Collaboration Examines the Application of Decision Science to Mental Health Policy
Matt Miller (Injury Center, HPM) and Jane Kim (CHDS) are awarded seed money from the Department of Health Policy and Management to explore the use of decision analytic tools to address questions in mental health policy that are characterized by complexity, uncertainty, and tradeoffs. The motivation for the project they have chosen emanates from the conflicting evidence of the effects of antidepressants on suicidal behavior in children from several clinical studies. In 2004, the FDA conducted a meta-analysis using clinical trial data and found that the risk of suicidal ideation among children on antidepressants was significantly higher than for those in placebo groups. As a result, the FDA ordered that all antidepressants carry a black-box warning indicating that their use may be associated with an increased risk of suicidal ideation in children. A separate meta-analysis, which assumed different model specifications in the statistical analysis, concluded that the risk of suicidal ideation among children on antidepressants was indeed increased but not significant, and that the overall benefits of antidepressant use in this population outweighed the risks.
Miller and Kim are conducting an analysis weighing the risks and benefits of antidepressant use in children and contrasting the cost-effectiveness of antidepressants when using different estimates of treatment effects on suicide risk. They have thus far developed a computer-based model to simulate the course of depression among U.S. children, carefully synthesizing the best available epidemiological and clinical data.
They have recruited a postdoctoral fellow in decision science, Djora Soeteman who has done her thesis work in the mental health area. The group of three plans to hold a meeting in the Fall with key experts in clinical psychiatry, suicide risk, health economics, and decision sciences to get feedback on their analytic assumptions. The long-term objective is to secure funding for a series of analyses that apply decision science methods to mental health policy.