Scientific texts – only for academics?
“There’s supposed to be a study that proves that…” – this is how sentences sometimes begin among divers, with which a rather crude assertion is to be presented as correct and scientifically sound. If you ask what study it is, most people only give vague information – and if you search, you can’t find any study far and wide that even remotely supports the claim.
If an assertion is too simple, too simplistic, too clear – then it is usually a rumor, the end of a broken telephone message, and a lot of misunderstanding. Why? Clear, simple findings are rare; usually, a study focuses on a very small, clearly defined area of a topic and never provides a simple solution to complex questions.
Research is being carried out on diving and decompression in many places, and yet this rarely results in clear guidelines and simple rules of thumb. If you want to delve deeper into a specific question, you sometimes won’t get much further with the simple articles in specialist magazines or, for example, the Alert Diver. However, you can often find references to sources that you can look at yourself.
In most cases, the publications are in English. Automatic translators can now help with this, and you can also ask ChatGPT to summarize a study. But be careful – AI seems to be prone to jumping to conclusions, like many people unfortunately. It simplifies things so much that it becomes wrong, or it invents things that are not true. If you really want to understand something, you should take a look at the scientific publications yourself. AI can help, but it does not replace your own thought process.
Reading studies made easy
Scientific studies look very bulky at first glance, often contain a lot of text passages that really only those who have studied and mastered this subject can understand, and can therefore have a deterrent effect. You don’t have to understand everything exactly. If you know how scientific texts are normally structured, you can specifically read the parts that are kept generally understandable and then consider whether you want to delve deeper.
How is a study that is published in a specialist journal normally structured?
- Short abstract
- Introduction: State of research
- Method: What exactly was done and how
- Discussion: The exciting part
- Result: Often disappointingly short
- Literature: Dig deeper
We’ll look at this using the example of a study listed in the Publications section of DAN Europe. Why this one in particular? Because it’s current, looks interesting, and DAN has such a good reputation that it could be a good recommendation.
It’s best to download the study, take a first look at it, get ready to highlight – and go!
Title, authors, journal: First classification
The first thing you notice is the title – and here, as with the other information about the article, you can recognize a few things.
In our example, it is about Deco stress, i.e. physical reactions to decompression that do not cause symptoms. And simulated profiles were tested, i.e. in a pressure chamber.
Some of the authors are associated with DAN Europe. The lead is the University of Sao Paolo, Brazil.
The text is published in a specialist journal, but not one that you “need to know”.
- What kind of journal? Anyone who does research would prefer to publish their findings in the largest, most respected publications. However, not every small study is considered relevant enough by them. Studies in the field of diving medicine are therefore often found in smaller specialist journals like this one. However, it meets the criteria that all papers go through a peer review before publication, i.e. are assessed by colleagues who do not know the author. Here we have a small, little-known medical journal specializing in a specific field that adheres to scientific standards – a reliable source.
- Who are the authors? The list of authors is often long. Usually, the first one mentioned is the one who did most of the work. The last one is the one with the highest position in the working group. In between are the others who have made relevant contributions.
Googling the first author can help. Here we learn that he has been regularly publishing on the same complex of topics since 2018, has a doctorate in philosophy (!), and is also a GUE Instructor.
First read the abstract
Every study begins with a summary. This is often the easiest part to understand, and you can get a good first overview here. What did the authors want to find out, what exactly did they do, and what was the result?
In our example, we learn here:
- Different Deco profiles with deeper and shallower stops were tested
- The search was on for markers for a reaction of the immune system, for oxidative stress and effects on heart rate variability (HRV)
- 23 participants who took part in two pressure chamber dives with different profiles
- Differences were found in both profiles
- The aim is to show how the ascent rate influences inflammatory and immune reactions, which could ultimately lead to DCS
Develop questions
If you want to continue reading after the abstract without being frustrated by the fact that you don’t understand all the details, it makes sense to think in advance about what you would like to get out of the text.
Here you might want to know the following, for example:
Ascent rate? He actually means different ascent profiles, right?
What exactly were the profiles that were compared?
What do the markers they measured mean?
Can you see from the result which profile was “safer”, or are the indications not so clear?
To do this, you have to go into the further text – and perhaps get a little help from ChatGPT.
The introduction: Good for classification
The introduction presents the current state of research on your own question and what this study is intended to contribute. Here you can get a good overview and find a compilation of the relevant literature on the topic. This part is usually still relatively understandable and explains what is being investigated and why.
Let’s look at our example again.
First, there is a general explanation of what DCS is. Since cases are so rare, it’s about other indicators – things that happen in the body that indicate that symptoms are more likely. And it’s about the debate about whether deeper or shallower stops make more sense in decompression.
This classification is followed by an explanation of the markers for decompression stress. Here it makes sense to see if you understand the explanation – this is what is being measured here, and everything else will be much more incomprehensible than the explanation in the introduction.
It is mentioned here that, in addition to gas bubbles, other markers are now being used to find indications of the extent to which the body is exposed to physiological stress. These include microparticles in the blood, which indicate activation of the immune system. Several of these are now mentioned and the mechanisms are described in more detail – but delving into these details would be a project larger than “I’ll just read this study”. We can simply accept that various microparticles provide clues – which ones exactly remains open for the time being.
Then there’s Heart Rate Variability (HRV) as an indicator. Here, too, there is a brief explanation of what it’s all about – but of course not in such a way that you can do anything with it if you hear the term for the first time. But Google helps here – just try it out. Maybe you’ll also find a blog article written by the author of the study.
