Dataset Information

Machine learning prediction of suicidal ideation, planning, and attempt among Korean adults: A population-based study.

ABSTRACT:

Background

Suicide remains the leading cause of premature death in South Korea. This study aims to develop machine learning algorithms for screening Korean adults at risk for suicidal ideation and suicide planning or attempt.

Methods

Two sets of balanced data for Korean adults aged 19-64 years were drawn from the 2012-2019 waves of the Korea Welfare Panel Study using the random down-sampling method (N = 3292 for the prediction of suicidal ideation, N = 488 for the prediction of suicide planning or attempt). Demographic, socioeconomic, and psychosocial characteristics were used to predict suicidal ideation and suicide planning or attempt. Four machine-learning classifiers (logistic regression, random forest, support vector machine, and extreme gradient boosting) were tuned and cross-validated.

Results

All four algorithms demonstrated satisfactory classification performance in predicting suicidal ideation (sensitivity 0.808-0.853, accuracy 0.843-0.863) and suicide planning or attempt (sensitivity 0.814-0.861, accuracy 0.864-0.884). Extreme gradient boosting was the best-performing algorithm for predicting both suicidal outcomes. The most important predictors were depressive symptoms, self-esteem, income, consumption, and life satisfaction. The algorithms trained with the top two predictors, depressive symptoms and self-esteem, showed comparable classification performance in predicting suicidal ideation (sensitivity 0.801-0.839, accuracy 0.841-0.846) and suicide planning or attempt (sensitivity 0.814-0.837, accuracy 0.874-0.884).

Limitations

Suicidal ideation and behaviors may be under-reported due to social desirability bias. Causality is not established.

Discussion

More than 80% of individuals at risk for suicidal ideation and suicide planning or attempt could be predicted by a number of mental and socioeconomic characteristics of respondents. This finding suggests the potential of developing a quick screening tool based on the known risk factors and applying it to primary care or community settings for early intervention.

SUBMITTER: Lee J

PROVIDER: S-EPMC9573904 | biostudies-literature | 2022 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning prediction of suicidal ideation, planning, and attempt among Korean adults: A population-based study.

Lee Jeongyoon J Pak Tae-Young TY

SSM - population health 20220914

<h4>Background</h4>Suicide remains the leading cause of premature death in South Korea. This study aims to develop machine learning algorithms for screening Korean adults at risk for suicidal ideation and suicide planning or attempt.<h4>Methods</h4>Two sets of balanced data for Korean adults aged 19-64 years were drawn from the 2012-2019 waves of the Korea Welfare Panel Study using the random down-sampling method (<i>N</i> = 3292 for the prediction of suicidal ideation, <i>N</i> = 488 for the pr ...[more]

PMID: 36263295

Similar Datasets

Project description:BackgroundGiven the relatively high rates of suicidal ideation and attempt among people with chronic pain, there is a need to understand the underlying factors to target suicide prevention efforts. To date, no study has examined the association between pain phenotypes and suicide related behaviors among those with mild traumatic brain injuries.ObjectiveTo determine if pain phenotypes were independently associated with suicidal ideation / attempt or if comorbidities within the pain phenotypes account for the association between pain phenotypes and suicide related behaviors.MethodsThis is a longitudinal retrospective cohort study of suicide ideation/attempts among pain phenotypes previously derived using general mixture latent variable models of the joint distribution of repeated measures of pain scores and pain medications/treatment. We used national VA inpatient, outpatient, and pharmacy data files for Post-9/11 Veterans with mild traumatic injury who entered VA care between fiscal years (FY) 2007 and 2009. We considered a counterfactual causal modeling framework to assess the extent that the pain phenotypes during years 1-5 of VA care were predictive of suicide ideation/attempt during years 6-8 of VA care conditioned on covariates being balanced between pain phenotypes.ResultsWithout adjustment, pain phenotypes were significant predictors of suicide related behaviors. When we used propensity scores to balance the comorbidities present in the pain phenotypes, the pain phenotypes were no longer significantly associated with suicide related behaviors.ConclusionThese findings suggest that suicide ideation/attempt is associated with pain trajectories primarily through latent multimorbidity. Therefore, it is critical to identify and manage comorbidities (e.g., depression, post-traumatic stress disorder) to prevent tragic outcomes associated with suicide related behaviors throughout the course of chronic pain and mild traumatic brain injury management.

Dataset Information

Machine learning prediction of suicidal ideation, planning, and attempt among Korean adults: A population-based study.

Background

Methods

Results

Limitations

Discussion

Publications

Machine learning prediction of suicidal ideation, planning, and attempt among Korean adults: A population-based study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets