U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
|This report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-RD-99-089
Date: December 1999
Research, Development, and Implementation of Pedestrian Safety Facilities in the United Kingdom
2. Pedestrian Safety in the United Kingdom
2.1 Casualty statistics
Road accidents and casualties are the conventional means of assessing safety. In Great Britain1 in 1996, there were 320,302 reported casualties of which 46,381 (14.5%) were pedestrians (DETR,1997a).
In certain areas and sub-groups pedestrians form a higher percentage of casualties. They make up approximately one in three of all road users killed, some 20 percent of casualties in built-up areas, and casualty rates for child pedestrians (particularly 11-13 years old) and elderly pedestrians are particularly high. Details of pedestrian accident statistics are provided annually in Road Accidents Great Britain (DETR, 1997a). Commentaries on the pedestrian statistics and trends can be found in the background paper the National Walking Strategy (DOT, 1996a), Lambert (1997) and, on child pedestrians, O'Reilly (1994).
The GB pedestrian fatality rate is around 2.0 per 100,000 population. In 1994, the GB (and the UK) rate was the eighth lowest when compared to other European Union member countries, and almost the same as the United States rate (2.2). Differences in exposure (the amount of walking) population profile, modal split, and other factors may explain many of the differences and need to be taken into account when making comparisons. The GB fatality rate (all modes) is 6.4 per 100,000 population; this is the lowest overall in Europe (Sweden, 6.7: Netherlands, 8.5; and Finland, 9.5) and considerably lower than some other countries (US, 15.6; Greece, 20.3; Portugal, 28.7) (DETR 1997a, Table 48).
There are significant regional differences in pedestrian casualty rates within the United Kingdom. Wales has the lowest rate (1.8 per 100,000) followed by England (2.0), Scotland (2.2), and Northern Ireland (2.7).
The number of GB pedestrian fatalities declined rapidly at the end of the second World War (1945) to 2063 in 1952, then rose during the 1960s to a peak of 3,153 in 1966, and then declined during the 1970s to 2,335 in 1976, with a steep decline from the late 1980s onwards. Each year since 1990 has seen a new record low number of pedestrian fatalities. Total pedestrian casualties have also declined but at a considerably lower rate from an average 1981-85 of 61,741 to 47,028 in 1996.
Calculating pedestrian casualty rates in relation to pedestrian exposure is difficult because adequate pedestrian activity data are generally lacking. A major study in the United Kingdom (Ward et al, 1994) did provide accident rates for a range of pedestrian environments and pedestrian groups. This found that, for example, there were 411 casualties per 100 million km (61 million mi) walked or 66 casualties per 100 million roads crossed.
Reported accidents — the "official"road accident statistics, recorded by the police on STATS 19 forms and collated by local and central government — are known to underestimate the severity of pedestrian injuries (Hopkin et al., 1993). Also, pedestrian accidents not involving a vehicle (falls, trips, etc.) are not classified as road accidents and are not reportable to the police. The number of footway falls by pedestrians is hard to determine accurately but some local authorities are now paying more for injury compensation claims than they are spending on footway maintenance (Kindred Associations, 1995). Accidents between pedestrians and cyclists are rarely reported, and some police forces instruct staff not to record them.
2.2 National road safety targets
Targets for 2000
In 1987, national road casualty reduction targets were set by the Government. A "headline" target of a one third reduction from the 1981-85 average was to be achieved by the year 2000. There were also subsidiary targets for certain groups. Targets for pedestrians were to:
The targets were based on trends and projections, plus estimates of what might be achieved through concerted efforts. To date the target for fatal pedestrian casualties has been achieved (table 1 and figure 1 show casualties during recent years). The DETR publishes an annual review of road safety which includes progress towards targets (DETR, 1997b). The projection for the period 1997-2001 for all road user casualties is for a slight fall in fatal and serious injuries and a slight rise in slight injuries (DETR, 1997c p26). Falls in pedestrian casualties are likely to be outweighed by rises in casualties to drivers and
Targets beyond 2000
Road safety targets for beyond the year 2000 are now being devised (PACTS 1995). It has been suggested that there should be separate targets for fatal/serious injuries and slight injuries; and that the targets should relate to levels of exposure, particularly for pedestrians and cyclists. Some organizations have argued that measures of danger, such as vehicle speeds and modal split, should be included in the targets in addition to casualty numbers. At this stage (early 1998), the new targets are, at least publicly, yet to be decided. However, the Government has announced that new targets will be adopted for the period 2000-2010 and that these will include a headline target and subsidiary targets for different severities (LTT 1997).
