Road Safety News
 

DfT stats show hike in casualties

Friday 8th August 2014

The latest road casualty statistics published yesterday (7 August) show a year on year increase for the 12 months ending March 2014 – and a sharp hike in Q1 2014 casualties compared with Q1 2013.

Often quarterly increases can be attributed to weather patterns (heavy rain, snow, extreme cold etc) but in this instance the DfT says the reasons for the increases “aren’t fully clear”.

The year on year stats show a 4% increase in road deaths (1,750) and a 2% increase in KSIs (24,160) during the year ending March 2014. There were 189,880 casualties of all severities, 1% higher than the 188,884 for the year ending March 2013.

The overall KSI increase was due to a rise in pedal cyclist and motorcyclist KSIs which both increased by 7%.

Traffic levels increased by 3% in the same period which means the overall casualty rate per vehicle mile decreased by 2% for the same period.

Looking at Q1 2014 (Jan-Mar), there were 380 road deaths, a rise of 13% compared with Q1 2013. KSIs and slightly injured casualties also increased by 17% and 15% respectively. The overall casualty rate for all severities increased by 10%.

Both child casualties of all severities and child KSIs increased, by 21% and 17% respectively. Child pedestrian casualties and KSIs also increased, by 15% and 9% respectively.

Over the same period, traffic levels increased by 5.1%, and as a result, the casualty rate per billion vehicle miles increased for all casualty severities compared to the same period in 2013.

However, the DfT points out that Q1 2013 was the lowest quarter for casualties since detailed reporting began in 1979. Fatalities in Q1 2014 were still at their second lowest level and KSIs, slight injuries and total casualties at their third lowest level since detailed records began.

Commenting on the Q1 figures, the DfT bulletin says: “The first quarter of 2013 was very cold which may have reduced the numbers of vulnerable road users (particularly pedal cyclists and motorcyclists) on the road, reducing their exposure to accidents.

“In comparison, the first quarter of 2014 was milder and above average in temperature, despite heavy rain.

“2014 marked the third wettest January and February since records began and several areas in Great Britain experienced severe flooding. We might have expected the heavy rainfall to have reduced vulnerable road user activity during Q1 and thus suppress casualty numbers.

“However, given the large increases in casualties for all road user types in this quarter (particularly pedal cyclists and motorcyclists) it is reasonable to assume that the high rainfall did not have the expected impact. It’s also possible that had rainfall been lower in this quarter there could have been even more casualties and thus a larger increase on the same quarter in 2013.

“In conclusion, at this stage it is difficult to say what has caused the increases in the latest quarter on quarter and rolling year comparisons. However, as more 2014 data becomes available later in the year, we will get a clearer picture of the longer-term trend.”

 

Comments

Comment on this story
Report a reader comment

What's your view - comment on this story:

I confirm that I have read and accept the moderation policy and house rules relating to comments posted on this website.
Your comment:
Your name and location:
Your email:

Thank you Idris for your support to my statements. However when I said that someone must doing something right, I didn't particularly mean the road engineers etc. who to my mind have made gigantic failures in places in the past and only now are we seeing to what extent they need to change and rethink their attitudes to accommodate all vehicular traffic of whatever mode.

Perhaps what I may say is that it's the general driving public that have improved and caused the lessening that we all desire. Was it not mooted on this site not long ago that education comes to nothing.....unless the recipient take it on and practise what has been learned. Good for the Public. They must have been listening.
bob craven Lancs

Agree (3) | Disagree (2)
+1

I am not sure what Idris's point and how it contradicts my comment. There may be 200 automatic speed sensing sites across country but whether they will show any changes in speed will be very much dependent upon road type, position, congestion, etc. Even so this does not provide any evidence that it is maintaining speeds that has reduced casualties or that a reduction in speed would not result in even lower casualties.
Rod King, 20's Plenty for Us

Agree (9) | Disagree (4)
+5

I'm inclined to agree that a year on year data comparison has very limited value - certainly within this forum - especially when a rich history of data is available. But everything has a customer.

Recently Devon has started producing annual collision reports that have the following pattern:

1. A 10 year trend to illustrate the extent to which our broad strategic approach is working.
2. A five year trend to help illustrate the extent to which our operational work is contributing to, or detracting from, that 10 year trend.
3. Tests for statistical significance to show the extent to which the last data year is or is not in line with the five year operational performance.

Each of these data sets has an audience - and despite all that, you'll not be surprised to learn that there is still a constituency that want to know how this year compares with last. Suffice to say we have macro shortcuts for the data health warnings...

So, DfT will have an audience for this - but I suspect not so much in here. We need something a little more substantial to satisfy our needs.
Jeremy, Devon

Agree (5) | Disagree (0)
+5

Wrong again Rod.
I have all the data that exists, since 1926. Casualty rates per vehicle mile have fallen steadily ever since 1950 when mileage records began. Since the late 1960s when mileage increases were trumped by safer roads and vehicles, casualties have fallen very substantially despite many more vehicles on the roads.

The record shows that average speeds across the country, monitored 24/7 by about 200 automatic systems spread across a representative cross-secrion of roads, have hardly changed for years. The record shows that average speeds across the country - monitored.
Idris Francis Fight Back With Facts Petersfield

Agree (7) | Disagree (6)
+1

Bob is right - I recently spotted in Stats19 data more than 30 serious injuries in a single accident.

