In this paper, we consider the case of efficient estimation of the risk dif
ference in a multicenter study allowing for baseline heterogeneity. We cons
ider the optimally weighted estimator for the common risk difference and sh
ow that this estimator has considerable bias when the true weights (which a
re inversely proportional to the variances of the center-specific risk diff
erence estimates) are replaced Ly their sample estimates. In addition, we p
ropose a new estimator for this situation of the Mantel-Haenszel type that
is unbiased and, in addition, has a smaller variance for small sample sizes
within the study centers. Simulations illustrate these findings.