided some quantitative data on suicide underreporting for general or specific populations. 14 of these studies provided data to be meta-analyzed. The global suicide underreporting rate was estimated to be 17.9% (95% CI: 10.9-28.1%) with large differences between countries with high and low/very low data quality. In this scenario, the last WHO estimates of suicide deaths - corrected for underreporting - would be more than one million (1,000,638; 95% CI: 859,511-1,293,006) and not 727,000 suicides per year. Underreporting was higher in low- and middle-income countries (LMICs) with incomplete death registration systems, such as India and China (34.9%; 95% CI 20.3-53%), while high-income countries exhibited lower rates (11.5%; 95% CI 6.6-19.3%). Contributing factors included stigma, religiosity, limited forensic resources, and inconsistent use of International Classification of Diseases (ICD) codes. Gender and age disparities were notable; Female suicides and those among younger or older individuals were more likely to be misclassified. Discussion: Addressing suicide underreporting requires improving death registration systems globally, particularly in LMICs. Standardizing ICD usage, improving forensic capacity, and reducing stigma are critical steps to ensure accurate data. Heterogeneity, geographical disparities, temporal biases, and invariance of suicide underreporting for countries with low-quality data demand further corroboration of these findings. Systematic Review Registration: https://osf.io/9j8dg.