We are delighted to announce the winners of the 2022 Statistical Excellence in the Pharmaceutical Industry Award, which is presented jointly by the RSS and Statisticians in the Pharmaceutical Industry (PSI).
A team of biostatisticians and statistical engineers from the healthcare company Roche in collaboration with the University of Bath has won this year’s prize for their entry ‘Standard and reference-based conditional mean imputation (methodology and open-source software)’. The team comprising of Marcel Wolbers, Alessandro Noci, Paul Delmar, Craig Gower-Page, Sean Yiu (of Roche) and Jonathan W. Bartlett (of the University of Bath) were praised by the judges for their innovative approach.
The project addresses the need of trial statisticians in the pharmaceutical industry to fully align their analysis strategy with the targeted estimand and flexible missing data assumptions for clinical trials with continuous longitudinal outcomes. The team present a novel imputation procedure which is fully deterministic (i.e. free of Monte Carlo error) and provides treatment effect estimates consistent with the Bayesian approach as well as reliable frequentist and inference with accurate standard error estimation and type I error control. The companion open-source R package ‘rbmi’ is a flexible toolbox to implement these methods in practice.
Chrissie Fletcher, Chair of PSI, said: ‘The PSI/RSS Statistical Excellence in the Pharmaceutical Industry Award is a significant achievement and recognition of a new statistical innovation created to advance drug development. Well done to the team at Roche who have applied innovative statistical thinking to a common issue in clinical trials relating to missing data and also provided a new R package to implement the methodology.’
Stian Westlake, RSS CEO, added: ‘The team are to be congratulated for this outstanding piece of work which could potentially be a big game-changer for the industry.’