That's one data point.

Let's look at the data in aggregate instead, in order to make a decision, using the link I already gave. It's an extraordinarily low risk. I'm not even sure how quickly a hiker could bring a rifle into battery when they're just hiking along. And a pistol isn't going to stop a large animal. Whole thing seems unrealistic to me.

Bear spray, easily reachable on one's person, will be much more effective than a pistol. But data-wise, you have to consider that most near-misses probably aren't reported, that successfully defended attacks may not even be reported, and that people in certain environments will have a much greater risk than the average.