The authors reply [to: Is the Severity of Illness a Confounder or Mediator for the Association Between Emergency Department Length of Stay and Hospital Mortality?]

We appreciate Sasaki et al and Ji et al for their comments on our recently published study in Critical Care Medicine. Both letters expressed their concern on the aspect of causality in the association (emergency department [ED] to ICU time and mortality) that we tested. They argue that Acute Physiology and Chronic Health Evaluation (APACHE) IV probability was not measured directly at ICU admission. The APACHE IV probability might have been affected by pretreatment and may therefore be considered more a mediator variable than a confounder. [...] Taken together, we considered APACHE IV probability a confounder and next an effect modifier and reported the results in our recent work. We decided to start a new study where we will collect ED triage score information. Together with prospective studies starting follow-up at ED admission, this new data will hopefully shed more light on the causal relationships between disease severity, ED to ICU time and mortality.