Effectiveness of speed cameras in preventing road traffic collisions a
http://www.100md.com
《英国医生杂志》
1 University of the West of England, Faculty of Health and Social Care, Bristol BS16 1DD, 2 Department of Social Medicine, University of Bristol, Bristol
Correspondence to: P Pilkington paul.pilkington@uwe.ac.uk
Abstract
Road traffic collisions are an important cause of death and disability worldwide. Every year around the world 1.2 million people are killed and up to 50 million are injured or disabled as a result of road traffic collisions.1 Morbidity from road traffic collisions is expected to increase in future years, and it is estimated that road traffic collisions will move from ninth to third place in the global burden of disease ranking, as measured in disability adjusted life years.2 3
Measures to reduce traffic speed are considered essential to reducing casualties on the road.1 4 5 Speed cameras are increasingly used to help to reduce traffic speeds in the belief that this will reduce road traffic collisions and casualties, and an expansion in the use of speed cameras is under way in many countries, most notably the United Kingdom.6 The use of speed cameras is controversial, however. Vociferous opponents, including some motoring associated organisations, oppose their use, and cameras are often criticised in the media.7-9 The lack of readily available evidence of the effectiveness of cameras has made it difficult for road safety and health professionals to engage in an informed debate about the effectiveness of speed cameras.
A previous small non-systematic review of six studies found a 17% reduction in collisions after introduction of speed cameras.10 Non-systematic reviews can, however, be limited by bias. We aimed, therefore, to systematically assess the evidence for the effectiveness of speed cameras in reducing road traffic collisions and related casualties.
Methods
Controlled trials and observational studies assessing the impact of fixed or mobile speed cameras on any or all of three outcomes (collisions, injuries, and deaths) were eligible for inclusion. We considered all published and unpublished material, with no restrictions on date or language. As the effect of co-intervention is difficult to exclude in interventions such as this, studies that did not have speed cameras as the major intervention were not eligible for inclusion.
Identification of primary studies
One reviewer (PP) searched for studies by using the search strategy outlined earlier and, together with a second reviewer (SK), selected studies to obtain for possible inclusion in the review, on the basis of titles and abstracts (where available). Both reviewers then independently extracted data from each study by using a data extraction form that was piloted before use. We extracted details of the study design, aim of the study, setting of the study and nature of the roads, study period, measurement of exposure, outcome and relevant confounders, and results. We also assessed the quality of the studies with a predefined quality scale, which, in the absence of pre-existing scales, we developed and piloted ourselves. The quality scale rated studies on the basis of representativeness of study areas to general population; control areas being representative of intervention areas; objective and valid outcome(s); results provided with estimates of uncertainty; main conclusions based on primary study hypotheses; and important confounders measured and controlled for. For each of the six quality criteria, we rated the studies on a three point scale (0-2). We rated studies scoring a total of 9-12 as good quality, 6-8 as average quality, and 0-5 as poor quality (see appendix for full data extraction form and quality scale). We combined the results from the data extraction forms of the two reviewers and made decisions on inclusion in the review. We resolved disagreements by consensus.
We also extracted data on the actual number of cases in the intervention and control areas for each time period and where appropriate combined them to produce summary statistics. We calculated risk ratios with confidence intervals for before-after and experimental-control comparisons where possible.
Meta-analysis
We selected 92 studies to review, on the basis of the title or abstract of the report. After reviewing the full articles, we identified 21 studies that were potentially suitable for inclusion. Of these, two studies did not consider the intervention or outcome of interest,11 12 one study reported only secondary results without details of the methods,13 two studies did not look at the effectiveness of the introduction of cameras,14 15 and two studies were preliminary reports that were updated in later publications.16 17 After excluding these studies, we included 14 studies in the final review (see figure on bmj.com).
