Road Safety News

Casualties rise on roads where streetlights are switched off

Wednesday 23rd April 2014

Figures published by The Times indicate that casualties have risen by 20% in the past four years on roads where street lights have been turned off.

The Times’ report (content only available to Times’ subscribers) shows that in 2011/12, 324 more people were killed or seriously injured in crashes at night on roads where street lights were off, than in the previous year. Deaths and serious injuries have risen by 39% and 27% respectively on roads where around 750,000 streetlights are being switched off or dimmed down to either save money or reduce emissions.

The Times says the figures are “raising fears that cost-cutting and carbon emissions targets are claiming lives”.

In an article on the same subject, the AA said its research shows that night-time accidents in bad weather on well-lit 30mph urban roads have fallen by 15.6% over the past five years, but where street lights have been switched off or are not present the fall is just 2%.

The AA also says that official statistics show that, on darkened 40 mph built-up roads, accidents in the wet, snow or ice are down 21.8% where there is lighting, but only 5.2% where there is not. Overall, from 2007 to 2012, the AA says a 19.6% reduction in road accidents along town and city roads where street lights were on "shrank to 8.8% where drivers, cyclists, bikers and pedestrians travelled in darkness".

Talking to the Daily Mail, Edmund King AA president, said: “Worse accident rates on roads with street lights turned off or not present is an insidious threat that has crept in literally under the cover of darkness.

“Many local authorities based their risk assessment on police accident profiles for the affected roads. This had two huge drawbacks.

“Firstly and fundamentally, roads that are safe when lit can become unsafe with the lights switched off, but that is only shown when drivers, cyclists, bikers and pedestrians start to get hurt and killed.

“Secondly, with an extra casualty here and there, it is difficult to spot a creeping overall trend that might suggest something is dangerously wrong with a blackout.”


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:

Uncertain about scientific trial, confused about regression to mean and concerned that ETP has to be proved as well as delivered, I just want to comment about keeping and using the information we should already have on engineering schemes before staff retire, IT systems change and files get deleted. We need to keep information ourselves and pool it in some way possibly in the replacement to UKMorse.

This week I chanced to read a 2003 accident investigation report. The accidents were all the same type; a car turning right into a slip road to a dual carriageway cutting across the path of an approaching car. On updating the accident record there were 18 accidents at this location between 1998 and 2007 and none at all between 2008 and 2013. The report said the speed limit was national in 2003 and the speed limit layer now shows a 40 mph limit as part of a wider 40, 50, 40 mph scheme introduced in 2008 between villages.

Not sure exactly what this tells us and this is a very simple site with stark results. For example, is this site evidence to roll out 40 mph speed limits over wider areas? Having been slowed down from 60 to 40, did accidents on the nearby main road increase as drivers tried to make up time? I suspect most sites are far more complicated than this with less clear cut results.

However keeping information we already have provides opportunity to look back and work out what might have happened at this site. The key insight made by the accident investigator was that some drivers turning right anticipate that the oncoming vehicles will turn left (as the majority do), and occasionally gets it wrong. The factor specific to this site is that the distance between the place where vehicles turn left and other vehicles turn right is about 40m. The Highway Code stopping distance at 40 mph is 36m.

So any measure that reduced the speed of the approaching vehicle so that it could stop in the distance available would have worked here. I think we were lucky with the 40 mph speed limit. Hindsight rather than a scientific trial appears more appropriate in this case.
Steven Cross Leicestershire

Agree (0) | Disagree (0)

The link given by David Wakefield (or maybe it's David from Wakefield) would seem to promise a more in-depth study of this, although it doesn't seem to say when the results might be published.
Hugh Jones, Cheshire

Agree (3) | Disagree (0)

May I suggest that you visit
Where you will see that research is being carried out on this topic!
David Wakefield

Agree (3) | Disagree (0)

I am surprised at some of the elementary mistakes made in the OECD report and it may be that sort of misunderstanding that Duncan is alluding to.
Dave Finney, Slough

Agree (4) | Disagree (3)

The 'Safe System Approach' is internationally recognised as the framework for road safety research and interventions. It is promoted through the UN Decade of Action, the ITF, the European Commission and has been adopted by many countries worldwide, including the UK.

To quote the OECD recommendations directly:
"It is recommended that all countries, regardless of their level of road safety performance, move to a Safe System approach to road safety. This approach builds on existing road safety interventions but reframes the way in which road safety is viewed and managed in the community. It addresses all elements of the road transport system in an integrated way with the aim of ensuring crash energy levels are insufficient to cause fatal or serious injury. It requires acceptance of shared overall responsibilities and accountability between system designers and road users and stimulates the development of innovative interventions and new partnerships."

The full OECD report is available here:
Matt Staton, Cambridgeshire

Agree (3) | Disagree (4)

Got to fully agree with Dave Finney on this one, but would add that not only do we need better research we also need a better theoretical framework on which to place the data.

Without a sound theory of road accident causation the results of any research will always be open to incorrect interpretation, but a universal theory would provide a great leap forward in our struggle to reduce the KSIs. Examples of what I mean can be found with the theories of evolution, plate tectonics and the big-bang all of which provide much needed structure to researchers.
Duncan MacKillop, Stratford on Avon

Agree (6) | Disagree (2)

We have now published a micro-study on our website that examines the issue in a little more detail -
Richard Owen, Banbury

Agree (7) | Disagree (2)

I completely agree with Richard that we need far higher quality research on the effects of switching off street lamps, and all road safety interventions generally.

Road safety is complex, a factor cannot be changed in isolation, others factors will change in response. Switch off street lamps and councils might repaint the lines and maintain the cats eyes, drivers might slow down, cyclists might buy lights, the Police may alter their patrols and refocus their priorities. In short, like all other interventions, people may respond to mitigate or in unexpected ways, and analysing the net effect could easily cost more than simply running scientific trials in the first place!

If Richard could produce a good quality report on the effects of switching off street lamps, perhaps the Times, the AA and the Daily Mail would like to publicise it?
Dave Finney, Slough

Agree (11) | Disagree (1)

We have been having a look at the figures quoted in The Times' article and can't seem to match them up to the figures from STATS19 / MAST.

KSI casualties arising from collisions where lighting conditions were recorded as 'Night - Streetlights present but Unlit' were as follows:

2008 - 133
2009 - 113
2010 - 81
2011 - 125
2012 - 125

Hopefully we can find out a little more from the journalist involved on how the actual analysis was carried out. A lot of surveys relating to road safety are published without a methodology or even correct references!

The jury's out on this one but we will provide an update if we get any more data.
Richard Owen, Banbury

Agree (13) | Disagree (0)

Switching street lights off is a good idea, have you seen earth from space at night?

I knew this would happen. I said when switching off street lights was first proposed that scientific trials must be used otherwise how will we know what effect that has on road safety (and other factors like crime, trips and falls etc).

Were street lights switched off where many crashes had occurred, or on relatively safe roads? I suspect where there have been few crashes therefore crashes could be expected to rise simply due to RTM. Also other changes were made such as reduced speed limit etc so the rise might not be due to the unlit roads.

It's simple, run all interventions in scientific trials where possible (it's dead easy) and then we will know what works and what doesn't:
Dave Finney, Slough

Agree (10) | Disagree (4)

After reading this AA report it appears the streets are not the only thing to be left in the dark? As I stumble around looking for the AA light switch of reason.

Are we talking about casualties linked to poor weather, or the dimmed and darkend streets? I am happy to be illuminated up with that light bulb moment!
Gareth, Surrey

Agree (4) | Disagree (0)