[DanBC]

You're counting flu and covid-19 deaths differently.

For flu you're using sophisticated statistical modelling of excess death combined with community surveillance of use of primary and secondary care. This will tend to overcount.

For covid-19 you're using "died after testing positive for covid, or with covid mentioned as the cause on the death certificate". This will tend to undercount (even when taking into account the slight overcount involved in people who die from other things after a +ve test).