It could be widespread and less lethal AND we might still see the explosions we've seen due to the sheer number getting infected all at once. I think there's almost no question this is more widespread than we've officially measured. At this point the only debate is how widespread.

I've been looking at the influenza-like illness (ILI) reports produced by the CDC. The numbers this year are strange to say the least. I'm going full armchair epidemiologist here so take it with a grain of salt.

The ILI reports draw data from a nationwide surveillance network. Doctors report data on patients with a fever over about 100 and a cough. My understanding is that this isn't confirmed flu, just influenza-like. I think they use this data to infer true flu cases. Probably something we should start doing to infer coronavirus cases but that's another post.

If you look at the portion of cases under 25 and the portion over 25 historically you see a 60/40 split. That is, about 60 percent of reported cases are under age 25 and 40 percent over. So given that coronavirus seems to be disproportionately impacted the elderly, I looked at the portions this year. What you see is that the portions started to flip. The most recent week had it at 35/65. This has NEVER happened in any week in the last 5 years (I didn't look further back). 4 of the top 10 weeks in the last 5 years occured this year. It starts trending upward in early February and then takes off.

Point is, for it to be hitting the flu surveillence network could mean there were a lot of cases already in early February we just didn't realize it. We were not testing ANYONE back then. How would we know?