the effect of reduced street lighting on road casualties and crime in england and wales: controlled interrupted time series analysis - street light price

Many local authorities in England and Wales have reduced street lighting at night in order to save money and reduce carbon emissions.
So far, there is no evidence of whether these cuts have an impact on public health.
We quantify the effects of four street lamp adaptation strategies (
Part-closed
Night lighting, dimming and white light)
Casualties and crimes in England and Wales
Methods based on the analysis of the geocoded police data of road traffic collisions and crimes of 62 local authorities, the study was conducted.
Using conditional Poisson model to analyze the longitudinal variation of night count
The time collision on the affected roads during the period of 3-2000, and the crime in the Super Output Area of the census during the period of 3-10.
Impact estimates were adjusted based on regional time trends in casualties and crimes.
As a result, there is no evidence that any street light adaptation strategy is associated with changes in night collisions.
It is estimated that there is significant statistical heterogeneity in the impact of the police force on crime.
In general, there is no evidence of an association between the total number of crimes and the closure (RR 0. 11; 95% CI 0. 01 to 2. 75)or part-Lighting at night (RR 0. 96; 95% CI 0. 86 to 1. 06).
Insufficient evidence to reduce and darken the total number of crimes (RR 0. 84; 95% CI 0. 70 to 1. 02)
And white light (RR 0. 89; 95% CI 0. 77 to 1. 03).
Conclusion This study found very little evidence of closure
Road collisions in England and Wales or changes in night lighting, dimming, or White Light/LEDs in crime.
Background many local authorities in England and Wales have reduced street lighting at night in order to save money and reduce carbon emissions.
So far, there is no evidence of whether these cuts have an impact on public health.
We quantify the effects of four street lamp adaptation strategies (
Part-closed
Night lighting, dimming and white light)
Casualties and crimes in England and Wales
Methods based on the analysis of the geocoded police data of road traffic collisions and crimes of 62 local authorities, the study was conducted.
Using conditional Poisson model to analyze the longitudinal variation of night count
The time collision on the affected roads during the period of 3-2000, and the crime in the Super Output Area of the census during the period of 3-10.
Impact estimates were adjusted based on regional time trends in casualties and crimes.
As a result, there is no evidence that any street light adaptation strategy is associated with changes in night collisions.
It is estimated that there is significant statistical heterogeneity in the impact of the police force on crime.
In general, there is no evidence of an association between the total number of crimes and the closure (RR 0. 11; 95% CI 0. 01 to 2. 75)or part-Lighting at night (RR 0. 96; 95% CI 0. 86 to 1. 06).
Insufficient evidence to reduce and darken the total number of crimes (RR 0. 84; 95% CI 0. 70 to 1. 02)
And white light (RR 0. 89; 95% CI 0. 77 to 1. 03).
Conclusion This study found very little evidence of closure
Road collisions in England and Wales or changes in night lighting, dimming, or White Light/LEDs in crime.
In the context of cost constraints and increased local accountability for climate change mitigation, technological innovation is changing the way street lights are supplied in England and Wales, which may have a significant impact on public health.
Emerging technologies such as LED lighting and computerized central management systems create opportunities for local authorities to adjust their street lighting strategies in ways that reduce energy costs and help reduce carbon emissions.
However, there is also concern that the reduction in street lights at night may increase road traffic injuries, crime and fear of crime.
When deciding on street lighting, local authorities can learn from the state's guidance on lighting 1 and professional standard 2, but in the end, they made their own assessment of the most appropriate lighting system on each road they were responsible.
Three street lamp adaptation strategies reduce the amount of light: turn off the lights permanently (‘switch off’)
, Reduce the number of hours when lights are turned on at night (‘part-Night lighting)
And reduce the power or output of the lamp (‘dimming’).
The fourth strategy is to replace the traditional sodium lamp (
Orange/yellow light)
Use more energy-efficient lamps or led (‘white light’).
Each of these strategies has drawn public and media attention to the negative impact of night risk
Traffic accidents, crime and fear of crime.
Three road collision incidents received special attention, and some Coroner concluded that reduced road lighting led to road deaths and investigations.
However, health and health
Benefit from reducing light pollution (
Improve sleep, see the night sky, for example)
Also noticed.
5, 6 the reduction in artificial lighting at night may also bring benefits, which are associated with a range of health outcomes affected by circadian disruption.
7-9 the public concern section on road safety and crime prevention needs Street lights is supported by the literature: the system review has identified some evidence that, as street lighting increases, road Safety 10 improved and crime11 overall decreased.
With regard to crime, however, studies have shown that the number of victims has decreased during the two hours of day and darkness, raising questions about the mechanism by which street lamp improvements may affect crime.
