stratified randomization

A method for ensuring that an equal number of participants with a characteristic that may influence the prognosis or the response to the intervention will be allocated to each of the compared groups, to reduce the potential for confusion bias.

Note: For example, in a breast cancer study, it may be important to have the same proportion for pre-menopausal and post-menopausal women in each group. To that end, the randomisation is done separately for each menopausal status stratum (often using random permuted blocks) or by means of minimization.

Alternate spelling: stratified randomisation.