Continuous Treatment Recommendation with Deep Survival Dose Response Function
This addresses the challenge of making personalized treatment recommendations in healthcare using observational data, which is incremental as it applies causal models to continuous treatments in a medical context for the first time.
The paper tackled the problem of learning continuous treatment effects from observational clinical survival data, proposing the Deep Survival Dose Response Function (DeepSDRF) to estimate treatment effects and develop recommender algorithms with bias correction, and tested it on simulations and the eICU database, finding similar performance between random search and reinforcement learning approaches.
We propose a general formulation for continuous treatment recommendation problems in settings with clinical survival data, which we call the Deep Survival Dose Response Function (DeepSDRF). That is, we consider the problem of learning the conditional average dose response (CADR) function solely from historical data in which observed factors (confounders) affect both observed treatment and time-to-event outcomes. The estimated treatment effect from DeepSDRF enables us to develop recommender algorithms with the correction for selection bias. We compared two recommender approaches based on random search and reinforcement learning and found similar performance in terms of patient outcome. We tested the DeepSDRF and the corresponding recommender on extensive simulation studies and the eICU Research Institute (eRI) database. To the best of our knowledge, this is the first time that causal models are used to address the continuous treatment effect with observational data in a medical context.