A comparison of the statistical power of different methods for the analysis of cluster randomization trials with binary outcomes

Cluster randomization trials are randomized controlled trials (RCTs) in which intact clusters of subjects are randomized to either the intervention or to the control. Cluster randomization trials require different statistical methods of analysis than do conventional randomized controlled trials due to the potential presence of within-cluster homogeneity in responses. A variety of statistical methods have been proposed in the literature for the analysis of cluster randomization trials with binary outcomes. However, little is known about the relative statistical power of these methods to detect a statistically significant intervention effect. We conducted a series of Monte Carlo simulations to examine the statistical power of three methods that compare cluster-specific response rates between arms of the trial: the t-test, the Wilcoxon rank sum test, and the permutation test; and three methods that compare subject-level response rates: an adjusted chi-square test, a logistic-normal random effects model, and a generalized estimating equations (GEE) method. In our simulations we allowed the number of clusters, the number of subjects per cluster, the intraclass correlation coefficient and the magnitude of the intervention effect to vary. We demonstrated that the GEE approach tended to have the highest power for detecting a statistically significant intervention effect. However, in most of the 240 scenarios examined, the differences between the competing statistical methods were negligible. The largest mean difference in power between any two different statistical methods across the 240 scenarios was 0.02. The largest observed difference in power between two different statistical methods across the 240 scenarios and 15 pair-wise comparisons of methods was 0.14.     

Examining the effect of linkage disequilibrium between markers on the Type I error rate and power of nonparametric multipoint linkage analysis of two-generation and multigenerational pedigrees in the presence of missing genotype data

Because most multipoint linkage analysis programs currently assume linkage equilibrium between markers when inferring parental haplotypes, ignoring linkage disequilibrium (LD) may inflate the Type I error rate. We investigated the effect of LD on the Type I error rate and power of nonparametric multipoint linkage analysis of two-generation and multigenerational multiplex families. Using genome-wide single nucleotide polymorphism (SNP) data from the Collaborative Study of the Genetics of Alcoholism, we modified the original data set into 30 total data sets in order to consider six different patterns of missing data for five different levels of SNP density. To assess power, we designed simulated traits based on existing marker genotypes. For the Type I error rate, we simulated 1,000 qualitative traits from random distributions, unlinked to any of the marker data. Overall, the different levels of SNP density examined here had only small effects on power (except sibpair data). Missing data had a substantial effect on power, with more completely genotyped pedigrees yielding the highest power (except sibpair data). Most of the missing data patterns did not cause large increases in the Type I error rate if the SNP markers were more than 0.3 cM apart. However, in a dense 0.25-cM map, removing genotypes on founders and/or founders and parents in the middle generation caused substantial inflation of the Type I error rate, which corresponded to the increasing proportion of persons with missing data. Results also showed that long high-LD blocks have severe effects on Type I error rates.