Improving health worldwidewww.lshtm.ac.uk

Phil EdwardsLANTERNS ProjectSenior Lecturer, Department of Population Health

Does street lighting adaptation impact onroad traffic injury or crime?

The LANTERNS project is funded by the NIHR Public Health Research Programme. The views expressed here are those of the authors and do not necessarily reflect those of NIHR or the Department of Health

EconomyAusterity

Sustainability

Evidence for street lighting

Welsh BC, Farrington DP.Effects of improved street lighting on crime. Campbell Systematic Reviews 2008:13 DOI: 10.4073/csr.2008.13

Beyer FR, Ker K.Street lighting for preventing road traffic injuries. Cochrane Database of Systematic Reviews 2009, Issue 1. Art. No.: CD004728. DOI: 10.1002/14651858.CD004728.pub2.

LANTERNS Project

Local Authority collaborators’ National Evaluation ofReduced Night-time Streetlight

1. Collate information on street lighting adaptation schemes nationally

2. Statistically examine whether reduced lighting has effects on traffic collisions and crime

3. Explore public opinion on street lighting provision, and potential for reducing levels

4. Investigate whether street lighting adaptation schemes offer value for money

Project aims

1. Collate information on street lighting adaptation schemes nationally

2. Statistically examine whether reduced lighting has effects on traffic collisions and crime

3. Explore public opinion on street lighting provision, and potential for reducing levels

4. Investigate whether street lighting adaptation schemes offer value for money

Project aims

METHODSControlled interrupted time series analysis

slope pre-

Monthly counts of collisions/ crime

Time (months)

slope post-

step change associated with the intervention

Lighting interventions introduced:•Part-night lighting•Dimming•White light/LEDs•Switch off

Example datasetColumn ID Easting Northing Change Regime Date change applied

58 466015 174494 White Light LED 03/02/201278 465885 173200 White Light LED 26/03/201383 466006 174470 White Light LED 03/02/201284 465977 174470 White Light LED 03/02/201285 465958 174450 White Light LED 03/02/201286 465973 174438 White Light LED 03/02/201287 466001 174448 White Light LED 03/02/201288 466034 174445 White Light LED 03/02/201289 465937 174420 White Light LED 03/02/201290 465927 174422 White Light LED 03/02/201291 465922 174389 White Light LED 03/02/201299 465544 174593 White Light LED 30/01/2011

100 465584 174600 White Light LED 30/01/2011101 465512 174571 White Light LED 30/01/2011

10207 464465 171217 Part-night lighting 00:00 to 05:30 24/02/201210208 464432 171206 Part-night lighting 00:00 to 05:30 24/02/201210209 464394 171191 Part-night lighting 00:00 to 05:30 24/02/201210210 464352 171170 Part-night lighting 00:00 to 05:30 24/02/201210211 464311 171145 Part-night lighting 00:00 to 05:30 24/02/201210212 464274 171119 Part-night lighting 00:00 to 05:30 24/02/201210213 464234 171085 Part-night lighting 00:00 to 05:30 24/02/201210214 464198 171049 Part-night lighting 00:00 to 05:30 24/02/201210207 464465 171217 Part-night lighting 00:00 to 05:30 24/02/201210208 464432 171206 Part-night lighting 00:00 to 05:30 24/02/2012

Road traffic collisionsSTATS19

– Easting and Northing of collision location– Time of day– Severity of injuries

Date: 24th Nov 2004Time: 17:20Casualty class: PedestrianAccident severity: Serious

Date: 6th Nov 2007Time: 18:02Casualty class: DriverAccident severity: Slight

Date: 29th Oct 2010Time: 01:20Casualty class: DriverAccident severity: Slight

CrimePolice.uk

– Monthly counts– Burglary, Violence, Vehicle crime, Robbery

Police.uk data spatial accuracy

Tompson, L., Johnson, S.D., Ashby, M., Perkins, C., and Edwards, P. (2014). UK open source crime data: Accuracy and possibilities for research. Cartography and Geographic Information Science, 42(2), 97-111.