Furthermore, to date, there is no empirical evidence of whether the implementation of reduced street lighting has had a negative impact on these important public health outcomes.
Therefore, it is timely for the state to study the impact of reducing street lights on road traffic collisions and crimes.
In theory, the adaptation strategy of street lamps may affect the risk of road traffic collision and crime from many aspects.
Reduce the lighting level by closing the part
Night lighting or dimming can reduce visibility in an area, which may increase the risk of collisions if road users are no longer able to detect hazards.
If fear of collision, falling in the dark, or fear of crime prevents people from making some trips, reduced visibility can also reduce mobility.
There are fewer people on the streets, which may reduce road casualties by reducing the possibility of collisions, but may reduce the "natural monitoring" of an area, resulting in an increase in crime.
If a criminal cannot identify a potential target, reduced visibility may also reduce the crime, such as theft of a vehicle or robbery.
The introduction of energy-saving lamps or led can change the quality and color of the lighting (
(From yellow to white)
In Don't reduce visibility of situation under
White light may improve vision by improving CCTV (CCTV)
Images can make criminals feel more visible, thus preventing certain types of crimes.
Street lamps as an integral part of modern lighting have a wider cultural significance (sub)
City life may also close some links
Night lighting and dimming strategies for crime.
Residents and tourists may feel that the target area for reducing lighting has depreciated;
This can reduce community pride and lead to behavioral changes that affect crime.
Alternatively, the introduction of new technologies such as White Light/LEDs in a region may indicate to local residents an increase in investment in the community, which may increase the pride and willingness of the community to use, and supervise their neighbors, thus reducing crime.
In conclusion, street light adaptation strategies may have different effects on the public health outcomes of casualties and crime.
Given the lack of evidence, we use the reduction of street lights by local authorities in England and Wales as a natural experiment to examine whether it is related to any changes in road traffic collisions and crimes.
This project is called Lantern (
National Assessment by local government partners on reducing night activity
Time street light)project.
Methods: we designed a controlled interrupt time series analysis to examine the association between reduced street lighting and road traffic collisions and crime and adjust it according to regional trends.
Traffic collisions were analyzed at the road level and crimes were analyzed at the regional level.
The analysis is based on a pattern of changes in the number of monthly collisions and crimes in each street or region.
Street lighting-data source
All local authorities in England and Wales were invited to participate in the study (
See other reports for recruitment methods).
3 from the local authorities where the street lamp supply has been changed in the evening, the data on the geographical location of each street lamp column, lighting changes are requested (eg, part-night lighting)
, And the date the change was introduced.
All local authorities providing available data by October 2014 were included in the analysis.
Traffic collision
Traffic collision data were obtained from STATS19, the official data set of collisions on British public highways collected by police.
Data were obtained on the date, time, geographical coordinates of the collision location and the severity of casualties of all collisions that occurred between 2013 and 2000. Crime—
Police have obtained criminal data.
UK website from December 2010 (
Earliest date available for data)
December 2013.
The data includes the type of crime, the year, the month, the name of the road where the crime occurred, and the approximate geographical coordinates.
Street lighting-
Linking data using a geographic information system (GIS)
Detailed Road segment database derived from Ordnance Survey general map integrated transport network.
12 during the period from January 2000 to December 2013, monthly depending on whether or not to close, part-
Night lighting, dimming, and white light have been implemented in the lighting bar of this section.
Combination of intervention measures implemented (eg, part-
Night lighting with white light)
Road sections are classified as two interventions.
Facilitating the region
Horizontal analysis, we also calculated the proportion of total road mileage in each area that was introduced and closed, part
Lighting, dimming and white light at night.
Result variable collision-
Collisions recorded in STATS19 are classified as "night-time’ (
Between sunset and sunrise)or ‘day-time’ (
Between sunrise and sunset)
Depending on the time and date of the collision.
To this end, we received the daily time of sunrise and sunset in each area of England and Wales from January 1-20, 2000 to December 31.
Using data on the severity of the injury, we classify each collision based on whether any casualty has suffered a fatal or serious injury.
If a person dies within 30 days of the collision, the casualty is classified as fatal.
If the person is admitted to hospital or suffered a fracture, concussion, internal injury, extrusion, non-injury, the casualty is defined as severe
Rubbing burns, severe wounds and tears, or severe systemic shock requiring treatment.
Using GIS, text descriptors for spatial overlay and analysis of each road location link collisions to road sections.
In short, the algorithm assigns collisions to the most recent road segments of the type shown in the STATS19 report.
Collisions occurring from the appropriate type of road section over 50 CUCM are excluded from the analysis.
We calculated the number of road crashes per month from January 2000 to December 2013. Crime—
To protect the anonymity of the victims, the public police.
