Is Survivorship Bias an important factor in the promotion of cycle helmets?

[Jdsk]

...
Nor does it account for the possibility that a proportion of the helmet wearers are more likely to be knowingly and willingly involving themselves in more risky forms of cycling, and are wearing helmets to back-off a bit of that risk. In short, the exact opposite of the much mentioned risk compensation.
...

Could be. And of course we can only discover what's happening by studying it.

Jonathan

[Nearholmer]

It is better called risk homeostasis.

Ah, yes; please get about renaming it, so that I don’t get confused in future.

[Mike Sales]

[/i]olmer post_id=1897161 time=1741683981 user_id=57112]

It is better called risk homeostasis.

Ah, yes; please get about renaming it, so that I don’t get confused in future.

'Risk homeostasis' has been long used.
Perhaps 'homeostasis' is avoided for those allergic to long, unfamiliar words.
I do not see why one cannot compensate for a reduction in risk by doing something previously avoided.
Some here reduce risk with a helmet and then ride on busier roads or hurtle downhill on a rough track.
All life has risks, but also rewards. Even staying in bed risks bed sores. Going downstairs risks a fall, but gets one breakfast. We balance the risks of crossing the road with the need of getting somewhere.
This balancing behaviour is seen in all animals. Even an amoeba must move away from a substance it senses as harmful. A rabbit will decide whether or not to venture out of its burrow to eat depending on the dangers it hears.
I cannot imagine that we do not do the same, and assess risks versus rewards when deciding actions.
If you are interested in the subject read Risk by John Adams, available online to download.

[Nearholmer]

Of course we do, both instinctively and consciously. And, sometimes we get it right, and sometimes we get it spectacularly wrong by vastly over-valuing either the risk, or the reward.

What I find odd about the ever-recirculating helmet argument here is that people frequently say “ah …., but wearing a helmet makes people take more risks”, and barely ever say “ah …… but people who have consciously and willingly decided to take risks in return for fun are more likely to wear helmets”.

[Jdsk]

...
If you are interested in the subject read Risk by John Adams, available online to download.

If you're interested in changes of behaviour caused by wearing helmets when cycling it would probably be wise to read the relevant studies of changes of behaviour caused by wearing helmets when cycling. "Risk" doesn't cite any and was published thirty years ago.

Jonathan

[Jdsk]

It is better called risk homeostasis.

Ah, yes; please get about renaming it, so that I don’t get confused in future.

Homeostasis means something like restoring to an optimal state. I'd suggest a more neutral wording for the relevant questions, something like:

What are the effects of wearing a helmet on the behaviour of the cyclist?
and
What are the effects of wearing a helmet on the behaviour of other road users?

Of course these questions are only relevant to an evidence-based approach of trying to find out what's known and what isn't known.

Jonathan

[Stevek76]

It more means staying the same than specifically 'optimal'.

Plenty of homeostatic processes in biology aren't exactly optimal from an objective standpoint, we're messy products of evolution where the criteria is rather more 'good enough' than optimal.

Though from the point of view of an individual's own risk/reward processing the target is optimal. It's just almost certainly not optimal for actual risk exposure as the aforementioned messily evolved monkey brain is really extrapolating well beyond the training dataset when it comes to modern technology.

We can already see that with the way plenty react to the concept of not using a helmet to cycle to the shops vs walking it. Similar actual risk exposure, massively differing popularly perceived risk exposure.

...
If you are interested in the subject read Risk by John Adams, available online to download.

If you're interested in changes of behaviour caused by wearing helmets when cycling it would probably be wise to read the relevant studies of changes of behaviour caused by wearing helmets when cycling. "Risk" doesn't cite any and was published thirty years ago.

Except the bulk of studies on the topic aren't really worth the paper they're written on, the state of the literature has been pretty thoroughly picked apart in previous threads.

Unconvinced such things are a particularly useful expenditure of research resources either. For sportsing a focus on better testing standards and consumer information would be far more useful. For transport policy, helmets are a severely harmful distraction embraced by those who want to discourage and marginalise cycling, of which the uncertainty over sub topics of risk compensation and efficacy are irrelevant rounding errors.

barely ever say “ah …… but people who have consciously and willingly decided to take risks in return for fun are more likely to wear helmets”.

I mention that element frequently to be fair, it's likely a significant part of the reason casualty rates go up with helmet wearing rates (particularly where laws get involved in aus/NZ etc) as you're just leaving the sportsing cyclists as the only ones doing the cycling, and they're the ones that crash all the time and/or willing to cycle on the high risk roads.

[Jdsk]

...
Except the bulk of studies on the topic aren't really worth the paper they're written on, the state of the literature has been pretty thoroughly picked apart in previous threads.
...

In this forum?

I totally disagree. We've rarely got to trying to establish the relevant questions. And the noise to signal ratio from the logical fallacies is massive.

But it's entirely up the forum participants to determine what sort of discussion they want.

Jonathan

[853]

What I find odd about the ever-recirculating helmet argument here is that people frequently say “ah …., but wearing a helmet makes people take more risks”, and barely ever say “ah …… but people who have consciously and willingly decided to take risks in return for fun are more likely to wear helmets”.

It's largely bigotry. There are many members here who won't study what data is available, because they have already made up their mind and will just parrot the same old line that we've read many, many times before. There are others who will spend time to carefully cherry-pick some scrap of data from 20+ years ago, because it supports their agenda. There are very people who actually want to study the data and what it might mean, which is why the 'discussion' is so poor and so few people read it

[mattheus]

What I find odd about the ever-recirculating helmet argument here is that people frequently say “ah …., but wearing a helmet makes people take more risks”, and barely ever say “ah …… but people who have consciously and willingly decided to take risks in return for fun are more likely to wear helmets”.

