Skip to main content
. Author manuscript; available in PMC: 2018 Jan 1.
Published in final edited form as: Crit Care Med. 2017 Jan;45(1):85–93. doi: 10.1097/CCM.0000000000002026

Figure 1. Conceptual model including instrumental variables analysis.

Figure 1

Patients choose whether to be treated in a rural emergency department (ED) or a top-decile sepsis volume ED, but that decision is influenced by several factors, including geography, demographics, comorbid conditions, and illness severity. After arriving to their ED of choice, patients may be admitted directly to the hospital or transferred to another hospital for admission. The care provided during the ED and inpatient stay influences mortality, but measurable and unmeasurable variables that influenced patient choice also influence mortality. Distance to a top-decile sepsis volume hospital is proposed as an instrumental variable because it is clearly associated with patient choice, but should not be associated with mortality except through the care and choices that patients make in where they receive their care. In that way, the instrument can be used to understand the role of patient choice in influencing care and ultimately the causal relationship between patient choice and mortality.