Is cycle helmet promotion warranted? DK v NL data
-
- Posts: 2749
- Joined: 4 Jan 2009, 4:31pm
Is cycle helmet promotion warranted? DK v NL data
The Universities Transport Study Group 54th Annual Conference was held 4-6 July at Edinburgh Napier University and included a paper by Colin Clarke, ‘Is cycle helmet promotion warranted?’.
The research details from 50 years of cycling the average person could incur about a 1% risk of being admitted to hospital for a serious head injury and mentions concerns with helmet wearing, such as the incidence of cyclists hitting their head/helmet was significantly higher for helmet wearers and in both low speed linear impacts and the most severe oblique cases, linear and rotational accelerations may increase for a helmeted head. Some data also indicates a higher rate of falling off for helmeted cyclists. Several reports comparing helmeted to non-helmeted are mentioned and it appears some of the “protective effect” of helmets may be due to difference in behaviour.
Denmark has promoted helmets for some years and is compared to the Netherlands over a 20 year period, cycling has increased in the Netherlands by 19% compared to a reduction of 15% for Denmark. It appears helmet promotion is a main reason for the difference.
In summary, the case has not yet been convincingly made for the promotion of cycle helmets. There is evidence they may help prevent some injuries, such as skull fractures, however some evidence suggest they are also a safety hazard, with extra accidents and impacts. There are serious disadvantages to promoting helmets and addressing alternative means to improving safety for cyclists could be the better approach.
https://www.researchgate.net/publicatio ... _warranted
The research details from 50 years of cycling the average person could incur about a 1% risk of being admitted to hospital for a serious head injury and mentions concerns with helmet wearing, such as the incidence of cyclists hitting their head/helmet was significantly higher for helmet wearers and in both low speed linear impacts and the most severe oblique cases, linear and rotational accelerations may increase for a helmeted head. Some data also indicates a higher rate of falling off for helmeted cyclists. Several reports comparing helmeted to non-helmeted are mentioned and it appears some of the “protective effect” of helmets may be due to difference in behaviour.
Denmark has promoted helmets for some years and is compared to the Netherlands over a 20 year period, cycling has increased in the Netherlands by 19% compared to a reduction of 15% for Denmark. It appears helmet promotion is a main reason for the difference.
In summary, the case has not yet been convincingly made for the promotion of cycle helmets. There is evidence they may help prevent some injuries, such as skull fractures, however some evidence suggest they are also a safety hazard, with extra accidents and impacts. There are serious disadvantages to promoting helmets and addressing alternative means to improving safety for cyclists could be the better approach.
https://www.researchgate.net/publicatio ... _warranted
Re: Is cycle helmet promotion warranted? DK v NL data
The full Methods section of that reads:
Method
Consideration is given to a wide range of information from articles, submissions, accident/cycling/population data to assess safety and social outcomes of helmet promotion. Risk and health calculations are provided to assess if cycle helmet promotion provides a benefit. The reliability of some helmet claims is also considered.
That doesn't meet any of the standards for what should be in the Methods section of a scientific paper: basically what was done to a level that would allow someone else to understand and replicate it.
It doesn't state which studies were included and which weren't and why. I'd encourage the author to use systematic methods in reviews rather than running all of the risks of cherrypicking.
And to consider submitting it for publication. The reviews can be very helpful in reducing the risks of bias.
Jonathan
Method
Consideration is given to a wide range of information from articles, submissions, accident/cycling/population data to assess safety and social outcomes of helmet promotion. Risk and health calculations are provided to assess if cycle helmet promotion provides a benefit. The reliability of some helmet claims is also considered.
That doesn't meet any of the standards for what should be in the Methods section of a scientific paper: basically what was done to a level that would allow someone else to understand and replicate it.
It doesn't state which studies were included and which weren't and why. I'd encourage the author to use systematic methods in reviews rather than running all of the risks of cherrypicking.
And to consider submitting it for publication. The reviews can be very helpful in reducing the risks of bias.
Jonathan
-
- Posts: 2749
- Joined: 4 Jan 2009, 4:31pm
Re: Is cycle helmet promotion warranted? DK v NL data
As with many articles they are limited in either the word count or pages. This was limited to 12 pages, so material included had to be selected to provide the best to explain the issues. Adding more details to the method section would detract from other parts and it explains a wide selection of material was considered. The paper explains how it reaches it conclusions and provides more than 50 references that can be checked. The method allowed for including material from published papers and from other sources, that allowed for a fully evaluation.
