Premium
Criteria for surrogate end points
Author(s) -
Chen Hua,
Geng Zhi,
Jia Jinzhu
Publication year - 2007
Publication title -
journal of the royal statistical society: series b (statistical methodology)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.523
H-Index - 137
eISSN - 1467-9868
pISSN - 1369-7412
DOI - 10.1111/j.1467-9868.2007.00617.x
Subject(s) - surrogate endpoint , surrogate model , surrogate data , point (geometry) , computer science , statistics , mathematics , medicine , physics , geometry , nonlinear system , quantum mechanics
Summary. A surrogate end point is often used to evaluate the effects of treatments or exposures on the true end point in medical researches. Various criteria for the statistical surrogate, principal surrogate and strong surrogate have been proposed. We first illustrate that, with a surrogate end point that is defined by these criteria, it is possible that a treatment has a positive effect on the surrogate, which in turn has a positive effect on the true end point, but the treatment has a negative effect on the true end point. We define such a phenomenon as a surrogate paradox. The surrogate paradox also means that the sign of the treatment effect on the true end point is unpredictable by the effect signs of both the treatment on the surrogate and the surrogate on the true end point. Then we propose two notions for a consistent surrogate and a strictly consistent surrogate to avoid the surrogate paradox. With the causal network that was presented by Lauritzen, we discuss the conditions for a strong surrogate to be a consistent surrogate and a strictly consistent surrogate.