COVID-19 Prediction Model - April 9, 2020

View the COVID-19 Prediction Model Graphic

Michael Roberts, M.D., is a hospitalist at EAMC and is the hospital's current Chief of Staff. Using Penn Medicine's COVID-19 Hospital Impact Model for Epidemics (CHIME) tool, Dr. Roberts is tracking COVID-19 to see how social distancing plays a role in the length of the virus in in our hospital and community. This model is being updated each week.

Q&A with Michael Roberts, M.D.

How did EAMC come up with these numbers?

Using the Penn Medicine COVID-19 Hospital Impact Model for Epidemics (CHIME) tool, we input our current assumption data (population, date of first COVID patient hospitalized at EAMC, current number of COVID patients and a couple of other required data). The tool then provides the results for when it predicts a hospital and its community will peak for COVID-19 hospitalizations and then begin to see those numbers decline.

How long should it take after we peak until we are able to return to normal activities?

That’s a very difficult question to answer. Just because we reach peak activity does not mean we are out of the woods. There will still be many sick patients in the hospital who will take time to recover. There is risk of a second wave of infections when people return to their usual patterns of movement and interaction. The timing of lifting a “stay at home” order, and the process by which that is done, is a decision that will require coordination between healthcare and civic leaders.

Why does it show April 9, 2020 as the peak date for 100% social distancing?

Because it assumes you have been—and will continue—with 100% social distancing. If that were the case, the current date (as this is being written) would be considered as the peak date of disease activity. This would be the worst it will be and it would be a downward trend until the virus reaches more manageable levels in the community.

Last week, the peak date for 75% social distancing was listed as June 15, 2020. How did it change to April 10 in one week?

As time passes, we are able to input more of our actual local data, rather than using default assumptions in calculations. This makes our predictions more accurate. Additionally, as we move further from the date of our first documented cases in the community, we can see the effects of social distancing measures that have already been in place. In the end, however, these are predictions and are not set in stone. The dates will change week-to-week based on our local activity and updates to the prediction tool. The predictions should become more accurate over time.

The peak date for 100% social distancing and 75% social distancing are just one day apart. Why?

100% means no contact outside your home. That is very difficult to accomplish. If we truly achieved 100% social distancing then, based on the timing of the arrival of COVID-19 in our community, we may have already reached our peak. At 75% social distancing, we could be very close to our peak. Maybe not April 10, but perhaps pretty close.

Why are 50% social distancing, as well as 25% and 0%, considered unsustainable?

EAMC has 314 acute care beds, and some of them are specifically for childbirth services and psychiatry. At 50% social distancing, we would have 577 patients. That number alone is proof that we would not be able to care for that many people at once. Even with 290 patients in the ICU and 268 of them on life support, our community could not handle this volume because we simply don’t have the critical care beds, the critical care staff or the number of ventilators needed to care for that many patients.

Are the efforts in our community making a difference?

In one word, yes. HOWEVER, it is not time to take our foot off the gas pedal as it’s very possible to have a second wave of cases and then have to start the whole process over again. We greatly appreciate our community for taking this public health issue seriously. We would also like to thank our area leaders and first responders who have helped encourage citizens to stay at home and not gather.