Re: Is cycle helmet promotion warranted? DK v NL data
It doesn't explain anywhere why some studies were included and others weren't. Unless I've missed the section where it does...Steady rider wrote: ↑7 Jul 2022, 12:39pm As with many articles they are limited in either the word count or pages. This was limited to 12 pages, so material included had to be selected to provide the best to explain the issues. Adding more details to the method section would detract from other parts and it explains a wide selection of material was considered. The paper explains how it reaches it conclusions and provides more than 50 references that can be checked. The method allowed for including material from published papers and from other sources, that allowed for a fully evaluation.
Jonathan
-
- Posts: 2749
- Joined: 4 Jan 2009, 4:31pm
Re: Is cycle helmet promotion warranted? DK v NL data
Method
It used information to assess safety and social outcomes of helmet promotion and included material most suitable for that purpose.Consideration is given to a wide range of information from articles, submissions, accident/cycling/population data to assess safety and social outcomes of helmet promotion. Risk and health calculations are provided to assess if cycle helmet promotion provides a benefit. The reliability of some helmet claims is also considered.
Re: Is cycle helmet promotion warranted? DK v NL data
Yes, I quoted the Methods section in full above.Steady rider wrote: ↑7 Jul 2022, 1:15pm MethodIt used information to assess safety and social outcomes of helmet promotion and included material most suitable for that purpose.Consideration is given to a wide range of information from articles, submissions, accident/cycling/population data to assess safety and social outcomes of helmet promotion. Risk and health calculations are provided to assess if cycle helmet promotion provides a benefit. The reliability of some helmet claims is also considered.
It doesn't say what criteria were used in that assessment of what was "most suitable" or who made that assessment. So a reader can't assess the risk of bias.
The current best ways of doing this and reducing the risk are well known. (Goldacre and Spiegelhalter have written and spoken extensively on this topic.)
Jonathan
-
- Posts: 2749
- Joined: 4 Jan 2009, 4:31pm
Re: Is cycle helmet promotion warranted? DK v NL data
Some aspects reported are provided with detailed calculation, such as the general safety assessment, that can be checked for bias from the references.
Comparing helmeted to non-helmeted, mainly refers to data from Olivier and Creighton and other references showing difference in behaviour in the two groups.
Risk of helmet impacts and extra falls refers to 5 sources of data where concerns can be seen in the results.
There are many references provided that can be viewed..
With thousands of helmet reports the author had to use his experience in deciding how to assess if cycle helmet promotion is warranted.
It is based on experience and considering a wide range of information. Other authors may not have the same experience and base their approach on selection procedures. The information for Denmark shows a 15% reduction compared with a 19% increase in cycling for the Netherlands, with data from both countries that can be check. The average distance cycled for the 10-17 age group in Denmark is approximately 1.75km compared with 6.42km for the 12-18 age group in the Netherlands. It clearly indicates helmet promotion could have seriously discouraged cycling.
Comparing helmeted to non-helmeted, mainly refers to data from Olivier and Creighton and other references showing difference in behaviour in the two groups.
Risk of helmet impacts and extra falls refers to 5 sources of data where concerns can be seen in the results.
There are many references provided that can be viewed..
With thousands of helmet reports the author had to use his experience in deciding how to assess if cycle helmet promotion is warranted.
It is based on experience and considering a wide range of information. Other authors may not have the same experience and base their approach on selection procedures. The information for Denmark shows a 15% reduction compared with a 19% increase in cycling for the Netherlands, with data from both countries that can be check. The average distance cycled for the 10-17 age group in Denmark is approximately 1.75km compared with 6.42km for the 12-18 age group in the Netherlands. It clearly indicates helmet promotion could have seriously discouraged cycling.
Re: Is cycle helmet promotion warranted? DK v NL data
The paper doesn't say that any other reports or studies were assessed and rejected from consideration. Let alone the criteria used to do that.Steady rider wrote: ↑7 Jul 2022, 3:30pmWith thousands of helmet reports the author had to use his experience in deciding how to assess if cycle helmet promotion is warranted.
