Ranking of average treatment effects with generalized random forests for time-to-event outcomes

TL;DR

This paper introduces a novel data-adaptive method using generalized random forests to estimate and rank treatment effects on time-to-event outcomes, effectively handling competing risks and censoring.

Contribution

It develops a two-step estimation procedure that distinguishes between crude and net probabilities, applying inverse probability weighting within a causal forest framework.

Findings

Effective ranking of treatments based on causal effects.

Successful application to Danish health registry data.

Identification of drugs with unexpected protective effects.

Abstract

In this paper we present a data-adaptive estimation procedure for estimation of average treatment effects in a time-to-event setting based on generalized random forests. In these kinds of settings, the definition of causal effect parameters are complicated by competing risks; here we distinguish between treatment effects on the crude and the net probabilities, respectively. To handle right-censoring, and to switch between crude and net probabilities, we propose a two-step procedure for estimation, applying inverse probability weighting to construct time-point specific weighted outcomes as input for the forest. The forest adaptively handles confounding of the treatment assigned by applying a splitting rule that targets a causal parameter. We demonstrate that our method is effective for a causal search through a list of treatments to be ranked according to the magnitude of their effect. We further apply our method to a dataset from the Danish health registries where it is of interest to discover drugs with an unexpected protective effect against relapse of severe depression.