The UK data only includes the approximate geographical coordinates of the crime and does not specify the time of the day of the crime.
The study of spatial accuracy of data shows that this is very good for large area units, such as the middle Super Output Area of the census (MSOA)
, Contains about 7500 people, 13 people, but there is no postal code or street segment level.
So we analyze crime at the MSOA level and assign each crime to the relevant MSOA.
We analyze the crimes most likely to occur at night or at night: theft, theft (or from)
Transportation, robbery, criminal injuries and violence, including sexual assault.
14 we calculated the number of each crime in MSOAs each month from December 2010 to December 2013.
The statistical approach focuses on describing the impact of street lighting changes on collisions and crime after taking into account potential trends.
The analysis is based on the change pattern of the number of monthly collisions within each road segment, and the change pattern of the number of monthly crimes within each MSOA.
In order to control the confusions caused by differences in selected and unselected intervention areas, we analyze these road levels (collision)
And MSOA level (crime)
Conditional on the total number in each road or MSOA, treat it as a "group" study so that any factor remains the same over time (Road design)
No information is provided for the analysis.
The conditional Poisson regression model is used instead of the standard Poisson model to allow rate changes for road sections and MSOA without having to directly estimate a large number of annoying parameters.
Further details of the model are given in Appendix 1 of the Web.
We perform sensitivity analysis by assuming a negative binomial rather than a Poisson model, by allowing zero inflation and fitting models that do not have a self-regression term.
Traffic collision
We estimate the association of each street light adaptation intervention with the night
Avoid time conflicts for a single model that is confused with each other.
To prevent deviations due to changes occurring at the same time as the lighting intervention affecting the whole (
Day and night)
Collisions, data sets include the count of daytime collisions and binary variables indicating day/night.
We estimate the impact rate changes associated with each street light adaptation intervention during the day and night from the regression model
Ratio of time to night
Change of time and day.
We believe that this "Day collision" adjustment measure is the most robust estimate of night change
Time collision after lighting intervention.
By fitting the various terms of the year and month, the model is adjusted according to seasonal changes and time trends.
Models were fitted for each region of England and Wales and effect estimates were concentrated in the fixed effect elementanalysis. Crime—
We estimate the association between changes in the total road mileage ratio in each MSOA that introduce each street light adaptation intervention and the count of each criminal offence (
And their aggregation)
Control crime trends over time.
We fitted the indicator variables for a few months after the start of the study (
That is, from December 2010, after a month's step function).
Given the different data collection systems and evidence of each police officer in different background time patterns in these systems, we installed different models for them.
We estimate the effect in a random effect meta-concentrationanalysis.
In view of the large number of police forces and types of crimes considered, we did not show yuan-
The analysis in the main text is only the means of England and Wales.
Street lighting-data source
All local authorities in England and Wales were invited to participate in the study (
See other reports for recruitment methods).
3 from the local authorities where the street lamp supply has been changed in the evening, the data on the geographical location of each street lamp column, lighting changes are requested (eg, part-night lighting)
, And the date the change was introduced.
All local authorities providing available data by October 2014 were included in the analysis.
Traffic collision
Traffic collision data were obtained from STATS19, the official data set of collisions on British public highways collected by police.
Data were obtained on the date, time, geographical coordinates of the collision location and the severity of casualties of all collisions that occurred between 2013 and 2000. Crime—
Police have obtained criminal data.
UK website from December 2010 (
Earliest date available for data)
December 2013.
The data includes the type of crime, the year, the month, the name of the road where the crime occurred, and the approximate geographical coordinates.
Street lighting-
Linking data using a geographic information system (GIS)
Detailed Road segment database derived from Ordnance Survey general map integrated transport network.
12 during the period from January 2000 to December 2013, monthly depending on whether or not to close, part-
Night lighting, dimming, and white light have been implemented in the lighting bar of this section.
Combination of intervention measures implemented (eg, part-
Night lighting with white light)
Road sections are classified as two interventions.
Facilitating the region
Horizontal analysis, we also calculated the proportion of total road mileage in each area that was introduced and closed, part
Lighting, dimming and white light at night.
Result variable collision-
Collisions recorded in STATS19 are classified as "night-time’ (
Between sunset and sunrise)or ‘day-time’ (
Between sunrise and sunset)
Depending on the time and date of the collision.
To this end, we received the daily time of sunrise and sunset in each area of England and Wales from January 1-20, 2000 to December 31.
Using data on the severity of the injury, we classify each collision based on whether any casualty has suffered a fatal or serious injury.
If a person dies within 30 days of the collision, the casualty is classified as fatal.
If the person is admitted to hospital or suffered a fracture, concussion, internal injury, extrusion, non-injury, the casualty is defined as severe
Rubbing burns, severe wounds and tears, or severe systemic shock requiring treatment.
