The Modular Assessment of Risk for Imminent Suicide (MARIS): A validation study of a novel tool for suicide risk assessment

Abstract

Background: Reliable diagnostic tools for the short-term suicide risk assessment are needed. The recently developed multi-informant Modular Assessment of Risk for Imminent Suicide (MARIS) includes four modules: two are patient-rated and two clinician-rated. The patient-rated modules assess a proposed pre-suicidal cognitive/emotional state (Module 1) as well as patients' attitudes towards suicide (Module 2). The clinician-rated modules assess traditional suicide risk factors (Module 3) and clinicians' emotional responses to the patient (Module 4).

Methods: With the aim of extending our previous preliminary proof of concept findings, the MARIS was administered to 618 psychiatric patients (167 inpatients, 451 outpatients) and their clinicians (N = 115). Patients were assessed with a battery including the Columbia-Suicide Severity Rating Scale. Four outcomes were considered: lifetime and past month suicidal thoughts and behaviors (STB) (0-10 point scale) and suicidal behaviors (SB) (0-5 point scale). Reliability and concurrent, convergent/divergent and incremental validity were assessed.

Results: Good internal consistency was found for modules 1 and 4 (Cronbach's α: 0.87 and 0.86, respectively) but not for the others. Module 1's total score positively correlated with lifetime STB/SB and past month STB (all p ≤ 0.003). Module 4's total score positively correlated with all four outcomes (all p < 0.001). Modules 1 and 4 showed additional capacity to detect patients' lifetime and past month STB/SB beyond other associated factors.

Limitations: Lack of prospective assessment. Inpatients were evaluated at discharge, whereas outpatients at intake.

Conclusions: These findings supported the utility of multiple data sources to identify patients at imminent suicide risk, and in particular clinicians' emotional responses.

Keywords: Assessment/diagnosis; Depression; Health services; Measurement/psychometrics; Suicide/self harm.