The outbreak of COVID-19 has by now been defined as a pandemic. Especially in Italy, it is causing serious problems for hospitals as COVID-19 patients are already crowding out the other patients. Doctors already have to make tough decisions. A large issue is the limited supply of critical care beds. In this analysis I combine the COVID-19 outbreak data from the Johns Hopkins University with information on the number of critical care beds from the following sources 1, 2 and 3
I have selected the eight countries with the highest percentage, and then also included Netherlands, since that is where I am from.
I have created a graph with on the x-axis the days after the first confirmed COVID-19 infection. The y-axis is the percentage of critical care beds from the entire country that should be required to care for the proportion of COVID-19 patients that may require critical care.
I have added a red solid line at 52%. Dutch data from Stichting Nice shows that 35% of critical care admissions is from planned surgery. That combined with a normal occupancy of 80% source, results in that at most 52% of the beds can be made available. Under the assumption that all of planned surgeries can be postponed, 52% is percentage of critical care beds could be available for COVID-19 patients.
I started with compiling the data about the number of critical beds per country. Unfortunately, this information was not easy to retrieve, so I copy pasted it into a CSV. I also manually fixed a couple of country name mismatches the with data compiled by the Johns Hopkins University.
The demand of critical care beds is determined not by the total confirmed cases, but only by the current number of "active" cases. This can be easily calculated from the time series by subtracting the number of recoveries and deaths from the number of confirmed cases. As of 24-03-2020 the recoveries are not available anymore, because there was no accurate source for that for most countries. (e.g. Netherlands had only 2 recovered cases, I am pretty sure that number must be higher.) The number of active cases is now defined as number of confirmed cases minus the number of reported deaths.
Luckily, not all COVID-19 patients require critical care. Estimates range from 5% (data from China) to 9-11% (data from Italy). So in the graph, we have the solid line on 8% (the average of 5 and 11) and a ribbon ranging from 5% to 11%.
I updated the number of critical care beds for the Netherlands based on a report from the NOS.