Story at a glance
- A free online program allows individuals to map out the effects COVID-19 will have on their local hospitals.
- Penn Medicine created the model to help public health officials and medical professionals predict and prepare for the needs of individual hospitals.
- The model has been updated to reflect recent information amid undercounting of admissions and diagnoses.
As the novel coronavirus continues to spread across the United States and a new study warns of a second wave, local hospitals are scrambling to prepare for more cases. But how do you plan for a situation you’ve never seen before?
Penn Medicine's Predictive Healthcare department and Code for Philly have created the COVID-19 Hospital Impact Model for Epidemics to help hospitals and public health officials predict their needs at capacity.
“It’s allowing hospital systems to assess their needs now, and if things don’t change, how they can get more resources, ICU beds, things like that,” Michael Becker, a senior data scientist at Penn Medicine, told Technical.ly.
The model provides estimates of the number of new daily cases of COVID-19, running totals of inpatient hospitalizations, ICU admissions and patients requiring ventilation using a standard epidemiological modeling technique: Susceptible, Infected, Recovered. The most important data point, the team said in a post, is the doubling time, or how rapidly the disease will spread.
READ MORE OF OUR BREAKING NEWS ABOUT CORONAVIRUS
"This value is particularly important because of the exponential nature of the spread of infectious diseases such as COVID-19. This is also why public health officials recommend measures like social distancing and hand washing: the more we can slow down the spread of COVID-19, the lower the peak demand on our health care system," reads the post.
While the default values reflect the COVID-19 outbreak in Philadelphia, anyone can access the model, input the number of known regional infections, hospitalized patients and the hospital's market share to represent their region. As of noon on March 27, there were 510 confirmed cases of COVID-19 in the Penn hospital system’s region, with the estimated number of infected individuals at 3,733.
Since making the model public on March 14, the team has revised it to reflect new admissions and census calculations, which had previously been undercounted. In a post, the team said they would continue to improve upon the model and welcomed input from the public.
SEE MORE ABOUT THE CORONAVIRUS PANDEMIC IN THE US