Aim: The aim of this study was to demonstrate the effect of paired data numbers on the win ratio statistic and its 95% confidence intervals for two composite endpoints. Material and Methods: This is a simulation study. Data was generated for 35 different scenarios of matched data pairs within the range 10≤𝑛≤1000 using the Python random library. A total of 34101 different data sets (286 for 𝑛=10, 815 for 𝑛=15, and 1000 data sets for each of the other 33 cases with 𝑛≥20) were used in the study. Results: As the number of paired data increases, the win ratio increases, although not regularly. The number of cases where the value of the win ratio statistic is greater than one is not affected by the number of paired data, and the win ratio numbers considered important increase as the number of paired data increases. As the number of paired data increases, the lower and upper limits of the win ratio approach each other, although not regularly, and the differences between the upper and lower limits also decrease. Conclusion: When evaluating the results of the study in general, it can be said that the increase in the number of paired data affects the number of significant win ratio and the maximum win ratio, and the confidence interval of the win ratio is more affected by the data structure than the size of the number of paired data.
İsmet Doğan (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: