English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Adapting the final sample size of a trial to the evidence accruing during thetrial is a natural way to address planning uncertainty. Designs with adaptivesample size need to account for their optional stopping to guarantee stricttype-I error-rate control. A variety of different methods to maintain type-Ierror-rate control after unplanned changes of the initial sample size have beenproposed in the literature. This makes interim analyses for the purpose ofsample size recalculation feasible in a regulatory context. Since the samplesize is usually determined via an argument based on the power of the trial, aninterim analysis raises the question of how the final sample size should bedetermined conditional on the accrued information. Conditional power is aconcept often put forward in this context. Since it depends on the unknowneffect size, we take a strict estimation perspective and compare assumedconditional power, observed conditional power, and predictive power withrespect to their properties as estimators of the unknown conditional power. Wethen demonstrate that pre-planning an interim analysis using methodology forunplanned interim analyses is ineffective and naturally leads to the concept ofoptimal two-stage designs. We conclude that unplanned design adaptations shouldonly be conducted as reaction to trial-external new evidence, operational needsto violate the originally chosen design, or post hoc changes in the objectivecriterion. Finally, we show that commonly discussed sample size recalculationrules can lead to paradoxical outcomes and propose two alternative ways ofreacting to newly emerging trial-external evidence.

Conditional Power and Friends: The Why and How of (Un)planned, Unblinded Sample Size Recalculations in Confirmatory Trials