Methods for the analysis of multiple endpoints in small populations: A review

Area of Science:

• Biostatistics
• Clinical Trial Design
• Pharmaceutical Research

Background:

• Current clinical trial guidelines often favor single primary endpoints.
• Analyzing multiple endpoints is crucial for comprehensive treatment effect assessment in specific settings.
• Combining estimates from multiple outcomes can boost statistical power, particularly in small population trials.

Purpose of the Study:

• To review statistical methods for analyzing multiple endpoints in clinical trials.
• To focus on approaches suitable for small sample sizes and non-asymptotic considerations.
• To identify methods that increase statistical power or allow confirmatory conclusions on individual endpoints.

Main Methods:

• Systematic literature search of Scopus database, supplemented by manual search.
• Identification and grouping of analysis methods for multiple endpoints relevant to small populations.
• Review of approaches combining endpoints and multiple testing procedures.

Main Results:

• Methods were categorized into those combining endpoints for increased power and those using multiple testing for individual endpoint conclusions.
• Focus on feasibility and applicability in small sample size trials.
• Consideration of both combined endpoint measures and individual endpoint testing procedures.

Conclusions:

• Statistical methods exist to effectively analyze multiple endpoints in small population clinical trials.
• These methods can enhance statistical power and provide robust evidence for treatment effects.
• The reviewed approaches support comprehensive conclusions on treatment efficacy across diverse outcomes.