Reading Between the Lines: A Pursuit of Estimating the Population Prevalence of Mental Illness Using Multiple Data Sources

Population-based prevalence estimates of mental illness are foundational to health service planning, strategic resource allocation, and the development and evaluation of public mental health policy. Generating valid, reliable, and context-specific population-level estimates is of utmost importance and can be achieved by combining various data sources. This pursuit benefits from the right combination of theory, applied statistics, and the conceptualization of available data sources as a collective rather than in isolation. We believe there is a need to read between the lines as theory, methodology, and context (i.e., strengths and limitations) are what determines the meaningfulness of a combined prevalence estimate. Currently lacking is a gold standard approach to combining estimates from multiple data sources. Here, we compare and contrast various approaches to combining data and introduce an idea that leverages the strengths of pre-existing individually linked population-based survey and health administrative data sources currently available in Canada.