Historically there are between ten and thirty deaths during instruction (amongst US Citizens, or in US waters) each year. The typical case involves a OW student who becomes separated from his or her instructor and is later found ... dead. When a good autopsy is done, the cause of death is typically an AGE. Lies on the medical form rarely have anything to do with this.
This seemed to me to be so totally incorrect that I first went through the last two DAN fatality reports, case by case, to see what they say about fatalities associated with instruction....(For this quote I omitted the statistics I found that contradict that.)
For what it is worth: I am the only one here on ScubaBoard that has ever put together a yearly diving fatality report. That was when I was part of the National Underwater Data Center. We only used fatalities that either involved US Citizens, worldwide, or Citizens of any country in US waters. DAN's numbers have always come out significantly lower than ours did, I chalk this up to a difference in methodology, the data for our reports was collected actively while DAN's is a rather passive approach. We had three people on the payroll pursuing leads, DAN has (as I understand it) none. There are some who feel that DAN grossly underestimates the number stemming from a lapdog to DEMA stance that it adopted during the Bennett years that has yet to be completely shed.
Do you have any links to published versions of these reports? Can you tell when it was you did this work?
All the NUADC reports are available on RUBICON.
I looked through Rubicon and read what I could find. I first of all learned that the early DAN reports are in fact the NUADC reports, and DAN took over their work some time ago. The NUADC-only reports are therefore relatively old. The newest NUADC document I could find was from 1995, and it was a comprehensive analysis of diving fatalities from 1970 to 1994.
McAniff, JJ (National Underwater Accident Data Center, Department of Physical Education, Health, and Recreation, College of Human Science and Services, University of Rhode Island, 1995)
You may be interested in its conclusions (these are direct quotes--emphasis added):
The population estimates presented by NUAOC are reasonable and conservative and fall between those presented by other studies.
Fatality rates per 100,000 have decreased considerably, (8.65 in 1976, ••• 2.67 to 3.44 in 1993).
Student deaths are at an all-time low, (1.50 per 100,000 in 1993).
With literally millions of dives, working scuba instructors as a group have suffered only three fatalities (two were heart related) in the 24 year period of this study.
It can be said that recreational scuba diving is a relatively safe activity with fatality rates very near ordinary swimming.
In summary, as far as I can see, the NUADC studies in which you participated indicated that diving in general was much safer in 1995 than in 1976, and that student deaths during instruction had dropped during that period of time to "an all time low."
I also found some studies by Edmonds on diving fatalities in Australia and New Zealand that used NUADC data. One comprehensive multi-year study found 5 deaths during scuba instruction, but it did not say what the reasons were.
The Rubicon search system was not particularly good at finding these reports for me, so I may have missed something that indicates the horrific rate of death in scuba instruction that you describe. I would be interested in reading the reports that indicate this, but I am going to need your help in finding them.