Registered adolescent suicide mortality in Colombia, 2015-2024: a Bayesian spatiotemporal mapping study

To describe temporal and geographic variation in registered adolescent suicide mortality in Colombia and its ecological association with multidimensional poverty.

This nationwide ecological study analyzed 4,234 presumed suicide deaths at ages 10-19 during 2015-2024 from the national forensic register (INMLCF). Bayesian Poisson models smoothed rates across neighboring departments and years. Aggregate rates combined four sex-age strata. Separate national vital-statistics (DANE) analyses compared suicide-coded deaths with two scenarios including undetermined intent.

Male-to-female rate ratios (RRs) were 2.00 (95% credible interval [CrI] 1.86-2.15) at ages 15-19 and 1.05 (0.93-1.18) at ages 10-14. Total posterior rates were 4.54 (4.22-4.87) per 100,000 in 2015 and 5.23 (4.86-5.60) in 2024; no adjacent-year change had a 95% CrI excluding zero. Putumayo, Arauca and Amazonas exceeded 10 per 100,000 in 2024. Fifteen INMLCF departments had a probability > 0.95 of rates above the national comparator in all ten years. The pooled Multidimensional Poverty Index (MPI) RR per standard deviation was 0.84 (0.74-0.94). Including usable DANE undetermined-intent deaths increased the strict count by 22.18%; the expanded pooled MPI RR was 1.04 (0.93-1.17).

Registered mortality varied by age, sex and department. MPI associations and geographic classifications depended on registration source and case definition. The findings do not establish protection from poverty and support strengthening adolescent surveillance and death classification.