Using GIS, text descriptors for spatial overlay and analysis of each road location link collisions to road sections.
In short, the algorithm assigns collisions to the most recent road segments of the type shown in the STATS19 report.
Collisions occurring from the appropriate type of road section over 50 CUCM are excluded from the analysis.
We calculated the number of road crashes per month from January 2000 to December 2013. Crime—
To protect the anonymity of the victims, the public police.
The UK data only includes the approximate geographical coordinates of the crime and does not specify the time of the day of the crime.
The study of spatial accuracy of data shows that this is very good for large area units, such as the middle Super Output Area of the census (MSOA)
, Contains about 7500 people, 13 people, but there is no postal code or street segment level.
So we analyze crime at the MSOA level and assign each crime to the relevant MSOA.
We analyze the crimes most likely to occur at night or at night: theft, theft (or from)
Transportation, robbery, criminal injuries and violence, including sexual assault.
14 we calculated the number of each crime in MSOAs each month from December 2010 to December 2013.
The statistical approach focuses on describing the impact of street lighting changes on collisions and crime after taking into account potential trends.
The analysis is based on the change pattern of the number of monthly collisions within each road segment, and the change pattern of the number of monthly crimes within each MSOA.
In order to control the confusions caused by differences in selected and unselected intervention areas, we analyze these road levels (collision)
And MSOA level (crime)
Conditional on the total number in each road or MSOA, treat it as a "group" study so that any factor remains the same over time (Road design)
No information is provided for the analysis.
The conditional Poisson regression model is used instead of the standard Poisson model to allow rate changes for road sections and MSOA without having to directly estimate a large number of annoying parameters.
Further details of the model are given in Appendix 1 of the Web.
We perform sensitivity analysis by assuming a negative binomial rather than a Poisson model, by allowing zero inflation and fitting models that do not have a self-regression term.
Traffic collision
We estimate the association of each street light adaptation intervention with the night
Avoid time conflicts for a single model that is confused with each other.
To prevent deviations due to changes occurring at the same time as the lighting intervention affecting the whole (
Day and night)
Collisions, data sets include the count of daytime collisions and binary variables indicating day/night.
We estimate the impact rate changes associated with each street light adaptation intervention during the day and night from the regression model
Ratio of time to night
Change of time and day.
We believe that this "Day collision" adjustment measure is the most robust estimate of night change
Time collision after lighting intervention.
By fitting the various terms of the year and month, the model is adjusted according to seasonal changes and time trends.
Models were fitted for each region of England and Wales and effect estimates were concentrated in the fixed effect elementanalysis. Crime—
We estimate the association between changes in the total road mileage ratio in each MSOA that introduce each street light adaptation intervention and the count of each criminal offence (
And their aggregation)
Control crime trends over time.
We fitted the indicator variables for a few months after the start of the study (
That is, from December 2010, after a month's step function).
Given the different data collection systems and evidence of each police officer in different background time patterns in these systems, we installed different models for them.
We estimate the effect in a random effect meta-concentrationanalysis.
In view of the large number of police forces and types of crimes considered, we did not show yuan-
The analysis in the main text is only the means of England and Wales.
As a result, by October 2014, we had 125 (72%)
Of the 174 local authorities in England and Wales, 71 submitted data, of which 62 provided available data (figure 1).
Due to the lack of information on the date of the change and the changes implemented after the study, data from nine authorities were excluded.
Local authorities in each region provided data.
According to the population density, the local authorities involved are evenly distributed in ten cities (
Number per hectare)
, Covers the range of levels of deprivation measured in the multiple deprivation index 2010 and covers each of the six urban and rural areas classification.
Download the process of local authority participants in the figureOpen project in the new tabDownload powerpoint figure 1. Street lamp-
Of the 62 local authorities with available data, 5 (8%)
Closure has been introduced, 30 (48%)
Introduction part
Lighting at night, 40 (65%)
Dimming is introduced, and 52 (84%)
White light was introduced.
From 2009, the introduction of street lighting adaptation strategies has increased steadily (figure 2).
By December 2013, the local authorities involved in the study had implemented white light on a total of 15 kilometres and 833 kilometres of roads (
Of the 62 participating local authorities, 7% of the total road mileage); part-
12 kilometers of roads and 10 kilometers of night lighting (5%);
Dimming road of 10 km/h 519 km/h (4%);
And closed on a 946-kilometer road (0. 4%).