Local authorities in England & Wales invited to participate

71 local authorities contributed data

Data from 62 authorities included in analysis

Part-night30/6248%

Dimming40/6265%

White light52/6284%

0

2000

4000

6000

8000

10000

12000

14000

16000

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Km of road

Switch offPart-night lightingDimmingWhite light

RESULTSDoes street lighting adaptation impact on road traffic injury?

Night-time collisions on roads (2000–13):

5,670 (4%) with part-night lighting11,634 (7%) with dimming12,423 (8%) with white light 1,202 (0.7%) with switch off

Overall(I2 = 0.0%, p =0.46) 0.95 (0.84, 1.07)

0.78 (0.50, 1.22)

1.82 (0.58, 5.75)

0.80 (0.60, 1.07)

0.99 (0.63, 1.54)

1.35 (0.85, 2.12)

0.97 (0.80, 1.17)

0.96 (0.74, 1.24)

Decrease with PNL Increase with PNL 1.5 1 2

Part night light Rate ratio (95% CI)

East

London

East Midlands

Wales

West Midlands

North East

South West

Yorkshire

North West

South East

Region

Overall(I2 = 1.7%, p =0.42)

North East

Yorkshire

Wales

London

East Midlands

South West

South East

East

North West

West Midlands

1.00 (0.91, 1.10)

0.87 (0.64, 1.19)

1.18 (0.82, 1.70)

0.98 (0.75, 1.28)

0.86 (0.64, 1.16)

0.86 (0.70, 1.07)

1.13 (0.94, 1.36)

1.15 (0.71, 1.84)

1.16 (0.78, 1.72)

Decrease with Dimming Increase with Dimming 1.5 1 2

Dimming Rate ratio (95% CI)Region

Overall(I2 = 38.1%, p =0.11) 1.01 (0.93, 1.09)

1.06 (0.88, 1.28)

0.84 (0.64, 1.10)

0.96 (0.81, 1.15)

1.10 (0.80, 1.51)

1.30 (1.03, 1.65)

0.90 (0.71, 1.14)

0.76 (0.48, 1.21)

1.21 (0.95, 1.55)

0.89 (0.73, 1.09)

Decrease with White Increase with White 1.5 1 2

White light Rate ratio (95% CI)

East

London

East Midlands

Wales

West Midlands

North East

South West

Yorkshire

North West

South East

Region

Overall(I2 = 0.0%, p =0.98)

East

London

East Midlands

Wales

West Midlands

North East

South West

Yorkshire

North West

South East

0.97 (0.82, 1.15)

0.98 (0.79, 1.22)

0.97 (0.58, 1.62)

Rate ratio (95% CI)

0.95 (0.68, 1.32)

Decrease with Switch off Increase with Switch off 1.5 1 2

Switch offRegion

RESULTS Does street lighting adaptation impact on crime?

Crime (Dec 2010 - Dec 2013):

581,837 Burglaries730,280 Violence475,657 Vehicle crime 67,470 Robberies

Aggregate

Burglary

Robbery

Vehicle

Violence

0.96 (0.86, 1.06)

0.92 (0.74, 1.13)

1.48 (0.99, 2.21)

0.91 (0.76, 1.09)

1.01 (0.92, 1.11)

Decrease with PNL Increase with PNL

1.5 1 2

Part night light Rate ratio (95% CI)

Aggregate

Burglary

Robbery

Vehicle

Violence

0.84 (0.70, 1.02)

0.92 (0.65, 1.30)

0.75 (0.50, 1.13)

0.97 (0.78, 1.21)

0.78 (0.60, 1.01)

Decrease with Dimming Increase with Dimming

1.5 1 2

Dimming Rate ratio (95% CI)

Aggregate

Burglary

Robbery

Vehicle

Violence

0.89 (0.77, 1.03)

0.80 (0.59, 1.07)

0.98 (0.61, 1.56)

0.95 (0.73, 1.23)

0.91 (0.77, 1.08)

Decrease with White Increase with White

1.5 1 2

White light Rate ratio (95% CI)

Aggregate

Burglary

Robbery

Vehicle

Violence

0.11 (0.00, 2.75)

0.10 (0.00, 11.52)

1.17 (0.20, 6.71)

