Machine Learning–Based Predictive Modeling of Anxiety and Depressive Symptoms During 8 Months of the COVID-19 Global Pandemic: Repeated Cross-sectional Survey Study

Abstract

Background The COVID-19 global pandemic has increased the burden of mental illness on Canadian adults. However, the complex combination of demographic, economic, and lifestyle factors and perceived health risks contributing to patterns of anxiety and depression has not been explored. Objective The aim of this study is to harness flexible machine learning methods to identify constellations of factors related to symptoms of mental illness and to understand their changes over time during the COVID-19 pandemic. Methods Cross-sectional samples of Canadian adults (aged ≥18 years) completed web-based surveys in 6 waves from May to December 2020 (N=6021), and quota sampling strategies were used to match the English-speaking Canadian population in age, gender, and region. The surveys measured anxiety and depression symptoms, sociodemographic characteristics, substance use, and perceived COVID-19 risks and worries. First, principal component analysis was used to condense highly comorbid anxiety and depression symptoms into a single data-driven measure of emotional distress. Second, eXtreme Gradient Boosting (XGBoost), a machine learning algorithm that can model nonlinear and interactive relationships, was used to regress this measure on all included explanatory variables. Variable importance and effects across time were explored using SHapley Additive exPlanations (SHAP). Results Principal component analysis of responses to 9 anxiety and depression questions on an ordinal scale revealed a primary latent factor, termed “emotional distress,” that explained 76% of the variation in all 9 measures. Our XGBoost model explained a substantial proportion of variance in emotional distress (r2=0.39). The 3 most important items predicting elevated emotional distress were increased worries about finances (SHAP=0.17), worries about getting COVID-19 (SHAP=0.17), and younger age (SHAP=0.13). Hopefulness was associated with emotional distress and moderated the impacts of several other factors. Predicted emotional distress exhibited a nonlinear pattern over time, with the highest predicted symptoms in May and November and the lowest in June. Conclusions Our results highlight factors that may exacerbate emotional distress during the current pandemic and possible future pandemics, including a role of hopefulness in moderating distressing effects of other factors. The pandemic disproportionately affected emotional distress among younger adults and those economically impacted.

Canadian perspectives of digital mental health supports: Findings from a national survey conducted during the COVID-19 pandemic

“…While recent federal efforts have resulted in affordable broadband access to 94% of Canadians, 57 , 58 which may reduce the potential bias of a web-based survey, other digital divide factors, such as a lack of digital literacy, may hinder participation and must be considered. Another source of bias is the low response rate; however, this rate is consistent with other web-panel studies 57 , 59 – 64 and is expected for population surveys of this length administered on the web and without financial incentive. 65 Lastly, the cross-sectional design provides a snapshot of the experience and needs of the population, especially with the dynamic nature of jurisdictional pandemic responses and their impact on Canadians.…”

Aligning the Goals Hybrid Model for the Diagnosis of Mental Health Quality

“…Not all decision-makers have decisionmaking power. Therefore, it is essential to distinguish the degree of influence of decisionmakers in the decision-making process [44,45].…”

“…The importance judgments between pairs of criteria (paired judgments) were used concerning the main focus in applying the AHP multi-criteria method to improve the results of the machine learning algorithms. Therefore, the value scale established by Saaty [45] was considered.…”

The role of booster vaccination in decreasing COVID-19 age-adjusted case fatality rate: Evidence from 32 countries

“…The complex factors involved in real-world health emergencies are more effectively analyzed with fast-evolving machine learning algorithms, such as Extreme Gradient Boosting (XGBoost) algorithm. XGBoost is a decision tree-based gradient boosting ensemble machine learning algorithm with improved performance based on other tree-based models such as Random Forest and Gradient Boosting Decision Tree (GBDT), which is well suited for solving classification and regression problems (22). It features several advantages that allow it to be effectively adapted to real-world studies:…”