Objective To compare the balance of simple randomization, stratified blocked randomization and minimization. Methods Monte Carlo technique was employed to simulate the treatment allocation of simple randomization, stratified blocked randomization and minimization respectively, then the balance of treatment allocation in each group and the balance for every prognostic factor were compared. Results The simulation demonstrated that minimization provides the best performance to ensure balance in the number of patients between groups and prognostic factors. Balance in prognostic factors achieved with stratified blocked randomization was similar to that achieved with simple randomization. Conclusion Minimization offers the best balance in the number of patients and prognostic factors between groups.
Objective To introduce the use of Central Randomization System in clinical trials. Methods We discussed the application of Central Randomization System in clinical trials from object management, drug management and user management, and made a brief description of minimization method. Results Central Randomization Systems can guarantee the nnplementation of the scheme of randomization, and can be used in clinical trials with minimization. Conclusion Central Randomization Systems are feasible in clinical trials especially in traditional Chinese medicine and open clinical trials.
ObjectiveTo introduce the theoretical foundation, application scenarios, and implementation in R language of covariate-adaptive randomization (CAR) and restricted randomization (RR). MethodsThis article initially expounds the significance of CAR and RR in clinical trials, particularly in balancing covariates between treatment groups on the basis of dynamic adjustment and pre-defined rules, in order to enhance the accuracy and reliability of trial outcomes. ResultsRR is applicable to large-scale trials, ensuring balance between groups but potentially inducing selection bias. CAR is suitable for small-sample and complex covariate trials, improving accuracy yet having complex implementation. In clinical trials of traditional Chinese medicine, CAR enables personalized group allocation, and RR ensures baseline balance. Dynamic randomization strategies enhance the flexibility of trials. ConclusionThrough code examples in R language, this study offers practical guidance for researchers to implement these randomization methods, ensuring the scientificity and rigor of data processing and analysis.