Email (record): On Estimation of the Survivor Average Causal Effect in Observational Studies when Important Confounders are Missing Due to Death