In the text of the study, you can jump to the last sentence in the paragraph: A lower HRV is associated with inflammatory reactions in the body and should provide additional information in addition to the microparticles.
It is expected that different values will be measured for both profiles – this can then show that the type of ascent is not irrelevant. Which variant is better is not clearly asked at first.
InDepth Magazine
Heart Rate Variability – What it is and Why it matters
Sergio Schirato, 2020
https://indepthmag.com/heart-rate-variability-what-it-is-and-why-it-matters/
Blog article by the author of the study on Heart Rate Variability (HRV)
The block: Methods
The point at which many attempts to read studies from foreign fields fail (and sometimes also from your own…) is the part on methodology.
This part is extremely important because it shows in detail what exactly was investigated. This makes it possible to understand how plausible the results are and whether the work was carried out cleanly – but for someone who is simply reading out of interest, this part is usually too much.
But you can easily skim through it here and simply pick out the understandable information. Let’s take a look:
- 23 healthy, fit, male divers. Male not as discrimination, but because cycle fluctuations could have a relevant influence.
- Pressure chamber dive with a rebreather
- Diluent: Trimix 18/45, Setpoint 1.2 bar pO2, in the Deco from 6m 1.4 bar pO2.
- Profiles: 20 min descent to reach 53, then another 15 min bottom time, ascent at 9 m/minute, Deco is shown in a table
- The Deco profile is explained in a very complicated way – more on that in a moment
- An EKG was performed before and after the dive to measure HRV
- Blood samples were taken before and after the dive and analyzed using various methods
- The data was then statistically evaluated
So far, so good. What we are perhaps most interested in: What exactly are the profiles that are being compared? Basically, it is already noticeable that the option “CCR in a pressure chamber and then breathing a trimix with a relatively large amount of helium” is not what you encounter in other studies. The reason for this particular setup is not given. Since helium and nitrogen have very different properties, this is a point that makes the matter more complicated than perhaps necessary.
In the explanation of the profile, it is mentioned that the Bühlmann algorithm ZHL16B was used for Deco planning. Today’s computers largely use ZHL16C, but at least it comes close to profiles that you would dive like this. If you play through the profiles in Subsurface with ZHL 16-C, the deep profile approaches GF 20/70, the shallow one is roughly 40/60. Rebuilding this is often easier than trying to understand the very confused explanation from the study…
Results and discussion
After the tiring methods, you can now look forward to the results. But what…?
Yes. These are also really not easy to read and often appear bulky and uninteresting. Here it is simply shown which data has now been collected. Well sorted, prepared, with graphics and tables, but still largely without interpretations. Without background knowledge, you can often do little with it. You can take a look at what was measured here and see if there is anything that immediately stands out.
The part of a study that is really exciting is usually in the discussion. Here, the results are evaluated, compared with the results of other studies, and an attempt is made to show what it could all mean.
In our example, it gets really complicated here: Which microparticles in the blood could mean what exactly? How meaningful is the HRV? What can you read from it?
Since this article is not about the content of this study, but only about how to read something like this, we will not go into more detail here. If you notice that you are burning with interest in exactly this topic, you can now spend hours, days, weeks trying to understand exactly what was done here – or you can look at the graphics, scan the text for understandable sentences and move on to the next section.
Especially in this example, you can sense in the discussion that the meaning of the individual measured values is not clear. And at exactly that point you can really lean back and decide that you don’t even have to understand the details that experts are discussing here. As long as we don’t want to do research in exactly this area ourselves, it is enough to concentrate on what can be regarded as reasonably secure knowledge. And for that we jump straight to the last section….
What does this "p" mean in all the graphics and tables?
A sign that appears in almost every publication on every single representation of results is the “p”.
The p-value is the probability of obtaining a result at least as extreme as the one observed, assuming that the null hypothesis is true.
What does that mean?
The null hypothesis is based on the assumption that the observed phenomenon has no effect. In this example, one could assume that there is no difference in the values before and after diving or no difference between the two profiles.
The p-value is a value between 0 and 1. It indicates how likely it would be to observe such a result or an even more extreme result if there were actually no effect.
So a p-value of 0.05 means: If there were no difference in reality, such a result would occur in about 5% of cases by chance alone.
The smaller the p-value, the worse the observed data fits the assumption of “no effect”. A threshold of 0.05 is often used to describe a result as statistically significant.
Mostly disappointingly little: Conclusion
After many pages full of numbers, graphics, discussions, you might like to see a really groundbreaking result at the end. But no – even studies that have a really big impact usually end with a very small, modest result. Perhaps one or two of you have already looked at the NEDU study on deeper or shallower stops during ascent – much cited, great importance in the scene, heralded the end of deep stops. But their result is just one sentence: For Deco dives with air Deco, deeper stops are not suitable for reducing bubble growth. Nothing more…
This is normal, because progress in science is always made up of many very small results. And so it is not surprising that our example also only provides a small detail: Different decompression profiles lead to different physiological reactions. Whether and how these are related to the DCS risk still needs to be researched…
No “this profile is safer”, no “this value means…” – simply that differences are measurable at all, and even with profiles that are not too far apart. And that is certainly a relevant result: It doesn’t matter how we decompress – but we still don’t know what the best way is.
Bonus: List of references
Every study you look at opens a gateway to infinitely many other studies. Of course, a reference is also provided for every question that is addressed. If you want to get into a topic, a study from the field is often the gateway to find out what there is on the topic. And since you don’t even have to run to a library these days, you can have a lot of new texts on your PC in a short time. And have them summarized by ChatGPT. But nobody can take away the understanding….
