What happens to casualties and collisions when a 20mph scheme is implemented?
It’s a question that has been debated endlessly in discussion threads on this newsfeed. Which prompted us to see if we could shed any light on the subject.
The scope of our investigations
We focused on what happens to collision and/or casualty rates after a 20mph signed limit or engineered zone has been implemented. Did casualties/collisions increase or decrease, and how does the rate of any increase or decrease compare with that experienced across a wider area?
While some are mentioned in passing, we did not focus on any of the following:
• Severity of casualties (fatal, serious etc)
• Type of casualties (pedestrians, motorcyclists etc)
• Casualty/collision rates on neighbouring roads.
• Vehicle speeds
• Traffic volumes
• Public perception/support etc.
What data did we find?
It will not come as a shock to anyone who is involved with the 20mph issue to hear that there is a distinct lack of data available across the UK. And the data that is available is not collected and/or collated in a uniform manner – which makes analysis very difficult.
There are a number of areas where 20mph schemes have been introduced but ‘before’ and ‘after’ casualty/collision data is not available. There is data from a very well-established scheme in Hull but it dates back to 2002, which in our view makes it too old to be considered in this article. And the scheme in Edinburgh is at a relative early stage and as such the ‘after’ data is not yet available.
We did, however, manage to obtain ‘before and after’ data for 20mph schemes in the following 10 areas: London, Portsmouth, Warrington, Bristol, Birmingham, Newcastle, Middlesbrough, Gateshead, Hertfordshire and Derbyshire.
Some of the schemes are relatively small and as such the number of casualties/collisions is also small. This should be borne in mind when analyzing the data.
What did the data tell us?
• In five areas - London, Birmingham, Portsmouth, Newcastle and Gateshead – collisions and/or casualties fell at a significantly greater rate after a 20mph scheme was implemented, than would have otherwise been expected.
• In Warrington, while collisions fell at a significantly greater rate after the scheme was implemented, the number of casualties increased slightly.
• In Middlesbrough there was little or no significant difference between collisions/casualty rates in the 20mph scheme and the rest of the town/city.
• In Bristol the rate of collision reduction in the 20mph pilot areas was lower than across the city as a whole.
• In Hertfordshire and Derbyshire, while average collision rates in the 20mph schemes fell by 45% and 36% respectively, comparison with rates across the rest of the county is not possible (see below for more detail).
The following table summarises the findings. It shows the areas where collision reductions were higher in 20mph schemes than elsewhere; those where the reductions were lower; and those areas where it is not possible to compare. Best practice suggests that schemes should be evaluated over a minimum six-year period (three years before and three years after) in order to have sufficient data - three of the schemes do not meet this time period. The table also shows the difficulties of comparing the current evaluations of 20mph schemes.
• There is an urgent need for a national initiative to record and report in a uniform way data relating to 20mph schemes. Given the investment that is being made in this area, it is perhaps very surprising that this is not already taking place.
• The lack of clearly-presented data makes it very difficult to conclusively assess the impact of 20mph schemes with regard to casualties and collisions.
• Having said that, in half of the 20mph schemes we looked at, casualties and/or collisions were significantly lower than would otherwise have been expected – and in only one of the schemes were collisions higher than would otherwise have been expected.
• While not suggesting that the 20mph schemes are the only reason for these reductions, it is perhaps reasonable to assume they have played a part in this.
Results area by area
Here is a summary of the data for each area, along with links to full reports where these are available.
London (self-enforcing engineered zones)
Summary: a large-scale study over an extended period which concluded that “20 mph zones are effective measures for reducing road injuries and deaths”.
A team of researchers from the London School of Hygiene and Tropical Medicine carried out a study titled the “Effect of 20 mph traffic speed zones on road injuries in London, 1986-2006”. Their objective was “to quantify the effect of the introduction of 20 mph zones on road collisions, injuries, and fatalities in London”, and the report was published in 2009.
The researchers analysed Police STATS19 data for the period 1986-2006, to identify changes in counts of road injuries. Estimates of the effect of introducing 20 mph zones on casualties within those zones and in adjacent areas were adjusted for the underlying downward trend in traffic casualties.
