Speed cameras for the prevention of road traffic injuries and deaths

asteric* denotes studies reporting both speed and crash outcomes.

About 70% reduction in speeding vehicles when police present. A time halo effect lasted at least 2 days.

Overall mean speed reduction of about 2‐3km/h seen at both experimental and control sites. Distance halo of up to 14km downstream was seen. Time halo for at least the day after enforcement.

Main findings= 5.8km/h reduction in mean speeds at 24 months at camera sites. For percentage exceeding speed limit, a 70% reduction at camera sites and a 20.9% reduction downstream at 12 months. These reductions were maintained at 24 months. For p...

Reduction in speed and narrowing of speed distribution for 60km/h, 70km/h and 100km/h zones at experimental sites. No further information given.

Sharp reduction of 5 km/h in median speed on experimental roads after program launch. No further reduction observed despite intensified enforcement. However reduction sustained throughout the 15 month trial period.

Significant average 3.6 km/h decrease in mean speed compared to before. No time halo effect seen.

Net 3.4% speed reduction after consideration of changes in average speeds that occurred at control site during the same before‐after periods. The proportion of drivers exceeding the 80km/h posted speed limit decreased by a net 66%. For speeders ov...

A reduction from 66% to 35% in the percentage of speeding drivers at experimental sites compared to before period.

A time halo effect lasted for 3 days following a single application of enforcement. With 5 days of consecutive enforcement a time halo of at least 6 days was noted. At the site of speed limit enforcement the average speed was reduced to around the...

2.8 km/h reduction in mean speed at monitoring site, 2km from experimental site.

Reductions in median speed by 40 km/h and 28km/h in left and middle car lanes and 23km/h in truck lane on autobahn, sustained through 10 years of continuous enforcement. Also reductions in 85th percentile speeds of 42 and 37km/h in car lanes and 3...

During pilot program an overall 2.4 km/h reduction in mean speed with only small differences between the 3 cities. A 10.4% reduction in proportion of drivers exceeding the speed limit and a 4.5% reduction in those exceeding the speed limit by 10km.

On 80 km/h sections an 8% reduction of speeding vehicles seen on experimental sites compared to control sites in year 1, with a further 2% decrease in year 2. On 100 km/h stretches a 5% reduction in number of speeders seen in year 1, with a furthe...

Reduction in mean speeds by an average of 4.4 mph, 85% percentile speed reduction of 5.9 mph and a 35% reduction in percentage over the speed limit. These reductions were comparable for all monitored lengths e.g. up to 250m, 500m and 1 km.

65% reduction in speeding vehicles in excess of 15km/h or more, over the speed limit.

Mean speed reduction of 4 km/h on experimental roads compared to 1.5 km/h on control roads. A 12% reduction in percentage of drivers over the allowed speed limit on experimental roads compared to 5% reduction on control roads.

Aggregation of data from the 4 roads in provinces (compared to phase 0) showed a reduction in average speed of 3km/h in phase 1 and 5km/h in phase 2, a reduction in 85 percentile speed of 3km/h in phase 1 and 8 km/h in phase 2. A reduction in prop...

Reduction in average speeds by 0.9 to 4.8 km/h for all times of day reported. 10% reduction in proportion of speeding drivers. A time halo effect of up to 8 weeks was shown.

Average reduction in mean speed 2.3 km/h (p = 0.05) at camera sites post 1st year. Average reduction in mean speeds 1.3 km/h (‐2.6 to 0.1)post 2nd year on experimental open roads

At the Scottsdale sites mean speeds decreased from 70 mph to about 65 mph. At the non‐Loop 101 sites (where no before data was available) mean speeds fluctuated between 68 and 69 mph. The proportion of vehicles in excess of the tolerated speed lim...

Mean speed at the 5 sites (where cameras were deployed) decreased by 10% (range 5% ‐18%) The proportion of drivers travelling > 10 mph above the speed limits declined by about 70% at locations with both warning signs and speed camera enforcement, ...

A 0.91mph significant reduction in mean speeds at experimental sites with a small insignificant reduction at control sites, 0.99mph significant reduction in 85th percentile speeds at experimental sites with a small insignificant reduction at contr...

A significant 14% reduction in mean speeds compared to control sites. The proportion of vehicles exceeding the speed limit by more than 10mph decreased by 82%.

23% crash reduction during the day and a 21% reduction at other times.

Aggregate changes in crashes showed an insignificant 19.7% reduction in all reported crashes with a 20.1% reduction in injury crashes and an 89.8% reduction in fatal crashes along camera segments.

A significant 11% reduction in total crashes outside of city. Estimated 31% reduction in fatal crashes.

Greatest crash reductions in areas within 2km of camera sites, with 31% (1100), 39% (2200), 19% (500) and 21% (1600) reductions estimated for hospitalisation, medically treated, other injury and non‐injury crashes respectively.

