GLUMIP 2.0: SAS/IML Software for Planning Internal Pilots.

Area of Science:

• Biostatistics
• Statistical Software Development

Background:

• Internal pilot designs allow sample size adjustments based on interim power analysis without analyzing interim data.
• Existing methods for internal pilot designs can inflate Type I error rates.
• The general linear univariate model (GLUM) framework is widely used in statistical analysis.

Purpose of the Study:

• To introduce GLUMIP 2.0, an updated free SAS/IML software for planning internal pilot studies.
• To provide a tool that avoids Type I error rate inflation associated with internal pilot designs.
• To facilitate exact power analysis for internal pilot studies under the GLUM framework.

Main Methods:

• Development of GLUMIP 2.0 software incorporating new analytic forms.
• Implementation within the general linear univariate model (GLUM) framework.
• Focus on independent Gaussian errors and fixed predictors.

Main Results:

• GLUMIP 2.0 offers solutions to problems in current internal pilot techniques for linear models.
• The software enables exact power analysis for internal pilot studies.
• Facilitates accurate sample size modification in study planning.

Conclusions:

• GLUMIP 2.0 provides a user-friendly and statistically sound approach to internal pilot study design.
• The software improves the reliability of power analysis in the GLUM framework.
• Researchers can confidently plan studies with adjusted sample sizes using GLUMIP 2.0.