Comparisons of the t test and Wilcoxon rank sum test for the time, late time and focus: CCTRN trials

Yiwen Cao, The University of Texas School of Public Health

Abstract

The purpose of this study was to determine if different statistical methods would provide consistent results from clinical trials on stem cells to confirm if a patient’s bone marrow could be transplanted to the injured parts and improves the heart's function following a heart attack and heart failure. There were three completed trials conducted in the Cardiovascular Cell Therapy Research Network (CCTRN), i.e. TIME, Late TIME and FOCUS that would provide data for my thesis. Existing, de-identified data only were used. In a clinical trial setting, the primary outcome variables may not be normally distributed. So, the data might require the use of nonparametric analysis methods to assess the difference between treatment and placebo groups. This study compared the Wilcoxon rank sum test (nonparametric method) and the t test (parametric method) when they were applied to TIME, Late TIME and Focus trials. We found the results from the Wilcoxon rank sum test were all consistent with that from the t test from published results. Our study confirmed the conclusions from original t test, that there was no significant difference in the change in LV function between cell therapy and placebo group at 6 months compared to baseline. It would be informative on cell therapy for treating coronary artery disease for the future study.^

Subject Area

Biostatistics

Recommended Citation

Cao, Yiwen, "Comparisons of the t test and Wilcoxon rank sum test for the time, late time and focus: CCTRN trials" (2015). Texas Medical Center Dissertations (via ProQuest). AAI1603949.
http://digitalcommons.library.tmc.edu/dissertations/AAI1603949

Share

COinS