The scattered anecdotes are scary: brain fog that's bad enough to prevent you from doing any mechanical or intellectual labour. The initial data was also scary, but I never saw anything with a particularly convincing survey size.

What initial data?

Near the beginning of the pandemics I've seen percentages of confirmed infections presenting symptoms X weeks after diagnosis. It seemed to not die off. But it both topped at a not too long interval and didn't see that many cases yet.

The US Census Bureau basically said as much recently. Their findings from October stated that 3.7 million people in the US were currently unemployed because they were "caring for someone or sick myself with coronavirus symptoms" [1]

[1] https://www2.census.gov/programs-surveys/demo/tables/hhp/202...

Edit: to clarify, a lot of those 3.7 million people are probably suffering from acute Covid, but if you look back at the older survey data, the number of people not working due to "coronavirus symptoms" still seems to remain pretty high even when cases are low.

It unfortunately does not describe how long the surveyed people were unemployed for which would be interesting to know. Thanks for the relevant link & data.

That data is not hard to find:

https://www.cdc.gov/mmwr/volumes/70/wr/mm7036a1.htm

https://journals.plos.org/plosmedicine/article?id=10.1371/jo...

https://www.beckershospitalreview.com/patient-safety-outcome...

Based on a quick scan, none of these studies back up the OP's claim. All these studies show is that long covid exists, but nobody disputes that.

Your scan was far too quick, in that case. You might want to scan again. :)