Is there data on how close to NDL undeserved hits occur?

[dmaziuk]

Understanding M-Values by Erik C

Understanding M-values By Erik C. Baker, P.E. In conjunction with an array of hypothetical tissue compartments, gas loading calculations and M-values

docslib.org

 

[scubadada]

@boulderjohn @dmaziuk

DAN Europe article from 2017, a interesting read. PDF can be downloaded from the link

Frontiers | Dive Risk Factors, Gas Bubble Formation, and Decompression Illness in Recreational SCUBA Diving: Analysis of DAN Europe DSL Data Base

Introduction: The popularity of SCUBA diving is steadily increasing together with the number of dives and correlated diseases per year. The rules that govern...

www.frontiersin.org

As there have been no additional comments regarding this publication, I will share a few thoughts I had after reading it that I jotted down in the margins of the paper.

The study includes 2,629 divers and 39, 099 dives. They had 320 cases of DCS, for a rate of 0.82% or 82 cases/10,000 dives. This is significantly higher than the rate of somewhere around 1-3 cases/10,000 dives in recreational dives usually cited. There is no discussion of this rate. (see comments in posts 33, 35 and 36 and 38)

Nearly 74% (236/320) of the cases of DCS occurred in divers with a GF of 0.7-0.9. Unfortunately, there are no denominators for the GFs. I would not be surprised if these GFs were also the most common in all of the dives.

As per @dmaziuk it was not surprising that most cases of DCS implicated the medium compartments.

There is interesting information here, as often, to be taken with a grain of salt.

 

[dmaziuk]

I remember now: it seemed to be the case having two unrelated studies that, while informative, would not merit individual publications. So they mashed them together into one confusing article: the analysis of 40K NDL dives appears to be completely unrelated to the analysis of the 320 DCS cases; as I recall there was no indication that there's any intersection in the 2 datasets in the paper.

So this is wrong:

The study includes 2,629 divers and 39, 099 dives. They had 320 cases of DCS, for a rate of 0.82% or 82 cases/10,000 dives. This is significantly higher than the rate of somewhere around 1-3 cases/10,000 dives in recreational dives usually cited. There is no discussion of this rate.

They had 320 cases of DCS in the database of 320 DCS incidents (i.e. 100%), and zero reported cases of DCS (i.e. 0%) in the 40K dives, is my reading of their paper.

 

[scubadada]

Understanding M-Values by Erik C

Understanding M-values By Erik C. Baker, P.E. In conjunction with an array of hypothetical tissue compartments, gas loading calculations and M-values

docslib.org

This is a great article.

A little more about M-values.

Many already know about Buhlmann ZH-L16A, B, and C. ZH-L16A was thought too aggressive, so the Buhlmann a coefficients were reduced for compartments 6, 7, 8, and 13. This reduced the Mo, which is the y-intercept, the inert gas pressure at surfacing. The a coefficients were further reduced in ZHL-16C in compartments 5, 6, 7, 8, 9, 10, 11, 12, and 14, resulting in reduction/further reduction of the Mo for those compartments. The slope of the M-value lines was not changed, so this resulted in shifting the M-value line to the right, more conservative. ZH-L16C is most commonly used in dive computers. The GF high and GF low values in your dive computers increase conservatism further, by shifting the intersection with the M-value line further to the right.

 

[scubadada]

I remember now: it seemed to be the case having two unrelated studies that, while informative, would not merit individual publications. So they mashed them together into one confusing article: the analysis of 40K NDL dives appears to be completely unrelated to the analysis of the 320 DCS cases; as I recall there was no indication that there's any intersection in the 2 datasets in the paper.

So this is wrong:



They had 320 cases of DCS in the database of 320 DCS incidents (i.e. 100%), and zero reported cases of DCS (i.e. 0%) in the 40K dives, is my reading of their paper.

Thanks, I did not get it that way, I would not have expected that. I'll have to go back and read it over again. Seems like that would have been very obviously stated in the methods, maybe it was.

Edit, you may be right, very poorly written, I will go back to it when I can tolerate reading it again. Perhaps that's why it is published in Frontiers in Psychology. I have emailed the corresponding author, for what it's worth.

 

[lowwall]

I remember now: it seemed to be the case having two unrelated studies that, while informative, would not merit individual publications. So they mashed them together into one confusing article: the analysis of 40K NDL dives appears to be completely unrelated to the analysis of the 320 DCS cases; as I recall there was no indication that there's any intersection in the 2 datasets in the paper.

