CUK dropping the ball with MTB advice

[Steady rider]

CUK/BCF have had years to arrange a joint meeting or venue to discuss helmets. I think even AGM motions along those lines were suggested, but no progress.

[pjclinch]

Here's a question that rarely gets asked:

Is there actually any signal - amongst all the noise - for us to find? Just how hard should we keep digging?

I think Jonathan's answer was fair, and it's also worth noting that this forum is for discussion of helmets, rather than specifically the state of the academic literature on helmets.
e.g. If someone turns up saying "what can I expect from a helmet" then no particular need to resort to research papers as the spec will give answers.

Pete.

[De Sisti]

sport https://www.google.com/search?q=define+ ... e&ie=UTF-8
Learn to pronounce
noun
1.

an activity involving physical exertion and skill in which an individual or team competes against another or others for entertainment.

they could also include money

People bimbling on their mountain bike in the local woods doesn't fit with that criteria.

[Steady rider]

Good to hear a point of view.

[Steady rider]

Jonathan wrote, page 7

Cherrypicking isn't one of them. Systematic review is.

All the cherry picking references provided does not seem to include cycle helmets.

Many reports comparing helmeted to those not wearing fail to provide full details of the differences in behaviour, more likely for those not wearing to have been drinking for example. Other examples can be found in; https://www.researchgate.net/publicatio ... le_helmets

Systematic reviews would need to be based on comparing similar groups in behaviour as well as comparing helmet use. Research on cycle helmets does not usually provide sufficient details to allow for comparing the outcome for cyclists behaving similarly apart from helmet use. Because of this, there appears insufficient studies to provide reliable systematic reviews.

This seems to leave considering individual studies, population based studies, studies with common aspects, social studies and other studies. It looks like all the comments about cherry picking are really red herrings, diverting attention from the important aspects.

[Jdsk]

Systematic reviews are the currently best available method for establishing what is known and what isn't known.

They don't only work if there is evidence for a signal. If there isn't enough evidence they're the best way of establishing that. And it's very common for systematic reviews to conclude that there isn't enough evidence to come to any conclusions.

If you don't know that you've made the best possible attempt to gather all of the relevant publications how can you know that there isn't something else that should have been in there. And why should anyone else trust your review if you don't even say what you've included in the review and what you haven't and why?

Jonathan

[Steady rider]

https://www.researchgate.net/publicatio ... le_helmets

The link provided considered about 40 published articles, proving consideration was given to reports covering a long time period and various outcomes.

Method
Consideration is given to the detailed reports and approaches used to evaluating cycle helmet. Most of the reports included in a meta-analysis use similar methods that have weaknesses in not always fully detailing the differences in behaviour of wearers v non-wearers and accident situations. In all cases they also do not have any measure of the accident rate per km of travel. The basic approach of considering helmet use v injury outcome is basically considering two groups and may be insufficient to fully evaluate helmet effects. Attention is given to the fatality claims, serious head injury, head injury, face and neck claims and the merits of the supporting evidence. Additional consideration is given to other research giving an indication of safety and societal health outcome when the effects of helmet legislation and cycling levels are combined.

(4) (PDF) Weaknesses with a meta-analysis approach to assessing cycle helmets. Available from: https://www.researchgate.net/publicatio ... le_helmets [accessed Nov 05 2022]. published link https://trid.trb.org/view/1491227

[Jdsk]

You prefer reviews with unstated methods of what has been included in the review and what hasn't and why.

I prefer reviews with stated methods of what has been included in the review and what hasn't and why.

Changing the subject from systematic review to meta-analyis doesn't affect that.

But as the topic of this particular thread was how to change other people's minds I'd warn against using cherrypicked reviews. There's quite a lot of people out there who are familiar with current best practice in reviewing evidence. Often through reading and listening to what Goldacre has written about it. There might even be some of them making these decisions.

Jonathan

[Steady rider]

I can see the methods often used have not provided the clear and conclusive results people may wish for.

There may be several approaches to making progress, not allowing the Highway Code advice to be used to reduce fair compensation. I cannot see you and I agreeing, judging by previous comments. CUK have failed to make progress on the helmet advice in the Code, they could have done better.
So I will keep on trying.
ps
https://www.headway.org.uk/media/5191/b ... alysis.pdf
is considered a-systematic-review-and-meta-analysis
and does have a method.

Systematic reviews are the currently best available method for establishing what is known and what isn't known.

Systematic reviews need more than one person, so individual authors cannot provide them.

[mattsccm]

Sorry that I am bit late to the party but I have just spotted something that just proves that everyones views are tainted by their own perceptions. I quote;
" Forest of Dean I wear a helmet and everyone else does"
Not true. I don't. I commute daily (20 + miles) around the skills centre in the FoD. A fave home run is down the red route on my commuter, mudguards, Carradice bag and all. No helmet and I am not the only one. However I do make a concession, a dynamo light as the risk of lighting failiure is greater than banging my head.
The point is that, yet again, a certain section of (cycling) society insists on ramming their views dow the throats of others. How nasty, selfish and unkind. All because they see things in a narrow way.
It is as bad as those who try to ram religion or politics.

