Could it also be that they are counting flu as something else?

The article raises that question, but it does not reach this conclusion. In particular, it states:

> [In the six countries with robust data] the total number of influenza tests has fallen by just 20%, while the share of tests that have come up positive has plummeted to record lows.

It appears to be a legitimate effect.