Can the Potential Benefit of Individualizing Treatment Be Assessed Using Aggregate Data Alone?

• Activity Number: 281 - New Methods with Applications in Mental Health Statistics
• Type: Contributed
• Date/Time: Tuesday, August 9, 2022 : 10:30 AM to 12:20 PM
• Sponsor: Mental Health Statistics Section
• Author(s): Nina Galanter* and Alex Luedtke and Marco Carone
• Companies: University of Washington and University of Washington and University of Washington
• Keywords: optimal treatment rules; mental health; causal inference

Abstract

Individualizing treatment assignment can improve outcomes for diseases with patient-to-patient variability in treatment effects. When pre- and post-treatment results from a clinical trial demonstrate that some patients improve on treatment while others do not, it is tempting to assume there is treatment effect heterogeneity. However, if variability in outcomes is mainly driven by factors other than treatment, then collecting further covariate data to predict treatment responders is a potential waste of resources. Motivated by recent meta-analyses assessing the potential of individualizing treatment for major depressive disorder, we provide a method that uses summary statistics widely available in published clinical trial results to bound the maximum benefit of optimally assigning treatment to each patient. We also provide alternate bounds for settings in which (i) trial results are stratified by another covariate, or (ii) investigators believe treatment effect heterogeneity is only present in a subset of the population. We illustrate and validate our approach using summary statistics from two depression treatment trials.