The proportion of total road mileage in each MSOA that turns off the lights ranges from 0% to 60% (median 0. 03%; IQR 0. 01–0. 05%). For part-
Lighting at night, the proportion varies from 0 to 84% (median 0. 2%; IQR 0. 1–18%);
For dimming, the ratio ranges from 0 to 93% (median 0. 14%; IQR 0. 07–5%);
For white light, the ratio ranges from 0 to 81% (median 1%; IQR 0. 15–11%).
Download figureOpen on the road of the new tabDownload powerpoint figure 2 kilometers and implement the lighting adaptation strategy at the participating local authorities.
Traffic collision
During the period of 3-859, 935 collisions occurred in 62 local authorities included in the analysis;
Of which 161 are 049 (19%)were night-Time Collision; 153u2005442 (18%)
The collision caused serious injury or death. Of the night-
Time Collision, 1202 (0. 7%)
Occurred on roads closed as of December 2013; 5670 (4%)
On the way with partsnight lighting; 11u2005634 (7%)
On the dim Road; and 12u2005423 (8%)
On the white road.
We found no evidence of any association between street light adaptation strategies and daily lifeadjusted night-
Time collision rate (figure 3).
There is no regional model or overall estimate indicating an association between closures (rate ratio (RR)0. 97; 95% CI 0. 82 to 1. 15), part-Lighting at night (RR 0. 95; 95% CI 0. 84 to 1. 07); or dimming (RR 1. 00; 95% CI 0. 91 to 1. 10).
In London, there is insufficient evidence on the association between the introduction of white light and the increase of night lighting
Time Collision (RR 1. 30; 95% CI 1. 03 to 1. 65);
However, the overall national estimate does not provide evidence for such associations (RR 1. 01; 95% CI 0. 93 to 1. 09).
Similarly, models that estimate the association between street light adaptation strategies and collisions that cause serious injury or death did not find evidence related to any lighting intervention (
See Appendix 2 of the web).
Download the association between figure open street light adaptation strategy and night in New tabDownload powerpoint Figure 3-
Road traffic collision. Crime—
In 62 a place authorities area the 581 u2005 837 the theft;
Theft of a vehicle or vehicle;
470 robbery;
Between December 2010 and December 2013, 486 per cent were criminal offences of 367 per cent and 7302280 per cent were violent.
Figure 4 shows the estimated average correlation between the change in the total road mileage ratio in each MSOA that introduces each street lighting adaptation intervention and the count of each crime in England and Wales.
The rate ratio indicates that the crime is expected to change if 100% of roads in an area are subject to lighting intervention.
Download the new tabDownload powerpoint figure 4 association between street light adaptation strategy and crime.
There is no evidence from the overall estimate that there is an association between the total number of crimes and the closure (RR 0. 11; 95% CI 0. 01 to 2. 75)or part-Lighting at night (RR 0. 96; 95% CI 0. 86 to 1. 06).
Insufficient evidence to reduce and darken the total number of crimes (RR 0. 84; 95% CI 0. 70 to 1. 02)
And white light (RR 0. 89; 95% CI 0. 77 to 1. 03).
We found significant heterogeneity between estimates at the police level (
See web Appendix 3);
There is strong evidence that part-
Lighting, dimming and white light at night, and a reduction in crime by some police forces, as well as parts-
Lighting at night, dimming and white light, and the increase in crime by others.
When considering a specific offence, the estimates provide evidence of implied evidence that is in part-
Night lighting may be related to increased robbery (RR 1. 48; 95% CI 0. 99 to 2. 21)
This dimming may be related to the reduction of violence (RR 0. 78; 95% CI 0. 60 to 1. 01).
At the national level, we find that the estimated association between crime and darkening is in the same direction, and all point estimates indicate a reduction in crime.
There is a similar consistency with the estimated Association of white light, all of which suggest a reduction in crime.
However, we again find that there is a significant statistical heterogeneity in the estimation of the association between all lighting adaptations of all crime types at the police force level (
See web Appendix 3).
Small differences are obvious in sensitivity analysis, but the results of collisions and crimes are not materially different from the main results.
We find no evidence of zero inflation.
The results of the discussion did not provide evidence of partial closure
Night lighting, dimming, or white light adaptation to street lights related to night
Traffic collision.
The results also did not provide evidence that these lighting strategies were related to increased crime at the regional level.
The results show that, in general, the darkening and white light systems are associated with reduced crime, although the estimates are inaccurate.
Limiting selection bias: This study was able to obtain available data on street lighting changes from 62 of the 174 local authorities.
Local authorities may refuse to participate due to changes in lighting, as collisions or offences in their area are expected or known to increase.
If the collision or crime changes in
Our study may have underestimated the impact of reduced lighting on collisions and crime.
Information bias: our study uses conventional data sources for road traffic collisions and crimes.
There are several limitations to these data sets;
In particular, they may be incomplete for the following reasons
Incident report.
