TY - GEN
T1 - A Bayesian Response-Adaptive, Covariate-Balanced and Q-Learning-Decision-Consistent Randomization Method for SMART Designs
AU - Dai, Tianjiao
AU - Shete, Sanjay
N1 - Funding Information:
Acknowledgements The study was funded by the National Cancer Institute (P30CA016672 to S. Shete), the Betty B. Marcus Chair in Cancer Prevention (to S. Shete), the Duncan Family Institute for Cancer Prevention and Risk Assessment (S. Shete) and the Cancer Prevention Research Institute of Texas (grant RP170259 to S. Shete).
Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
PY - 2022
Y1 - 2022
N2 - In clinical trials, various randomization strategies have been developed to allocate subjects to different interventions or treatment arms. The preferred strategy is one that balances the distribution of covariates across the treatment arms and assigns more subjects to the treatments associated with better outcomes. Sequential multiple assignment randomized trial (SMART) designs involve an initial stage in which participants are randomized to a set of intervention options, followed by subsequent stages in which some or all of the individuals are re-randomized to the intervention options available at that stage. For SMART designs, the intervention for each subject can be optimized individually using a Q-learning-based optimization algorithm. However, such response-adaptive or Q-learning-based randomization strategies lead to covariate imbalance that can result in biased inference when the imbalanced covariates are associated with the outcome of interest, particularly with small to moderate sample sizes. To combine the advantages of Q-learning-decision-consistent strategies and response-adaptive designs while controlling for covariate balance, we propose a Bayesian response-adaptive, covariate-balanced and Q-learning-decision-consistent randomization method (RCQ) for SMART designs. Simulation studies to illustrate the performance of the proposed method show that the RCQ randomization method assigned the lowest percentage of subjects to the inferior intervention arms and the highest percentage of subjects to optimal Q-learning-decision-consistent strategies that maximize the long-term outcome for the individual while exhibiting well-controlled covariate balance. The alternative randomization strategies showed pronounced covariate imbalance or assigned higher percentages of subjects to inferior interventions or were not consistent with the Q-learning-based optimal decision strategy.
AB - In clinical trials, various randomization strategies have been developed to allocate subjects to different interventions or treatment arms. The preferred strategy is one that balances the distribution of covariates across the treatment arms and assigns more subjects to the treatments associated with better outcomes. Sequential multiple assignment randomized trial (SMART) designs involve an initial stage in which participants are randomized to a set of intervention options, followed by subsequent stages in which some or all of the individuals are re-randomized to the intervention options available at that stage. For SMART designs, the intervention for each subject can be optimized individually using a Q-learning-based optimization algorithm. However, such response-adaptive or Q-learning-based randomization strategies lead to covariate imbalance that can result in biased inference when the imbalanced covariates are associated with the outcome of interest, particularly with small to moderate sample sizes. To combine the advantages of Q-learning-decision-consistent strategies and response-adaptive designs while controlling for covariate balance, we propose a Bayesian response-adaptive, covariate-balanced and Q-learning-decision-consistent randomization method (RCQ) for SMART designs. Simulation studies to illustrate the performance of the proposed method show that the RCQ randomization method assigned the lowest percentage of subjects to the inferior intervention arms and the highest percentage of subjects to optimal Q-learning-decision-consistent strategies that maximize the long-term outcome for the individual while exhibiting well-controlled covariate balance. The alternative randomization strategies showed pronounced covariate imbalance or assigned higher percentages of subjects to inferior interventions or were not consistent with the Q-learning-based optimal decision strategy.
KW - Covariate-balanced
KW - Q-learning
KW - Response-adaptive
KW - SMART designs
UR - http://www.scopus.com/inward/record.url?scp=85128984111&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85128984111&partnerID=8YFLogxK
U2 - 10.1007/978-981-16-7932-2_13
DO - 10.1007/978-981-16-7932-2_13
M3 - Conference contribution
AN - SCOPUS:85128984111
SN - 9789811679315
T3 - Springer Proceedings in Mathematics and Statistics
SP - 199
EP - 213
BT - Applied Statistical Methods - ISGES 2020
A2 - Hanagal, David D.
A2 - Latpate, Raosaheb V.
A2 - Chandra, Girish
PB - Springer
T2 - International Conference on Importance of Statistics in Global Emerging, ISGES 2020
Y2 - 2 January 2020 through 4 January 2020
ER -