Considerate Cycling 51: Bristol Cyclists, Pedestrians, Police Data and Emergency Admissions to Hospital


This blog was co-authored by Rob Harding (who did all the real work)

 In Bristol, as in the UK as a whole, there are two main sources of information on road traffic incidents causing injury and involving cyclists. Each has its strengths and weaknesses.

Police incident records, known as STATS19 reports, give a specific location for the incident, a series of context details and a brief account of what happened. The reporting Officer also ticks one or more “contributory factors” from a long list to give some sense of “why” things had gone wrong. In some cases, from reading these reports, it’s possible to form a view on who was most at fault. We know, however, that there is a degree of under- or even mis-reporting and that police themselves have some concerns as to the accuracy and consistency of the reports. Some are completed at the scene, others might be done some time later and it is not hard to imagine that on busy days some might not be reported at all.

Bristol Council’s Highway and Transport Management Department use the STATS19 data to analyse trends and to rank contributory factors within the city (“failed to look properly” is far and away the most commonly cited factor). They also add the location and summary details of each incident to the map displayed on the Bristol – Pinpoint Local Information page. National data can be explored here on the UK Government data website or here on the excellent CycleStreets collision map, one incident at a time, with lots of detail.

Augmenting the Police data, Bristol City Council’s Directorate of Public Health produce figures for Bristol emergency admissions to hospital due to transport related accidents. These are records of Bristolian residents who arrive at hospital A&E Departments following transport related “accidents” (their choice of word). These figures comprise allBristol residents’ admissions whereever in the country the traffic-related injuries were incurred. They are thought to be more comprehensive than the Police reports but incidents are not identifiable by street location and, while they identify the types of road user, they don’t identify the more detailed factors that might lie behind the road traffic injuries.

So what does the Bristol Emergency Admissions data tell us about cyclists and road traffic incidents in Bristol?

In 2014 44 Bristol cyclists were admitted to hospital as a result of collisions with motorised vehicles. Separate studies, by BCyC members of incidents in Gloucester Road and Cabot Ward, suggest that a majority of these are associated with motor vehicle drivers “failing to look properly” or some other factor that relates to motor vehicle driver behaviour. As well as these casualties, 113 more cyclists were admitted to hospital as a result of non-collision transport incidents. Given the lack of narrative or the identity of specific locations, it is difficult to pin down causes and propose remedial actions for these non collision accidents but road surfaces (that could be remedied with prompt action) are associated with a significant proportion.

Plotting them against time of year shows peaks in December and June. A 2010 survey, biased towards commuters, suggested that 25% of incidents were the result of ice but that still leaves a large number unaccounted for.

As far as trends are concerned, in the last five years an average of just under 40 Bristol cyclists were hospitalised each year as a result of collisions with motorised vehicles. This compares with an average of just over 27 per year between 2003 and 2007 (the earliest years for which we have figures). While it is difficult to accurately adjust these figures to allow for the increased level (perhaps a doubling) of cycling in the city, they would not seem to suggest that cycling has got more dangerous. On the other hand it cannot be said that they demonstrate a clear “safety in numbers” effect either.

With regard to pedestrians, in 2014 73 Bristol people were injured as a result of collisions with motorised vehicles compared with 5 injured in collisions with cyclists. Put another way, 93.5% of pedestrians admitted to hospital were injured in collisions with motorists compared to 6.5% injured in collisions with cyclists. The latter figure includes those injured when stepping off the pavement and into the path of cyclists as well as those injured by pavement cyclists (it is not possible to break the figures down any further). The figures do show that the overwhelming number of pedestrian casualties are caused by collisions with motorised vehicles and, because lower collision speeds cause lesser injuries, provide clear evidence of the need for 20 mph enforcement, even if there was no reduction in the number of collisions. Injury increases with mass and with velocity and every day in Bristol there are, on average, 5 Police reports of road traffic casualty incidents of one sort or another.

Considerate Cycling 42: Some Simple Numbers

Figure 1 (source: DfT)

I was interviewed on BBC Radio Bristol recently about the relationship between cyclists and buses in the city. I talked about cyclists’ respect and bus drivers’ professionalism. However, I did suggest that putting cyclists into a shared lane with buses was a bit like putting the meerkats in with the elephants. I also said that compared with goods vehicles and cars, buses were much less of a problem than people might assume.

I was talking about Bristol, nowhere else. I had checked some stats before I went in to talk, so I was fairly happy that what I was saying had some grounding in what had already been recorded and counted. Afterwards I started thinking about whether the simple numbers I had could support the hypothesis that despite their lower numbers, buses and goods vehicles were in some way safer or else more dangerous than cars. I showed the statistics to my son who is a proper mathematician and a data manager by trade. He pointed out my basic problems with standardisation of such different populations and gradually one thing became very clear.

The clear point was that at some level I was trying to find statistical evidence to confirm my prejudice that cars were in some way to blame even though vans, lorries and buses seem so dangerous. I was looking at small differences of probability where none were to be found. What I should have been staring at was Figure 1. The plain fact is that there are over 230,000 cars being kept in Bristol. That’s roughly one car to every 2 people. In comparison, there are just 39,200 goods vehicles, less than 1 van or lorry to every 5 cars. Buses and coaches add up to just 2,400 vehicles, only about one to every 10 cars. What about some more simple numbers? Here is Figure 2

Figure 2 (Source: correspondence with Bristol City Council)

Looking at this, what can we do at a local level to reduce bicycle casualties? What would a modest 10% improvement look like for each vehicle type? Haul back injuries for car-related incidents and we would avoid 47 injuries. 10% of the goods vehicle total  would save 3. 10% of bus-related injuries would save 1.

So what might reduce car-related injuries? Some more simple numbers offer a clue:

Figure 3 (Source: DfT)

The number of cars in Bristol has gone from 184,019 in 2004 to 231,853 in 2013. That’s a 26% increase in what was already a large number. Most of that has been in the last four years.

Here’s a photograph of a Bristol street. Recent political turmoil in the City has raged over Mayor George Ferguson’s attempt to restrict car parking and reduce speed limits.

Picture 1

What does it look like to you? What, if you had the power, would you want to control if a better and safer city was your aim? What would you prioritise if you wanted to reduce road traffic casualties and improve public health? How you do it is up to you, but I would say that the answer stares us in the face.