[Suicide epidemiology in the province of Málaga (Spain) [2024]: a retrospective analysis using machine learning techniques].

OBJECTIVE

Suicide constitutes a Public Health challenge whose demographic and geographical heterogeneity demands targeted interventions. The paucity of subnational studies integrating advanced data analytics restricts the precise identification of risk patterns. We therefore aimed to characterise the sociodemographic, temporal, and forensic profiles of suicides in Málaga during 2024, through integration of data from the Institute of Legal Medicine and Forensic Sciences (IMLCF).

METHODS

A retrospective observational study of all suicide cases (N=197) registered by the IMLCF in 2024 was carried out. Sociodemographic, temporal, geographical, and circumstantial variables were analysed using multivariate statistical techniques and machine learning algorithms (hierarchical clustering and k-means) to identify risk patterns and distinctive profiles.

RESULTS

There was male predominance (80.2%; male-female ratio=4:1) with a mean age of 54.51 years (SD=17.00). The most frequent methods were hanging (40.61%), jumping from height (26.90%) and substance ingestion (16.24%), with statistically significant sex differences (p<0.001). Geographical distribution showed higher rates in inland municipalities (mean 53.79/100,000 inhabitants) compared with coastal areas (mean 11.56/100,000). Cluster analysis revealed four profiles: middle-aged adults in inland areas, young adults in coastal regions, jumping-related deaths in coastal areas, and older adults in coastal regions.

CONCLUSIONS

This study represents the first detailed epidemiological mapping of suicide in Málaga, highlightingcontextual vulnerabilities and specific risk patterns, and supports stratified interventions in accordance with the WHO Mental Health Action Plan 2013-2030.