Risk and Freedom

drossall
Posts: 4934
Joined: 5 Jan 2007, 10:01pm
Location: North Hertfordshire

Re: Risk and Freedom

Postby drossall » 14 Sep 2020, 5:33pm

mikeymo wrote:
Jdsk wrote:
Mike Sales wrote:I think that if helmets work, it should be easy to tell. Certainly benefits "too modest to capture" are not good enough to force us to wear them.

The existing studies might be too weak to show even a large effect, especially in particular settings or for particular groups of riders.


The existing studies most frequently quoted to support anti-helmeters' opinion don't study the effect of cycle helmets on safety. They study the effect of legislation. A statute is not the same thing as a helmet.

If you read the piece from which the quotation is taken, it's certainly referring to effects of legislation, but also to the experience in the Low Countries, where helmet-wearing rates are low, and so are injuries. It's hard to defend a view that the quotation is considering only legislative effects. It's talking about the benefits of helmets, not the benefits of legislation.

Nor is it a quotation from an anti-helmeteer, but from a statistician whose interest is in any example of how statistics are used in medicine, epidemiology and similar fields, and for whom this seems to be just one more example. I'm not aware that the author has any other interest in helmets or cycling.

mikeymo
Posts: 1321
Joined: 27 Sep 2016, 6:23pm

Re: Risk and Freedom

Postby mikeymo » 14 Sep 2020, 5:51pm

drossall wrote:
mikeymo wrote:
Jdsk wrote:The existing studies might be too weak to show even a large effect, especially in particular settings or for particular groups of riders.


The existing studies most frequently quoted to support anti-helmeters' opinion don't study the effect of cycle helmets on safety. They study the effect of legislation. A statute is not the same thing as a helmet.

If you read the piece from which the quotation is taken, it's certainly referring to effects of legislation, but also to the experience in the Low Countries, where helmet-wearing rates are low, and so are injuries. It's hard to defend a view that the quotation is considering only legislative effects. It's talking about the benefits of helmets, not the benefits of legislation.

Nor is it a quotation from an anti-helmeteer, but from a statistician whose interest is in any example of how statistics are used in medicine, epidemiology and similar fields, and for whom this seems to be just one more example. I'm not aware that the author has any other interest in helmets or cycling.


I was responding to Jdsk really, rather than the quote in a quote. The studies most commonly quoted here are from antipodean sources, where the studies centre around the periods before and after the introduction of a law about cycle helmets.

drossall
Posts: 4934
Joined: 5 Jan 2007, 10:01pm
Location: North Hertfordshire

Re: Risk and Freedom

Postby drossall » 14 Sep 2020, 6:30pm

But you can't measure the effect of a helmet directly without reproducing the same accident with and without one. This is difficult - impossible in the real world, and simulated accidents often seem unrepresentative. And anyway, "Does this make a difference in the real world?" is a sensible question.

So you need to look at both - what does an individual helmet do, and what happens when helmets are more widely worn in populations. Which is often before and after legislation, because that produces a step change that gives something to measure.

slowster
Posts: 1586
Joined: 7 Jul 2017, 10:37am

Re: Risk and Freedom

Postby slowster » 14 Sep 2020, 8:16pm

I doubt that risk compensation behaviour is likely to feature in work related safety in the same way that studies have identified it outside of work.

Implicit in the idea of risk compensation on the road is that people must be deriving some benefit from behaving in a more risky way because they are relying on some safety measure that offsets the increase in risk. For speeding that benefit might be getting home earlier/not being late, or even just the thrill of driving fast. In the workplace those 'rewards' for risk compensation are less likely to exist at an individual level. The motivations for individuals to engage in more risky behaviour at work are probably quite different, i.e. employees taking shortcuts with risks not because of any compensating safety measure, but because of, for example, a macho culture or for financial gain if paid for piecework and productivity can be increased by cutting corners.

mikeymo
Posts: 1321
Joined: 27 Sep 2016, 6:23pm

Re: Risk and Freedom

Postby mikeymo » 14 Sep 2020, 8:22pm

drossall wrote:So you need to look at both - what does an individual helmet do, and what happens when helmets are more widely worn in populations. Which is often before and after legislation, because that produces a step change that gives something to measure.