All the studies were observational studies; we found no randomised controlled trials. Five studies had control areas distinct from the areas where the cameras were introduced.18-22 One study used the same areas at times when cameras were not operating as a control,23 and eight studies used the same areas before introduction of cameras as the comparison group (before-after studies).10 24-30 The studies were published between 1992 and 2003. All studies were in high income countries. Six studies assessed the effect of fixed cameras,10 18-20 29 30 four studied the effect of mobile cameras,21-23 26 and four studied the effect of a combination of fixed and mobile cameras.24-28 Outcome measures in the studies were diverse and included various measures of collisions, deaths, and injuries. Three studies had a follow up period of one year following the introduction of cameras,22 26 29 nine studies had a follow up period of one to three years,18-21 23-25 27 28 and one study had a follow up period of four years.10 One study stated only that follow up data of at least one year were used.30 See table A on bmj.com for details of the studies. In terms of methodological quality, we classified no studies as being good quality, seven as average, two as average-poor, and five as poor.
All studies reported a reduction in road traffic collisions and casualties. The reduction in adverse outcomes in the immediate vicinity of camera sites varied considerably across studies, with ranges of 5-69% for collisions, 12-65% for injuries, and 17-71% for deaths at camera sites. Smaller reductions in adverse outcomes were seen over a wider area. See table B on bmj.com for full results.
Discussion
Peden M, Scurfield R, Sleet D, Mohan D, Hyder AA, Jarawan E, et al, eds. World report on road traffic injury prevention. Geneva: World Health Organization, 2004.
Murray CJL, Lopez, AD. Alternative projections of mortality and disability by cause 1990-2020: global burden of disease study. Lancet 1997;349: 1498-504.
Roberts I. War on the roads. BMJ 2002;324: 1107-9.
House of Commons Transport, Local Government and the Regions Committee. Road traffic speed: ninth report of session 2001-2002, vol 1. London: Stationery Office, 2002.
Crombie H. The impact of transport and road traffic speed on health. London: Health Development Agency, 2002.
DTLR press release. Life-saving cameras to be made more visible—Spellar, August 2001. www.press.dtlr.gov.uk/pns/DisplayPN.cgi?pn_id=2001_0359 (accessed 5 Jan 2005).
Association of British Drivers. www.abd.org.uk/ (accessed 29 April 2004).
Massey R. New camera blitz on cars. Daily Mail 2002 April 11: 1,6.
Barker P. Speed cameras: costing us millions, and costing us lives. Autocar Magazine 2003 Nov 11: 18-23.
Elvik E. Effects of accidents of automatic speed enforcement in Norway. Transportation Research Record 1997;1597: 1-19.
Ali SY, Al-Saleh O, Koushki PA.. Effectiveness of automated speed-monitoring cameras in Kuwait. Transportation Research Record 1997;1595: 20-6.
Brownfield DJ. Environmental areas—interim report on a before-and-after accident study. Traffic Engineering and Control 1980;21: 278-82.
Traffic: speed cameras prove to be lifesavers. Law and Order 1996;44: 75-6.
Keall MD, Povey LJ, Frith WJ. The relative effectiveness of a hidden versus a visible speed camera programme. Accid Anal Prev 2001;33: 277-84.
Keall MD, Povey LJ, Frith WJ. Further results from a trial comparing a hidden speed camera programme with visible camera operation. Accid Anal Prev 2002;34: 773-7.
Swali L. The effect of speed cameras in West London. In: Traffic management and road safety: proceedings of a PTRC seminar at Manchester University, September 1993. PTRC Education and Research Services Ltd, 1993.
Department for Transport. A cost recovery system for traffic safety cameras—first year report. London: Department for Transport, 2002.
Highways Agency. West London speed camera demonstration project: analysis of accident data 36 months before and after implementation. London: London Accident Analysis Unit, 1997.
Chen G, Meckle W, Wilson J. Speed and safety effect of photo radar enforcement on a highway corridor in British Colombia. Accid Anal Prev 2002;34: 129-38.
