NIDA awarded $1.57 million grant to Dr. Ertefaie to study the effect of partial treatment compliance in constructing individualized treatment strategies. The grant aims to develop methodologies to adjust for partial compliance in constructing individualized treatment strategies using sequential multiple assignment randomized trials data. Existing tools that estimate the treatment effects using intention-to-treat analyses ignore information on patients’ compliance. We fill this important gap by providing a set of analytical tools that consider the noncompliance in the setting of sequential clinical decision making. The proposal studies treatment efficacy and personalized medicine under complex compliance behaviors. These behaviors may be manifested through unobserved potential compliance values that can be a function of measured or unmeasured patients’ characteristics.
Robust Q-learning
Ertefaie A, McKay J. R., Oslin D., and Strawderman R. L. (2020). Journal of the American Statistical Association, DOI: 10.1080/01621459.2020.1753522 Q-learning is a regression-based approach that is
December 8, 2020
