A new and improved confidence interval for the Mantel–Haenszel risk difference

Writing the variance of the Mantel-Haenszel estimator under the null of homogeneity and inverting the corresponding test, we arrive at an improved confidence interval for the common risk difference in stratified 2 × 2 tables. This interval outperforms a variety of other intervals currently recommended in the literature and implemented in software. We also discuss a score-type confidence interval that allows to incorporate strata/study weights. Both of these intervals work very well under many scenarios common in stratified trials or in a meta-analysis, including situations with a mixture of both small and large strata sample sizes, unbalanced treatment allocation, or rare events. The new interval has the advantage that it is available in closed form with a simple formula. In addition, it applies to matched pairs data. We illustrate the methodology with various stratified clinical trials and a meta-analysis. R code to reproduce all analysis is provided in the Appendix.