However,
Reporting affects the results of our analysis, which will require different changes over time in recording crimes and collisions on streets where lighting has changed, this seems unlikely compared to the street without changing the street lights.
Addressing potential deficiencies
Based on the deviation estimate of the impact of the collision, we analyzed separately the collisions where casualties were fatal or seriously injured, as these collisions were more likely to be reported to the police.
These analyses did not find evidence of a correlation between street light change and night timeTime collision.
We also analyzed the crimes that were more likely to be reported to the police (
That is, insurance claims require theft and vehicle theft reports).
The estimated impact of these two types of crimes is not always greater.
We are not able to specifically review the crimes that took place during the night with the police. uk data.
However, our analysis is limited to crime investigations in England and Wales, which show that crimes are more likely to occur at night or at night.
Another dataset is unlikely to be able to address this limitation because of the exact time of many crimes (
Such as property or vehicle theft)
The victim did not know. The Police.
There is also a limitation in the UK's crime data: to protect the anonymity of victims, the data is geographically ambiguous.
In order to solve this problem, we analyze the data at the regional level, and although this statistical capability is weakened, the spatial accuracy of known data is good.
Mixed factors: We did not take into account the potential impact of other road safety or crime prevention initiatives such as improved road marking, security intervention or CCTV.
If such measures are taken more often than elsewhere on streets that change lighting, in areas where lighting changes, some changes in crime may be due to these other measures.
Despite the limitations of this study, we use two open large data sets to provide evidence on the relationship between recent street light adaptation strategies, road traffic collisions and crimes.
A systematic review of evidence of increased street lighting effects on road traffic collisions and crime shows that the relative reductions were 32% and 38%, respectively.
Current studies have shown that CIs is narrow enough to exclude changes in collisions of this size.
The results of the crime are highly heterogeneous and with limited power due to the need to aggregate to the MSOA level, however, the average impact of the overall estimate does not indicate an increase in crime at the district level where lighting is reduced.
Similarly, the scope of the CIS is narrow enough to rule out the anticipated increase in crime on such a large scale.
However, the estimate of the closure is inaccurate because there are very few areas where the closure is implemented and should therefore be treated with caution.
Although 14 years of road traffic collision data have been used by 62 local authorities, we have not found any convincing evidence of an association between street light adaptation and road traffic collisions.
Since we do not directly measure the number of trips or the way of travel before and after the implementation of street lamp intervention, we do not know if this reflects mobility reduction or pattern change (
Travel from walk to car)
In closed or partially closed areas-
Night lighting was implemented.
While street lighting is reduced, the number of pedestrians, cyclists, bikers and car drivers may decline, resulting in fewer collisions on these streets.
If so, due to low lighting conditions, an increase in harm to road users may be overshadowed by a decrease in the number of people exposed to road injury risks at night.
Qualitative and investigative evidence 3, 16, however, suggests that changes in overall liquidity and pattern selection in affected areas may be minimal.
However, this study does indicate the link between some street light adaptation and crime.
While there is a significant statistical heterogeneity in the effect estimated at the level of the police force, the overall results indicate a correlation between crime reduction and reduction, particularly with respect to violent crime.
These results may help support the hypothesis of linking lower visibility to difficulties in identifying "appropriate" targets on the street at night.
The results also show that there is a correlation between white light and crime reduction, especially burglary, which may support mechanisms that link increased visibility in local communities or increased investment in local communities to crime reduction
If street lighting is reduced, pedestrian activities can be better replaced.
Illuminate the streets, which may reduce the risk of victims and interpersonal crimes on these streets and increase the custody of better peoplelit streets.
Different causal mechanisms may apply in different cases, which need to be evaluated through further studies.
This study does not support concerns about the impact of switchesoff, part-
Night lighting, dimming and white light on crime and road safety.
Informal Local Government (
Sometimes formal.
Assess the risks of local adaptation to street lights, draw on local experiments and experiments, 17 and national and professional guidance for lighting course selection.
1. the Two considerations of the appropriate "lighting grade" of the road reflect many factors, including the speed limit, traffic flow, composition of motor and non-motor vehicles
Motor traffic, density of intersections, parking vehicles, ambient brightness (
Lighting from other sources such as shops)
And how easy the road is to navigate.
1 The results of this study show that when the risk is carefully considered, local authorities can safely reduce street lights by turning off part of the power supply, thus saving costs and energy
Night lighting, dimming and white light strategies do not necessarily have a negative impact on road traffic collisions and crime.
The local authorities involved included many communities in terms of population density, geographical location and economic resources.
Therefore, considering the strategy of reducing street lights, our results may be extended to other communities.
This study was able to reveal the effect of reducing street lighting at night.
