Surrogate marker evaluation from an information theory perspective.

摘要: The last 20 years have seen lots of work in the area surrogate marker validation, partly devoted to frame evaluation a multitrial framework, leading definitions terms quality trial- and individual-level association between potential true endpoint (Buyse et al., 2000, Biostatistics 1, 49-67). A drawback is that different settings led measures at individual level. Here, we use information theory create unified definition surrogacy with an intuitive interpretation, offering interpretational advantages, applicable wide range situations. Our method provides better insight into chances finding good given situation. We further show some previous proposals follow as special cases our method. illustrate methodology using data from clinical study psychiatry.