Journal Article

Predicting mental health problems in adolescence using machine learning techniques

Larsson, Henrik, Lundström, Sebastian, McCabe, Ryan C., Lichtenstein, Paul, Tate, Ashley E., Kuja-Halkola, Ralf

2020

Overview

Predicting which children will go on to develop mental health symptoms as adolescents is critical for early intervention and preventing future, severe negative outcomes. Although many aspects of a child's life, personality, and symptoms have been flagged as indicators, there is currently no model created to screen the general population for the risk of developing mental health problems. Additionally, the advent of machine learning techniques represents an exciting way to potentially improve upon the standard prediction modelling technique, logistic regression. Therefore, we aimed to I.) develop a model that can predict mental health problems in mid-adolescence II.) investigate if machine learning techniques (random forest, support vector machines, neural network, and XGBoost) will outperform logistic regression. In 7,638 twins from the Child and Adolescent Twin Study in Sweden we used 474 predictors derived from parental report and register data. The outcome, mental health problems, was determined by the Strengths and Difficulties Questionnaire. Model performance was determined by the area under the receiver operating characteristic curve (AUC). Although model performance varied somewhat, the confidence interval overlapped for each model indicating non-significant superiority for the random forest model (AUC = 0.739, 95% CI 0.708-0.769), followed closely by support vector machines (AUC = 0.735, 95% CI 0.707-0.764). Ultimately, our top performing model would not be suitable for clinical use, however it lays important groundwork for future models seeking to predict general mental health outcomes. Future studies should make use of parent-rated assessments when possible. Additionally, it may not be necessary for similar studies to forgo logistic regression in favor of other more complex methods.

• Publisher: Public Library of Science,Public Library of Science (PLoS)
• Subject: Adolescence / Adolescent / Adolescents / Age / Analysis / Artificial neural networks / Biology and Life Sciences / Child / Child & adolescent psychiatry / Child development / childhood / Children / Computer and Information Sciences / Confidence intervals / Datasets / disorders / Engineering and Technology / Epidemiology / Female / Future predictions / Health problems / Health risks / Humans / impulsivity / Learning algorithms / Machine learning / Male / Medicine and Health Sciences / Mental disorders / Mental Disorders - diagnosis / Mental Disorders - psychology / Mental Health / Mental health care / Methods / Models, Psychological / Neural networks / outcomes / People and Places / Prediction models / Predictive Value of Tests / Psychiatry / Psychological aspects / Psykiatri / Questionnaires / Regression / Research and Analysis Methods / Science & Technology - Other Topics / Statistical analysis / strengths / Studies / subthreshold / suicide / Suicides & suicide attempts / Support Vector Machine / Support vector machines / Sweden / symptoms / Teenagers / twin / Twins / Variables / Youth