The researchers concluded that the introduction of 20 mph zones was associated with a 41.9% reduction in road casualties, after adjustment for underlying time trends. The percentage reduction was greatest in younger children and greater for the category of killed or seriously injured casualties than for minor injuries. There was no evidence of casualty migration to areas adjacent to 20 mph zones, where casualties also fell slightly by an average of 8%.
Given the scale of the study, it was possible to remove data for three, four and five years before the introduction of the zones in order to account for any high casualty rates in site selection periods. The removal had little effect on the results of the casualty outcome and suggested that regression to the mean is not the explanation for the observed effects.
The general trend in casualties and collisions over time in London, an annual decline of 1.7%, was equivalent to a 15.8% reduction over 10 years or a 29.0% reduction over 20 years. Thus, in broad terms, the additional effect of the 20 mph zones was that of a step reduction in casualties and collisions by an amount that has taken over 20 years to achieve on roads without 20 mph zones.
The study concluded that: “20 mph zones are effective measures for reducing road injuries and deaths”.
Portsmouth (signed speed limits)
Summary: mature, area-wide scheme under which collisions have reduced by 18.8% on the roads where a new 20mph limit has been implemented, compared with 10.3% across the rest of the city’s streets.
Portsmouth City Council was the first local authority in England to implement an extensive area-wide 20 mph speed limit scheme covering most of its residential roads which previously had a 30 mph speed limit. The scheme, which covers 94% of the city’s residential road length, was implemented in 2008.
Analysis has been undertaken of all injury collisions which occurred in Portsmouth for the three years before the 20mph scheme was implemented (2005-2007) and for the three years after (2009-2011). Collisions occurring in 2008 have been excluded as this was the implementation year.
All injury collisions were included to ensure a large sample size and to avoid any skewing of the data where severity levels might have increased with higher speeds.
The analysis shows that in the three-year period prior to the scheme there was a total of 505 collisions. The corresponding figure for the three-year period following implementation is 410 collisions, which equates to a reduction of 18.8%. Across the rest of Portsmouth’s roads, during the same period the number of collisions fell by 10.3% - from 1,618 to 1,451.
Warrington (signed speed limits)
Summary: 18-month pilot during which collisions reduced significantly but casualties increased slightly.
Warrington Borough Council carried out a 20mph pilot in three areas in the town where there was a history of higher casualties among vulnerable road user groups. The pilot ran for an 18-month period from February 2009 to August 2010.
In order to determine the true impact of the 20mph limits, the project team looked at the borough-wide collisions and casualty trend using the same ‘before’ and the ‘actual’ trial period. In percentage terms this showed an authority wide reduction of 11.9% in the number of collisions and 8.1% reduction in the number of casualties.
Taking the borough-wide reductions into account, the pilot resulted an overall reduction of 13.7 collisions (25%) over and above that which might have been expected – from 53.7 to 40. However, with regard to casualties there was an increase of three – from 59 to 62. The casualty increase was in one of the trial areas – the town centre. In the other two areas, casualties were slightly down.
Bristol (signed speed limits)
Summary: two pilots covering 500 roads and 30,000 households, in which collisions in the pilot areas reduced at a slower rate than across the city as a whole.
Bristol City Council ran two area-wide ‘signs only’ 20mph pilots; the Inner South scheme was implemented in May 2010 and the Inner East scheme in October 2010. The two schemes covered 500 roads and 30,000 households.
In the two years before the pilots there were 382 collisions across both areas. In the two years after implementation there were 364 collisions, which equates to a reduction of 4.7%.
With regard to collisions across Bristol, looking first at the South pilot, in the two years before the pilot there were 2,201 collisions across the city. The ‘after figure is 2,034 which equates to a reduction of 7.54%.
Looking at the East pilot period, the number of pre-pilot collisions was 1,981 and the number of post collisions was 1,784, which equates to a reduction of 9.9%.
It must be concluded, therefore, that the rate of collision reduction in the 20mph pilot areas was not as great as it was across the whole of the city.
North East England (all signed speed limits)
The North East Regional Road Safety Resource has produced reports analyzing 20mph schemes in three of the region’s conurbations – Newcastle, Middlesbrough and Gateshead.
The data for Newcastle and Gateshead led the authors to conclude that “whilst it should not be implied that the reduction in speed limits is the only cause, these statistics show that it is probable that these 20mph areas have helped to further reduce the number of collisions in these areas, over the reductions seen through the rest of the city in the same period”.