Tay R. Do speed cameras improve road safety? In: Traffic and transportation studies: international conference on traffic and transportation studies. Beijing, China, July 2000: 44-57.
Christie SM, Lyons RA, Dunstan FD, Jones SJ. Are mobile speed cameras effective? A controlled before and after study. Inj Prev 2003;9: 302-6.
Cameron M, Cavallo A, Gilbert A. Crash-based evaluation of the speed camera program in Victoria 1990-1991. Phase 1: general effects. Phase 2: effects of program mechanisms. Victoria, Australia: Monash University Accident Research Centre, 1992 (report 42).
Diamantopoulou K, Cameron M. An evaluation of the effectiveness of overt and covert speed camera enforcement achieved through mobile radar operations. Victoria, Australia: Monash University Accident Research Centre, 2002 (report 187).
Department for Transport. A cost recovery system for speed and red light cameras—two year pilot evaluation. London: Department for Transport, 2003.
Bourne M, Cooke R. Victoria's speed camera program. In: Clarke RV, ed. Crime prevention studies 1. New York: Criminal Justice Press, 1993: 177-92.
Chen G, Wilson J, Meckle W, Cooper P. Evaluation of photo radar program in British Colombia. Accid Anal Prev 2000;32: 517-26.
Mara MK, Davies RB, Frith WJ. Evaluation of the effects of compulsory breath testing and speed cameras in New Zealand. Proceedings—Australian Road Research Board 1996;18: 269-82.
Hooke A, Knox J, Portas A. Cost benefit analysis of traffic light and speed cameras. London: Police Research Group, 1996 (police research series paper 20).
Hook D, Kirkwood A, Evans D. Speed cameras in Oxfordshire. Highways and Transportation 1995;42(2): 11.
Hess S. An analysis of the effects of speed limit enforcement cameras with differentiation by road type and catchment area . London: Centre for Transport Studies, Imperial College, August 2003.
Morton V, Torgerson DJ. Effect of regression to the mean on decision making in health care. BMJ 2003;326: 1083-4.(Paul Pilkington, lecturer in public heal)
Correspondence to: P Pilkington paul.pilkington@uwe.ac.uk
Abstract
Road traffic collisions are an important cause of death and disability worldwide. Every year around the world 1.2 million people are killed and up to 50 million are injured or disabled as a result of road traffic collisions.1 Morbidity from road traffic collisions is expected to increase in future years, and it is estimated that road traffic collisions will move from ninth to third place in the global burden of disease ranking, as measured in disability adjusted life years.2 3
Measures to reduce traffic speed are considered essential to reducing casualties on the road.1 4 5 Speed cameras are increasingly used to help to reduce traffic speeds in the belief that this will reduce road traffic collisions and casualties, and an expansion in the use of speed cameras is under way in many countries, most notably the United Kingdom.6 The use of speed cameras is controversial, however. Vociferous opponents, including some motoring associated organisations, oppose their use, and cameras are often criticised in the media.7-9 The lack of readily available evidence of the effectiveness of cameras has made it difficult for road safety and health professionals to engage in an informed debate about the effectiveness of speed cameras.
A previous small non-systematic review of six studies found a 17% reduction in collisions after introduction of speed cameras.10 Non-systematic reviews can, however, be limited by bias. We aimed, therefore, to systematically assess the evidence for the effectiveness of speed cameras in reducing road traffic collisions and related casualties.
Methods
Controlled trials and observational studies assessing the impact of fixed or mobile speed cameras on any or all of three outcomes (collisions, injuries, and deaths) were eligible for inclusion. We considered all published and unpublished material, with no restrictions on date or language. As the effect of co-intervention is difficult to exclude in interventions such as this, studies that did not have speed cameras as the major intervention were not eligible for inclusion.
