Identifying subtypes of suicidality: a second larger consensus study in emergency clinical psychiatric practice

TL;DR

This study tests a model to identify four subtypes of suicidality in emergency psychiatric patients, showing strong agreement among professionals using a new tool.

Contribution

The study validates the usability of the (h)4ME model and SUICIDI-3 tool in a larger clinical sample.

Findings

Excellent agreement was found for Perceptual Disintegration and Primary Depressive Cognition subtypes.

The SUICIDI-3 tool showed good feasibility for assigning patients to subtypes.

Replication in diverse populations is needed to confirm the model's consistency.

Abstract

A model has been developed to distinguish subtypes of pathways to entrapment leading to suicidality in clinical mental health practice. The (hypothetical) 4-type Model of Entrapment ((h)4ME) delineates four subtypes of suicidality: I) Perceptual Disintegration (PD), II) Primary Depressive Cognition (PDC), III) Psychosocial Turmoil (PT) and IV) Inadequate Coping/Communication (IC).

To examine the model’s usability and feasibility in a larger cohort of suicidal patients (n= 75) following a pilot study.

Consultation reports to general practitioners of 75 suicidal emergency patients were independently allocated to subtypes by three psychiatrists and three nurses using the SUICIDI-3 tool. This tool describes the proposed subtypes. Interrater agreement was assessed by calculating Intraclass Correlation Coefficients (ICCs). Absolute and dimensional type agreement was established to assess the model’s usability and the SUICIDI-3 tools feasibility.

All raters were able to assign cases to subtypes. Excellent absolute Type Agreement (aTA) was observed for PD (0.96) and PDC (0.92), and good aTA for PT (0.83) and IC (0.83). For dimensional Type Agreement (dTA) the ICC was excellent for PD (0.97), PDC (0.95) and IC 0.92), and good for PT (0.88).

The (h)4ME demonstrates promising usability and feasibility when tested by staff of psychiatric emergency services. Replication studies of samples of various clinical, demographic or ethnic origin and of diverse professional background and contexts are needed to confirm the consistency of these findings.

Figures