5 minutes estimated total reading time
Chances are you saw one of the various news reports last year on the following paper:
It makes pretty interesting, and in parts, shocking, reading – The life expectancies of popular musicians are considerably shorter than that of the general population, with those of punk, metal, rap and hip hop artists being lucky to make it past their mid 30s!
The Conversation made a nice graphic to display all the causes of death by musical genre, highlighting in green if a cause of death was significantly below the average, blue if it was above, or red if it was significantly above. Hopefully they won’t mind me reproducing it here:
As you can see, punk and metal music both greatly increase your risk of accidental death (predominantly car accidents and drug over doses), and suicide, rap and hip hop have a very good chance of getting you killed! Blues it appears can break your heart, and Jazz and Folk can lead to cancer…
I’ve recently taken up Jazz Flute, should I be worried?
I decided to write about this study, as I think it is a fantastic example of how correlation does not equal causation.
I think there are few people who would deduce from these findings, that if you were to start playing heavy metal music, you would suddenly develop the urge to kill yourself (however much your neighbours might wish that you would).
I think fewer people yet, would think that taking up Jazz or Blues would significantly elevate your risk of cancer or heart disease… Unless maybe it’s something in the relative frequency of the blue notes and minor diminished chords which disrupts the chakras…
So Jazz won’t kill me, but what about Milk?
Though people (or scientists or the press) are unlikely to interpret these correlations as probable causations, it is easy for us to fall into this trap with other variables.
How many headlines have you seen along the lines of “Red Meat/Milk/Coffee/Eggs/Grapefruits may cause Cancer/Heart Disease/Bear Attacks”?
Or “Vegetarianism/Tomatoes/Olive Oil/Being French may reduce your risk of Cancer/Heart Disease/Alien Abduction”?
It was a recent Daily Mail (oh oh!) article I saw shared to Facebook stating that “Daily Saunas May Reduce Risk of Dementia and Alzheimer’s Disease” which prompted me to write this post.
The good old DM isn’t shy in inferring a likely causative link, with the opening line:
Frequent sauna bathing can reduce the risk of dementia, a new study has claimed
Plus of course, liberal use of the “66% less likely” statistic, knowing people won’t realise or understand that this is a reduction in relative risk, which isn’t anywhere near as impressive.
The scientists themselves, are of course a bit more careful in their choice of words:
In this male population, moderate to high frequency of sauna bathing was associated with lowered risks of dementia and Alzheimer’s disease. Further studies are warranted to establish the potential mechanisms linking sauna bathing and memory diseases.2
Though of course, despite their more careful choice of words, there still appears to be the implicit suggestion that there is likely some causative effect of sauna bathing which brings about this reduction in risk, and they most certainly knew when they gave out their press release how the media were going to report it, and the conclusions the general public were likely to draw.
As a species, we human beings are amazing at spotting patterns, and through this pattern recognition, determining the underlying causes. This ability is a major factor in our success as a species. But sometimes we get a bit carried away, and no matter how much we repeat “correlation does not mean causation”, it can be very hard not to jump to erroneous conclusions!
Do saunas actually have some kind of special property or effect which reduce the risk of degenerative brain diseases? Well, it’s not impossible I suppose, but I’d say it’s unlikely, and this study certainly does not prove it. No more than the study of Popular Musicians proves that Jazz Music increases your risk of cancer.
People that do this, probably do that as well
People are complex beings, for this reason, we’re very hard to study.
It is typically IMPOSSIBLE to draw any kind of conclusions from these kind of studies.
All you can say is that “The type of people who regularly perform activity A tend to have increased/decreased risk of outcome Z”.
Before jumping to the conclusion that A causes Z, however, you have to bear in mind that people who regularly perform activity A, probably also regularly perform activities B, C and D too, or perhaps abstain from E, F and/or G, or are from a particular socio-economic-ethnic-demographic.
Jazz and Blues musicians are for example more likely to be from a less privileged socio-economic background, more likely to smoke and drink, spend more time in smoke-filled bars, work unsociable hours, etc, etc. Jazz music itself doesn’t likely have any deleterious effects, it’s the lifestyle that goes with it.
Of course, epidemiologists are aware of this, and make attempts to “adjust” for as many of these variables as they can, but in reality, the web is usually just far too tangled to be unpicked.
When it comes to saunas, while there is a slim chance it’s something to do with the heat or the steam, it’s more likely that it’s because people who spend more time in the sauna are spending more time socialising, less time in the office, have more disposable income and/or leisure time, etc.
People who follow a vegetarian diet tend to have lower risks of heart disease and cancer than those that don’t. This doesn’t mean that eating meat will raise your risk of either disease however, any more than it will reduce your risk of growing a beard, wearing sandles or bad knitwear. It just means that this behaviour correlates with a host of other behaviours, and socio-economic traits, which are in turn associated with better health and longevity.
(Before any vegetarians get offended, I’m an ex veggy myself, and still have a beard and a penchant for open toed footwear)
The Take Home Message is that There is No Take Home Message
It is impossible to draw any conclusions from these type of population based surveys.
They are only useful for generating hypotheses, which then need to be tested by more rigorously designed studies.
Unfortunately, as I’ve written about before, such studies (i.e. double blind randomised controlled trials) are often expensive, impractical, or even impossible to conduct. Doh.
So the next time you see an alarmist headline in the press stating something along the lines of “Bacon will kill you”, don’t get too stressed out over your BLT, and rest assured that it’s probably actually no more dangerous that a bit of Miles Davis…
This post was written by Simon Whyatt and first appeared at www.livenowthrivelater.co.uk
1) Life Expectancy and Cause of Death in Popular Musicians: Is the Popular Musician Lifestyle the Road to Ruin?
Kenny DT, Asher A.
2) Sauna bathing is inversely associated with dementia and Alzheimer’s disease in middle-aged Finnish men
Tanjaniina Laukkanen Setor Kunutsor Jussi Kauhanen Jari Antero Laukkanen