They had 320 cases of DCS in the database of 320 DCS incidents (i.e. 100%), and zero reported cases of DCS (i.e. 0%) in the 40K dives, is my reading of their paper.

My reading as well. Moreover, 39k of the 40k dives weren't used for anything other than a description of the dives. They did Doppler studies on 970 of the dives and the largest section of the paper was looking at bubble grades versus various factors related to those dives and divers.

And then there was a section summarizing the completely separate set of dives that made up their DCS database.

All in all, it looks like they'd hoped to get some DCS cases as part of the big study, but when they didn't, they scrambled to come up with something they could get published somewhere. I'm going to assume it's no coincidence that it appeared in Frontiers in Psychology, a pay to play journal in an unrelated field, instead of one of the normal landing spots for decompression research.

To me the most interesting part of this was the lack of DCS cases amongst the 40k dives. I will hypothesize that this is attributable in large part to self-selection and observer effects. Basically, that people volunteering for such a study and aware that they are going to be reporting all aspects of a dive are going to be much more careful and conservative in that dive than the average diver.

 

[scubadada]

...The Saul statistics listed above appear not to make a distinction between Type 1 and Type 2.

Hi @FishWatcher747

I heard back from Saul. The probabilities of DCS do not differentiate between Type 1 and Type 2 DCS.

 

[scubadada]

As there have been no additional comments regarding this publication, I will share a few thoughts I had after reading it that I jotted down in the margins of the paper.

The study includes 2,629 divers and 39, 099 dives. They had 320 cases of DCS, for a rate of 0.82% or 82 cases/10,000 dives. This is significantly higher than the rate of somewhere around 1-3 cases/10,000 dives in recreational dives usually cited. There is no discussion of this rate. (see comments in posts 33, 35 and 36)

Nearly 74% (236/320) of the cases of DCS occurred in divers with a GF of 0.7-0.9. Unfortunately, there are no denominators for the GFs. I would not be surprised if these GFs were also the most common in all of the dives.

As per @dmaziuk it was not surprising that most cases of DCS implicated the medium compartments.

There is interesting information here, as often, to be taken with a grain of salt.

I remember now: it seemed to be the case having two unrelated studies that, while informative, would not merit individual publications. So they mashed them together into one confusing article: the analysis of 40K NDL dives appears to be completely unrelated to the analysis of the 320 DCS cases; as I recall there was no indication that there's any intersection in the 2 datasets in the paper.

So this is wrong:



They had 320 cases of DCS in the database of 320 DCS incidents (i.e. 100%), and zero reported cases of DCS (i.e. 0%) in the 40K dives, is my reading of their paper.

My reading as well. Moreover, 39k of the 40k dives weren't used for anything other than a description of the dives. They did Doppler studies on 970 of the dives and the largest section of the paper was looking at bubble grades versus various factors related to those dives and divers.

And then there was a section summarizing the completely separate set of dives that made up their DCS database.

All in all, it looks like they'd hoped to get some DCS cases as part of the big study, but when they didn't, they scrambled to come up with something they could get published somewhere. I'm going to assume it's no coincidence that it appeared in Frontiers in Psychology, a pay to play journal in an unrelated field, instead of one of the normal landing spots for decompression research.

To me the most interesting part of this was the lack of DCS cases amongst the 40k dives. I will hypothesize that this is attributable in large part to self-selection and observer effects. Basically, that people volunteering for such a study and aware that they are going to be reporting all aspects of a dive are going to be much more careful and conservative in that dive than the average diver.

@dmaziuk and @lowwall are correct. This was 2 separate analyses reported in one publication.

A subset of the larger group was analyzed for bubble production and was largely unremarkable. These subjects may have come from DAN Project Dive Exploration Project Dive Exploration (PDE)

A separate database of DCS cases was analyzed and was also largely unremarkable. The most common GFs of 0.7-0.9 are probably the most common surfacing GFs in recreational divers. No denominators are available. It has already been stated that it is not a surprise that medium tissue compartments were largely responsible for DCS.

This is not the well designed and executed trial you might want.

Thank you to the two posters for spotting my errors, I am usually a more careful reader

 

[inquis]

The slope of the M-value lines was not changed, so this resulted in shifting the M-value line to the right, more conservative.

@scubadada it's interesting you say the M-value line shifts right going to ZHL-C. I picture it shifting down.

 

[scubadada]

@scubadada it's interesting you say the M-value line shifts right going to ZHL-C. I picture it shifting down.

The intercept, Mo, shifts down, as the slope does not change, the line shifts right. I'll share that beer with you.