ABSTRACT
Healthcare quality is impacted by structure and organization. Approximately 6 million Americans are admitted
to an intensive care unit (ICU) yearly and many require mechanical ventilation for acute respiratory failure.
Understanding optimal ICU organization is essential to ensure high quality care is delivered to these, our
sickest patients. Over the past 20 years as the number of US ICU beds and the complexity of care provided
has increased dramatically, intensivists (physicians trained specifically in critical care) working with
interprofessional ICU teams (including nurses, respiratory therapists, clinical pharmacists, etc.) have become
the norm in many ICUs. Studies highlight the positive impact of having an intensivist care for critically ill
patients; for this reason, the Society of Critical Care Medicine recommends “high intensity intensivist staffing”.
Similarly, the positive impact of a multidisciplinary team on patient outcomes is well established. What is not
known, however, is how ICU care providers impact patient care in the context of team structure and workload.
Our recent work in the United Kingdom suggests there is a significant relationship between the number of
patients each intensivist cares for and their patients’ mortality; whether this relationship is the same in the US
and how it is affected by other ICU care providers is unknown. Finally, since a landmark study in 2000
highlighted the gap between intensivist demand and supply in the US, it has become clear that updated ICU
workforce projections are needed to aid in resource planning; however, these will fail if they are limited by
“siloing” (e.g., projecting intensivist need without considering the mitigating effect of other care providers) and
an underappreciation of how optimal staffing structures may differ from what is in use today. In this study we
will use primary surveys linked with existing patient- level data across multiple US ICUs and a novel
methodology of System Dynamics Modeling to address 3 aims: (1) determine detailed staffing models currently
used across the US; (2) quantify the association of patient-to-care provider ratio with patient outcomes across
selected ICUs; and (3) estimate current and future ICU workforce need. This project will yield critical insights
into the best staffing models for ICU care delivery and how resources must be allocated in the future to close
ICU care provider gaps. While focused on ICU care, this project will create a replicable framework for (1)
quantifying provider workload by daily patient census in light of other staffing availability and (2) using System
Dynamics Modeling to simulate workforce supply and demand which will be useful to plan for any aspect of
healthcare. We will provide clinicians and policy makers with key information on ICU staffing to improve patient
safety.