I published a post in French on the coronavirus yesterday that, ironically, went completely viral. I don’t have time to translate the whole thing, and there are already many great resources on the topic available in English, but if you want to read it anyway I’ve been told by several people that Google Translate does a pretty good job at translating it. However, there is one calculation I did in that post that I haven’t seen anywhere else and I actually think it’s the most interesting part of the post, because it gives a sense of how much worse than the seasonal flu COVID-19 is and the impact it could have on health care systems around the world. As many people have pointed out, the problem with COVID-19, in addition to a relatively high fatality rate even when people have access to all the medical care they need, is that it seems to require that people be admitted in ICU far more often than in the case of the seasonal flu. So I was wondering how much more often exactly and I set out to find out. I had a bit more time to do research for this post, so I think the estimate is more reliable, but it’s very similar to what I found in my previous post in French.
According to the CDC, between 2010-2011 and 2016-2017 (the estimates for more recent flu seasons are still preliminary, so I have ignored them), between 1% and 2% of symptomatic flu patients were hospitalized.
But this doesn’t tell us how many of them developed symptoms that were acute enough to require ICU admission at some point.
Judging from the quick search I did, it’s surprisingly difficult to find reliable data on the proportion of flu patients who require ICU admission, but the spreadsheet-based model developed by the CDC to help hospital administrators forecast the surge in demand for hospital-based services during a flu pandemic assumes that 15% of people who are hospitalized for the flu will require ICU admission, so I’m going to use this estimate. What this means is that, among people who are infected by the seasonal flu and develop symptoms, between 0.15% and 0.3% require ICU admission. In contrast, according to the WHO report on the coronavirus epidemic in China, 13.8% of those infected developed severe disease and 6.1% of them were critical, which means they developed respiratory failure, septic shock and/or multiple organ failure. The report doesn’t actually say that, but given the description of the symptoms for patients who developed critical forms of the disease, it seems clear to me that people in that condition had to be admitted to intensive care.
Thanks to the Italian government, which uploads on GitHub and regularly updates the data on the epidemic since February 24, we have another source of information on this question. According to those data, during this period, the proportion of people currently in ICU among those who were diagnosed as infected with the coronavirus and who had neither died nor recovered yet fluctuated between 7.7% and 11.8 %.
This is even higher than the estime from China, but since we’re still early in the pandemic and the estimates might differ between countries for a variety of reasons, I think it’s difficult to determine which estimates are more reliable. So let’s be optimistic and assume that only 5% of people infected by COVID-19 that develop symptoms require ICU admission at some point.
This would mean that the proportion of symptomatic COVID-19 patients who require ICU admission at some point is between 16 to 33 times higher than the proportion of symptomatic flu patients who do. As I just noted, this estimate is very rough because we don’t have very good data on COVID-19 yet, but it’s based on relatively optimistic assumptions and, even if we assume that the real figure is only half as high because many symptomatic cases were missed and that a much larger proportion of people infected by COVID-19 are asymptomatic, it’s clear that if it’s allowed to spread as much as the seasonal flu it will impose a burden on the health care system that is vastly greater. This is all the more true that, from what I understand, it seems that COVID-19 patients who develop acute symptoms need to remain in ICU for a longer period of time than seasonal flu patients who end up in ICU. I was really shocked when I first did this calculation, but no matter how I look at it, I reach the same conclusion. I think it really makes vivid how much worse than the flu COVID-19 is and I hope that it will help people who still think it’s not so bad that actually it’s really bad.