Age at onset of first suicide attempt: Exploring the utility of a potential candidate variable to subgroup attempters

Abstract

Purpose: Our objective was to explore the utility of age at first suicide attempt in identifying subgroups of suicide attempters.

Methods: In a retrospective study design, we collected information from the clinical charts of 895 patients assessed over a seven-year period. Admixture analysis was used to determine the best fitting theoretical model for distribution of age at first attempt that divided the sample. Subsequently, multivariate analysis was performed to identify variables that distinguished the subgroups identified.

Results: The theoretical solution that best explained the observed distribution of age at first suicide attempt was a mixture of two Gaussian distributions with a cut-off of 31 years for the two subgroups. In logistic regression analysis, male gender (Odds ratios [OR] 3.047, 95% Confidence Interval (CI) 1.818-5.106), fewer years of formal schooling (OR 3.384, 95% CI 1.701-6.734) and being married (OR 23.36, 95% CI 10.753-50.000), were more commonly associated with the late onset subgroup (age at first attempt >31 years). Further, the late onset subgroup had poorer global functioning (OR 0.980, 95% CI 0.962 to 0.998).

Conclusion: Age at onset of first suicide attempt is a useful candidate marker to delineate an early and late onset subgroup among suicide attempters. These results are likely to inform customization of suicide prevention strategies.

Keywords: Age at onset; Attempted suicide; Mixture analysis; Psychiatry; Suicide.