Deep Stops Increases DCS

[Ayisha]

Sigh. You guys really confuse me.

 

[rjack321]

Depends on the dive. For 300 gas were using 10/85. Not much nitrogen in that mix.

What gas do you use for a 200ft OC dive?

Yes but you'd still be ongassing N2 by switching to EAN40 (a lot actually). EAN 50 at least has a decently shallow EAD. IBCD is not a concern of mine, extra N2 is!

 

[PfcAJ]

Yes but you'd still be ongassing N2 by switching to EAN40 (a lot actually). EAN 50 at least has a decently shallow EAD. IBCD is not a concern of mine, extra N2 is!

You won't see me using 40% or 80%.

 

[Dr. Lecter]

85% He for an OC dive at less than 100m? I guess the answer is "it's in a cave"...but, I'm going to leave this one to you guys now.

 

[PfcAJ]

85% He for an OC dive at less than 100m? I guess the answer is "it's in a cave"...but, I'm going to leave this one to you guys now.

The gue standard gas is 12/65, fwiw.

 

[Slamfire]

Are you guys still doing extra O₂-window-time at the 70' gas switch?

 

[PfcAJ]

Are you guys still doing extra O₂-window-time at the 70' gas switch?

Depends who you ask and what day it is. Generally linear from 70-20.

 

[Dr Simon Mitchell]

What you are failing to understand is that these studies you refer to were DESIGNED to bend divers...if there was 0 incidence of DCS then the study would be worthless. But when you deliberately change protocol to induce DCS, or at least make it more likely, you invalidate the findings.

Hello Tom,

I have no desire to get into yet another debate about deep stops and current evidence. So, leaving that specific issue aside, I want to say that I don't think it is Michael who is struggling with comprehension around research methodology in diving medicine.

Let me ask you this. Assume you have two decompression profiles of exactly the same length from exactly the same depth and bottom time, but the profiles distribute their stops in very different ways (to avoid argument, forget about deep and shallow for the moment - just assume they are very different in some way). Now lets say you perform 200 dives run according to each of these 2 profiles under completely standardised conditions. Remember all the dives are exact same depth, exact same bottom time, exact same total decompression time, and add in exact same work at the bottom on all dives, exact same temperature on all dives, exact same thermal protection on all dives. The ONLY thing that is different is the decompression stop pattern. Now, let's assume that 10% of divers get DCS when they ascend according to one decompression stop pattern, and 5% of divers get DCS when they ascend according to the other. What do you conclude?

I would conclude that a decompression stop pattern that produces 20 cases of DCS in 200 dives is inferior to a decompression stop pattern that produces 10 cases of DCS in 200 dives.

It sounds like you would conclude that the experiment is invalid simply because the proportion of divers suffering DCS is higher than occurs in general technical diving. But what is your justification for ignoring the obvious difference in outcome between the two patterns of decompression? Surely there is a signal in the results that one pattern is inferior to the other? While I agree that it is ideal to calibrate studies to produce DCS rates reasonably close to those seen in the real world, from the point of view of judging which decompression pattern is superior do we really care whether the results are 1% vs 2 %; 2% vs 4%; 3% vs 6%; 5% vs 10% etc? The advantage of the studies with the higher rates is that they require less dives to produce a statistically significant result and are therefore "doable".

A second point I would raise with you is that you seem to believe that a 3.4% incidence of DCS is so high as to be un-interpretable in terms of "normal" technical diving. What data have you drawn on to form that conclusion? Most data describing DCS rates in technical diving comes from retrospective voluntary reporting studies, and these are confounded by reporting bias (divers might not want to talk about their DCS problems). In contrast, the DCS rates in the studies you are so critical of are collected prospectively through direct observation, interrogation, and examination of divers in the immediate post dive period by diving medicine experts. The dives in these experimental studies often involve more work than we would typically do, and always test the limits of whatever algorithm is employed (unlike general technical diving in which we frequently extend / pad stops etc). For these reasons the formal experiments will always result in the recording of higher rates than estimated for larger "general" populations from voluntary reporting data. Sorry to be long winded, but my point is that for all you know the DCS rates for profiles in general technical diving might be very similar to those in the studies if the data were gathered in the same way, the case definitions were the same, and the profiles were genuinely dived to the letter with no padding or other strategies that might increase safety. There seems little basis for dismissing these studies merely because the reported rates of DCS are higher than you perceive is normal for technical diving.

