Mental distress is an epidemic that endangers global well-being and contributes to various illnesses. In the United States, the prevalence of mental distress has risen rapidly in recent years. However, this topic is understudied in spatial information research, as current literature lacks focus on spatially varying relationships between mental distress and relevant factors, which leads to impediment of prevention and mitigation efforts. Therefore, this study aims for investigating the spatiotemporal relationships of mental distress with crime, housing cost, poverty, air quality. Using the space–time scan statistic, we illustrate the spatiotemporal distribution of mental distress in Chicago, IL. In addition, we employ geographically and temporally weighted regression (GTWR) to find the varying relationships between aforementioned factors and mental distress. Lastly, we compare GTWR to a linear ordinary least squares model to assess the effect of spatial and temporal dependence in found relationships. Our findings indicate that, while the crime rate, housing costs, and poverty explain the prevalence of mental distress over time and space, the space–time variation of PM
2.5
is not a predominant determinant of mental distress in Chicago. The practical implications of our work are that planners and policymakers are encouraged to identify spatiotemporal patterns of mental distress so that resources can be directed to the most vulnerable communities. Spatiotemporal modelling, the identification of geographic patterns and relationships, enables novel understanding of societal issues, and is an integral part of spatial information science.