Nonparametric testing methods based on relative effect in non-inferiority clinical trial with multiple experimental...

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Rare Disease Research

Background:

• Conventional non-inferiority (NI) trials often require large sample sizes and parametric statistical methods.
• Rare disease research faces challenges due to small patient populations and non-normal data distributions.
• Existing methods may be inadequate for NI trials with limited data or multiple experimental drugs.

Purpose of the Study:

• To review existing parametric and nonparametric NI testing methods.
• To propose a novel nonparametric NI method utilizing a rank-based relative effect measure.
• To evaluate the performance of the proposed method against existing approaches.

Main Methods:

• Review of parametric and nonparametric NI statistical methods.
• Development of a new nonparametric NI test based on relative effect (rank-based).
• Simulations and analysis of real clinical trial data to assess method performance.

Main Results:

• The proposed nonparametric method demonstrated superior performance in scenarios with small to moderate sample sizes.
• The novel method was effective even when data did not follow a normal distribution.
• Outperformance was observed compared to existing NI testing methods in specific conditions.

Conclusions:

• Nonparametric approaches offer a viable and effective alternative for NI clinical trials, particularly in rare diseases.
• The proposed relative effect-based nonparametric method shows significant potential for improving NI trial efficiency and reliability.
• This research highlights the importance of considering nonparametric methods for NI trials facing data limitations.