Robust estimation of optimal dynamic treatment regimes for sequential treatment decisions. (English) Zbl 1284.62508

Zhang, Baqun; Tsiatis, Anastasios A.; Laber, Eric B.; Davidian, Marie

Biometrika 100, No. 3, 681-694 (2013).

Summary: A dynamic treatment regime is a list of sequential decision rules for assigning treatments based on a patient’s history. Q- and A-learning are two main approaches for estimating the optimal regime, i.e., that yielding the most beneficial outcome in the patient population, using data from a clinical trial or observational study. Q-learning requires postulated regression models for the outcome, while A-learning involves models for that part of the outcome regression representing treatment contrasts and for treatment assignments. We propose an alternative to Q- and A-learning that maximizes a doubly robust augmented inverse probability weighted estimator for population mean outcome over a restricted class of regimes. Simulations demonstrate the method’s performance and robustness to model misspecification, which is a key concern.

MSC:

62L10	Sequential statistical analysis
62C99	Statistical decision theory
62P10	Applications of statistics to biology and medical sciences; meta analysis
68T05	Learning and adaptive systems in artificial intelligence
62G35	Nonparametric robustness
65C60	Computational problems in statistics (MSC2010)

Keywords:

A-learning; double robustness; outcome regression; propensity score; Q-learning

Software:

qLearn