Amongst rock climbers there is a naturally occurring population of free climbers, and it is possible to note the type and severity of injuries between different types of rock climbers (although i'm guessing your n is going to be low).
The safety of parachutes themselves is an interesting question because they're not used as a preventative mechanism, I think that's the primary difference between them and climbing ropes/harnesses/pitons.
Edit: I just noticed that I didn't address the randomized component. I'd still argue that evidence based medicine can actually include arguments that aren't randomized trials. That said, randomized trials are still one of the strongest sort of argument that can be made. As noted above, seatbelts are another sort of device that I don't think we need randomized trials to test the efficacy of.
p.s. Just finished (and enjoyed) the Apocalypse Codex. The Laundry Files is one of my favorite series of books as someone who's always believed that empiricism and existentialism are reconcilable. Please write more :)