OK, I'll bite: I'm happy to agree with both of those hypotheses ...
but may I ask, how do you think the 2nd one affects anything in these debates?

[Nearholmer]

It’s frequent absence is the problem, so often it doesn’t affect these debates, when perhaps it should.

As with all of this topic, things are complex, potentially conflicting, bordering on imponderable, because different people may wear helmets for different reasons, for instance: knowingly engaging in risky activity, mirroring peers, timidity (risk aversion), compulsion, habit formed during childhood under parental compulsion, etc.

So when some people here, as they do, spot correlations between helmet-wearing and being involved in accidents, then leap to the conclusion that there is a causal relationship in a particular direction, they may be seriously off-beam.

[pjclinch]

What I find odd about the ever-recirculating helmet argument here is that people frequently say “ah …., but wearing a helmet makes people take more risks”, and barely ever say “ah …… but people who have consciously and willingly decided to take risks in return for fun are more likely to wear helmets”.

It's largely bigotry. There are many members here who won't study what data is available, because they have already made up their mind and will just parrot the same old line that we've read many, many times before. There are others who will spend time to carefully cherry-pick some scrap of data from 20+ years ago, because it supports their agenda. There are very people who actually want to study the data and what it might mean, which is why the 'discussion' is so poor and so few people read it

The above doesn't actually address Nearholmer's point particularly well.
Like stevek76 I'm quite surprised to see a suggestion that there's little mention of using PPE for active risk management. I'm pretty sure I say that a lot, and it's certainly something I actively do (so helmet for white water and surf but not for lochs, helmet for mountain cragging but not for indoor overhangs, helmet for MTB trail centres but not for easy gravel, and so on). I'd say we're in the realm of the bleedin' obvious that PPE is used for active risk management because that's a lot of the point of PPE.

The description of various poster characteristics seems to have left out those who have studied the field in considerable detail for years (decades even!) and have come to the conclusion that it's hard to come to a conclusion. That that is widely seen as "anti helmet" does conform to some of the other characteristics noted above.

Pete.

[pjclinch]

...
If you are interested in the subject read Risk by John Adams, available online to download.

If you're interested in changes of behaviour caused by wearing helmets when cycling it would probably be wise to read the relevant studies of changes of behaviour caused by wearing helmets when cycling. "Risk" doesn't cite any and was published thirty years ago.

One of the main points of Risk is it is a generalised, multi-faceted view of how "obvious" engineering safety interventions tend to be undermined by human behaviour. That's just as true now as it was 30 years ago.

Individual studies are tightly focused and subject to numerous methodological shortcomings, often caused by the same factors of human behaviour they're trying to quantify.

The book is qualitative, the studies attempt to be quantitative. Done well, quantitative is better, but where it's actually damn hard to do it right it's often wrong, and that's where a more general, qualitative approach can actually tell you more.

If you want a general understanding of the problems human behaviour cause for risk management then you're probably better off reading a general treatise on the subject than attempts to quantify one aspect which is widely acknowledged to be very hard to do well.

Pete.
(edit to introduce a bit more appropriate vagueness)

[pjclinch]

It’s frequent absence is the problem, so often it doesn’t affect these debates, when perhaps it should.

As with all of this topic, things are complex, potentially conflicting, bordering on imponderable, because different people may wear helmets for different reasons, for instance: knowingly engaging in risky activity, mirroring peers, timidity (risk aversion), compulsion, habit formed during childhood under parental compulsion, etc.

So when some people here, as they do, spot correlations between helmet-wearing and being involved in accidents, then leap to the conclusion that there is a causal relationship in a particular direction, they may be seriously off-beam.

Yes. A huge amount of "work" is undermined by the simplistic assumption that there are broadly two sorts of cyclists, those who wear helmets and those who don't. It shouldn't be too surprising that it's a bit more complicated than that, but looking at the black/white standpoints you see you might be forgiven for assuming otherwise.

(It turns out that the reasons that typically undermine efforts to show helmets are Amazingly Wonderful also undermine efforts to show they're Really Terrible.)

Pete.

[853]

What I find odd about the ever-recirculating helmet argument here is that people frequently say “ah …., but wearing a helmet makes people take more risks”, and barely ever say “ah …… but people who have consciously and willingly decided to take risks in return for fun are more likely to wear helmets”.

It's largely bigotry. There are many members here who won't study what data is available, because they have already made up their mind and will just parrot the same old line that we've read many, many times before. There are others who will spend time to carefully cherry-pick some scrap of data from 20+ years ago, because it supports their agenda. There are very people who actually want to study the data and what it might mean, which is why the 'discussion' is so poor and so few people read it

The above doesn't actually address Nearholmer's point particularly well.
Like stevek76 I'm quite surprised to see a suggestion that there's little mention of using PPE for active risk management. I'm pretty sure I say that a lot, and it's certainly something I actively do (so helmet for white water and surf but not for lochs, helmet for mountain cragging but not for indoor overhangs, helmet for MTB trail centres but not for easy gravel, and so on). I'd say we're in the realm of the bleedin' obvious that PPE is used for active risk management because that's a lot of the point of PPE.

The description of various poster characteristics seems to have left out those who have studied the field in considerable detail for years (decades even!) and have come to the conclusion that it's hard to come to a conclusion. That that is widely seen as "anti helmet" does conform to some of the other characteristics noted above.

Pete.

I was specifically answering the question raised by Nearholmer, and you've said my answer doesn't address their point particularly well. But you have then launched off on a list of your non-cycling active pursuits and whataboutery. Anything in fact but answer the question of Nearholmer, or why you don't think my reply doesn't address their point particularly well.

Thanks for proving my point for me.