Is the answer to ignore anecdotes until long term data has been done? Sometimes science doesn't have the needed solution quickly enough to minimize damage. Being overly cautious in times of uncertainty has always been a human risk minimalization technique, might as well wield it, in moderation of course.