Promoting child well-being and reducing mental health difficulties at the population level
Lydia M. Li is a doctoral student in the Department of Applied Psychology and Human Development at the University of Toronto, and recipient of the Leong Centre Studentship Award. She provides an update on her ongoing project to identify the top statistical predictors of mental well-being and mental health difficulties among school-aged children. Lydia also highlights the most up-to-date methodologies she has employed to answer her research question and the preliminary results her analysis has shown.
In Canada, approximately one out of every five children experience a mental health problem, with evidence of even higher rates since the COVID-19 pandemic began. Children spend most of their time either at school or at home, making these environments centrally important to their overall health and well-being. Previous research has identified a wide range of school, home, and demographic factors that contribute to mental health outcomes of children. However, due in part to problems with small samples and limited analytical approaches, it is unclear which factors provide the best predictive power for mental health well-being mental health difficulties. This is a critical gap, as effective public health strategies should target those factors that best predict mental health and well-being. To address this, my doctoral research uses machine learning to increase precision in the prediction of Canadian children’s mental health.
Specifically, my program of research aims to identify the top factors for general mental well-being, as well as general mental health difficulties, internalizing problems, externalizing problems, and individual mental health problems (e.g., oppositional defiance disorder, conduct problems, ADHD, anxiety, depression) in a large Canadian survey of school-aged children’s well-being (the School Mental Health Survey; SMHS). To identify the factors linked to disparities in mental health between different groups, I also stratify based on gender, socio-economic background (i.e., educational differences), and immigration status. Under the supervision of my PhD advisor, Dr. Mark Wade, and in collaboration with Dr. Kathy Georgiades from the McMaster University and Dr. Hause Lin from the Massachusetts Institute of Technology, I am leveraging the latest big data analytical methods, including LightGBM and generalized random forest, as well as multiple feature analysis methods, to ensure robustness of the statistical analysis. You can review my complete analytical strategy at https://osf.io/skndp
Most recently, I presented my preliminary results at the Association for Psychological Science 2024 Annual Convention in May. You can view my poster presentation here. The results are fascinating, showing that school belonging, lifestyle (e.g., sleep, physical activity), and friendship quality best predicted general mental well-being, whereas lifestyle, peer bullying, and class preparedness best predicted general mental health difficulties. This suggests that promoting well-being and reducing mental health difficulties in school-aged children requires partially distinct strategies. Additionally, I identified several factors that best predicted mental health inequalities between groups, which can increase targeted intervention to address the mental health needs of unique populations. The next step is to use a larger dataset to test the reliable and generalizability of findings and ultimately disseminate the findings through publication and conference presentations.
The Leong Centre Studentship has equipped me with crucial computational equipment for conducting extensive data analysis and sharing my research results at a prestigious international conference. I also greatly appreciate that the Edwin S.H. Leong Centre for Healthy Children values conducting research with the involvement of local communities and mobilizing knowledge gained through research to better serve communities. I am humbled and inspired by many researchers who are so dedicated to tackling child mental health inequities through the Centre.