Applications of Meta-Analytic Approaches to Inform Estimates for the Global Burden of Mental Disorders

Thesis (Ph.D.)--University of Washington, 2021University of Washington Abstract Applications of Meta-Analytic Approaches to Inform Estimates for the Global Burden of Mental Disorders Modhurima Moitra Chair of the Supervisory Committee:Pamela Y. Collins, MD, MPH Department of Global Health Department of Psychiatry and Behavioral Sciences Mental disorders are some of the most important contributors to disability and indirectly to mortality worldwide. Despite the existence of effective interventions, treatment coverage for common and highly prevalent mental disorders such as major depressive disorder (MDD) is remarkably low. Given the important role mental disorders play in contributing to health loss, it is important to develop improved population-level estimates of the burden of mental disorders that may better inform resource allocation and prioritize prevention and treatment for mental health. The process of developing population-level estimates such as those from the Global Burden of Disease (GBD) study typically involves thorough reviews of the available evidence and application of appropriate methods to arrive at usable estimates of the global burden of mental disorders. However, certain challenges exist in this process. Available data may be derived from a variety of study settings or constrained to select locations that may contribute to bias in existing population-level estimates. Additionally, important components of MDD care such as treatment coverage and efficacy that will likely alter the known burden of MDD have not been systematically quantified in the recent literature. To this end, this work utilizes a meta-analytic measurement framework to address these limitations in three areas that are important for developing better estimates but remain relatively unexplored to date. First, a comparative assessment of the risk of suicide associated with mental disorders is conducted via an updated systematic review and updated meta-regression methods. Second, the efficacy of available interventions for MDD are explored in a novel network meta-analytic approach to better account for between-study heterogeneity. Third, an updated systematic review and Bayesian meta-regression analysis of treatment coverage data between 2000 and 2019 was conducted to obtain treatment coverage estimates for six modalities of MDD care across income levels and geography. The work reported here collectively provide more accurate estimates that may inform the larger process of developing burden measures for mental disorders. There are several important implications of this work. Updated estimates of the risk of suicide associated with mental disorders may better inform population attributable fractions for suicide and subsequently fatal burden associated with mental disorders. Incorporating estimates of treatment efficacy and coverage for MDD may potentially inform us about the burden of MDD that can be reduced or altered in the presence of adequate and effective treatment. Overall, this work represents emerging areas in global mental health that can inform evidence-based benchmarking and priority setting to reduce the global burden of mental disorders