Clinical trials with rescue medication applied according to a deterministic rule

TL;DR

This paper proposes a new analysis method for clinical trials with rescue medication administered based on a deterministic rule, aiming to accurately estimate the biological effect despite confounding from additional treatments.

Contribution

It introduces an approach that treats rescue medication as a study outcome rather than a covariate, enabling clearer biological effect estimation in such trials.

Findings

Provides a practically unbiased estimator for normally distributed data.

Compared with ITT and non-rescue analyses, showing improved accuracy.

Addresses confounding issues in rescue medication administration.

Abstract

Clinical trials in specific indications require the administration of rescue medication in case a patient does not sufficiently respond to investigational treatment. The application of additional treatment on an as needed basis causes problems to the analysis and interpretation of the results of these studies since the effect of the investigational treatment can be confounded by the additional medication. Following-up all patients until study end and capturing all data is not fully addressing the issue. We present an analysis that takes care of the fact that rescue is a study outcome and not a covariate when rescue medication is administered according to a deterministic rule. This approach allows to clearly define a biological effect. For normally distributed longitudinal data a practically unbiased estimator of the biological effect can be obtained. The results are compared to an ITT analysis and an analysis on all patients not receiving rescue.