Related Experiment Video
Updated: Aug 12, 2026

11:00
Biaxial Mechanical Characterizations of Atrioventricular Heart Valves
Published on: April 9, 2019
Spreadsheet method for determining sample sizes for heart valve studies
1Natural Science Division, Pepperdine University, Malibu CA 90263, USA.
The Journal of Heart Valve Disease
|January 1, 1995
Summary
This study implements a heart valve clinical trial sample size calculation method on a spreadsheet. It ensures the most powerful statistical test for precise significance and power, optimizing study design.
Area of Science:
- Biostatistics
- Medical Devices
- Clinical Trials
Background:
- Accurate sample size calculation is crucial for clinical study validity.
- Previous methods for heart valve study sample sizes exist.
- Implementing these methods efficiently is key for researchers.
Purpose of the Study:
- To provide a practical spreadsheet implementation of the Grunkemeier, Johnson, and Naftel sample size method.
- To compute the optimal sample size for the most powerful statistical test.
- To facilitate the generation of all necessary graphs and tables within a spreadsheet environment.
Main Methods:
- Utilized a computer spreadsheet to implement a known sample size calculation method.
- Focused on identifying the sample size for the most statistically powerful test at a given significance level.
- Ensured all graphical and tabular outputs are generated directly on the spreadsheet.
Main Results:
- A functional spreadsheet tool for calculating sample sizes for heart valve clinical studies.
- Demonstrated that adherence to the most powerful test ensures optimal statistical power.
- Eliminated the need for specialized statistical software or functions.
Conclusions:
- The developed spreadsheet provides an accessible and efficient tool for clinical trial sample size determination.
- This implementation aids researchers in designing more powerful and statistically sound studies.
- The method ensures that the calculated sample size supports the most powerful statistical test, avoiding larger sample sizes for less powerful tests.

