|
Abstract:
|
Ratio of means and difference of means are often used for continuous data in a superiority, non-inferiority (NI), or average bioequivalence (ABE) test to evaluate whether the test mean is superior , NI, or equivalent to the reference (placebo or active control) mean. Previous literature provided some recommendations regarding how to choose between the two measures, e.g., whether the biological effect of the treatment is additive or relative, whether the outcome of interest is measured in identical units across trials (Friedrick et al 2008, 2012), whether different shifts (location shift, scale shift, a combination of location and scale shifts, or other shifts) are applied to the scoring system of the same outcome (Sun et al 2016). In this paper, we evaluate the two measures from other perspectives focusing on the impact of adjustment for covariates/factors in the model on the treatment effect in a NI or ABE test when using ratio of means vs. difference of means as the measure. Our conclusion is when covariates or factors are adjusted in the model, treatment effect is usually more complicated and arbitrary for a ratio of means NI or ABE test than a difference of means NI or ABE t
|
