Information-adaptive clinical trials with selective recruitment and binary outcomes

TL;DR

This paper introduces information-adaptive clinical trial designs that selectively recruit and allocate treatments based on statistical informativeness, leading to increased power and efficiency in trials.

Contribution

It defines and evaluates four methods for quantifying information, demonstrating their effectiveness in improving trial power and reducing required sample sizes.

Findings

Selective recruitment increases statistical power.

Information-adaptive allocation improves efficiency.

Potential for fewer recruits in successful trials.

Abstract

Selective recruitment designs preferentially recruit individuals that are estimated to be statistically informative onto a clinical trial. Individuals that are expected to contribute less information have a lower probability of recruitment. Furthermore, in an information-adaptive design recruits are allocated to treatment arms in a manner that maximises information gain. The informativeness of an individual depends on their covariate (or biomarker) values and how information is defined is a critical element of information-adaptive designs. In this paper we define and evaluate four different methods for quantifying statistical information. Using both experimental data and numerical simulations we show that selective recruitment designs can offer a substantial increase in statistical power compared to randomised designs. In trials without selective recruitment we find that allocating individuals to treatment arms according to information-adaptive protocols also leads to an increase in statistical power. Consequently, selective recruitment designs can potentially achieve successful trials using fewer recruits thereby offering economic and ethical advantages.