...Does this BMI argument have anything to do with the scuba death?
Yes and no. There are two issues at play here. One whether the BMI charts are a dependable measure whether someone is "overweight" or not, and the other is whether overweight folks are a "higher risk" for scuba mishaps based on that weight/height ratio from a BMI table, and to use a "free statistic" like "Last year, 75% of folks who died scuba diving are overweight" to back it up.
What do I mean by a "free statistic?" It is a statistic that is free of other facts. For example, here are two true statistics:
(1) In the last decade, 99% of people who died before they reached fifty ate carrots.
(2) In the last decade, 99% of people who lived to 100 ate carrots.
Can you draw an accurate conclusion about the relationship between eating carrots and longevity from either of those?
Here's a third, along the lines of the DAN statistic:
(3) In the last decade, 99% of the people who died scuba diving ate carrots.
You can no more draw any conclusion in the overweight/scuba death statistic than you can the carrot/scuba death stat unless you buttress it with more information.
In our society we have this obsession with fat, that fat must be bad in all cases and in all places, and that overweight people must *not* have the same mortality rate as skinny folks because, well, it just wouldn't be fair! Thin is in!
So when a pesky statistic like "3 out of 4 scuba divers are overweight" makes our "75% of scuba deaths are to overweight people" meaningless, we just ignore the first one and present the second one so it'll fit the mantra.
Across the population, obesity carries a higher mortality rate than "normal," and it's a bit of a surprise that scuba death stats don't follow the same ratio as sitting on the couch. There are reasons for that, and I can safely say that on the whole, all other things being equal, obese divers are at higher risk of a fatal debilitating event than divers who are not obese, but all other things are not equal, so don't expect a raw statistic like the DAN one to support that statement, because it just simply does not.
Rick