A statistical evaluation of decision-making methods and the efficiency of Bayesian multi-arm multi-stage trials

Abigail McGrory, Haolun Shi, Anna Heath · 2026

Read A statistical evaluation of decision-making methods and the efficiency of Bayesian multi-arm multi-stage trials by Abigail McGrory, Haolun Shi, Anna Heath (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13624359.txt.

Description

Abigail McGrory, Haolun Shi, and Anna Heath published a 2026 statistical study of how decision rules affect the efficiency of Bayesian multi-arm multi-stage trials. These trials test several treatments at once and check results midway, dropping losers and promoting winners early. Each check needs a cutoff: how strong results must be before declaring a treatment superior or futile Trialists set those cutoffs cautiously, which drains power.

The team used an algorithm to maximize the chance of detecting real effects while keeping the false-alarm rate under five percent. Simulations compared standard judging of several treatments against three alternatives, varying effect sizes and arm counts. Every method reached similar power, yet the best cutoffs differed between methods and sat below the conservative values used in practice. The authors argue that tuning the thresholds is free efficiency, and flag non-normal outcomes and platform designs as next tests.

Catalog record

CREATORAbigail McGrory, Haolun Shi, Anna Heath
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE34792
LICENSEhttps://creativecommons.org/licenses/by/4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/article/MED/PMC13624359

Browse the full library · About the Tower of Babel