Stanford Develops Method to More Accurately Predict ER Wait Times

The fear of a sudden trip to the emergency room is all-encompassing, but scattered and incorrect waiting times can often make a bad situation even worse. The Stanford Graduate School of Business recently explored this unfortunate issue, finding ways to bridge the gap between the ER’s intention and execution.

In a study published in the Manufacturing & Service Operations Management Journal, Stanford professors Mohsen Bayati and Erica L. Plambeck, with the help of graduate students Sara Kwasnick and Erjie Ang, and San Mateo Medical Center member Michael Aratow, looked at how four hospitals and their emergency departments estimated “wait times using actual patient data.” The researchers found that the “most commonly used method proved extremely unreliable in all cases.” For much of the time at SMMC, their estimates were “off by as much as an hour and a half.”

Bayati explains the difficulty of accurately assessing wait times and their desire to find a simple, accurate measurement in spite of the “complicated reality on the ground.”

“It turns out that it’s hard to tell people exactly what the wait time will be, because by the time they arrive, things will have changed. At the same time, there is a lot of uncertainty about how the number is being generated,” they explained.

Spikes in demand are difficult to predict and no patient is the same. They arrive erratically with a wide range of conditions—anything from injuries sustained from a car accident to heart attacks to indigestion. “Since critically endangered patients are usually seen instantly, this skews the predicted wait times significantly,” which is to say nothing of the capacity of the available hospital staff.

Dubbed “Q-Lasso,” the team developed a new method to assess wait times that “cut the margin of error by as much as 33 percent,” but was still off between 17 minutes and an hour. The researchers drew upon queuing theory, which explains “how people and things move through lines,” to assess a large number of potential factors from which the program could then select the best option(s).

The study suggests that Q-Lasso could accommodate “any complex system with a lot of variability in both the people or things that need to be addressed as well as the providers available to address them, [such as] passenger boarding and deboarding on airlines, and complicated manufacturing processes.”

Kwasnick concludes that he hopes their research helps “close the gap, to where people’s expectations are more in line with reality in more complicated situations.”

regions: San Francisco

About the Author