More research is needed to understand how different lighting systems affect opportunities for crime and crime prevention, as well as other effects of light changes at night on public health, given the suggestive evidence that artificial lighting may be related to multiple health outcomes such as obesity, sleep and cancer.
What is already known on this issue is that there is evidence that the introduction of street lights at night is related to the reduction of road traffic accidents and crimes.
Many local authorities in England and Wales are reducing street lighting at night in order to save energy costs and reduce carbon emissions.
The study adds that there is no evidence that road traffic collisions or increased crime are associated with a reduction in street lighting.
Dimming the light or switching to a White Light/LEDs can reduce crime in one area.
When considering risks carefully, local authorities can safely reduce street lights, save energy costs and reduce carbon emissions without having a negative impact on traffic collisions and crimes.
Limiting selection bias: This study was able to obtain available data on street lighting changes from 62 of the 174 local authorities.
Local authorities may refuse to participate due to changes in lighting, as collisions or offences in their area are expected or known to increase.
If the collision or crime changes in
Our study may have underestimated the impact of reduced lighting on collisions and crime.
Information bias: our study uses conventional data sources for road traffic collisions and crimes.
There are several limitations to these data sets;
In particular, they may be incomplete for the following reasons
Incident report.
However,
Reporting affects the results of our analysis, which will require different changes over time in recording crimes and collisions on streets where lighting has changed, this seems unlikely compared to the street without changing the street lights.
Addressing potential deficiencies
Based on the deviation estimate of the impact of the collision, we analyzed separately the collisions where casualties were fatal or seriously injured, as these collisions were more likely to be reported to the police.
These analyses did not find evidence of a correlation between street light change and night timeTime collision.
We also analyzed the crimes that were more likely to be reported to the police (
That is, insurance claims require theft and vehicle theft reports).
The estimated impact of these two types of crimes is not always greater.
We are not able to specifically review the crimes that took place during the night with the police. uk data.
However, our analysis is limited to crime investigations in England and Wales, which show that crimes are more likely to occur at night or at night.
Another dataset is unlikely to be able to address this limitation because of the exact time of many crimes (
Such as property or vehicle theft)
The victim did not know. The Police.
There is also a limitation in the UK's crime data: to protect the anonymity of victims, the data is geographically ambiguous.
In order to solve this problem, we analyze the data at the regional level, and although this statistical capability is weakened, the spatial accuracy of known data is good.
Mixed factors: We did not take into account the potential impact of other road safety or crime prevention initiatives such as improved road marking, security intervention or CCTV.
If such measures are taken more often than elsewhere on streets that change lighting, in areas where lighting changes, some changes in crime may be due to these other measures.
Despite the limitations of this study, we use two open large data sets to provide evidence on the relationship between recent street light adaptation strategies, road traffic collisions and crimes.
A systematic review of evidence of increased street lighting effects on road traffic collisions and crime shows that the relative reductions were 32% and 38%, respectively.
Current studies have shown that CIs is narrow enough to exclude changes in collisions of this size.
The results of the crime are highly heterogeneous and with limited power due to the need to aggregate to the MSOA level, however, the average impact of the overall estimate does not indicate an increase in crime at the district level where lighting is reduced.
Similarly, the scope of the CIS is narrow enough to rule out the anticipated increase in crime on such a large scale.
However, the estimate of the closure is inaccurate because there are very few areas where the closure is implemented and should therefore be treated with caution.
Although 14 years of road traffic collision data have been used by 62 local authorities, we have not found any convincing evidence of an association between street light adaptation and road traffic collisions.
Since we do not directly measure the number of trips or the way of travel before and after the implementation of street lamp intervention, we do not know if this reflects mobility reduction or pattern change (
Travel from walk to car)
In closed or partially closed areas-
Night lighting was implemented.
While street lighting is reduced, the number of pedestrians, cyclists, bikers and car drivers may decline, resulting in fewer collisions on these streets.
If so, due to low lighting conditions, an increase in harm to road users may be overshadowed by a decrease in the number of people exposed to road injury risks at night.
Qualitative and investigative evidence 3, 16, however, suggests that changes in overall liquidity and pattern selection in affected areas may be minimal.
However, this study does indicate the link between some street light adaptation and crime.
While there is a significant statistical heterogeneity in the effect estimated at the level of the police force, the overall results indicate a correlation between crime reduction and reduction, particularly with respect to violent crime.
These results may help support the hypothesis of linking lower visibility to difficulties in identifying "appropriate" targets on the street at night.
The results also show that there is a correlation between white light and crime reduction, especially burglary, which may support mechanisms that link increased visibility in local communities or increased investment in local communities to crime reduction
If street lighting is reduced, pedestrian activities can be better replaced.
