But this sort of trouble with contamination occurs with other elements, giving older ages to recent things?
In theory, yes. In practice, not very often if ever.
"Simple" methods such as carbon dating and K-Ar dating are susceptible to contamination errors. This is known, as are methods that are nearly guaranteed to avoid contamination.
"Age-diagnostic" methods such as isochrons, Ar-Ar, and U-Pb concordia-discordia are extremely unlikely to yield an older-than-actual age. These methods either produce a date
and an assessment of the reliability of that date, or they don't produce a date at all. They are by far the most widely-used methods.
Then you start dating things by different methods and getting the same age from each method. And you compare results to non-radioisotope methods, and again they agree.
It is not possible that
all radiometric age determinations are tremendously wrong, as the YECs would have us believe. That would be like you winning the lottery grand prize a billion times in a row. Neither is going to happen.
See
Correlation Among Various Radiometric Ages and
Radioisotope dating links and information.