NHS says "cyclists are slightly more to blame for fatal collisions"

[clarion]

That report is a very interesting exercise in the different ways of misinterpreting statistics.  A gift to Maths teachers to set as homework.

[hatler]

Paging jo, paging jo.

Critical statistical analysis please.

[Pickled Onion]

To understand the contributory causes to fatal cycle accidents, the Transport Research Laboratory has analysed data from 2005 to 2007.
<snip>
On average, there were 1.82 contributory factors associated with cyclists involved in a fatal collision and 1.60 contributory factors for drivers.
This suggests that cyclists are slightly more to blame for fatal collisions.

The Contributory factors in a road collision are the key actions and failures that the
police officer judges to have led directly to the actual impact and, as mentioned earlier,
should be viewed as indicative only as they result from a subjective view of the accident.
Under the STATS19 system, factors can be attributed to any of the vehicles or casualties
involved. This provides the opportunity in the case of cycle collisions to investigate the
extent to which the police reporting officers judged that primary contributory factors
were related to the cyclist or to others involved in the collision (this does not necessarily
mean that the cyclist or other road user was in fact responsible).

Incidentally, the TRL report does not discuss the number of contributory factors at all.

It goes on to say

For children, contributory factors were more likely to be attributed to the cyclist (in over
three-quarters of serious collisions). This proportion decreases with age, with cyclist
casualties aged over 30 being more likely to have the contributory factor assigned to the
driver. It is not clear whether this means children are more likely than adults to behave
in ways that result in a collision or whether the police are simply more likely to attribute
contributory factors to a child.

[Pickled Onion]

Another thing that caught my eye in that NHS thing:

The latest study into London's cycle hire scheme found that women were twice as likely to be involved in a fatal collision with an HGV, despite making up just 30% of the scheme’s participants.

This cannot possibly be a finding, given that there was only a single fatality amongst participants in the cycle hire scheme.

Between 2005 and 2011, injury rates in the cycle hire zone were generally highest by a substantial margin for motorcycling, followed by cycling, walking, car or van, and bus, and finally train, underground, and taxi (table 4). One notable observation was that the fatality rate among female cyclists was high relative to both other modes of travel and relative to men. This reflected a higher number of cycling fatalities in women involving heavy goods vehicles (weighing >3.5 tonnes). Twice as many women as men were killed in collisions involving heavy goods vehicles (15 women v 7 men), despite women only accounting for an estimated 30% of total cycling time and having half as many fatalities not involving heavy goods vehicles (5 women v 10 men, P=0.04 for a sex difference in probability that a cycling fatality involved a heavy goods vehicle).

[David Martin]

From one of the forum statistical literates.
After a considered and careful analysis, thorough peer review and literature search:

(click to show/hide)

It's bollocks. It is lamp-posting with digits, not even attempting to use calculations with a pretence of credibility.
Stop even trying to pick over the bones, it is a cockatrice.

[clarion]

It's the equivocation I can't stand, David

[jsabine]

After careful consideration of DM's posting, I got his first spoilered sentence exactly right (well, I'd expanded the contraction) before clicking to reveal.

[hubner]

OK but what if some moton posts this link in some non-cycling forum as evidence for whatever they're trying proof, eg cyclists deserve what they get if they're stupid enough to ride on the roads, ie cyclists are to blame if they kill themselves?

That is how I came across the link. So we need to put the facts right, in as concise and simple language as possible. Remember it's the NHS (ffs), so it must be true, right?

And the NHS should remove or completely rewrite that article.

[Steph]

Si - that's what the Highway Code tells peds to do - don Hi Viz. I'd imagine in a car park, doubly so.

And, yeah, we've been here loads of times before.

I know Pancho, and I just think its a sad reflection of society, boom-tish

A variant of the knee-jerk hi-viz for cyclists only attitude is that leaflet Stena dished out at Harwich, where cyclists were told to wear helmets and hi-viz on the vehicle deck, where they were required to walk amongst all that vehicle passengers heading for the stairs, who weren't required to wear helmets or hi-viz.

[David Martin]

OK but what if some moton posts this link in some non-cycling forum as evidence for whatever they're trying proof, eg cyclists deserve what they get if they're stupid enough to ride on the roads, ie cyclists are to blame if they kill themselves?

That is how I came across the link. So we need to put the facts right, in as concise and simple language as possible. Remember it's the NHS (ffs), so it must be true, right?

And the NHS should remove or completely rewrite that article.

Ask them to retract that article. It is not possible to criticise constructively, in the same way it is not possible to build a bridge with the music from a song about a river.

[Asterix, the former Gaul.]

The fact is that we no longer know  who the NHS is. Private companies are allowed to operate under the NHs 'banner' without revealing who they are. So we simply don't know  who wrote this report. It certainly wasn't the NHS as we used to know it.  We have to realise that.

[tonycollinet]

.......so we are saying that drivers are several times more likely than drivers to be at fault, and that is hard to understand in terms of any discussion on levels of training or road competence).

Not that surprising at all. Drivers who happen to be cycling are at that point very very much more motivated not to do stuff that'll get them KSId.


And re: the article. NHS core competence is (or should be) making people well. It is NOT in judging blame in RTA's, and unsurprisingly they've (or more likely one individual writing a stupid conclusion) have got it spectacularly wrong.

[drossall]

Quite. My point was that the general assumption is that accidents are caused by careless, incompetent, untrained cyclists. Actually, many are trained as drivers. And, as you point out, they are motivated not to have accidents in a way that drivers cannot be. All of which helps to explain why the stats don't support the presumption.

