Causal Inference in Hybrid Intervention Trials Involving Treatment Choice

• Activity Number: 99
• Type: Invited
• Date/Time: Monday, August 9, 2004 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Roderick J. Little*+ and Qi Long and Xihong Lin
• Companies: University of Michigan and University of Michigan and University of Michigan
• Keywords: clinical trials ; doubly randomized preference trial ; EM algorithm ; partially randomized preference trial ; randomization ; selection bias

Abstract

Randomized allocation of treatments is a cornerstone of experimental design, but has drawbacks when a limited set of individuals are willing to be randomized, or the act of randomization undermines the success of the treatment. Choice-based experimental designs allow a subset of the participants to choose their treatments. We discuss causal inferences for experimental designs where some participants are randomly allocated to treatments and others receive their treatment preference. The paper was motivated by the "Women Take Pride" study (Janevic et al., 2001), a doubly randomized preference trial to assess behavioral interventions for women with heart disease. We propose a model for inference about preference effects and develop an EM algorithm to compute maximum likelihood estimates of the model parameters, using data from the Women Take Pride study. We also expand these methods to handle a broader class of designs involving treatment preference, and discuss alternative designs from the perspective of the strength of assumptions needed to make causal inferences.