Also there are several police areas including Thames Valley and South Wales where SI shot up by 60% or so in a single year. They didn't of course, all that happened was that someone noticed that the ratio of SI to slight in those areas was very different from elsewhere so the police officers were re-trained in how to differentiate the two severities. Difficult and highly subjective as it is.

Bob is also right about trials - no conceivable trial could produce more than a tiny fraction of the data which already exists and would therefore be a total waste of money.

As for "someone doing something right" - yes, primarily vehicle and component design engineers and road design and construction engineers. But most certainly not those who maintain the roads.
Idris Francis Fight Back With Facts Petersfield

Agree (4) | Disagree (8)
-4

Just chasing statistics won't do a great deal of good as the DfT says, because studies like this cannot make any clear links between causes and effects. Rather than continuing to gather ambiguous statistics that can be misinterpreted wouldn't it be a good idea to start gathering statistics that can't? Florence Nightingale showed that this is perfectly achievable when she showed that more soldiers were dying of infection than they were of wounds.

I would start by getting rid of KSI as the baseline as it's a quite meaningless figure. Much better to have impacts as the baseline as they are not reliant on post-crash interventions or the vagaries of luck. It would then be possible to compare all the hundreds of variables such as weather and economic conditions for example to the hard baseline and see what effect each one of them has.
Duncan MacKillop, Stratford on Avon

Agree (6) | Disagree (4)
+2

Lies, damn lies and statistics. Lots of number and confusion as to what they mean to whom and why. Maybe that is why the DfT don't like targets.

The penultimate paragraph states "It’s also possible that had rainfall been lower in this quarter there could have been even more casualties and thus a larger increase on the same quarter in 2013."

Should we pack it all in to learn how to do a rain dance?

Basic Guidelines on How To Rain Dance
(everyone/tribe is unique, be creative, remembering to honor the earth. nothing can be taken less something given.)

1. Never do a rain dance on a hill.
2. Make sure you have a lot of room so you don’t run into anything.
3. Spin around in clockwise circles.
4. Make up your own rain chant. It should be rhythmical and easy to say fast.
5. Yell your rain chant while spinning around in circles.
6. If you are trying to get rid of rain, spin in counterclockwise circles and say your chant backwards.

As taken from a book of Native American Activities.
Peter, London

Agree (10) | Disagree (0)
+10

It ain't rocket science. I think there is plenty of evidence as to what works to reduce road danger:-

Fewer motor vehicles
Slower motor vehicles
Less aggression on the roads
More segregated facilities
More focus by police on road bullying, speeding and intimidation
Realistic sentencing in courts for offenders

And whether we like it or no the data isn't available to track all the individual aspects that anyone would wish:-

DfT doesn't track changes to any speed limit or total miles
DfT doesn't any more track school travel with an annual census
DfT is loading more and more on traffic authorities resulting in inconsistent interpetations of national guidelines which themselves are fuzzy and imprecise.

Local and national government need to get on with doing what it knows works with rather more conviction than is being done at the moment.
Rod King, 20's Plenty for Us

Agree (9) | Disagree (5)
+4

Wouldn't do much good for job satisfaction if we went back 10 years as each year we seem to get less and less injuries, accidents, killed etc. So someone somewhere must be doing something right. But it appears that nobody knows what? I believe that we have all the stats that we need and interventions are required. As for the argument of having scientific trials, how long would they go on for, 3 yrs, 5 yrs maybe - and would they reach a conclusion accepted by all? I think not.
bob craven Lancs

Agree (8) | Disagree (5)
+3

I would be very surprised if the increase is in the younger sector ie below 24 year olds and would put it down to three changes to the motorcycle test discouraging youngsters to have further training and pass their bike test as there were 3000 less 18 year olds passing their test in 2014 than in 2013.
Chris White. Rainham kent

Agree (7) | Disagree (3)
+4

"However, as more 2014 data becomes available later in the year, we will get a clearer picture of the longer-term trend.” No. All the data for trend analysis exists - look back at the last 10-20 years! I agree with Idris and others regarding year-on-year reporting - please plot graphs over the last 10 years. And let's plot casualties against roll-out of 20mph limits.
Eric Bridgstock, Independent Road Safety Research, St Albans

Agree (8) | Disagree (9)
-1

Can I also just say that one incident, say in fog on a motorway, involving numerous cars and HGVs etc driving too close together (tailgating) can have a devastating effect on one quarters stat returns. Just one incident can colour the whole year. Circumstances like that have to be identified and therefore born in mind in any calculations and statistics.
bob craven Lancs

Agree (14) | Disagree (0)
+14

Just a blip? Can reach no conclusions? I don't like seeing motorcycle accidents up other than saying it was quite a good early spring. I did say early on that if this weather continued there would be more incidents involving bikers.
bob craven Lancs

Agree (8) | Disagree (2)
+6

Changes of this order from one quarter to another a year later have no statistical significance. End of story.
Idris Francis Fight Back With Facts Petersfield

Agree (15) | Disagree (6)
+9