Randomization of subjects between treatment groups in a clinical trial is an essential statistical element that allows one to claim that a study will be unbiased.
In this webinar, we look at the benefits that randomization provides and examine the properties of three well known algorithms for generating randomization lists: block randomizationcomplete randomization and Efron’s biased coin randomization.
More About The Webinar
A pivotal component for most clinical trials is randomization i.e. the random assignment of patients to receive either the experimental treatment(s) or controls. Without randomization, it can be difficult to ensure statistical comparability for the treatment effect between the different groups being compared. Blinded randomization also prevents operational bias due to trialists’ expectations influencing treatment assignment. Both of these could strongly distort the results of the study.
Many algorithms exist for implementing randomization in a clinical trial. While Complete Randomization (equivalent to assigning treatment via coin toss) or its variants (such as Efron’s Biased Coin) are sometimes used, in many clinical trials other algorithms are usually used to ensure that the influence of time-varying treatment effects or important covariates can be accounted for. For example the most widely used randomization algorithm, Block Randomization, prevents long runs of single group assignment and can account for stratified randomization on important covariates such as age.
In this webinar, we discuss the reasons why randomization is such an important part of clinical trial design, compare the three algorithms available for randomization (Block Randomization, Complete Randomization, Efron’s biased coin randomization) and provide a practical tutorial on how software can be used to generate randomization lists in a quick and flexible manner.
Download both sets of example data below, to use on your desktop version of nQuery.
|Efron's Biased Coin Randomization|
Looking for more resources?