Table 1. UK Pedestrian and Car Casualties for 1981-1996
Figure 1A. Car occupant casualties.
Figure 1B. Pedestrian casualties.
Figure 1C. Pedestrian fatalities.
Figure 1. Casualties and fatalities.
2.3 Explaining the trends
Describing changes in accident numbers and trends is relatively straightforward. It is far more difficult to say, with hard evidence, the reasons for the reduction in casualty numbers and equally difficult to say whether declining accidents signify improvements in road safety in a broader sense.
All road users
Commenting on the fall in all accident numbers compared to the 1981-85 baseline (figure 2), the DETR states:
"It is not possible to say with certainty why most casualty rates and numbers have reduced. There are many factors in play, but it is likely that the principal reasons are:
exposure: people are walking, cycling and motorcycling less, so the number of casualties is falling. The number of car occupant slight injuries is rising at least partly because of the increase in traffic;
safer cars: because cars are more robustly constructed and seat belt wearing rates have risen, car casualties are less severe;
safer roads : new road construction and local safety measures (e.g., traffic calming) have contributed to preventing accidents and reducing casualty severity;
anti drink drive: the number of fatal accidents in which an involved driver had been drinking over the limit has fallen by 57 percent since the baseline;
changing attitudes: there is less tolerance of road accidents than there was. In 1987, when the target was set, there was deep concern about public acceptance of road casualties. " (DETR, 1997b p5)
Over the period of 30 years, there has been a major shift from walk trips to car trips and this has accelerated over the past few years as car ownership has increased. The average distance walked declined by 18 percent between the 1981-85 baseline and 1995 (DETR 1997c) (figure 3). There has been a particularly large decline in walking by children, one of the pedestrian groups with the highest casualty rate. There has also been increased concern about personal security, and it is possible that this has reduced walk journeys after dark, which would also tend to have a disproportionate effect on accident numbers. Demographic changes also need to be considered. "It is not clear that the rate [per 100,000 km walked] of pedestrian casualties has fallen, and an upturn in serious injuries to child pedestrians since 1993 suggests that, for this subgroup, the risk is increasing." (DETR, 1997c p11)
Figure 2. Changes in pedestrian and car user casualties.
Figure 3. Average distance walked per person per year (1975-1995). (1 mile = 1.61 km)
Ninety-five percent of pedestrian casualties occur on "built-up" roads — defined as roads with a speed limit of 64 km/h (40 mi/h) or less. Most of these have a 48 km/h (30 mi/h) limit. Several research projects have attempted to address the fundamental question of why accidents occur, including pedestrian accidents. These include Carsten et al (1989) — all road users; Lawson (1990) — child pedestrians; and Davies and Winnett (1993) — all pedestrians.
Among the common factors identified by these studies were:
Conclusions that are drawn from these and other studies are that ordinary speeds and the 48 km/h (30 mi/h) speed limit are often too high for pedestrian safety and that it is unrealistic to expect child pedestrians to observe the same standards of traffic behaviour as adult road users. More fundamental assessments of society's propensity for accidents — inevitable results of its willingness to accept a certain amount of risk — are provided by Adams (1985).
