Binary Outcome Equivalence Trial
A Binary Outcome Equivalence Trial is a clinical trial method that aims to prove that two treatment methods are equivalent in terms of a binary outcome. A binary outcome can be, for example, “success” vs. “failure” or “survival” vs. “death”.
Novustat is your experienced partner and advises you on all aspects of statistics, from data analysis to the interpretation of results, and helps you to gain maximum insight from your data sets. Contact us without obligation for further information!
The aim of such a study is to show that the difference in outcome between two groups is not clinically significant. In contrast to superiority studies, in which one treatment is to be shown to be better than the other, the focus here is on demonstrating equivalence. This means that the new treatment has no significant advantage or disadvantage compared to the existing treatment.
Calculating the sample size for a binary outcome equivalence trial is crucial. It ensures that the study provides a high degree of statistical certainty. This involves defining an acceptable difference between the two groups, within which the results are considered equivalent. Our calculator for such studies helps determine this difference and the corresponding sample size.
Use our online calculator to quickly and easily determine the required sample size for your equivalence study.
A typical example of a binary outcome equivalence trial would be the comparison of two medications for the treatment of a disease where the primary outcome is the cure rate, e.g. “cured” vs. “not cured”). If a difference of 5% is considered clinically irrelevant, the trial would show that the two drugs are equivalent in this respect if the cure rates in both groups are within this 5% range.
Overall, conducting a binary outcome equivalence trial requires careful planning and precise definition of equivalence boundaries as well as sound statistical analysis to ensure that the results are valid and clinically relevant.
To ensure that a binary outcome equivalence trial is statistically valid, the sample size must be calculated to achieve the desired significance level and statistical power. The calculation is based on a specified significance level and a desired power. In addition, a defined equivalence margin is used, which specifies the acceptable difference between the two treatment groups.
The significance level, usually referred to as alpha (α) , indicates the probability that a result is falsely considered significant. In equivalence research, an alpha of 0.05 is often specified, which corresponds to a 5% probability of a first type error.
The power of a study, represented as 1-β, indicates how likely it is that the study will detect a true equivalence if it actually exists. A power of 95% means that the probability of correctly detecting an equivalence is 95%. The remaining beta (β) of 5% represents the probability of a second type of error.
The success rate, shown here with the symbol π, indicates the percentage that achieves the desired treatment success in both groups. In many studies, an assumed success rate of 10% is used for both groups.
The equivalence threshold defines the maximum acceptable difference between the success rates of the two groups within which the treatments are considered equivalent. A value of 5% (d = 5) means that a difference of up to 5% is considered clinically irrelevant.
The calculation of the sample size is based on the following formula:
Here, π represents the success rate in both groups. f(α, β/2) refers to the Z-scores, which are determined by the significance level and the power. This formula ensures a precise calculation of the sample size in order to obtain meaningful results in an equivalence study.
Frequently asked questions
What is a binary outcome equivalence trial?
Such a trial examines whether two treatments can be considered equivalent for a binary outcome within predefined margins.
How does an equivalence trial differ from a superiority trial?
A superiority trial aims to demonstrate an advantage. An equivalence trial instead examines whether a possible difference is not clinically meaningful.
What determines the required sample size?
Key factors are the expected success rate, equivalence margin, significance level, and desired statistical power.
What does the equivalence margin represent?
It defines the maximum permitted difference between success rates for the treatments to still be considered equivalent.
What is particularly important when planning?
The binary outcome and equivalence margin must be justified and predefined; sample size and analysis method also require careful planning.
Plan equivalence trials with binary outcomes accurately.
NOVUSTAT supports you in defining robust equivalence margins, planning sample size, preparing the statistical analysis plan, and interpreting binary outcomes soundly.
