Predicting a young person's risk of distress | UniSC | University of the Sunshine Coast, Queensland, Australia

Accessibility links

Predicting a young person's risk of distress

15 Jul 2021

LABS researchers have built a system that predicts a young person’s mental health.

Specifically, the model calculates the likelihood of psychological distress, which is when someone has deeply unpleasant feelings that are overwhelming.

To develop the model, our researchers studied ‘causal relationships’ between aspects of an adolescent’s life and their mental health levels.

This means they examined LABS data to determine how much things like sleep, food, and socialising influenced a young person’s mental health.

They then used a special method of data analysis, called a ‘Bayesian Network’, to create a system that determines probabilities.

Researchers can enter information about a young person and the system will say how likely it is that they will experience psychological distress.

The research found:

  • Unhealthy eating increases the chance of severe distress by 600%
  • A lack of social connection increases the chance of severe distress by 200%

A lack of physical activity and quality sleep also increased the chance of distress.

The model is still in early stages. As data collected from LABS grows, the accuracy of the model’s predictions increases.

You can view the published research paper here.