A new method helps hospitals prepare to promptly treat the most severely injured patients right after a major earthquake.
By, Elizabeth Goldbaum, Science Writer (@EFGoldbaum)
Citation: Goldbaum, E, 2020, Hospitals can coordinate to save lives after an earthquake, Temblor, http://doi.org/10.32858/temblor.140
When Luis Ceferino was a teenager in 2007, a magnitude-8.0 earthquake hit the central coast of Peru. During the aftermath, “we saw how hospitals lost all ability to function,” Ceferino said.
Over a decade later, as a doctoral student in civil engineering at Stanford University, Ceferino published a scientific paper aimed at helping hospitals not only function after a major earthquake, but effectively coordinate with each other to ensure the most severely affected patients receive timely treatment. He and his collaborators found that if hospitals are highly coordinated, they can cut patients’ wait times by more than half.
The team created a method that helps hospitals maximize coordination. It considers how a major earthquake will cause injuries and a subsequent surge in demand on the buildings, ambulances and other aspects of the healthcare system, as well as each aspect’s capacity to withstand major damage and continue to function.
The team’s method recommends that city officials first use a statistical risk analysis model to estimate where deaths are most likely to occur. Next, they need to apply building-specific performance assessment techniques to estimate the extent of damage to hospitals. Finally, they need to map out the best routes between hospitals so that patients are able to get timely treatment.
“One of the key contributions of our paper is that we’re looking at what happens on a regional scale, rather than focusing on what happens to a single hospital,” Ceferino said. The paper highlights how cities and local governments need to create a framework for hospitals to coordinate their emergency response after a major earthquake, he said.
A holistic approach to emergency management
In 2010, two to three thousand patients had to be transferred between hospitals after a magnitude-8.8 earthquake in Chile. In 1999, the magnitude-7.6 İzmit earthquake in Turkey caused 50,000 injuries and affected 10 major hospitals, forcing those hospitals to relocate their patients. Severe earthquakes can not only injure thousands of people, they can also damage the infrastructure designed to take care of the injured.
If hospitals do not coordinate with each other by sharing information and resources, they are liable to make potentially fatal decisions. For instance, after the 1994 magnitude-6.7 Northridge earthquake in Southern California, two hospitals transferred their patients to a nonfunctioning hospital, prompting yet another transfer.
In studying past earthquake responses and simulating potential future scenarios in Lima, Peru, the team learned that there is a spatial mismatch between where people live and where hospitals are, Ceferino said. While many people live on the periphery of the city in sprawling suburban neighborhoods, hospital resources are generally in the center of the city.
This “islanding” effect is not unique to Lima. In Seattle, Washington, for instance, some neighborhoods are located far from hospitals that are primarily concentrated in the city’s center. Residents might not be able to reach emergency medical services after a major earthquake, especially if there were damage to transportation infrastructure, Nicole Errett, an assistant professor researching disaster and public health policy at the University of Washington in Seattle who was not affiliated with this study, said.
“This is one of the first efforts to really think about this outside the scope of a single hospital,” Errett said. If an entire community is going to be hit by an event, local decisionmakers need to look at the entire hospital system and plan together to form a response that will ensure the community gets the healthcare it needs, Errett said.
An effective plan of action
To create a method for a city’s hospital network to effectively respond to earthquakes, the team
simulated four alternative emergency response plans to treat patients in Lima, a city with close to 10 million people, after a magnitude-8.0 earthquake at night, when most people would be inside vulnerable residential buildings.
The team included city-wide data on the seismic vulnerability of over 1.5 million buildings in Lima, including over 700 buildings on 41 public hospital campuses. The team used a probabilistic model that considers high-resolution building seismic vulnerability data, population distribution and soil conditions to estimate that, on average, nearly 4,700 people would require surgical procedures in operating rooms for injuries like punctured organs and compound bone fractures after the earthquake. The team also projected that 87 of the 182 hospital operating rooms would be functional based on prior experiences as well as the World Health Organization’s Hospital Safety Index of each of Lima’s 41 healthcare campuses.
The four simulated emergency response plans each have a different level of coordination among hospitals. In the first plan, hospitals would use their own ambulances to transfer their patients to the closest functioning hospital only if all of their operating rooms were nonfunctional. The average waiting time to receive treatment under this plan would range from 29 to 64 days. In the second plan, everything stays the same, but hospitals would send their patients to the hospital with the largest number of functional operating rooms, lowering the average wait time range to between 24 and 44 days.
The third and fourth plans move to higher levels of coordination. In the third plan, hospitals would share ambulances, lowering wait times to between 11 and 20 days. In the fourth plan, hospitals would share ambulances and deploy 15 additional mobile operating rooms to treat areas in high demand, lowering wait times to between eight and 16 days.
“We studied these four scenarios as extreme cases,” Ceferino said. Although they represent the best- and worst-case scenarios, in reality, the emergency response will likely be something in between, he said.
For instance, after the 2010 earthquake in Chile, hospitals deployed operating rooms close to existing hospitals to leverage additional resources like personnel, power generators and backup water, similar to what is outlined in the fourth, best-case, scenario. However, in its simulations for Lima, the team found that strategically locating additional operating rooms in the periphery of a city allows a hospital to treat more patients.
A first step of many
“This paper is foundational for a lot of work that we need to do in the future,” Errett said. The more information we have, the more we can prepare.
In the future, Ceferino is planning to see how his team’s method would work in other urban layouts, including for cities in the United States. He’s planning to expand his simulations to include earthquakes of lower magnitude and less severe impacts at different times of the day. He’s also interested in adding extra layers to see how patient wait times correlate with socioeconomic variables.
Although this study is an excellent first step to understand the system, we need to keep traveling down this path to improve it, Errett said.
“It takes extra effort to go outside your environment and work with people [at other hospitals] you’re normally in competition with,” Errett said. However, Ceferino’s paper makes clear the benefit of coordination. “It gives you an incentive to go out and do the hard work,” Errett said.
Ceferino, L., Mitrani-Reiser, J., Kiremidjian, A., Deierlein, G., & Bambarén, C. (2020). Effective plans for hospital system response to earthquake emergencies. Nature communications, 11(1), 1-12. https://doi.org/10.1038/s41467-020-18072-w