Font Size: a A A

Statistical Comparison Among Simple Randomization, Stratified Blocked Randomization And Minimization In Clinical Trials

Posted on:2007-10-24Degree:MasterType:Thesis
Country:ChinaCandidate:S Y YanFull Text:PDF
GTID:2144360185970795Subject:Epidemiology and Health Statistics
Abstract/Summary:PDF Full Text Request
As one of the three basic principles in experiment design, the important role of randomization was known well by everyone. The principle of randomization is that everyone has the same chance to be chosen, includes random sampling and random assignment Random sampling is that each unit in population has the same chance to enter the study; and the later is that each unit enters the study with the equal chance to be assigned to one of two or more treatments. Because of the peculiarity of clinical trials, it is almost impossible to do random sampling in clinical trials. The application of random assignment is more popular.There are many randomization methods in clinical trials. Simple randomization, blocked randomization and stratified randomization are the most popular. When sample size is small, simple randomization takes the risk of imbalanced treatment totals, and the possibility that important prognostic factors may also be imbalanced by chance. To avoid imbalance and improve the statistical power, blocked randomization or stratified randomization is usually employed. Blocked randomization can avoid imbalance in treatment totals and stratified randomization can ensure that the distribution of important prognostic factors are balanced among...
Keywords/Search Tags:Randomization, stratified blocked randomization, simple randomization, dynamic randomization, central randomization system, minimization
PDF Full Text Request
Related items