I think you're either missing or proving an implicit point.

I'll try to explain my reasoning:

- there are lots of ILI

- COVID-19 is a more severe ILI

- more severe ILI requires more hospital resources

- COVID-19 is an infectious disease, more disease, more infections

- depending on how much more severe COVID-19 is set to be, a conclusion can be drawn whether it was more or less likely that you had COVID-19 versus any other ILI based on increased hospital resource usage.

The important thing here is removing the perceived illness severity (i.e. worst flu ever, etc.) from the calculation, because it's highly subjective, as well of being a sample size of one.

The thing that usually happens is that someone writes that they had the worst flu ever, and soon somebody responds with something that they feel is relevant to the post, e.g. "I also had the worst flu ever", these comments often imply the possibility that their "worst flu" was their "COVID-19", which is exponentially less likely the closer to the start of the pandemic you go.

I'll call this phenomenon COVID-19+1 unless there's already a better name for it. And I won't claim immunity to it.

And by late February and early March when medical professionals were hunting for signs of the novel coronavirus and people were about to start being hospitalised and dying in large numbers, there'd be an enormous amount of hospital inpatients with respiratory symptoms to test. We know this wasn't the case until later...