2.4 Promoting walking and pedestrian safety
The issue of pedestrian safety has been given a new significance in the past 5 years because of increased concern about congestion and the environmental effects of traffic. Various influential reports have addressed this including that of the Royal Commission on Environmental Pollution (1994) which called for pedestrian fatality rates to be reduced from 2.2 to 1.5 by 2000. National land-use planning policy now requires that new developments are located so that they are accessible to pedestrians (DOE and DOT, 1994). The UK government, in cooperation with other agencies and voluntary bodies is currently drawing up a National Walking Strategy intended to halt and possibly reverse the decline in the amount of journeys walked, as part of an integrated transport policy. Some local authorities have already devised walking strategies for their areas. A comprehensive walking strategy has been drawn up for London (London Planning Advisory Council, 1996) although this does not deal in detail with the safety of pedestrian facilities (figures 4 and 5).
The decline in children walking to school is particularly relevant. Although few accidents occur on the journey to and from school, parents regularly cite traffic danger as the main reason for not allowing their children to walk (or cycle) to school. Perception of danger is difficult to measure and even harder to reliably compare over time or between places and individuals. However, it is probably true to say that the ordinary person in the street would not agreed with the professional's view that the UK's roads are now safer for pedestrians. There is some policy tension between targets to reduce road accidents and policies to encourage walking (and other sustainable but vulnerable modes such as cycling).
Source: National Travel Survey 1993/1995 (Excludes journeys under one mile (1.61 km))
Figure 4. Distance walked per person per year by journey purpose.
Source: National Travel Survey: 1994/1996
Figure 5. Distance travelled per person per year by car by journey purpose.
Although the United Kingdom probably has as much walking as most other comparable countries, and a lower pedestrian fatality rate than most, there is still some concern that more walking may lead to more casualties. "It would not be sensible to encourage people to walk more if this led to more casualties." (Lambert, 1997). Increased amounts of walking may lead to more pedestrian casualties but there are likely to be offsetting reductions, particularly if increased walking is at the expense of trips made by other modes and results in changes in journey lengths and patterns. The experience of cities such as York, which have achieved their casualty reduction targets while promoting walking and cycling, suggest that a simple comparison of casualty rates per 100,000 km is misleading (White, 1994).
This issue was addressed by the House of Commons Transport Committee (1996a) in its report on road safety for vulnerable road users. It concluded that both safety and use should be promoted and that there should be greater emphasis on reducing the dangers to pedestrians and cyclists and not restricting their movement. (For the Government's response, see House of Commons Transport Committee, 1996b.)
2.5 Child and elderly pedestrians
Whilst the UK pedestrian casualty rate of 2 per 100,000 pop is low, the rate for child pedestrians is acknowledged to be higher than in most other European countries: 1.3 fatalities per 100,000 children, nearly one third higher than the European Community average (Lambert 1997). The child pedestrian accident problem is addressed in DETR (1997c) and in section 9 of this report. A detailed study was undertaken by Lawson (1990). The causes of child pedestrian accidents are also analysed by Lynam and Harland (1992) and O'Reilly (1994). The higher rate of child pedestrian accidents amongst poorer families has also been recently investigated (Christie, 1995). The DETR has recently revised its strategy for improving child pedestrian safety. In remarkably plain terms, it states that, from now on, more responsibility will be placed on drivers to avoid accidents with children (DOT, 1996b).
Elderly pedestrians are also acknowledged to be at a higher risk than average; these over 60 years old make up only one fifth of the population yet half of all pedestrian fatalities.
Pedestrian casualties are declining but so is the amount of walking. Whether pedestrians perceive improvements in their safety is not clear. A common perspective seems to be that, as the roads get busier with motorized traffic, with higher speeds and acceleration, safety is reduced.
1The main source of published road accident data is Road Accidents Great Britain, published annually. Consequently, most of the accident data presented in this report are for Great Britain (not the United Kindgom) and exclude Northern Ireland. In 1995 the total number of road accidents in Northern Ireland was 11,725.