Counterfactual Survival Q Learning for Longitudinal Randomized Trials via Buckley James Boosting
This work addresses the need for robust and unbiased treatment decision-making in clinical trials, particularly for survival outcomes, though it appears incremental as it builds on existing Q learning and boosting methods.
The authors tackled the problem of estimating optimal dynamic treatment regimes for right-censored survival data in longitudinal randomized trials by proposing a Buckley James Boost Q learning framework, which integrates accelerated failure time models with boosting techniques to avoid restrictive assumptions and improve accuracy, as demonstrated in simulations and an HIV trial analysis.
We propose a Buckley James (BJ) Boost Q learning framework for estimating optimal dynamic treatment regimes under right censored survival data, tailored for longitudinal randomized clinical trial settings. The method integrates accelerated failure time models with iterative boosting techniques, including componentwise least squares and regression trees, within a counterfactual Q learning framework. By directly modeling conditional survival time, BJ Boost Q learning avoids the restrictive proportional hazards assumption and enables unbiased estimation of stage specific Q functions. Grounded in potential outcomes, this framework ensures identifiability of the optimal treatment regime under standard causal assumptions. Compared to Cox based Q learning, which relies on hazard modeling and may suffer from bias under misspecification, our approach provides robust and flexible estimation. Simulation studies and analysis of the ACTG175 HIV trial demonstrate that BJ Boost Q learning yields higher accuracy in treatment decision making, especially in multistage settings where bias can accumulate.