Psychometric properties, predictive utility and diagnostic capacity of the Persian version of the Scale of Resilience to Suicide Attempts in army conscripts.

Abstract

Introduction: Suicide among army conscripts represents a poorly understood and complex public health issue that has escalated in recent decades. The early identification of individuals at risk holds the potential to significantly contribute to the effective prevention of suicide attempts. To address this, the Scale of Resilience to Suicide Attempts-18 (SRSA-18) has been developed to assess protective factors related to suicide attempts. The present study aimed to develop a Persian version of SRSA-18 (P-SRSA) and examine its psychometric properties and predictive utility within a sample of the Iranian Army.

Methods: The initial study sample consisted of 400 male conscripts from the Iranian Army grand forces (M=19.86, SD=1.40). Subsequently, for the follow-up stage, 296 participants (M=19.83, SD=1.39) were recruited. The participants were assessed using various measures of resilience, social support, suicide and hopelessness at baseline and a 6-month follow-up.

Results: In contrast to the original version, both exploratory and confirmatory factor analyses supported an 18-item two-factor model of the P-SRSA (root mean square error of approximation=0.076; 95% CI (0.069 to 0.086), comparative fit index=0.943, Tucker-Lewis index=0.936). P-SRSA strongly correlated with convergent/divergent measures and demonstrated satisfactory internal consistency (α=0.82). Furthermore, regression analysis revealed that the P-SRSA significantly predicted suicide reattempts at baseline and during a 6-month follow-up period.

Conclusion: The study confirms that the P-SRSA, comprising a two-factor, 18-item solution, is a reliable measure of resilience, displaying robust discriminant and predictive validity. These findings provide substantial support for implementing P-SRSA in samples from the Iranian Army, highlighting its potential to identify suicidal conscripts effectively.