Simon Mitchell

 

[tomfcrist]

I have no desire to get into yet another debate about deep stops and current evidence. So, leaving that specific issue aside, I want to say that I don't think it is Michael who is struggling with comprehension around research methodology in diving medicine.

That's cool, lets disregard the deep stops altogether and talk about the way the studies are conducted.

Let me ask you this. Assume you have two decompression profiles of exactly the same length from exactly the same depth and bottom time, but the profiles distribute their stops in very different ways (to avoid argument, forget about deep and shallow for the moment - just assume they are very different in some way). Now lets say you perform 200 dives run according to each of these 2 profiles under completely standardised conditions. Remember all the dives are exact same depth, exact same bottom time, exact same total decompression time, and add in exact same work at the bottom on all dives, exact same temperature on all dives, exact same thermal protection on all dives. The ONLY thing that is different is the decompression stop pattern. Now, let's assume that 10% of divers get DCS when they ascend according to one decompression stop pattern, and 5% of divers get DCS when they ascend according to the other. What do you conclude?

I would conclude that a decompression stop pattern that produces 20 cases of DCS in 200 dives is inferior to a decompression stop pattern that produces 10 cases of DCS in 200 dives.

I would conclude the same thing, if the algorithms used hadn't been modified. Lets not forget that these profiles were not raw versions of either model...they were bastardized.

"The ideal characteristics of a test dive profile pair are (1) a large difference in estimated P​

DCS (deep stops dive profile − shallow stops dive profile) undereach probabilistic model; (2) the sign of this PDCS difference being opposite under theBVM(3) and NMRI98 models; and (3) all estimated PDCS being less than 7%. The first requirement was to enhance the likelihood of observing a significant difference in DCS​

incidence between the two profiles within a practical number of man-dives."

It sounds like you would conclude that the experiment is invalid simply because the proportion of divers suffering DCS is higher than occurs in general technical diving. But what is your justification for ignoring the obvious difference in outcome between the two patterns of decompression? Surely there is a signal in the results that one pattern is inferior to the other?

Simon....They literally said in the published draft that they deliberately increased the amount of DCS incidents...what more can be said?

While I agree that it is ideal to calibrate studies to produce DCS rates reasonably close to those seen in the real world, from the point of view of judging which decompression pattern is superior do we really care whether the results are 1% vs 2 %; 2% vs 4%; 3% vs 6%; 5% vs 10% etc? The advantage of the studies with the higher rates is that they require less dives to produce a statistically significant result and are therefore "doable".

When you add deco time all willy nilly to a given model, you are no longer following that model...so how can you invalidate the model that you are not even testing?

A second point I would raise with you is that you seem to believe that a 3.4% incidence of DCS is so high as to be un-interpretable in terms of "normal" technical diving. What data have you drawn on to form that conclusion?

What data do I have? Well I have the fact that I haven't been bent yet, nor have the majority of my dive buddies. Honestly, if the rate of DCS was really 3.4%, charter captains would quit running tech charters...especially in this litigious society, where you can be sued and lose over something you had nothing to do with.

Besides the point, the study deliberately induced DCS at a higher rate than normal...and admits to it.

Most data describing DCS rates in technical diving comes from retrospective voluntary reporting studies, and these are confounded by reporting bias (divers might not want to talk about their DCS problems). In contrast, the DCS rates in the studies you are so critical of are collected prospectively through direct observation, interrogation, and examination of divers in the immediate post dive period by diving medicine experts.

Ok, let me help you out with this one. These divers self reported their pre existings...what if they are too macho to admit that they were sore, or too afraid of being sent for a fit for duty to admit they went out drinking the night before? What were their hydration levels for each dive? Were all things created equal?

"Immediately before each experimental dive, diver-subjects reported any current injury or illness and their amounts of exercise and sleep, any alcohol consumed, and anymedications used in the previous 24 hours. On the bases of this self-report and a brief interview, a Diving Medical Officer either cleared or disqualified diver-subjects for​

participating in each experimental dive."

Additionally, there is nothing about a DMO that indicates "Expert" status. They are basically qualified Navy divers, that spend an additional 8 weeks getting lectures about dive medicine. Ask 100 Sailors about the quality of medical care at Portsmouth or if they perceive their DOC to actually know what the F they are doing.....that would be an interesting study.

The dives in these experimental studies often involve more work than we would typically do, and always test the limits of whatever algorithm is employed (unlike general technical diving in which we frequently extend / pad stops etc). For these reasons the formal experiments will always result in the recording of higher rates than estimated for larger "general" populations from voluntary reporting data.

