Predictive modelling of stress, anxiety and depression: A network analysis and machine learning study

Area of Science:

• Psychology
• Public Health
• Data Science

Background:

• The COVID-19 pandemic significantly impacted global mental health.
• Understanding predictors of stress, anxiety, and depression is crucial for public health interventions.

Purpose of the Study:

• To identify key predictors of stress, anxiety, and depression during the COVID-19 pandemic.
• To assess the relative importance of demographic, COVID-19 context, and psychological variables.

Main Methods:

• Utilized data from 741 adults participating in the Boston College daily sleep and well-being survey.
• Employed Gaussian graphical models (GGM) for feature selection and Random Forest (RF) for prediction.
• Analyzed demographic, COVID-19 context, and psychological variables.

Main Results:

• Psychological variables were significant predictors of stress, anxiety, and depression.
• Demographic and COVID-19 related factors showed minimal predictive value.
• Negative affect and subjective sleep quality emerged as common predictors for all three outcomes.

Conclusions:

• Psychological factors are primary drivers of adverse mental health outcomes during pandemics.
• Findings inform targeted interventions to mitigate pandemic-related mental health impacts.
• Highlights the importance of addressing negative affect and sleep quality.