Generally bias cannot be adjusted for by increasing sample size. However in non differential misclassification of outcome or exposure, increasing sample size can reduce the bias. Non differential misclassification of exposure would be unlikely in a clinical trial.