Abstract
Methods are presented for the estimation of a treatment effect based on before- and after-treatment values, where for ethical reasons all and only those patients are treated whose before-treatment values exceed a given constant.
Full text
PDF

