adaptr

adaptr simulates adaptive clinical trials using Bayesian statistical frameworks, incorporating adaptive stopping, arm dropping, and response-adaptive randomisation.

Key Features:

• Adaptive stopping: Allows trials to stop early based on Bayesian analysis of accumulating outcome data.
• Adaptive arm dropping: Enables dropping of inferior intervention arms during the trial based on interim analyses.
• Response-adaptive randomisation: Adjusts allocation probabilities to favour more promising interventions as data accumulates.
• Bayesian statistical framework: Provides analysis models and adaptation rules within a Bayesian paradigm.
• Statistical simulation: Supports evaluation of trial performance through simulation across multiple clinical scenarios.
• Performance metrics and calibration: Includes assessment of operating characteristics and sensitivity analyses.

Scientific Applications:

• Clinical trial design: Supports clinical trialists, methodologists, and biostatisticians in designing advanced adaptive randomised clinical trials.
• Trial performance evaluation: Enables evaluation of trial performance metrics including probability of conclusive results and allocation efficiency.
• Methodological research: Facilitates comparison of adaptive versus conventional trial designs and assessment of adaptation rules.

Methodology:

The package implements Bayesian statistical simulation of adaptive clinical trials with specification of interventions and common controls, outcome type and generation, analysis timing and outcome-data lag, allocation rules, analysis models, and adaptation rules for stopping and arm dropping.