# The population-attributable fraction for time-dependent exposures and competing risks - A discussion on estimands

## TL;DR

This paper discusses the challenges of defining and estimating the population-attributable fraction for time-dependent exposures with competing risks, proposing a new estimand and evaluating it through simulations and real data application.

## Contribution

It introduces a novel estimand based on dynamic prediction by landmarking to better quantify public health impact in complex time-to-event data.

## Key findings

- The proposed estimand improves interpretability in presence of competing risks.
- Simulation results show better performance of the new estimand under certain conditions.
- Application to ICU data estimates the health impact of ventilator-associated pneumonia.

## Abstract

The population-attributable fraction (PAF) quantifies the public health impact of a harmful exposure. Despite being a measure of significant importance an estimand accommodating complicated time-to-event data is not clearly defined. We discuss current estimands of the PAF used to quantify the public health impact of an internal time-dependent exposure for data subject to competing outcomes. To overcome some limitations, we proposed a novel estimand which is based on dynamic prediction by landmarking. In a profound simulation study, we discuss interpretation and performance of the various estimands and their estimators. The methods are applied to a large French database to estimate the health impact of ventilator-associated pneumonia for patients in intensive care.