Issues in Randomization Outline
Issues in Randomization Outline
Issues in Randomization Outline
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
Adaptive ?<strong>Randomization</strong>?Same Title, Different Mean<strong>in</strong>gs• Basel<strong>in</strong>e Covariate– M<strong>in</strong>imization/Dynamic allocation– Pocock & Simon; Brad Efron (biasedco<strong>in</strong>)• Adaptive <strong>Randomization</strong>/Allocation– Us<strong>in</strong>g <strong>in</strong>terim outcome data– Play the w<strong>in</strong>ner or 2-armed bandit– BayesianBasel<strong>in</strong>e Covariate Adaptive<strong>Randomization</strong>/Allocation• M<strong>in</strong>imization/Dynamic Allocation– Balance on the marg<strong>in</strong>s• Table 1 looks pretty– Does not promise overall treatmentarms balanced <strong>in</strong> #• Pocock & Simon (biased co<strong>in</strong>)– Basel<strong>in</strong>e covariates– Weighted probability (not 50/50)Why not just stratify?• Typically, many many variables• Will not have people <strong>in</strong> each “cell”if do traditional stratification– How many participants• Pittsburgh Site, Male, 40-64,• AND Grade 2, hormone therapy, 6-18 mopost treatment,• AND……8