Experience levels & diving deaths

[bsee65]

Seems that a correllation with numbers of logged dives and/or number of dives in the X number of months prior to the accident would be useful as well. That wouldd show how current and din-practice the victims were. A guy who has been diving for ten years might have fewer dives and less experience than someone who has only been in the water a year.

Another stat might be related to location familiarity. Are divers more likely to have an issue in an unfamiliar environment?

How about purpose of the dive? I would also think that divers who are trying to shoot pictures or fish, or who have some other extra activities in their plan might be more apt to make a mistake.

Finally, the presence of a buddy or other divers in close proximity. What percentage of these victims were within sight and reach of a didving partner and still didn't come home? If the numbers show a correllation between separation and death, that would certainly be a useful statistic to drive that message home.

 

[davidbaraff]

Good discussion and thread.


TOTAL DEATHS FOR THE 3-YEAR PERIOD: 266

AGE OF THE DEAD DIVER:
10-17 - 1 (>1%)
18-29 - 28 (11%)
30-39 - 37 (14%)
40-49 - 76 (29%)
50-59 - 93 (35%)
60-69 - 19 (7%)
70+ - 6 (2%)

Those are interesting numbers, but if in fact, the total number of divers diving matches the above, it says you're equally likely to die no matter what age you are.

Said a different way, is the total diving population ages 50-59 more or less than 35%? If way more, that means agers 50-59 actually have less accidents per diver diving than some other groups. I guess I'd like to see the above figured normalized by the the segment of the diving population each age group represents.

Otherwise, it says you should wait till age 60 to start diving to be safe .

 

[Thalassamania]

Good discussion and thread.

Ken,
Nice workup and approach. The one thing that you're "overlooking" (the quotes are because I'm sure you've thought about it) are the medical cases that would not have been fatal had they occurred on land, or that would not have occurred if the victim had not been physically stressed by the dive. McAniff and I debated this issue ad infinitum and never really came to a good integrative solution.

 

[Saturation]

ianr is correct. That graph has an artifact error and is incorrectly grouped.

You cannot present different groupings with a percent fatality and equate them.

< 1 year is 1 year in width, it should better read 0-1

so is 1 should be 1-2 years, which is still 1 year wide

2 should be 2-3 years, ditto

but 3-5 is 3 years wide, presenting 3->4->5->6 years of diving

6-10 is 4 years wide, representing 6-7-8-9-10

and > 10 is undefined width.

Its simply cannot be lumped that way. The whole graph is meaningless.

I did bring that up to DAN some years ago, but the updated graph still makes the same error.

In the newer version, the graph is fine until 6 years. That grouping is incorrect again for the same reasons. If you divide 6-10 value of ~ 8 percent by 4, it might mean something, but that's an average for 4 year span, not a single value, so again its erroneous.

My problem with those graphs is that the time spans for each bar are not the same.

>10 years of diving covers a lot of ground. Anyone in diving for the long haul might be a diver for 30 years. Divide the height of the >10 bar by 30 and it will be tiny.

Lets look at it another way. Just split the graph into 2 bars:
1)Less than one years experience and
2)More than one years experience.

the second bar will be much higher . Therefore "proving" that more experience makes you less safe.

IMHO it is very shoddy reporting/statistics and DAN should know better.

 

[Ken Kurtis]

. . . medical cases that would not have been fatal had they occurred on land, or that would not have occurred if the victim had not been physically stressed by the dive.

You're correct. We've batted this back-and-forth. Need a lot more data to come up with any even remotely meaningful conclusions. And the real problem is demonstratng that something would (or wouldn't) have happened in a situation in which it never occured.

It's sort of like when I'm doing expert witness work and I hear opposing counsel say, "Well, if only X had been done, this never would have happened." You can't prove a negative.

That all being said, when I give the talk, the slide that follows the medical statistics asks these three questions:

1. Would this have happened on this date and time no matter what the deceased had been doing? (If so, it's just coincidence that they were diving at the time.)

2. Is this something that might have happened eventually, but the cumulative physiological stresses of accumulated diving accelerated the timeline? (In other words, is it sort of like playing Russian Roulette and if we stop playing - i.e. stop diving - soon enough we avoid the problem?)

3. Is this something that might never have been a problem, but the physiological stresses of diving caused it to be a problem when if the deceased had not taken up diving, the medical probably would never have occured. (Again, can't prove a begative but basically wonders if diving causes problems to manifest themsevles that otherwise would lay dormant.)

The questions are ones that hopefully will be more closely addressed by the medical community but there will be a variety of answers and conclusions and the reality is that there's probably no way to definitively determine accurate answers to any of the three questions.

Bottom line is that you have to do what you feel comfortable doing and let the chips fall where they may.

- Ken

 

[RJP]

< 1 year is 1 year in width, it should better read 0-1

so is 1 should be 1-2 years, which is still 1 year wide

2 should be 2-3 years, ditto

Of course the ranges also need to be mutually exclusive. You cannot have 0-1, 1-2, 2-3, because a person with 1yr or 2yr of diving would fall into multiple groups.

 

[Saturation]

Of course the ranges also need to be mutually exclusive. You cannot have 0-1, 1-2, 2-3, because a person with 1yr or 2yr of diving would fall into multiple groups.

Yes, absolutely right. What I meant was more 0-0.99 years, 1.0-1.99 years, 2 to 2.99 years etc. referencing the bolded item above.

 

[deco_martini]

The number of dive fatalities in the compiled Dan report is so small that there isn't a great statistical significance in the reading.

If you know how to crunch numbers, then you know that looking for meaning in the experience chart is akin to reading tea leaves.

It is interesting. But ultimately, it is not definitive or enough evidence of anything more than simple correlation.

 

[rstofer]

I'm not even sure it shows correlation. There are a certain number of incidents in an age group. OK, how many dives are made by people in that age group? I would expect 10 times as many young people to be diving as people > 60. Maybe 100 times as many. I wouldn't be surprised if, as a group, they make 1000 times as many dives as the old timers.

When I walk up and down the beach, I am the old man (63+). All around are kids in their 20's and 30's. Well enough established to be able to afford to dive but nowhere near getting Social Security to pay for it.

Richard

 

[Zosimus]

Thought this post was interesting due to my instructor telling us some stats in class. I do not know what his source was. He mentioned that most deaths to divers were either with-in the first year of certification or after 10 years (which i see that some of those graphs point out). We didn't go in to much detail, but the general reasoning was mainly due to new divers getting certified and flying off to a high-end dive location and going way past their limits. Or the vet divers that have been diving over 10 years, but have not been diving actively for some years feel "invincible" and may not realize that some of their limits may have changed.