In this episode, we look at data trying to get at which parts of DBT are necessary and sufficient to decrease suicide attempts and non-suicidal self injury in patients with borderline personality disorder. Is less more or is more more? In working to understand how to approach this question, we consider non-inferiority designs and one of our hosts gets worked up about the use of a one-sided p value. Not sure what a one-sided p value is? Now is your chance to learn! The article we discuss in this episode is part of the ABPN Psychiatry MOC self-study. The podcasts hosts have no knowledge of the questions or assessments used as part of MOC. (Adriane dela Cruz, M.D., Ph.D., is an Assistant Professor of Psychiatry at UT Southwestern Medical Center and with the Peter O’Donnell Jr. Brain Institute who focuses on treating patients with drug and alcohol addictions. The opinions expressed are her own and do not reflect those of UT Southwestern, the O’Donnell Brain Institute, the UT System, or the state.)
The paper we discuss is M Linehan et al. Dialectical Behavior Therapy for High Suicide Risk in Individuals with Borderline Personality Disorder: A Randomized Clinical Trial and Component Analysis. JAMA Psychiatry 2015; 72(5):475-482.
We LOVE the JAMA Guide to Statistics and Medicine for quick, clinically relevant overviews of important issues in clinical trials design and analysis. Here's there article on non-inferiority designs.