New approaches for testing non-inferiority for three-arm trials with Poisson distributed outcomes

Area of Science:

• Biostatistics
• Clinical Trial Design
• Pharmaceutical Research

Background:

• Randomized controlled trials (RCTs) are crucial for discovering new treatments, but resource limitations necessitate innovative statistical methods.
• Non-inferiority (NI) trials assess if a new treatment is not unacceptably worse than an existing one, considering factors beyond just efficacy.
• Traditional NI trials face ethical challenges and require strong assumptions; three-arm trials offer a robust alternative by including placebo, reference, and experimental arms.

Purpose of the Study:

• To propose novel Frequentist and Bayesian statistical procedures for analyzing three-arm non-inferiority trials with Poisson distributed count outcomes.
• To provide improved methods for the design and analysis of randomized controlled trials (RCTs) when count data is the primary outcome.
• To explore the advantages of Bayesian approaches in active-control trials, particularly when historical data is available.

Main Methods:

• Development of a novel Frequentist approach for non-inferiority testing in three-arm trials with count data.
• Extension of the Frequentist approach to a Bayesian framework, leveraging historical data for improved inference.
• Discussion of sample size calculations for both Frequentist and Bayesian methods in this context.

Main Results:

• The study presents new Frequentist and Bayesian procedures tailored for three-arm non-inferiority trials with count outcomes.
• The proposed methods offer enhanced statistical power and flexibility for analyzing count data in clinical trials.
• Bayesian methods are highlighted for their utility when incorporating prior information from historical studies.

Conclusions:

• The developed Frequentist and Bayesian methods provide valuable tools for the design and analysis of three-arm non-inferiority trials with count outcomes.
• These innovative statistical approaches can aid in the efficient discovery and evaluation of new therapeutic interventions.
• The study emphasizes the practical application of advanced statistical techniques in modern clinical research.