Predicting perceived and unmet mental health needs in the populations

COVID-19 is unprecedented in terms of its magnitude and impacts on population mental health and health services delivery. The closure of in-person mental health services may have exacerbated the issues of unmet mental health needs and accessibility. Perceived and unmet mental health needs constitute the central part of demand which is a critical element in the process of health resources allocation. However, we have a little knowledge about perceived and unmet mental health needs at the nation, provincial/territorial, and health region levels, and how the needs may have changed in the context of COVID-19. Building upon our strong expertise in risk predictive analytics and machine learning, we proposed to: 1) estimate perceived and unmet mental health needs pre- and during the pandemic in the populations, and 2) develop innovative tools for estimating and projecting perceived and unmet mental health needs, using both statistical and machine learning approaches. For the objectives, data from the Canadian Community Health Survey and Canadian Urban Environment Research Consortium will be linked and analyzed. The research team involves expertise of psychiatric epidemiology, mental health services research, population health, machine learning, in collaboration with knowledge users from federal and regional health agencies. The proposed study will produce evidence about how the pandemic may have affected perceived and unmet mental health needs, and how the needs vary by demographic, socioeconomic and geographic characteristics, and neighborhood social determinants. We will build predictive models for estimating and projecting perceived and unmet mental health needs at health regional levels, assisting decision makers and mental health service planners in allocating healthcare resources in a timely, equitable and efficient way.