My quick 2-min synopsis is that is looks well-constructed and well analysed.

There is an effect of being a cyclist (I didn't look in detail to determin how this was determined) but the number of cyclists is small.

So small in fact, that I would assume that the cyclists are do other things lifestyle-wise that keep CVD low and overall healthiness high. The authors do NOT dwell on being a cyclist meaning anything ... they keep it observative.

I haven't watched the YT video as I don't go for so much popsci stuff but the title: "Study Finds Biking to Work HALVES Risk of Early Death."

Um, well, I doubt the authors would write that.

Maybe they'd say "behaviours that include being an active cyclist, reduced early mortality (death.)" They wouldn't stress any causation like the YT video. I would assume that nearly all cyclists are outliers in several lifestyle behaviours.

That's why I don't like these kinds of studies myself, but can understand the value when not over-inflated.