It is well-known that the power of Cochranâ??s Q test to assess the presence of\nheterogeneity among treatment effects in a clinical meta-analysis is low due\nto the small number of studies combined. Two modified tests (PL1, PL 2) were\nproposed by replacing the profile maximum likelihood estimator (PMLE) into\nthe variance formula of logarithm of risk ratio in the standard chi-square\ntest statistic for testing the null common risk ratios across all k studies\n( i 1, , k ). The simply naive test (SIM) as another comparative candidate\nhas considerably arisen. The performance of tests in terms of type I error rate\nunder the null hypothesis and power of test under the random effects hypothesis\nwas done via a simulation plan with various combinations of significance\nlevels, numbers of studies, sample sizes in treatment and control arms,\nand true risk ratios as effect sizes of interest. The results indicated that for\nmoderate to large study sizes ............
Loading....