OPINION: laser-like focus on the numbers that matter needed to reduce road deaths

10.29 | 19 October 2022 | | 12 comments

In this opinion piece, David Davies, executive director of PACTS, outlines how the DfT’s new ministerial team can quickly make “serious inroads” into reducing the number of road deaths.

We have (when I last looked) four ministers at the Department for Transport with some responsibility for road safety. Under the Secretary of State Anne-Marie Trevelyan, who I’m told has been asking searching questions about road safety, there is:

  • Katherine Fletcher MP, Parliamentary Under Secretary of State. She has responsibility for roads and lots more. Road safety is not listed in her portfolio but that’s nothing new. It seems to be the new love that dare not speak its name. She will lead on road safety. She is also Minister for Women.
  • Rt Hon Lucy Frazer KC MP, Minister of State, has responsibility for the future of transport and freight, decarbonisation and more that is road safety related. this will presumably include autonomous vehicles, e-scooters, electric vehicles and more.
  • Baroness Vere of Norbiton, Parliamentary Under Secretary of State, remains at the Department for Transport. She previously held the road safety brief and deserves credit for establishing the Road Safety Investigation Board. She now has responsibility for local transport and active travel and will answer on all transport matters in the House of Lords.

In fact, only one DfT minister, Kevin Foster MP, does not have a related responsibility. His portfolio is all rail.

Like any new ministerial team, they have a huge workload and will be wondering where to start. Will they pick up the (almost published) Road Safety Strategic Framework or start anew?

PACTS and other road safety groups promotes the Safe System. We want to see it firmly established in the UK. But it’s a lot to digest and will take time to put in place. There are endless road safety statistics. If ministers want to make a difference quickly, here are some they might look at.[1]

  • 1,558 people were killed on the roads in Great Britain in 2021.
  • Car occupants (682), pedestrians (361) and motorcyclists (310) make up 87% of all road deaths. Any serious strategy has to include ways to reduce casualties for these groups.
  • Per billion passenger miles, HGVs and motorcycles were by far most likely to involve the death of another road user.[2]
  • Around one third of people who died on the roads were driving for work. Working with employers would be more efficient and effective than with millions of private citizens.
  • In 25% of fatal collisions, “Exceeding speed limit or Travelling too fast for conditions” was recorded as a contributory factor.[3]
  • 30% of those who died in cars were not wearing a seat belt. That’s 200 deaths, many of which would have been prevented.
  • In 14% of all road deaths (220 people) the driver was over the legal blood alcohol limit. Deaths resulting from drug driving are harder to calculate but the DfT suggests this may be on the same scale or higher.[4]

The Secretary of State is a chartered accountant by trade and sat on the Public Accounts Committee, so obviously good with numbers. We need a laser-like focus on the numbers that matter and not a lengthy action plan with a bit of something for everyone. We know a lot about what works and also, if we’re honest, what does not. The interventions needed are all part of the Safe System. Many of the most effective actions will reduce danger and benefit all road users.

[1] Mostly from DfT, Reported Road Casualties Great Britain 2021, September 2022

[2] Reported road casualties Great Britain: road user risk 2021 – GOV.UK (www.gov.uk)

[3] RAS0704 Exceeding speed limit or Travelling too fast for conditions, 314/1339

[4] Developing drug driving statistics: initial feasibility study (updated) – GOV.UK (www.gov.uk)


Comment on this story

Leave a Reply

Your email address will not be published. Required fields are marked *

Report a reader comment

Order by Latest first | Oldest first | Highest rated | Lowest rated

    Hi David,

    Yes I know that the DfT get their data from traffic surveys. I’ve had this discussion with a variety of civil servants in Northern Ireland and in the DfT. Given all the issues I raised i.e. seasonality etc, you need to couple this with the admission of the DfT that what is critical in the analysis is – “critical distance travelled estimates, derived from a sample survey, will have a higher level of uncertainty for small sub-groups of the population such as motorcyclists” – “further pooled to help minimize any random variation”.

    Furthermore, the issue I have, is that when information like this is transferred to the real world, the “interpretation” of those data offers politicians and road safety campaigners a stick to beat the huddled masses with – I include your good self (I refer to your comments about risky motorcyclists in London). Please find the link to an article I wrote in 2015, (though still valid) about this very issue of identifying motorcyclists as risk takers –


    You cannot realistically tell me that you are in agreement that motorcyclists are as culpable as HGV drivers in causing the death of other road users. You and your road safety colleagues have to have a good think about using these data if they are intended to identify “risky road users”. It doesn’t work like that. There have to be solutions that actually make a difference.

    So if you wish to point to the DfT for their brilliant analyses of data (??) – do so. But all you are doing is just kicking the can down the road. I am happy to have a dialogue with you about these issues, but I WILL point out errors and what I consider incorrect interpretations when I see them as in this case.

    Finally, as I wrote – If you put your head above the parapet, you should expect a response. No? I would expect the same. As per my request – why not invite me to one of your meetings, I would be quite happy to present a more realistic interpretation of motorcycle crash statistics.

    Kind regards,

    Elaine Hardy, Belfast and Cabanas de Tavira
    Agree (3) | Disagree (3)

    From now on then, I suggest we only talk about reducing collisions per se, not just ‘collisions which result in death’ or ‘collisions which result in injuries (however serious)’. The common factor to all, is the road user behaviour which brings about the collision in the first place, the consequences of which cannot be predicted.

    Hugh Jones, Cheshire
    Agree (4) | Disagree (4)


    The stats per billion passenger miles I quote (ref 2 in the article), are from DfT. They do traffic surveys to get the mileage and divide it into the number of casualties. They know the other casualty involved from the police Stats19 report.

