How Beth Israel Deaconess Medical Center predicted COVID-19 surges 5 days before national models

Many health systems have built tools and dashboards to track the spread of COVID-19 and project where new outbreaks will occur.

Harvard Business Review featured Boston-based Beth Israel Deaconess Medical Center's efforts to predict where their resources will be needed next during the pandemic. The health system appointed a research group within the Center for Healthcare Delivery Science to apply epidemiology, machine learning and causal inference to forecast where COVID-19 would surge next.

The 670-bed academic medical center began to formulate its response in February and relied on national models to predict how the virus would spread. However, national models didn't consider local hospital and community decision-making or socioeconomic factors that could put communities more or less at risk. One example outlined in the report is that Beth Israel decided to admit patients with COVID-19 instead of sending them home for virtual monitoring: "Thus we needed a dynamic hyper-local model," wrote the report authors.

Beth Israel's research team created a hyper-local alert system, integrating a preliminary Susceptible, Infected and Recovered people model into the hospital's incident command structure. The researchers also relied on machine learning to take real-time data from the EHR to examine disease characteristics like incubation time, infection period and transmissibility.

The system was able to predict peaks and declines five days before the national models.

"Had leadership relied on national models, they would have expected a sharper peak and decline, and a peak two weeks earlier than the actual peak," the article authors wrote. "Our modeling affected key decisions, including the need to bolster personal protective equipment supplies; to gauge the necessity of even urgent procedures and postpone them if necessary in order to assure we had the capacity to absorb the peak; and to establish staffing schedules that continued farther into the future than originally planned."

However, as Massachusetts moves to resume more normal business, the hospital needed to take the increased interactions into considerations for its projections. It developed a risk index for local businesses as they reopened.

More articles on healthcare:
Texas Medical Center hit 100% ICU bed occupancy, then didn't report data for 3 days
University of Texas, MD Anderson Cancer Center launch cancer data initiative
CommonSpirit develops predictive models for next COVID-19 surge: 5 details

© Copyright ASC COMMUNICATIONS 2020. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 

Featured Content

Featured Webinars

Featured Whitepapers