Yes, but not in the way you're implying. Safety measures don't exist in a vacuum, they exist in a whole system of other similar safety measures. In aggregate, this system of safety measures does slow science down (or anything else). One might argue that it's worth it to slow down output for the sake of safety, but I don't think one can reasonably argue that output doesn't slow down.

It is possible to construct a system of safety measures so absolute that almost no work can get done, and it is possible (in a sufficiently dangerous field) to be so reckless that injuries to and deaths of would-be contributors stymie progress. Even ignoring the value of avoiding death and injury for its own sake, it is likely that optimal productivity lies somewhere in between.