    Simples! This reporting is increasingly common practice.

    David Davies, London
    Agree (2) | Disagree (2)

    I think Hugh has it the wrong way round (but bear in mind my experience is that of an engineer). I would suggest that any intervention aimed at reducing fatalities would probably have a beneficial effect on injuries of lesser degree. I would hesitate to suggest proportions, because, in treating black sites, one very quickly finds that some sites will produce little other than fatal accidents, others little other than slight accidents. It is, therefore, most important to identify and treat the latter first (regardless of the casualty class involved, of course). As to accidents resulting in life-changing injuries, I am not sure whether we can actually identify these or whether the contributory factors are the same as those leading to fatalities. Perhaps the unfortunately named “CRASH” data collection system will help, there, as I understand that its casualty classification system is rather more fine-grained. Perhaps not, because whether injuries are life-changing or not may not be clear immediately.

    Fraser Andrew, STIRLING, UK
    Agree (3) | Disagree (0)

    It would appear that the opinion of the author has cherry picked the data. PACTS own report https://www.pacts.org.uk/wp-content/uploads/PACTS-What-kills-most-on-the-roads-Report-15.0.pdf has some stats to cherry pick pick from as well.

    “Pedal cyclists and small motorcycles were involved in very few collisions where pedestrians were killed”

    “Very few vulnerable road user deaths resulted from collisions with pedal cycles or
    motorcycles 50cc and under”

    “More cars are involved in collisions in which motorcyclists are killed than any other single vehicle type”

    “Far more pedestrians, motorcycle riders and car drivers/passenger are killed in collisions with HGVs than cyclists”

    “Most deaths associated with walking, cycling or motorcycling are of the pedestrians or riders themselves”

    So while not wishing to be rude, one persons opinion on stats to fit the Safe System agenda they are promoting does not make it so!

    Trevor Baird, Belfast
    Agree (3) | Disagree (1)

    Me again, Sorry David, I should have included a link to my report “The Dynamics of Motorcycle Crashes. Here:


    On page 60 and 61 is a breakdown of what the motorcycle crashed with. You may find this information useful.

    Elaine Hardy, Belfast and Cabanas de Tavira
    Agree (3) | Disagree (0)

    Hi David,

    The Safe System approach, as all perspectives, has its flaws. In this case using billion miles travelled. For instance, due to the different methodologies of gathering data throughout the world, as suggested in the report, it is difficult to grasp the overall problem of collision fatalities which is exacerbated by the use of million (or billion) kilometres (or miles) travelled as a means of identifying risk.

    The reason for this is that for example in Northern European countries, PTW usage tends to be far less than car usage and seasonal and therefore incompatible with other forms of transport and with PTW usage in southern European and Low Income countries.

    Furthermore, these trends or extrapolation of figures, do not explain the reasons behind the collisions that lead to injuries – slight, serious or a fatality, because as we all know, each road traffic collision is unique.

    But in particular regarding the analysis you have presented, I would like to understand how on earth somebody came up with “Per billion passenger miles, HGVs and motorcycles were by far most likely to involve the death of another road user”. Really?

    I can understand why HGVs would be more likely to involve the death of pedestrians, cyclists, motorcyclists and even car occupants – but motorcycles? How do you square that? Or are you including cats, dogs and sheep? In all my research – which as you are fully aware, is considerable regarding motorcycling – there is no evidence that I am aware of that motorcyclists in European countries are any more likely to kill other road users than even cyclists. By implication you are suggesting that motorcyclists kill more other road users that car drivers??

    Returning to the issue of billion miles travelled. Perhaps this video may help to give a different perspective.



    Elaine Hardy, Belfast and Cabanas de Tavira
    Agree (3) | Disagree (3)

    My translation of laser-like is a short waffle-free set of actions that link the action (date bound) to a numerical outcome. The statistics provided are a good set of data to get the ball rolling.

    The NHS is in crisis and road collisions need to be factored into this. Simple actions like NHS investing in average speed cameras would be a start.

    Peter Whitfield, Liverpool
    Agree (4) | Disagree (4)

    The PACTS vision IS a transport system free from death or life-changing injury. https://www.pacts.org.uk/achieving-our-vision/

    In this piece I was trying to keep it simple.

    If you are correct that
    “The causation factors for both consequences are the same anyway, so any interventions to reduce collisions as a whole, are inevitably going to reduce deaths and injuries at the same time.”

    why complicate matters by going over all those stats too?

    David Davies, Westminster
    Agree (2) | Disagree (3)

    Sorry to read that we are no longer considering road traffic accidents. Not every death results from a collision, as these are normally understood. Concentrating on collision means that the reporting officer miss the most important factor in an accident. (Reporting officers often do, of course, as anyone working with their data will know.)

    Actually, I have yet to hear of a sound argument for the change that has been foisted upon us.

    Fraser Andrew, STIRLING
    Agree (2) | Disagree (4)

    Again – speeding again is the focus despite being number 6 in the list of contributory factors, with other causes remaining untackled because using speed cameras is easy and profitable. In most counties road policing is next to non existent. Regarding motorcycles, unlicensed commercial trainers are still being allowed to train riders for profit, due to a loophole in the law which needs to be closed.

    Mike Abbott, Newark
    Agree (7) | Disagree (5)

    I’ve asked this before: why is preventing road deaths seemingly the only thing that matters in road safety? Is preventing life-changing injuries somehow not important? The causation factors for both consequences are the same anyway, so any interventions to reduce collisions as a whole, are inevitably going to reduce deaths and injuries at the same time. It’s a bit short-sighted to only talk about reducing deaths on the road.

    Hugh Jones, Cheshire
    Agree (8) | Disagree (2)

By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.