Related Experiment Videos
Advanced statistics: how to determine whether your intervention is different, at least as effective as, or
Joshua H Tamayo-Sarver1, Jeffrey M Albert, Maritza Tamayo-Sarver
1Department of Epidemiology and Biostatistics, Case Western Reserve University School of Medicine, Cleveland, OH, USA. josh@tamayo-sarver.com
Summary
Many emergency medicine studies aim to show intervention differences but often fail. This paper reviews hypothesis testing and introduces methods to demonstrate intervention superiority or equivalence, enhancing result interpretation.
Area of Science:
- Emergency Medicine
- Clinical Research Methodology
- Biostatistics
Background:
- Many emergency medicine studies aim to detect differences between interventions.
- However, failing to detect a difference does not equate to proving equivalence or non-inferiority.
Purpose of the Study:
- To review classical hypothesis testing.
- To introduce methods for demonstrating that one intervention is at least as effective as another.
- To introduce methods for demonstrating the equivalence of two interventions.
Main Methods:
- Review of classical hypothesis testing principles.
- Introduction to non-inferiority testing methodologies.
- Introduction to equivalence testing methodologies.
Main Results:
- Distinguishing between failing to find a difference and demonstrating non-inferiority or equivalence.
- Understanding the statistical underpinnings of these distinct outcomes.
Conclusions:
- Appreciating the nuances between failing to find a difference and demonstrating non-inferiority or equivalence is crucial.
- This understanding can lead to a more accurate interpretation of study findings in emergency medicine.
- Improved statistical interpretation enhances the clinical significance of research results.