Mental Health Status of Bangladeshi University Students: A Quantitative Analysis and Machine Learning-Based Prediction of Depression, Anxiety, and Stress

Abstract

Rising mental health concerns within University student have created an unprecedented global public health problem in lower to moderate income countries. In this context, this study provides a quantitative assessment of the mental health of Bangladeshi university students, focusing on levels of depression, anxiety, and stress. We are able to provide an analytical breakdown of descriptive data ("descriptive") using both demographic information and correlation across mental health outcomes with sleep quality and social support, based on an analysis of 10,000 Bangladeshi university students. The report summarizes the findings of our study and discusses the implications for future research in the area of mental health within Bangladesh. Our primary findings demonstrate the prevalence of common issues related to mental health, the importance of certain demographic risk factors for developing mental health issues, as well as the role of machine learning (ML) techniques, such as logistic regression, Scikit-learn MLP, and TabNet, in providing tools for the early identification and early treatment of mental health issues. Finally, we create a normative framework for university-based mental health services and discuss the methodological limitations of our study and future research opportunities