Identification of primary studies
One reviewer (PP) searched for studies by using the search strategy outlined earlier and, together with a second reviewer (SK), selected studies to obtain for possible inclusion in the review, on the basis of titles and abstracts (where available). Both reviewers then independently extracted data from each study by using a data extraction form that was piloted before use. We extracted details of the study design, aim of the study, setting of the study and nature of the roads, study period, measurement of exposure, outcome and relevant confounders, and results. We also assessed the quality of the studies with a predefined quality scale, which, in the absence of pre-existing scales, we developed and piloted ourselves. The quality scale rated studies on the basis of representativeness of study areas to general population; control areas being representative of intervention areas; objective and valid outcome(s); results provided with estimates of uncertainty; main conclusions based on primary study hypotheses; and important confounders measured and controlled for. For each of the six quality criteria, we rated the studies on a three point scale (0-2). We rated studies scoring a total of 9-12 as good quality, 6-8 as average quality, and 0-5 as poor quality (see appendix for full data extraction form and quality scale). We combined the results from the data extraction forms of the two reviewers and made decisions on inclusion in the review. We resolved disagreements by consensus.
We also extracted data on the actual number of cases in the intervention and control areas for each time period and where appropriate combined them to produce summary statistics. We calculated risk ratios with confidence intervals for before-after and experimental-control comparisons where possible.
Meta-analysis
We selected 92 studies to review, on the basis of the title or abstract of the report. After reviewing the full articles, we identified 21 studies that were potentially suitable for inclusion. Of these, two studies did not consider the intervention or outcome of interest,11 12 one study reported only secondary results without details of the methods,13 two studies did not look at the effectiveness of the introduction of cameras,14 15 and two studies were preliminary reports that were updated in later publications.16 17 After excluding these studies, we included 14 studies in the final review (see figure on bmj.com).
All the studies were observational studies; we found no randomised controlled trials. Five studies had control areas distinct from the areas where the cameras were introduced.18-22 One study used the same areas at times when cameras were not operating as a control,23 and eight studies used the same areas before introduction of cameras as the comparison group (before-after studies).10 24-30 The studies were published between 1992 and 2003. All studies were in high income countries. Six studies assessed the effect of fixed cameras,10 18-20 29 30 four studied the effect of mobile cameras,21-23 26 and four studied the effect of a combination of fixed and mobile cameras.24-28 Outcome measures in the studies were diverse and included various measures of collisions, deaths, and injuries. Three studies had a follow up period of one year following the introduction of cameras,22 26 29 nine studies had a follow up period of one to three years,18-21 23-25 27 28 and one study had a follow up period of four years.10 One study stated only that follow up data of at least one year were used.30 See table A on bmj.com for details of the studies. In terms of methodological quality, we classified no studies as being good quality, seven as average, two as average-poor, and five as poor.
All studies reported a reduction in road traffic collisions and casualties. The reduction in adverse outcomes in the immediate vicinity of camera sites varied considerably across studies, with ranges of 5-69% for collisions, 12-65% for injuries, and 17-71% for deaths at camera sites. Smaller reductions in adverse outcomes were seen over a wider area. See table B on bmj.com for full results.
Discussion
Peden M, Scurfield R, Sleet D, Mohan D, Hyder AA, Jarawan E, et al, eds. World report on road traffic injury prevention. Geneva: World Health Organization, 2004.
Murray CJL, Lopez, AD. Alternative projections of mortality and disability by cause 1990-2020: global burden of disease study. Lancet 1997;349: 1498-504.
Roberts I. War on the roads. BMJ 2002;324: 1107-9.
House of Commons Transport, Local Government and the Regions Committee. Road traffic speed: ninth report of session 2001-2002, vol 1. London: Stationery Office, 2002.
Crombie H. The impact of transport and road traffic speed on health. London: Health Development Agency, 2002.
DTLR press release. Life-saving cameras to be made more visible—Spellar, August 2001. www.press.dtlr.gov.uk/pns/DisplayPN.cgi?pn_id=2001_0359 (accessed 5 Jan 2005).
