Causal Identification

Cross-source consensus on Causal Identification from 1 sources and 5 claims.

1 sources · 5 claims

How it works

Task I identification requires consistency, randomization, sequential ignorability, positivity, and baseline covariate overlap. — Efficient estimation of cumulative incidence curves via data fusion with surrogates: application to integrated analysis of vaccine trial and immunobridging data
The framework adds assumptions of approved-vaccine exchangeability across trial settings and no controlled direct effect between investigational and approved vaccines conditional on covariates and immune marker. — Efficient estimation of cumulative incidence curves via data fusion with surrogates: application to integrated analysis of vaccine trial and immunobridging data
No controlled direct effects is described as the most important and biologically fragile assumption. — Efficient estimation of cumulative incidence curves via data fusion with surrogates: application to integrated analysis of vaccine trial and immunobridging data
Several key assumptions cannot be tested when immunobridging trials lack clinical endpoints. — Efficient estimation of cumulative incidence curves via data fusion with surrogates: application to integrated analysis of vaccine trial and immunobridging data
Poor overlap may require restricting the target population or extrapolating beyond observed support. — Efficient estimation of cumulative incidence curves via data fusion with surrogates: application to integrated analysis of vaccine trial and immunobridging data