Agent-Based Modeling and Simulation of Depression and Its Impact on Stu- dent Success and Academic Retention

In the U.S., major depressive disorder affects approximately 14.8 million American adults. Furthermore, depression can lead to a several other illnesses and disabilities. Economic burden of depression is estimated to be $53 billion annually in the U.S. alone. Depression can reach high levels that can lead to suicide, the third leading cause of death among the U.S. college-aged population. Studies show a direct relation between mental health and academic success. In particular, depression is a significant predictor of lower GPA and increased drop out rate. A 15 point increase on the depression scale correlates with a 0.17 drop in GPA and corresponds to a 4.7 percent increase in probability of dropping out. High dropout rates also adversely impact both universities and society. In this work, we construct and exercise an agent-based model (ABM) of the evolution of depression among a population of roughly 19,000 college students. This model includes within-agent interactions among depression symptoms and agent-to-agent interactions defined by a college student social network. We conduct simulation studies to identify (model) parameters and initial conditions that most influence population outcomes. Connectivity among within-agent symptoms is demonstrated to have a large effect on population levels of depression. ————————————————————