You are absolutely right from a statistical standpoint. (And maybe I could have phrased it better.) We know the numerator, but we don't know the denominator.
The big problem here is that we, as an industry, don't track close calls. If you run out of air and make it successfully to the surface, that's not reported nor entered in any database.
However, there's another way to look at it. There ought to be some direct relationship between how often (%) something happens in the genral population vs. how often (%) it produces whatever neagtive effect we're monitoring.
My assumption, based on the number of dives that I personally monitor over the course of a year, is that out-of-air is a very rare occurence, as a % of total dives. (I have no idea what the actual number is but I certainly hear more stories of "Diver ran out of air and died" than I do of "Diver ran out of air and survived.")
And the simple point is that if something that is fairly rare (OOA) as an occuraence on a dive, can priduce a very high (41%) proportion of the accident triggers, then it says a couple of things (IMHO) about running out of air:
1. It's likely far more dangerous and likely to be life-threatening that we currently teach.
. . . and . . .
2. The "options" that we teach - alternate air, pony, octo ascent, buddy-breathing, free ascent - don't work (either because they're forgotten, not practiced, or panic prevents their use).
Bottom line is that I personall ythink OOA is far more danegrous that we, as an industry, preach/teach and it's high time we changed the culture and mindset of how we approach this because OOA incidents are easily and emminently preventable.
- Ken