Clarification on Expression indexes

Greetings all.

The following page:
https://www.postgresql.org/docs/11/indexes-expressional.html

States the following:

Index expressions are relatively expensive to maintain, because the derived
> expression(s) must be computed for each row upon insertion and whenever it
> is updated
>

I'd like to get an idea on "relatively expensive". For instance, compared
to a partial index, which is split on one or more boolean values. As
opposed to making a function for this that serves the same identical
calculation.

Let's say that over the lifetime of a row, it rarely gets updated more than
2000 times, and the majority of this is right after creation?

Is this a concern?

Regards,
Koen De Groote