Mental illness is the leading cause of disability in the United States, costing roughly $300 billion annually. Despite high societal burdens, population mental health surveillance for scientific inquiry and tracking progress toward goals like those described in Healthy People remains weak. Nonetheless, the Internet has been shown to be the most utilized health resource in the world, and ground-breaking work in infectious disease epidemiology demonstrates how aggregate Internet search queries may be used to validly estimate population prevalence trends. Capturing mental health related search queries could serve as a novel, cheap, and real- time data source for the development of more robust population mental health surveillance. Thus, we aim to validate adult mental health search query trends (specifically mentally unhealthy days and major depressive disorders) against BRFSS survey measures in the US. This includes identifying a criterion to validate against, selecting candidate queries, building a validation model, adjusting for potential bias, and implementing the query trend. The utility of these trends can be demonstrated by applying our metrics retrospectively to explore potential subseasonal rhythms (i.e., monthly, weekly, daily) in mental health and evaluate the mental health impacts of the Great Recession, just two of the many questions for which these data may be leveraged. Our validated trends will also serve as an important toolkit for federal agencies and advocates going forward. We will release our mental health metrics in a free application, Mental Health Trends, to be available at www.mentalhealthtrends.org. Running live on Google search data, Mental Health Trends will give investigators, clinicians, and policy makers the ability to rapidly detect changes in population mental health, with new trends released each day.
This application addresses the lack of real-time population mental health surveillance to support the public health evidence-base. This application uses a unique resource-Internet search queries-and develops methods to monitor population trends for mentally unhealthy days and major depressive disorders in real time, leveraging these trends to improve our understanding of population mental health, and making these trends available online to the community of clinicians, scientists and policy makers for their own analyses.
| Caputi, Theodore L; Leas, Eric C; Dredze, Mark et al. (2018) Online Sales of Marijuana: An Unrecognized Public Health Dilemma. Am J Prev Med 54:719-721 |
| Ayers, John W; Althouse, Benjamin M; Leas, Eric C et al. (2017) Internet Searches for Suicide Following the Release of 13 Reasons Why. JAMA Intern Med 177:1527-1529 |
