Machine Learning Models Predict the Emergence of Depression in Argentinean College Students during Periods of COVID-19 Quarantine (preprint)

INTRODUCTION: The COVID-19 pandemic has exacerbated mental health challenges, particularly depression among college students. Detecting at-risk students early is crucial but remains challenging, particularly in developing countries. Utilizing data-driven predictive models presents a viable solution to address this pressing need. AIMS: 1) To develop and compare machine learning (ML) models for predicting depression in Argentinean students during the pandemic. 2) To assess the performance of classification and regression models using appropriate metrics. 3) To identify key features driving depression prediction. METHODS: A longitudinal dataset (N = 1492 college students) captured T1 and T2 measurements during the Argentinean COVID-19 quarantine. ML models, including linear logistic regression classifiers/ridge regression (LogReg/RR), random forest classifiers/regressors, and support vector machines/regressors (SVM/SVR), are employed. Assessed features encompass depression and anxiety scores (at T1), mental disorder/suicidal behavior history, quarantine sub-period information, sex, and age. For classification, models' performance on test data is evaluated using Area Under the Precision-Recall Curve (AUPRC), Area Under the Receiver Operating Characteristic curve, Balanced Accuracy, F1 score, and Brier loss. For regression, R-squared (R2), Mean Absolute Error, and Mean Squared Error are assessed. Univariate analyses are conducted to assess the predictive strength of each individual feature with respect to the target variable. The performance of multi- vs univariate models is compared using the mean AUPRC score for classifiers and R2 score for regressors. RESULTS: The highest performance is achieved by SVM and LogReg (e.g., AUPRC: 0.76, 95% CI: 0.69, 0.81) and SVR and RR models (e.g., R2 for SVR and RR: 0.56, 95% CI: 0.45, 0.64 and 0.45, 0.63, respectively). Univariate models, particularly LogReg and SVM using depression (AUPRC: 0.72, 95% CI: 0.64, 0.79) or anxiety scores (AUPRC: 0.71, 95% CI: 0.64, 0.78) and RR using depression scores (R2: 0.48, 95% CI: 0.39, 0.57) exhibit performance levels close to those of the multivariate models, which include all features. DISCUSSION: These findings highlight the relevance of pre-existing depression and anxiety conditions in predicting depression during quarantine, underscoring their comorbidity. ML models, particularly SVM/SVR and LogReg/RR, demonstrate potential in timely detection of at-risk students, enabling preventive measures for improved mental health outcomes.

Empathy, Fear of Disease and Support for COVID-19 Containment Behaviors: Evidence from 34 Countries on the Moderating Role of Governmental Trust (preprint)

The current study investigated the motives that underlie support for COVID-19 preventive behaviorsin a large, cross-cultural sample of 12,758 individuals from 34 countries. We hypothesized that the associations of empathic prosocial concern and fear of disease, with support towards preventive COVID-19 behaviors would be moderated by the individual-level and country-level trust in the government. Results suggest that the association between fear of disease and support for COVID-19 preventive behaviors was strongest when trust in the government was weak (both at individual and country-level). Conversely, the association with empathic prosocial concern was strongest when trust was high, but this moderation was only found at individual-level scores of governmental trust. We discuss how both fear and empathy motivations to support preventive COVID-19 behaviors may be shaped by socio-cultural context, and outline how the present findings may contribute to a better understanding of collective action during global crises.

Changes in healthcare workers’ anxiety during two time points of the COVID-19 pandemic: Evidence from a longitudinal study (preprint)

Evidence on the within-person changes of healthcare workers’ mental health during the COVID-19 pandemic is absent. This study aimed to examine the within-person changes of anxiety in Argentinean healthcare workers during this pandemic, adjusting for main demographic factors, region, mental disorder history, and COVID-19 contagion. A longitudinal web survey (N = 305) was conducted during two time points of the pandemic, one of which was an infection peak. Anxiety significantly increased across time. However, there were significant interaction effects modulating anxiety levels. The largest anxiety increases occurred in healthcare workers who were wondering if they had contracted COVID-19 while symptomatic. Irrespective of the time point, anxiety was the highest in healthcare workers from a region inside the country who were wondering if they had contracted COVID-19, either asymptomatic or symptomatic. An interaction effect between the mental disorder history and the COVID-19 contagion suggested that the anxiety outcomes were mainly due to the concern about the COVID-19 contagion, rather than due to pre-existing mental health vulnerabilities. An increasing anxiety outcome may be expected among healthcare workers as the pandemic progresses. The uncertainty regarding COVID-19 contagion is a preventable and modifiable interacting factor to produce the worst anxiety outcomes among healthcare workers.