|
Activity Number:
|
138
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #307119 |
|
Title:
|
Statistical and Clinical Significance: a Practical Example Utilizing ST-Segment Monitor Endpoints in Acute ST Elevation MI (STEMI)
|
|
Author(s):
|
Cynthia Green*+
|
|
Companies:
|
Duke Clinical Research Institute
|
|
Address:
|
2400 Pratt Street, Durham, NC, 27705,
|
|
Keywords:
|
clinical trials ; medical devices ; biomarkers ; clinical meaning ; statistical significance
|
|
Abstract:
|
While mortality is the best measure of treatment benefit, thousands of patients often are required to show a statistically measurable reduction. Nonmortality biomarkers, capturing data at different times using different data collection methods and representing different physiological information, can show a significant treatment effect in smaller populations. A biomarker should be statistically and clinically significant. Statistical significance provides >95% likelihood that the observed biomarker change is the result of treatment, not chance or confounding. Clinical meaning denotes the biomarker change is associated with improved clinical outcome. Statistically significant benefits are not always clinically meaningful. This study shows a statistical model defining a threshold for clinical meaning for ST-segment endpoints used to measure therapeutic benefit in acute MI patients.
|