However, in Middlesbrough the 20mph schemes have yielded no collision reduction benefits, with identical reduction of 7% recorded in the 20mph areas and across the rest of the city.
There have been two phases of 20mph speed limit deployment in Middlesbrough; phase one ran March-June 2012 and phase two March-June 2013. The second phase has not yet been analysed.
Comparing the 14 months prior to implementation (Jan 2011 – Mar 2012) with the 14-month period post implementation (July 2012 – August 2013), the number of collisions in the new 20mph areas fell by 7%. However, the number of collisions right across Middlesbrough in the same period also fell by 7%.
A combined analysis of phases one and two is scheduled for 2014 when there will be at least one full year’s collision data available for each phase.
There have been six phases of 20 mph deployment in Newcastle, each containing several 20 mph zones that have been in force for at least a year up to the end of June 2012.
The data shows that overall collisions in these areas dropped by a yearly average of 25% following the introduction of the 20 mph zones.
The North East Regional Road Safety Resource says that “if collision reduction in 20 mph areas had matched the average reduction in Newcastle, we would have expected to see 21 fewer collisions per year between the before and after periods. However, we actually saw 59 fewer collisions per year, showing that there are 38 fewer collisions across the 20 mph areas each year than would probably have been the case if they had not been in place”.
In Gateshead, there are 21 20mph zones that have been in force for at least a year up to the end of March 2012.
Overall, there was an annual reduction of 7.5 collisions (13%) across all of the zones following their introduction. This equates to three fewer collisions per year than would be expected given the overall collision reductions across the rest of Gateshead.
Birmingham (engineered zones)
Summary: small sample, collisions and casualties both down, but comparison with pan-Birmingham casualty trend in same period is difficult.
As part of the Inner City Safety Demonstration Project, Birmingham City Council collected ‘before and after’ data in four areas where 20mph zones were introduced. The data relates to the three years prior to introduction, and three years post introduction. All of the data was collected during the period January 2005 – September 2012.
In total, across the four areas there were 44 collisions and 56 casualties during the pre-20mph period. The corresponding figures for post-20mph installation were 23 collisions and 30 casualties.
In terms of severity, there were no fatalities either before or after 20mph installation. Of the casualties in the pre-20mph period, four were serious and 52 were slight. The corresponding figures in the post-20mph period were two serious and 28 slight.
Drawing a direct comparison with collision/casualty reductions across Birmingham as a whole is not easy, because the 20mph trials were conducted at different times.
The reductions in collisions and casualties in the post-20mph period equate to 47% and 46% respectively. Across Birmingham, the number of collisions in the three-year period 2010-12 decreased by 35% compared with the period 2005-7.
Hertfordshire (engineered zones)
Hertfordshire’s highways’ team has provided data for 17 20mph zones in five conurbations across Hertfordshire – Welwyn Garden City, Stevenage, Watford, Ware, Hitchen and Hoddesdon.
However, only 11 of the sites have been operating for a minimum period of three years. Of the remaining six, three were only implemented in January 2013, and as such have been discounted from the analysis. A further three will reach three years in the first quarter of 2014, and we have adjusted the number of collisions to reflect this.
Across the 14 zones, the number of collisions in the three years prior to implementation of the 20mph zone was 30. The number of collisions in the three years post-implementation (including the adjustment referred to above) was 16.44, which represents a reduction of 45.2%.
Because the schemes were implemented at different times between December 2001 and March 2011, it is impossible to compare this with overall collision figures across Hertfordshire.
Derbyshire (engineered zones)
In Derbyshire, 26 20mph zones have been implemented between the years 2000 – 2012, in towns and villages across the county. We have excluded the two most recent schemes (in Matlock and Dronfield) on the grounds of insufficient ‘after’ data. Another scheme (Brimington) was introduced in November 2010, and while we do not quite have a full three years of data we have included this in the analysis.
In the three years prior to installation there were 106 collisions in areas where the zones were introduced. In the three years after installation the number of collisions was 68 which equates to a reduction of 36%.
As with Hertfordshire, because the schemes were implemented at different times over an extended period, it is impossible to compare this with overall collision figures from across Derbyshire.
For more information contact Nick Rawlings, editor Road Safety News, on 01379 650112.