Identifying the consequences of dynamic treatment strategies: A decision-theoretic overview

By A. P. Dawid, Vanessa Didelez. 2010

111 citations recorded by the source corpus.

We consider the problem of learning about and comparing the consequences of dynamic treatment strategies on the basis of observational data. We formulate this within a probabilistic decision-theoretic framework. Our approach is compared with related work by Robins and others: in particular, we show how Robins's 'G-computation' algorithm arises naturally from this decision-theoretic perspective. Careful attention is paid to the mathematical and substantive conditions required to justify the use of this formula. These conditions revolve around a property we term stability, which relates the probabilistic behaviours of observational and interventional regimes. We show how an assumption of 'sequential randomization' (or 'no unmeasured confounders'), or an alternative assumption of 'sequential irrelevance', can be used to infer stability. Probabilistic influence diagrams are used to simplify manipulations, and their power and limitations are discussed. We compare our approach with alternative formulations based on causal DAGs or potential response models. We aim to show that formulating the problem of assessing dynamic treatment strategies as a problem of decision analysis brings clarity, simplicity and generality.

Subfields: Economics and Econometrics, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Statistics and Probability

Citations in the corpus

Causal identifiability via Chain Event Graphs (2013, background): “Pearl’s Front Door theorem [14,15] for BNs can be used in cases where the Back Door theorem conditions do not hold or where the events needing to be observed for the Back Door theorem have too large an observational cost.”
Single World Intervention Graphs ( SWIGs ) : A Unification of the Counterfactual and Graphical Approaches to Causality (2013, background): “…and factorization properties encoded by our transformed graph are even satisfied by causal models that do not explicitly involve potential outcomes, such as ‘causal Bayesian networks’ (Pearl, 2000) and causal influence diagrams (Dawid and Didelez, 2008, 2010); see §9 for further details.”
Statistical Causality from a Decision-Theoretic Perspective (2014, background): “This computation, which can be effectively restructured as a recursion (Dawid and Didelez 2010), reduces to the g-computation formula of Robins (1986).”
Doubly Robust Estimation of Optimal Dynamic Treatment Regimes (2013, methodology): “Recent work includes Arjas and Saarela [1], Dawid and Didelez [6], Moodie et al. [13], Zhao et al. [21, 22].”
A Bayesian view of doubly robust causal inference (2016, background): “Obtaining (16) also required assuming that pE(yi | zi, xi;φ) = pO(yi | zi, xi;φ) ≡ p(yi | zi, xi;φ) and and pE(xi;ψ) = pO(xi;ψ) ≡ p(xi;ψ), which corresponds to the stability assumption discussed by Dawid and Didelez (2010).”
Causal Bounds and Observable Constraints for Non-deterministic Models (2012, methodology): “The model has been successfully applied in defining direct effects (Geneletti, 2007) and dynamic treatment strategies (Dawid and Didelez, 2010), to name a few.”