A significant 58% reduction in serious or fatal injury crashes.

An estimated 7% reduction in overall injuries, 10% reduction in serious injury, and a 13% reduction in fatal crashes was reported

A net reduction of 71.3 % (p = 0.064) in injury crashes up to and including 4 days after the presence of enforcement. A net reduction of 73.9% (p =0.045) on the day when a mix of overt/covert enforcement employed.

A 25% reduction in expected crash rate, 11% reduction in injury crashes and a 17% reduction in daytime crash fatalities.

A reduction of 16% (+/‐ 7%) in expected crashes along study corridor. A 2.8km/h reduction in mean speed at monitoring site.

Personal injury crash frequency reduced by a ratio of 1:1

A 22% reduction in injurious crashes in first year and 20% in second year post intervention compared to before.

A non‐significant 19% reduction in crashes (not defined) A 20% reduction in injury crashes 2 years post intervention

After accounting for all potential confounders, the overall average effect of speed cameras on crashes was a reduction of 25% in personal injury crashes (PIA's) and an average 11% reduction in fatal and serious crashes for the 500m monitoring leng...

At camera sites a reduction of 19% for all crashes and a reduction of 44% for fatal and serious crashes. The authors state that the reduction in total crashes was significantly greater than that expected from the effect of RTM in 12 out of the 20 ...

Reduction of 50% in personal injury accidents (0‐500m route)

A significant 56% reduction in fatal crashes relative to controls, a 12% reduction for all crashes and a 25% for serious crashes.

23% reduction in number of injury crashes. 66% reduction in fatal crashes.

Aggregate change in crashes was a significant 12% reduction for experimental roads.

A reduction of 35% in the total number of crashes for phase 1 and 2 compared to the control roads and the same period 3 years previously.

Phase 3 ‐ conducted in 1994 on one road in one province Noord‐Brabant. The reduction in crashes found in 1991 had been maintained.

A significant 20% reduction in injury crashes equating to about 62 injury crashes per year on all 64 road sections combined.

Reduction of 9.17% in crashes and a significant 32.4% in serious injury compared to New Zealand overall.

Total crash reduction at enforcement zones of 44 to 55%, injury crash reduction of 46 to 56% with no reduction seen in rear‐end crashes

Total aggregate crash reduction of 12% on experimental corridors

Phase 1 ‐ General effects ‐ effect of speed camera program on crash rate and severity covering first twelve month post intervention period. Phase 3, 4 and 5, of this 5‐phase study did not have before‐after data, hence only phase 1 and 2 tabled.

Phase 2 ‐ Impact of program mechanisms i.e. number of traffic infringement notices (tin's) hours of speed camera operations, amount of paid TV publicity, demerit points in relation to, injury crash incidence and severity during low alcohol times o...

Proportion of speeding vehicles in May 1996 (before) at photo radar sites was 66%. By July 1997(after) it was about 35%. At control sites speeding vehicles in September 1995 was about 73%, in November 1996 it was about 61%.

About 25% reduction in crashes to that expected. A significant 11% reduction in number of victims, with an average decrease of 139 daytime traffic crashes requiring an ambulance. A 17% reduction in daytime crash fatalities.

Outcome measures were numbers of road crashes, numbers of people injured (combined fatal and non‐fatal) and number of vehicles involved in crashes. The relative risk of a road crash after implementation was 0.73 (5% CI = 0.63, 0.85) This effect wa...

A 45.74% reduction in weighted injury crashes within 250 metres from camera sites (total monthly before count of 23.4 compared to total monthly after count of 12.7) with lesser but still significant decreases observed in the wider surrounding area...

Speed outcomes measured: Mean speed, 85th percentile speed, percentage of vehicles over the speed limit and percentage exceeding the speed limit by more than 15mph. Speed results across all new camera sites included a 2.3mph reduction (6% reductio...

A 30% reduction in vehicles exceeding the speed limit, and a 43% reduction in vehicles exceeding the speed limit by more than 15 mph. Time over distance cameras were shown to be particularly effective (around 100% reduction) at reducing speeds exc...

Crash outcomes measured: Killed and seriously injured (KSIs) and personal injury crashes (PICs). Across all sites KSI's decreased by 42%. This equated to about 1700 fewer KSI's per annum at these camera sites. Fixed cameras in both urban (47% redu...

Over the first 2 years of the study mean speeds fell by 1.3 km/h (‐2.6 to 0.1) and 85th percentile speeds reduced by 4.3 km/h on open roads generally,compared with the rest of the country.

Having controlled for season and general trends, injury crash outcomes in experimental areas relative to control areas were a net reduction of 11% (p= 0.03) in all crashes and 19% (p=0.002) in injury crashes on open roads across the region associa...