0.06 (0.00, 10.09)

0.84 (0.55, 1.29)

Decrease with Switch off Increase with Switch off

1.01 .1 1 5

Switch off Rate ratio (95% CI)

RESULTSCrime results showed heterogeneity between police forces

Overall(I2 = 65.4%, p<0.001)

14

16

31

1

26

4

9

28

3

11

27

24

8

29

20

5

6

12

0.96 (0.86, 1.06)

0.42 (0.01, 22.46)

0.99 (0.74, 1.34)

1.11 (0.39, 3.18)

1.26 (1.06, 1.48)

1.64 (1.26, 2.14)

0.99 (0.80, 1.24)

0.56 (0.26, 1.21)

0.91 (0.80, 1.04)

0.72 (0.45, 1.15)

0.90 (0.83, 0.98)

7.26 (0.72, 73.33)

0.83 (0.71, 0.97)

0.95 (0.46, 1.95)

0.93 (0.69, 1.24)

1.10 (0.73, 1.66)

0.67 (0.56, 0.81)

0.84 (0.38, 1.86)

0.96 (0.86, 1.06)

Decrease with PNL Increase with PNL 1.01 .1 1 10 100

Part night light Rate ratio (95% CI)Police force

Overall(I2 = 75.1%, p<0.001)

31

25

4

21

29

26

17

12

6

27

22

30

1820

16

10

13

1

15

0.84 (0.70, 1.02)

1.21 (0.13, 11.58)

0.69 (0.42, 1.16)

1.00 (0.86, 1.16)

0.71 (0.19, 2.65)

0.57 (0.18, 1.78)

1.89 (1.48, 2.40)

0.04 (0.00, 4.05)

0.45 (0.22, 0.93)

0.01 (0.00, 0.32)

0.41 (0.21, 0.82)

0.62 (0.49, 0.80)

0.89 (0.53, 1.48)

1.40 (0.03, 75.94)1.14 (0.73, 1.77)

0.30 (0.09, 0.99)

0.80 (0.69, 0.94)

4.79 (0.15, 156.20)

0.73 (0.34, 1.60)

1.05 (0.99, 1.11)

Decrease with Dimming Increase with Dimming 1.01 .1 1 10 100

Dimming Rate ratio (95% CI)Police force

Overall(I2 = 75.0%, p<0.001)

1820

13

23

21

30

11

31

25

8

5

28

4

29

109

16

22

17

1514

26

6

27

3

12

1

0.89 (0.77, 1.03)

0.81 (0.01, 44.39)1.07 (0.94, 1.23)

0.49 (0.08, 2.93)

1.29 (0.98, 1.71)

0.94 (0.25, 3.57)

0.62 (0.48, 0.81)

0.85 (0.59, 1.22)

0.55 (0.12, 2.51)

1.18 (0.68, 2.05)

0.61 (0.32, 1.17)

0.48 (0.22, 1.04)

2.20 (1.30, 3.72)

0.65 (0.33, 1.26)

2.35 (1.41, 3.93)

0.92 (0.76, 1.11)0.45 (0.33, 0.61)

0.90 (0.59, 1.37)

1.09 (0.87, 1.36)

1.20 (0.39, 3.66)

0.83 (0.74, 0.92)0.91 (0.74, 1.11)

0.61 (0.49, 0.75)

1.52 (0.09, 25.14)

0.25 (0.13, 0.49)

1.31 (0.69, 2.50)

1.08 (0.51, 2.27)

1.65 (0.77, 3.51)

Decrease with White Increase with White

1.01 .1 1 10 100

White light Rate ratio (95% CI)Police force

Limitations:-Selection bias?-Information bias?-Confounding?

Conclusions-Reduced street lighting over recent years has been done safely, without increasing road traffic casualties or crime.-Part-night lighting and dimming may be considered after careful assessment by qualified lighting professional, using a risk-based analysis.

Further research

AcknowledgementsThe authors wish to acknowledge the advice and support of the Institution of Lighting Professionals and the London Lighting Engineers Group.

The authors also thank: Denise Kendrick, Emily Conner and Mark Norris for advice, and the local authority street lighting managers who provided data for the project.