With all due respect, that demonstrates an incredibly simplistic view of the effect of mandatory helmet laws. Do you really believe that if 40% of cyclists pre legislation wear helmets, and after legislation 80% of cyclists wear helmets, that the only change is numerical? That the number of cyclists wearing helmets is the only change? Not the type of cyclists, or where it happens, the type of cycles ridden, the attitude of the remaining/new cyclists to risk?

Here are a few confounding variables:

1. Cautious cyclists think that enforced helmet wearing might mean that cycling is dangerous, so don't take up cycling.

2. Reckless cyclists who didn't wear helmets before legislation carry on cycling, and carry on not wearing helmets (compliance in Victoria was 75%).

3. Fast sports cyclists who wore helmets carry on cycling, fast, and carry on wearing helmets.

4. Cyclists who don't want to wear helmets buy cheap ones and don't wear them properly.

All of these are certainly things that will have an influence on the conclusions that can be drawn. But I fail to see how a helmet law will change "what an individual helmet [does]", as you put it. Unless that law is about the actual design and construction of the the helmet of course.

"...what happens when helmets are more widely worn in populations". That's the point, the change in population is more than just in number, or more accurately, proportion.

The places where mandatory helmet laws are introduced have been seized upon by anti-helmeteers because they imagine that somehow they can be used as subject for study. As though the before and after nature of helmet laws gives us a control group and an experimental group. The examples I give above (and I daresay there are others you could posit) demonstrate how deeply flawed the study of the introduction of mandatory helmet laws is as a way of examining the safety benefit of helmets. What is despicable is the leap that is sometimes made from "we can't be sure if they work or not", to "helmets and helmet promotion are bad". And even more extraordinarily - "helmets should be banned" (I kid you not).

mikeymo
Posts: 1321
Joined: 27 Sep 2016, 6:23pm

Re: Risk and Freedom

Postby mikeymo » 14 Sep 2020, 9:12pm

slowster wrote:I doubt that risk compensation behaviour is likely to feature in work related safety in the same way that studies have identified it outside of work.

Implicit in the idea of risk compensation on the road is that people must be deriving some benefit from behaving in a more risky way because they are relying on some safety measure that offsets the increase in risk. For speeding that benefit might be getting home earlier/not being late, or even just the thrill of driving fast. In the workplace those 'rewards' for risk compensation are less likely to exist at an individual level. The motivations for individuals to engage in more risky behaviour at work are probably quite different, i.e. employees taking shortcuts with risks not because of any compensating safety measure, but because of, for example, a macho culture or for financial gain if paid for piecework and productivity can be increased by cutting corners.


Have you ever worked in manufacturing or construction, with your hands?

drossall
Posts: 4934
Joined: 5 Jan 2007, 10:01pm
Location: North Hertfordshire

Re: Risk and Freedom

Postby drossall » 14 Sep 2020, 10:15pm

mikeymo wrote:With all due respect, that demonstrates an incredibly simplistic view of the effect of mandatory helmet laws. Do you really believe that if 40% of cyclists pre legislation wear helmets, and after legislation 80% of cyclists wear helmets, that the only change is numerical? That the number of cyclists wearing helmets is the only change? Not the type of cyclists, or where it happens, the type of cycles ridden, the attitude of the remaining/new cyclists to risk?

Not for a moment. Agree with everything you said.

Mike Sales
Posts: 5162
Joined: 7 Mar 2009, 3:31pm

Re: Risk and Freedom

Postby Mike Sales » 18 Sep 2020, 12:53pm

mikeymo wrote:
Mike Sales wrote:Still trying to evade the subject.


Yes, you certainly are.

Mike Sales wrote:If I imagined Spiegelhalter to be a Jewish name and so stumbled, what do you think that tells you about the murky depths of my psyche?


I'll leave that to you to consider.


I know what I think, but I wonder why you think a error like this is important.

Mike Sales
Posts: 5162
Joined: 7 Mar 2009, 3:31pm

Re: Risk and Freedom

Postby Mike Sales » 18 Sep 2020, 12:55pm

I am disappointed but not surprised to find that some posters feel able to dismiss Risk Homeostasis whilst clearly not bothering to find out what the theory means.