RePAIR

RePAIR leverages historical control data within a Bayesian framework to optimize sample size and increase statistical power and precision in animal experiments.

Key Features:

• Bayesian Framework: Utilizes prior information from historical control data to inform current study parameters and priors.
• Sample Size Optimization: Calculates the minimum sample size required to achieve a specified statistical power by incorporating priors from historical controls.
• Power Enhancement: Increases effective statistical power for a fixed number of animals by reducing variance through integration of historical control data.

Scientific Applications:

• Preclinical animal studies: Improves power and sample size calculation in animal research, including rodent experiments.
• Analysis of cognitive effects of early-life adversity: Applied to a dataset from seven independent rodent studies examining cognitive outcomes following early-life adversity.
• Method validation: Assesses methodological performance via simulation studies to evaluate power and sample size effects.

Methodology:

Implements a Bayesian analysis that incorporates prior distributions derived from historical control groups, integrates control-group data to reduce variability and improve precision, performs sample size calculations to determine minimum n for desired power, and is validated using simulation studies and application to a dataset of seven independent rodent studies on cognitive effects of early-life adversity.