Wrong again.....they padded the **** out of ALL of these stops...not just the ones on the Haldanian end.

Sorry to be long winded, but my point is that for all you know the DCS rates for profiles in general technical diving might be very similar to those in the studies if the data were gathered in the same way, the case definitions were the same, and the profiles were genuinely dived to the letter with no padding or other strategies that might increase safety. There seems little basis for dismissing these studies merely because the reported rates of DCS are higher than you perceive is normal for technical diving.

You obviously haven't read the study. Tell me how padded a 30/70 profile is compared to the deep stops version run in this study.

Stop depth- 80 70 60 50 40 30 20 10

NEDU DS- 12 17 15 18 23 17 72

ZHL/16Z 30/70- 2 2 3 4 7 10 19 35


Basically on the deep stops model, they added 44 minutes of dive time below 30ft while only adding 48 minutes on the haldanian end of the deco.

Lets take this off the deep stops issue and look at the us navy air deco tables.. This same 170/25min dive results in a total ascent time of 58 min. That is a far cry from 174min.

Kinda blows your Extra Padding theory to pieces. to get even close to the profile in the study, I would have to change to the GF's to 40/50.


Do I dispute the fact that staying at a deep stop for 17 minutes is bad? Hell no. Follow the damn profile.

Are deep stops effective if used properly? Absolutely, show me one piece of evidence to the contrary.

 

[LiteHedded]

That's cool, lets disregard the deep stops altogether and talk about the way the studies are conducted.



I would conclude the same thing, if the algorithms used hadn't been modified. Lets not forget that these profiles were not raw versions of either model...they were bastardized.

"The ideal characteristics of a test dive profile pair are (1) a large difference in estimated P​

DCS (deep stops dive profile − shallow stops dive profile) undereach probabilistic model; (2) the sign of this PDCS difference being opposite under theBVM(3) and NMRI98 models; and (3) all estimated PDCS being less than 7%. The first requirement was to enhance the likelihood of observing a significant difference in DCS​

incidence between the two profiles within a practical number of man-dives."




Simon....They literally said in the published draft that they deliberately increased the amount of DCS incidents...what more can be said?



When you add deco time all willy nilly to a given model, you are no longer following that model...so how can you invalidate the model that you are not even testing?




What data do I have? Well I have the fact that I haven't been bent yet, nor have the majority of my dive buddies. Honestly, if the rate of DCS was really 3.4%, charter captains would quit running tech charters...especially in this litigious society, where you can be sued and lose over something you had nothing to do with.

Besides the point, the study deliberately induced DCS at a higher rate than normal...and admits to it.



Ok, let me help you out with this one. These divers self reported their pre existings...what if they are too macho to admit that they were sore, or too afraid of being sent for a fit for duty to admit they went out drinking the night before? What were their hydration levels for each dive? Were all things created equal?

"Immediately before each experimental dive, diver-subjects reported any current injury or illness and their amounts of exercise and sleep, any alcohol consumed, and anymedications used in the previous 24 hours. On the bases of this self-report and a brief interview, a Diving Medical Officer either cleared or disqualified diver-subjects for​

participating in each experimental dive."

Additionally, there is nothing about a DMO that indicates "Expert" status. They are basically qualified Navy divers, that spend an additional 8 weeks getting lectures about dive medicine. Ask 100 Sailors about the quality of medical care at Portsmouth or if they perceive their DOC to actually know what the F they are doing.....that would be an interesting study.




Wrong again.....they padded the **** out of ALL of these stops...not just the ones on the Haldanian end.




You obviously haven't read the study. Tell me how padded a 30/70 profile is compared to the deep stops version run in this study.

Stop depth- 80 70 60 50 40 30 20 10

NEDU DS- 12 17 15 18 23 17 72

ZHL/16Z 30/70- 1 1 3 4 7 9 14 33


Basically on the deep stops model, they added 46 minutes of dive time below 30ft while only adding 56 minutes on the haldanian end of the deco.

Lets take this off the deep stops issue and look at the us navy air deco tables.. This same 170/25min dive results in a total ascent time of 58 min. That is a far cry from 174min.

Kinda blows your Extra Padding theory to pieces. to get even close to the profile in the study, I would have to change to the GF's to 40/50.


Do I dispute the fact that staying at a deep stop for 17 minutes is bad? Hell no. Follow the damn profile.

Are deep stops effective if used properly? Absolutely, show me one piece of evidence to the contrary.

i'm pretty sure Dr. Mitchell has read the study...