Construct a more deeply-tailored adaptive intervention using data from a SMART

About This Code

This code is used to investigate whether (and, if so, how) a baseline or time-varying covariate can be a useful tailoring variables in an adaptive intervention.

How can a behavioral scientist use this code?

Behavioral intervention scientists can use this code to generate hypotheses about more deeply-tailored adaptive interventions using data from a SMART.

What method does this code implement?

This code implements an easy-to-use version of Q-learning (a common reinforcement learning algorithm) and a method for estimating confidence intervals. The method can be viewed as an extension of moderated regression analysis for data arising from a SMART. In a two-stage SMART, Q-learning implements two regressions, starting at the second stage: The first regression selects the best second-stage treatment. The second regression selects the best first-stage treatment given that the best second stage treatment has been assigned to each participant. The backward ordering of the regressions avoids selecting treatment options that appear optimal in the short term, but may lead to a less desirable outcome in the long run.

Access Links

Download R Package
Download SAS Procedure

Upcoming Events

View All Events
Presentation

Introduction to MAISYs

Prevention Science and Methodology Group Virtual Grand Rounds
January 24, 2023
Presentation

Introduction to MAISYs

University of Florida Health Outcomes and Biomedical Informatics Grand Rounds
February 13, 2023
Training Institute

Getting SMART

Comprehensive Program for Adaptive Interventions Training in Education (CATIE)
March 14, 2023
View All Events

LET’S STAY IN TOUCH

Join the d3center Mailing List

Keep up to date with the latest news, events, software releases, learning modules, and resources from the d3center.