[Utility of the classification analysis in the typology of the the suicide attempt in a hospital population.]

Abstract

Objective: Subjects who attempt suicide are not a homogeneous group. Numerous studies accredit different subtypes of individuals with suicide attempts. One of the main applications of classification analysis is the segmentation and selection of the subject's characteristics. The objective of this study is to identify groups of subjects with a suicide attempt and to discover the relationships between them in the hospital population of an area of the Community of Madrid.

Methods: In a case-control study, the classification tree technique based on the CHAID (Chi-square automatic interaction detector) algorithm, included in the SPSS 23.0 program, was applied to a sample of 201 subjects admitted to a highly complex hospital of Madrid. Impulsiveness, couple conflict, psychiatric history and depression were evaluated.

Results: Three subtypes or differentiable profiles of the subject related to the suicide attempt have been obtained, (high profile 144 cases 71.64%, medium 35 cases 17.41% and low 22 cases 10.94% of basic risk patterns) with high application value to the clinical setting.

Conclusions: The typologies found allow us to adapt preventive measures on the suicide attempt and carry out focused clinical interventions of a preventive and predictive nature.