Illuminate the streets, which may reduce the risk of victims and interpersonal crimes on these streets and increase the custody of better peoplelit streets.
Different causal mechanisms may apply in different cases, which need to be evaluated through further studies.
This study does not support concerns about the impact of switchesoff, part-
Night lighting, dimming and white light on crime and road safety.
Informal Local Government (
Sometimes formal.
Assess the risks of local adaptation to street lights, draw on local experiments and experiments, 17 and national and professional guidance for lighting course selection.
1. the Two considerations of the appropriate "lighting grade" of the road reflect many factors, including the speed limit, traffic flow, composition of motor and non-motor vehicles
Motor traffic, density of intersections, parking vehicles, ambient brightness (
Lighting from other sources such as shops)
And how easy the road is to navigate.
1 The results of this study show that when the risk is carefully considered, local authorities can safely reduce street lights by turning off part of the power supply, thus saving costs and energy
Night lighting, dimming and white light strategies do not necessarily have a negative impact on road traffic collisions and crime.
The local authorities involved included many communities in terms of population density, geographical location and economic resources.
Therefore, considering the strategy of reducing street lights, our results may be extended to other communities.
This study was able to reveal the effect of reducing street lighting at night.
More research is needed to understand how different lighting systems affect opportunities for crime and crime prevention, as well as other effects of light changes at night on public health, given the suggestive evidence that artificial lighting may be related to multiple health outcomes such as obesity, sleep and cancer.
What is already known on this issue is that there is evidence that the introduction of street lights at night is related to the reduction of road traffic accidents and crimes.
Many local authorities in England and Wales are reducing street lighting at night in order to save energy costs and reduce carbon emissions.
The study adds that there is no evidence that road traffic collisions or increased crime are associated with a reduction in street lighting.
Dimming the light or switching to a White Light/LEDs can reduce crime in one area.
When considering risks carefully, local authorities can safely reduce street lights, save energy costs and reduce carbon emissions without having a negative impact on traffic collisions and crimes.
The authors would like to thank the Association of Lighting Professionals and the group of lighting engineers in London for their suggestions and support.
The authors thank the members of the project advisory group: Dennis Kendrick, Emily Connor and Mark Norris for their suggestions throughout the project, as well as the local authority Street Light Manager who provided data for the project.
British Standards Agency.
BS 5489 code of implementation for road lighting design.
Lighting of roads and public facilities.
British Standards Association, 2013.
Lighting professional organization.
PLG 08: Guide to adaptive lighting applications in the public domain.
Professional Lighting Guide.
Rugby in Warwick County: ILP, 2013.
Samuel Perkins C, Steinbach R, Thompson L, etc.
What is the impact of reducing street lights at night on crime and road traffic injuries?
Research on mixed methods.
Report to the National Institutes of Health, 2015.
Automobile Association.
Turn off the street lights: The survey shows that there is a mode of death on the road where the city councillors have a power outage.
Automobile Association, 2014.
The House of Commons on science and technology.
Light pollution and astronomy
London: fixed Office 2003.
Falfalchi F, Cinzano P, Elvidge CD, etc.
Limit the effects of light pollution on human health, the environment, and stellar visibility.
2011 Environmental Management92:2714–22. doi:10. 1016/j. jenvman. 2011. 06.
029 E Jones, sumeikel MJ, of openurlcross refpmedweb Science Limited McFadden, etc.
The relationship between obesity and light exposure at night: Crossover
Segmentation analysis of more than 100,000 women in breakthrough generation studies.
Am J Epidemiol 2014; 180:245–50. doi:10.
1093 Al Jazeera English channel/kwu117OpenUrlAbstract/free full Text radio Navara KJ, Nelson Crystal Head.
Dark side of light at night: Physiology, epidemiology, and ecological consequences.
J. Songguo Res 2007; 43:215–24. doi:10. 1111/j. 1600-079X. 2007. 00473.
D, Yansheng of XOpenUrlCrossRefPubMedWeb Science GmbH Shuboni.
Dim light at night changes the reaction of the circadian system.
Neuroscience 2010; 170:1172–8. doi:10. 1016/j. neuroscience. 2010. 08.
F, mark of 009 openurlcross refpmedweb Science archibeyer.
Road lighting to prevent road traffic damage.
The Cochrane Database Syst Rev 2009; (1):CD004728. doi:10. 1002/14651858.
B. Wales, Farrington D.
Improve the impact of street lights on crime.
Rev 2008 of Campbell system; 13:1–51. doi:10. 4073/csr. 2008.
Integrated transport network layer [13 openurlos OS MasterMap]
[Geospatial data]
Coverage: England and Wales, updated on November 2013, Ordnance Survey, GB.
Use: EDINA Digimap Ordnance Survey service,