Association of British Drivers. www.abd.org.uk/ (accessed 29 April 2004).
Massey R. New camera blitz on cars. Daily Mail 2002 April 11: 1,6.
Barker P. Speed cameras: costing us millions, and costing us lives. Autocar Magazine 2003 Nov 11: 18-23.
Elvik E. Effects of accidents of automatic speed enforcement in Norway. Transportation Research Record 1997;1597: 1-19.
Ali SY, Al-Saleh O, Koushki PA.. Effectiveness of automated speed-monitoring cameras in Kuwait. Transportation Research Record 1997;1595: 20-6.
Brownfield DJ. Environmental areas—interim report on a before-and-after accident study. Traffic Engineering and Control 1980;21: 278-82.
Traffic: speed cameras prove to be lifesavers. Law and Order 1996;44: 75-6.
Keall MD, Povey LJ, Frith WJ. The relative effectiveness of a hidden versus a visible speed camera programme. Accid Anal Prev 2001;33: 277-84.
Keall MD, Povey LJ, Frith WJ. Further results from a trial comparing a hidden speed camera programme with visible camera operation. Accid Anal Prev 2002;34: 773-7.
Swali L. The effect of speed cameras in West London. In: Traffic management and road safety: proceedings of a PTRC seminar at Manchester University, September 1993. PTRC Education and Research Services Ltd, 1993.
Department for Transport. A cost recovery system for traffic safety cameras—first year report. London: Department for Transport, 2002.
Highways Agency. West London speed camera demonstration project: analysis of accident data 36 months before and after implementation. London: London Accident Analysis Unit, 1997.
Chen G, Meckle W, Wilson J. Speed and safety effect of photo radar enforcement on a highway corridor in British Colombia. Accid Anal Prev 2002;34: 129-38.
Tay R. Do speed cameras improve road safety? In: Traffic and transportation studies: international conference on traffic and transportation studies. Beijing, China, July 2000: 44-57.
Christie SM, Lyons RA, Dunstan FD, Jones SJ. Are mobile speed cameras effective? A controlled before and after study. Inj Prev 2003;9: 302-6.
Cameron M, Cavallo A, Gilbert A. Crash-based evaluation of the speed camera program in Victoria 1990-1991. Phase 1: general effects. Phase 2: effects of program mechanisms. Victoria, Australia: Monash University Accident Research Centre, 1992 (report 42).
Diamantopoulou K, Cameron M. An evaluation of the effectiveness of overt and covert speed camera enforcement achieved through mobile radar operations. Victoria, Australia: Monash University Accident Research Centre, 2002 (report 187).
Department for Transport. A cost recovery system for speed and red light cameras—two year pilot evaluation. London: Department for Transport, 2003.
Bourne M, Cooke R. Victoria's speed camera program. In: Clarke RV, ed. Crime prevention studies 1. New York: Criminal Justice Press, 1993: 177-92.
Chen G, Wilson J, Meckle W, Cooper P. Evaluation of photo radar program in British Colombia. Accid Anal Prev 2000;32: 517-26.
Mara MK, Davies RB, Frith WJ. Evaluation of the effects of compulsory breath testing and speed cameras in New Zealand. Proceedings—Australian Road Research Board 1996;18: 269-82.
Hooke A, Knox J, Portas A. Cost benefit analysis of traffic light and speed cameras. London: Police Research Group, 1996 (police research series paper 20).
Hook D, Kirkwood A, Evans D. Speed cameras in Oxfordshire. Highways and Transportation 1995;42(2): 11.
Hess S. An analysis of the effects of speed limit enforcement cameras with differentiation by road type and catchment area . London: Centre for Transport Studies, Imperial College, August 2003.
Morton V, Torgerson DJ. Effect of regression to the mean on decision making in health care. BMJ 2003;326: 1083-4.(Paul Pilkington, lecturer in public heal)