Impact of Non-Informative Censoring on Propensity Score Based Estimation of Marginal Hazard Ratios

TL;DR

This paper investigates how non-informative censoring affects the bias in estimating marginal hazard ratios using propensity score methods in observational studies, proposing a bias correction approach.

Contribution

It quantifies the bias introduced by non-informative censoring in propensity score-based MHR estimation and introduces a bias correction method for improved accuracy.

Findings

Bias increases with higher censoring rates.

Bias correction reduces estimation bias.

Proposed method outperforms conventional PS methods.

Abstract

In medical and epidemiological studies, one of the most common settings is studying the effect of a treatment on a time-to-event outcome, where the time-to-event might be censored before end of study. A common parameter of interest in such a setting is the marginal hazard ratio (MHR). When a study is based on observational data, propensity score (PS) based methods are often used, in an attempt to make the treatment groups comparable despite having a non-randomized treatment. Previous studies have shown censoring to be a factor that induces bias when using PS based estimators. In this paper we study the magnitude of the bias under different rates of non-informative censoring when estimating MHR using PS weighting or PS matching. A bias correction involving the probability of event is suggested and compared to conventional PS based methods.