[pjclinch]

They don't only work if there is evidence for a signal. If there isn't enough evidence they're the best way of establishing that. And it's very common for systematic reviews to conclude that there isn't enough evidence to come to any conclusions.

My choice of signal/noise terminology may not be the best. Wheat and chaff would perhaps be better, and done right a SR will sort out the chaff with a quality threshold.
This is where we start to hit problems with helmets though, as typical "objective" quality measures (e.g. multiple authors, published in a recognised "quality" journal , oft-cited) don't really work. By most such standards Thompson, Rivara & Thompson's '89 opus is good work, while a lot of folk who've been looking at the debate for any length of time see quoted figures of 85% as shorthand for "hasn't done their homework properly".
So we are left with a quality threshold problem of what the SR author likes. I'd suggest with e.g. Jake Olivier amongst others that may not be quite as objective as we'd ideally like.
We also have the issue of where do we get data for our studies, and in this realm we have two main sources, the first of which is hospital data which appears to over report effectiveness with problems applying case/control to self selected cohorts who have already crashed. If you systematically review a body of work dominated by such studies you'll systematically over-report effectiveness, because the dataset you're reporting from is very probably skewed by the nature of the cohorts. Set a quality threshold to get rid of self-selecting, already crashed cohorts and there's significantly less to review!
The other source is population data which is very broad brush and used in most part as it's freely available, not because it's been designed to underpin helmet studies. This makes it hard to draw detailed conclusions, and concerning the effect at individual level is going to be very poor.

So doing a systematic review of the available literature where that has deep problems in it can't magic out the truth, and won't necessarily tell you what you have is biased (that's the SR author's job)

Pete.

[Jdsk]

They don't only work if there is evidence for a signal. If there isn't enough evidence they're the best way of establishing that. And it's very common for systematic reviews to conclude that there isn't enough evidence to come to any conclusions.

My choice of signal/noise terminology may not be the best. Wheat and chaff would perhaps be better, and done right a SR will sort out the chaff with a quality threshold.
...

Yes, I used signal because it was already in the thread. My preferred term outside the specialist domain is what's known and what isn't known. That's because establishing what isn't known is so important at the discovery phase, in the subsequent production of policy, and in directing future research,

Jonathan

[mattheus]

Sorry that I am bit late to the party but I have just spotted something that just proves that everyones views are tainted by their own perceptions. I quote;
" Forest of Dean I wear a helmet and everyone else does"
Not true. I don't. I commute daily (20 + miles) around the skills centre in the FoD. A fave home run is down the red route on my commuter, mudguards, Carradice bag and all. No helmet and I am not the only one. However I do make a concession, a dynamo light as the risk of lighting failiure is greater than banging my head.
The point is that, yet again, a certain section of (cycling) society insists on ramming their views dow the throats of others. How nasty, selfish and unkind. All because they see things in a narrow way.
It is as bad as those who try to ram religion or politics.

Well said! I've no problem with others wearing whatever clothes (or PPE) they choose on a bike. But please don't try to force others to do things just like you do them!

[Zulu Eleven]

Apologies for being even later to the party - but it’s worth noting that helmet use has been (as, at the least, a strong advisory) part of pretty much every inception of every code of conduct and other messaging regards MTB, dating back to at least 1987 (mountain bike code of conduct developed by countryside commission and the sports council, and endorsed by BMBF & CTC) So it seems a bit strange for people to get their knickers in a twist about it now, when it’s been such a consistent part of CTC/CUK and all other groups message regarding MTB for so many years.

It also seems obvious that the risk profile for head injury and impact of helmet use is clearly massively different between MTB and road riding, be that from different dangers (like overhanging branches) to speeds of impact and mode of injury x so one would question the relevance of generalised injury data and analyses that would also include road traffic injuries rather than purely off-road/MTB.

[pjclinch]

Some while ago on this I'd heard back from CUK and they'd asked did I want to add my comment to the piece originally brought up, and I said it would be better if they asked someone like Roger Geffen and I didn't hear any more.

On a whim I checked back on the piece today, some time later, and now it says...

Mountain biking is a challenging sport, so you want to make sure you’re properly protected. Knee pads and helmets might reduce the likelihood of injury. However, Cycling UK policy is very much that wearing a helmet is personal choice. Whether you need one depends on your own needs and the type of off-road cycling you’re doing.

The following advice is the personal opinion of the people we spoke to.

Shane advises: “A helmet is vital! Don’t even think about riding off road without a helmet. Even the smallest of crashes can cause significant head injuries, so before anything, get yourself a decent helmet.”

So that's at least a bit better from CUK (though they'd have got an extra point if they'd actually let me know!)

Pete.