It is based on experience and considering a wide range of information. Other authors may not have the same experience and base their approach on selection procedures. The information for Denmark shows a 15% reduction compared with a 19% increase in cycling for the Netherlands, with data from both countries that can be check. The average distance cycled for the 10-17 age group in Denmark is approximately 1.75km compared with 6.42km for the 12-18 age group in the Netherlands. It clearly indicates helmet promotion could have seriously discouraged cycling.
How do you know that he did that?
Thanks
Jonathan
Re: Is cycle helmet promotion warranted? DK v NL data
Meta analysis - unless the signal to noise is high then little is gained from these sorts of studies.
Are there are enough cyclists with equipoise for a prospective randomised controlled trial ? Not easy to blind it though.
Are there are enough cyclists with equipoise for a prospective randomised controlled trial ? Not easy to blind it though.
Re: Is cycle helmet promotion warranted? DK v NL data
Systematic reviews are much better than cherrypicking regardless of the level of evidence. And you can't assess the levels of evidence and the current state of knowledge and ignorance without adequate reviews.
Yes, *RCTs are very difficult in this area. But there are grades of evidence below RCTs and we should assess studies according to those grades, and use that assessment in the synthesis.
And in designing studies there's an excellent working principle: design them so that they are one level higher than you first thought possible. This is surprisingly often attainable.
Jonathan
* But there is one for yellow jackets worn by cyclists. And several pages of discussion in this forum where no-one has cited it!
-
- Posts: 3931
- Joined: 26 Mar 2022, 7:13am
Re: Is cycle helmet promotion warranted? DK v NL data
Having read several long threads on here about this, and a good selection of the cited papers, my one conclusion is that a really thoroughgoing set of studies, fresh ones, is needed to address four different questions:
- what affect does promoting helmet wear have on the take-up of cycling? (Possibly needing a “by sector” analysis, because it may affect sport, casual leisure, and utility cycling differently)
- does wearing a cycle helmet of the currently available kinds increase the probability of the wearer being involved in an accident? (Making sure to untangle non-causal correlations)
- how effective in mitigating head injury are currently available cycle helmets, and can they exacerbate injury, considering a wide range of credible incident scenarios?
- what, if any, improvements could practicably be made to helmet design to improve injury mitigation, and/or reduce injury exacerbation?
In the absence of really solid, back to basics, studies in all of these areas, it seems to me that the answers to every one of those questions are shaded with confusion, contradictory indicators, and hence doubt. I’m particularly suspicious of what I suspect to be non-causal correlations being cited in several directions.
How you would set about answering some of the questions in a meaningful way, I’m anything but sure, because I don’t think that the necessary data is being gathered, or possibly ever could be gathered. As a daft instance, I don’t file a report every time I get thwacked on the helmet by a low branch when cycling in the woods, but that is a credible incident type in which my helmet saves me at least an uncomfortable bruise.
Maybe it needs a huge population study like the Zoe health reporting thing that began with covid.
TBH, I’m personally not greatly impressed by meta-studies in a field where a lot of the underlying material seems to be narrowly particular, or presents big questions around possibly non-causal correlations.
- what affect does promoting helmet wear have on the take-up of cycling? (Possibly needing a “by sector” analysis, because it may affect sport, casual leisure, and utility cycling differently)
- does wearing a cycle helmet of the currently available kinds increase the probability of the wearer being involved in an accident? (Making sure to untangle non-causal correlations)
- how effective in mitigating head injury are currently available cycle helmets, and can they exacerbate injury, considering a wide range of credible incident scenarios?
- what, if any, improvements could practicably be made to helmet design to improve injury mitigation, and/or reduce injury exacerbation?
In the absence of really solid, back to basics, studies in all of these areas, it seems to me that the answers to every one of those questions are shaded with confusion, contradictory indicators, and hence doubt. I’m particularly suspicious of what I suspect to be non-causal correlations being cited in several directions.
How you would set about answering some of the questions in a meaningful way, I’m anything but sure, because I don’t think that the necessary data is being gathered, or possibly ever could be gathered. As a daft instance, I don’t file a report every time I get thwacked on the helmet by a low branch when cycling in the woods, but that is a credible incident type in which my helmet saves me at least an uncomfortable bruise.
