An Ordinal Mediation Analysis to Assess the Extent to Which a Change in Suicide Risk Is Mediated by or Independent of Improvement in Depression Symptoms

• Activity Number: 94 - Suicidal Ideation, Behavior, and Risk: Assessing Mediating Factors and Designing Interventions
• Type: Invited
• Date/Time: Monday, July 31, 2017 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Yun Zhang* and Rosanne Lane and Pilar Lim
• Companies: Janssen Research & Development, LLC and Janssen Research & Development, LLC and Janssen Research & Development, LLC
• Keywords: suicide risk ; ordinal mediation analysis ; counterfactual approach

Abstract

There is a high unmet medical need to reduce the symptoms of depression, including suicidal ideation, in subjects with major depressive disorder (MDD) who are at imminent risk of suicide. In conducting clinical research in this patient population, there are challenges to appropriately measuring improvement in suicidal ideation risk. It is difficult to differentiate how much of the improvement is due to the improvement in symptoms of depression and how much of the improvement is an independent effect of treatment on suicidal ideation. There is then interest in examining the mediating role of depressive symptoms on assessment of suicide risk. In a recent study of an antidepressant for the reduction of the symptoms of MDD, including suicidal ideation, in subjects assessed to be at imminent risk for suicide, suicide risk was measured using a 7-item rating scale. Since the change in the suicide risk assessment is an ordinal variable, a novel ordinal mediation analysis using simulation-based counterfactual approach was conducted to assess the extent to which a change in suicide risk is mediated by or independent of improvement in depression symptoms.