Data-driven simulations help ease ED crowding

Researchers from the University of Florida are touting the success of a data-driven simulation model that helps reduce crowding in the emergency room.

There are a variety of factors that contribute to ED crowding, the authors note in their paper, published at BMC Medical Informatics and Decision Making. This model helps determine in advance which intervention could be most effective in particular circumstances.

The model looks at two scenarios: patient flow in the average U.S. emergency department and patient flow in an academic hospital ED. Using public data, it looks at factors such as patient door-to-event times, propensity to leave without being seen, occupancy level, staffing and resource use.

With a shortage of ED physicians, finding more appropriate sources of care for patients with less acute issues could help more than reducing wait times. A shortage of beds is a bigger problem at academic hospitals, however, so adding doctors doesn't necessarily improve patient flow at those facilities.

The model was also able to identify a point of diminishing returns. It found that adding one doctor in the average ED reduces mean length of stay by one hour, but adding a second doctor provides no further improvement.

The national average wait time in emergency departments is 28 minutes, with EDs becoming more crowded since the Affordable Care Act went into effect in January.

A University of Florida Health hospital in Gainesville has reduced ED visits among its most "frequent fliers"--those who had been hospitalized more than eight times in the past year--by using a clinic-based multidisciplinary team. In addition to the doctor, the team included a social worker, a pain and addiction psychiatrist and a pharmacist.

To learn more:
- here's the research (.pdf)