On the Individual Surrogate Paradox

Linquan Ma, Yunjian Yin, Lan Liu, Zhi Geng

Research output: Contribution to journalArticle

Abstract

When the primary outcome is difficult to collect, surrogate endpoint is typically used as a substitute. It is possible that for every individual, treatment has a positive effect on surrogate, and surrogate has a positive effect on primary outcome, but for some individuals, treatment has a negative effect on primary outcome. For example, a treatment may be substantially effective in preventing the stroke for everyone, and lowering the risk of stroke is universally beneficial for a longer survival time, however, the treatment may still cause death for some individuals. We define such paradoxical phenomenon as individual surrogate paradox. The individual surrogate paradox is preposed to capture the treatment effect heterogeneity, which is unable to be described by either the surrogate paradox based on average causal effect (ACE) (Chen et al., 2007) or that based on distributional causal effect (DCE) (Ju and Geng, 2010). We investigate existing surrogate criteria in terms of whether the individual surrogate paradox could manifest. We find that only the strong binary surrogate can avoid such paradox without additional assumptions. Utilizing the sharp bounds, we propose novel criteria to exclude the individual surrogate paradox. Our methods are illustrated in an application to determine the effect of the intensive glycemia on the risk of development or progression of diabetic retinopathy.
Original languageEnglish (US)
JournalarXiv
StatePublished - 2017

Fingerprint

Paradox
Causal effect
Progression
Treatment effects
Substitute

Keywords

  • Statistics - Methodology

Cite this

Ma, L., Yin, Y., Liu, L., & Geng, Z. (2017). On the Individual Surrogate Paradox. arXiv.

On the Individual Surrogate Paradox. / Ma, Linquan; Yin, Yunjian; Liu, Lan; Geng, Zhi.

In: arXiv, 2017.

Research output: Contribution to journalArticle

Ma, L, Yin, Y, Liu, L & Geng, Z 2017, 'On the Individual Surrogate Paradox', arXiv.
Ma, Linquan ; Yin, Yunjian ; Liu, Lan ; Geng, Zhi. / On the Individual Surrogate Paradox. In: arXiv. 2017.
@article{54365769eb4e465b8f4c6d3fc5ab5e6d,
title = "On the Individual Surrogate Paradox",
abstract = "When the primary outcome is difficult to collect, surrogate endpoint is typically used as a substitute. It is possible that for every individual, treatment has a positive effect on surrogate, and surrogate has a positive effect on primary outcome, but for some individuals, treatment has a negative effect on primary outcome. For example, a treatment may be substantially effective in preventing the stroke for everyone, and lowering the risk of stroke is universally beneficial for a longer survival time, however, the treatment may still cause death for some individuals. We define such paradoxical phenomenon as individual surrogate paradox. The individual surrogate paradox is preposed to capture the treatment effect heterogeneity, which is unable to be described by either the surrogate paradox based on average causal effect (ACE) (Chen et al., 2007) or that based on distributional causal effect (DCE) (Ju and Geng, 2010). We investigate existing surrogate criteria in terms of whether the individual surrogate paradox could manifest. We find that only the strong binary surrogate can avoid such paradox without additional assumptions. Utilizing the sharp bounds, we propose novel criteria to exclude the individual surrogate paradox. Our methods are illustrated in an application to determine the effect of the intensive glycemia on the risk of development or progression of diabetic retinopathy.",
keywords = "Statistics - Methodology",
author = "Linquan Ma and Yunjian Yin and Lan Liu and Zhi Geng",
year = "2017",
language = "English (US)",
journal = "arXiv",

}

TY - JOUR

T1 - On the Individual Surrogate Paradox

AU - Ma, Linquan

AU - Yin, Yunjian

AU - Liu, Lan

AU - Geng, Zhi

PY - 2017

Y1 - 2017

N2 - When the primary outcome is difficult to collect, surrogate endpoint is typically used as a substitute. It is possible that for every individual, treatment has a positive effect on surrogate, and surrogate has a positive effect on primary outcome, but for some individuals, treatment has a negative effect on primary outcome. For example, a treatment may be substantially effective in preventing the stroke for everyone, and lowering the risk of stroke is universally beneficial for a longer survival time, however, the treatment may still cause death for some individuals. We define such paradoxical phenomenon as individual surrogate paradox. The individual surrogate paradox is preposed to capture the treatment effect heterogeneity, which is unable to be described by either the surrogate paradox based on average causal effect (ACE) (Chen et al., 2007) or that based on distributional causal effect (DCE) (Ju and Geng, 2010). We investigate existing surrogate criteria in terms of whether the individual surrogate paradox could manifest. We find that only the strong binary surrogate can avoid such paradox without additional assumptions. Utilizing the sharp bounds, we propose novel criteria to exclude the individual surrogate paradox. Our methods are illustrated in an application to determine the effect of the intensive glycemia on the risk of development or progression of diabetic retinopathy.

AB - When the primary outcome is difficult to collect, surrogate endpoint is typically used as a substitute. It is possible that for every individual, treatment has a positive effect on surrogate, and surrogate has a positive effect on primary outcome, but for some individuals, treatment has a negative effect on primary outcome. For example, a treatment may be substantially effective in preventing the stroke for everyone, and lowering the risk of stroke is universally beneficial for a longer survival time, however, the treatment may still cause death for some individuals. We define such paradoxical phenomenon as individual surrogate paradox. The individual surrogate paradox is preposed to capture the treatment effect heterogeneity, which is unable to be described by either the surrogate paradox based on average causal effect (ACE) (Chen et al., 2007) or that based on distributional causal effect (DCE) (Ju and Geng, 2010). We investigate existing surrogate criteria in terms of whether the individual surrogate paradox could manifest. We find that only the strong binary surrogate can avoid such paradox without additional assumptions. Utilizing the sharp bounds, we propose novel criteria to exclude the individual surrogate paradox. Our methods are illustrated in an application to determine the effect of the intensive glycemia on the risk of development or progression of diabetic retinopathy.

KW - Statistics - Methodology

M3 - Article

JO - arXiv

JF - arXiv

ER -