[P Walsh]

On the other hand, there have always been careless cyclists who do pretty stupid stuff. We all see them and cringe.I saw a middle aged man on a mountainbike, yesterday, who cycled around a corner on the road, mounted a dropped curb onto the pavement, cycled along the pavement for fifty metres to the next side road where he rejoined the road without a glance over his shoulder. Most of us don't do that sort of stuff, and I suspect that this dangerous minority of cyclists have a big impact on accident statistics. Without their "contribution" I think most of the injuries to cyclists on the road would more clearly be the result of driving errors.

[Regulator]

The fact is that we no longer know  who the NHS is. Private companies are allowed to operate under the NHs 'banner' without revealing who they are. So we simply don't know  who wrote this report. It certainly wasn't the NHS as we used to know it.  We have to realise that.

You don't half come out with some bollocks, you know...

NHS Choices is run on behalf of the Department of Health.  I've flagged the article to the Editorial Board and to the Department's PPS (who happens to be a cyclist)

[mrcharly-YHT]

The fact is that we no longer know  who the NHS is. Private companies are allowed to operate under the NHs 'banner' without revealing who they are. So we simply don't know  who wrote this report. It certainly wasn't the NHS as we used to know it.  We have to realise that.

You don't half come out with some bollocks, you know...

NHS Choices is run on behalf of the Department of Health.  I've flagged the article to the Editorial Board and to the Department's PPS (who happens to be a cyclist)

That's rudely unfair, Greg.
You might know which bits are private companies, which bits are NHS core, etc, but then you work in the system. We don't, and are exposed to a bewilderingly complex healthcare system.

[drossall]

On the other hand, there have always been careless cyclists who do pretty stupid stuff... I think most of the injuries to cyclists on the road would more clearly be the result of driving errors.

Possibly, but we need to be more evidence based. If safety measures are based on what people think, instead of what is actually happening, they will fail whatever they are.

There has been some evidence and speculation in the opposite direction. There's some reason to think that steady, predictable cycling leads to drivers passing closer and faster, and generally putting you more at risk. Whilst throwing yourself under a lorry will always be very dangerous, therefore, it's quite possible that having an air of unpredictability is a significant contribution to personal safety.

But I think I'd want to be more sure before proceeding on that basis.

[jsabine]

it's quite possible that having an air of unpredictability is a significant contribution to personal safety.

Nothing wrong with a tactical wobble.

[David Martin]

That's rudely unfair, Greg.
You might know which bits are private companies, which bits are NHS core, etc, but then you work in the system. We don't, and are exposed to a bewilderingly complex healthcare system.

That is not the case. Am I now getting a bill from a private company when I turn up for an X-ray? The answer is no, because, to put it in computer terms, the NHS API for the general public remains the same.
I turn up. I get treated. Free at the point of need.

It makes no difference to the treatment whether the syringes they use were bought from a mega-global earth-raping industrial behemoth, or locally hand crafted by a hippy collective out of organic recycled material. That is the delegated job of the NHS managers implementation.

[hatler]

Yebbut, we're not discussing treatment here, but the production of a report.

[mrcharly-YHT]

That's rudely unfair, Greg.
You might know which bits are private companies, which bits are NHS core, etc, but then you work in the system. We don't, and are exposed to a bewilderingly complex healthcare system.

That is not the case. Am I now getting a bill from a private company when I turn up for an X-ray? The answer is no, because, to put it in computer terms, the NHS API for the general public remains the same.
I turn up. I get treated. Free at the point of need.

It makes no difference to the treatment whether the syringes they use were bought from a mega-global earth-raping industrial behemoth, or locally hand crafted by a hippy collective out of organic recycled material. That is the delegated job of the NHS managers implementation.

I wish you were correct. But you aren't.
It's not who is supplying the syringes, it is who is delivering each part of a complex organisation. Communications between departments, people running machines, GPS is all made more complex by them being different companies.

Anyway, this is straying close to POBI.

[P Walsh]

It would be good to have a serious, in depth study of the causes of accidents involving cyclists because then we could have a properly informed approach to remedies. This report, though probably well meaning, is confused and confusing.

As for the attention grabbing assertion that "cyclists are slightly more to blame for fatal collisions", it may be true if it means that cyclist error initiated more than 50% of fatal accidents. If this is true, it means that for the more cautious amongst us the statistical risks are halved. I avoid making mistakes, so I am only half as likely to finish up being killed on the road. Does that make sense?

[drossall]

Sort of, but it's still out of keeping with other reports, as discussed much earlier in this thread, and therefore probably not true. Not sure that the 50% bit would follow, even if it were true.

[Bledlow]

They've clearly just added up the contributary factors numerically to get the 1.82:1.60 figure. Even if you could do that, which you can't, there would have to be some sort of scaling factor included regarding the size of the risk factor.

Indeed. That must have been written by someone who didn't have the faintest understanding of how to quantify risk.

Hypothetically, the number of risk factors for cyclists could be 10 times the number for motorists, but one of the motorist risk factors could account for 90% of the overall risk. The "count the number of factors" method used here would give something close to the exact opposite of the true risk ratio, in that case.

[Bledlow]

It's the unofficial hierarchy of the road:  People who think they're more important than you, as determined by your respective vehicles, expect you to give way in response to bullying.  This goes both ways, as I discovered the first time I drove a Man Car™ rather than a little hatchback, and found people being unexpectedly nice at roundabouts.

Mrs B commented that she started getting more courtesy from other motorists after changing from a 2006 Daihatsu Sirion to a 2012 Honda Jazz.    Just that little bit up the hierarchy ...