Assessment and analysis of factors influencing suicidal ideation in young adults: a large cohort study using an elastic network logistic regression model.

OBJECTIVES

This study aims to review and analyze factors associated with suicidal ideation to provide a rationale for subsequent effective interventions.

METHODS

Data from this study were obtained from the Assessing Nocturnal Sleep/Wake Effects on Risk of Suicide (ANSWERS). The University of Arizona evaluated 404 young adults aged 18-25 years using different scales. Then, general demographic data was recorded. An elastic network (EN) was used to optimize feature selection, combined with logistic regression, to determine the influencing factors associated with SI in young adults.

RESULTS

The EN regression retained 11 potential influencing factors with nonzero coefficients. In the multivariate logistic regression analysis, INQ-15 perceived burdensomeness (PB) scores (OR: 1.10, 95% CI: 1.04-1.17), CESD depression mood scores (OR: 1.16, 95% CI: 1.07-1.26), and age (OR: 0.72, 95% CI: 0.55-0.94) were significant factors for SI.

CONCLUSIONS

This is the first study to use an Elastic Network logistic regression model to assess the factors affecting suicidal ideation in young adults. Perceived Burdensome, depression, and age play an important risk role and are the best predictor combination of suicidal ideation in young adults, with depression being the most significant risk factor. Increased focus on Perceived Burdensome and negative emotions, along with simultaneous interventions for other potentially influential factors, can be more effective in preventing suicidal behavior in young adults.