Maybe it needs a huge population study like the Zoe health reporting thing that began with covid.
TBH, I’m personally not greatly impressed by meta-studies in a field where a lot of the underlying material seems to be narrowly particular, or presents big questions around possibly non-causal correlations.
Last edited by Nearholmer on 7 Jul 2022, 5:24pm, edited 2 times in total.
Re: Is cycle helmet promotion warranted? DK v NL data
Yes. Identifying specific questions and addressing them is crucial to making any progress.
I may well have said the same in other threads! : - )
Jonathan
I may well have said the same in other threads! : - )
Jonathan
-
- Posts: 2749
- Joined: 4 Jan 2009, 4:31pm
Re: Is cycle helmet promotion warranted? DK v NL data
Trying to answer some of the points, the paper includes- what affect does promoting helmet wear have on the take-up of cycling? (Possibly needing a “by sector” analysis, because it may affect sport, casual leisure, and utility cycling differently)
- does wearing a cycle helmet of the currently available kinds increase the probability of the wearer being involved in an accident? (Making sure to untangle non-causal correlations)
- how effective in mitigating head injury are currently available cycle helmets, and can they exacerbate injury, considering a wide range of credible incident scenarios?
Bryan-Brown and Taylor, (1997) reported, “Eleven Local Authorities had however held a helmet campaign when their activities were focused solely on the promotion of helmets. In these Local Authority areas, a larger increase in helmet wearing was found than in the areas which had not held such a campaign. However, this increase was found to be strongly linked to a decrease in the numbers of cyclists observed: in those areas where a campaign had been held and the numbers of cyclists had increased, helmet wearing fell.”
Cycling UK state, ‘Some evidence suggests they may in fact increase the risk of cyclists having falls or collisions in the first place or suffering neck injuries’.
In brief, it probably depends on the level of promotion and cohesion but it appears helmet promotion can have a negative influence.StClair and Chinn (2007) reported ‘However, in both low speed linear impacts and the most severe oblique cases, linear and rotational accelerations may increase to levels corresponding to injury severities as high as AIS 2 or 3, at which a marginal increase (up to 1 AIS interval) in injury outcome may be expected for a helmeted head.’
There is evidence connecting helmet use and a higher accident rate. For concussions they do not seem to be generally effective but in some impacts they can increase the injury level. ps they probably lower skull fractures that are about 1%-4% of injuries compared with about 40% for concussions.
http://www.cycle-helmets.com/us-cycling ... juries.pdf
Re: Is cycle helmet promotion warranted? DK v NL data
Cherrypicking specific sentences from specific papers from a cherrypicked review isn't going to answer any of Nearholmer's questions. They demand systematic review that covers all of the relevant evidence no matter which way it points.Steady rider wrote: ↑7 Jul 2022, 6:34pm Trying to answer some of the points, the paper includesBryan-Brown and Taylor, (1997) reported, “Eleven Local Authorities had however held a helmet campaign when their activities were focused solely on the promotion of helmets. In these Local Authority areas, a larger increase in helmet wearing was found than in the areas which had not held such a campaign. However, this increase was found to be strongly linked to a decrease in the numbers of cyclists observed: in those areas where a campaign had been held and the numbers of cyclists had increased, helmet wearing fell.”Cycling UK state, ‘Some evidence suggests they may in fact increase the risk of cyclists having falls or collisions in the first place or suffering neck injuries’.StClair and Chinn (2007) reported ‘However, in both low speed linear impacts and the most severe oblique cases, linear and rotational accelerations may increase to levels corresponding to injury severities as high as AIS 2 or 3, at which a marginal increase (up to 1 AIS interval) in injury outcome may be expected for a helmeted head.’
Jonathan
Last edited by Jdsk on 7 Jul 2022, 6:45pm, edited 1 time in total.
Re: Is cycle helmet promotion warranted? DK v NL data
What did the preceding and following sentences in that abstract say?Steady rider wrote: ↑7 Jul 2022, 6:34pmStClair and Chinn (2007) reported ‘However, in both low speed linear impacts and the most severe oblique cases, linear and rotational accelerations may increase to levels corresponding to injury severities as high as AIS 2 or 3, at which a marginal increase (up to 1 AIS interval) in injury outcome may be expected for a helmeted head.’
Thanks
Jonathan