The Scottish Government

« Previous | Contents | Next »

Listen

Commentary

Figure 1 reported Accidents by severity, 1966 to 2006

1. Trends in the reported numbers of Injury Road Accidents and Casualties

1.1 Main Points

Table 1 shows the long-term trends in the reported numbers of injury road accidents and casualties, the population of Scotland, the number of vehicles licensed, the length of the road network and the volume of traffic. Information on the severities of the accidents, and of the injuries suffered by the casualties, is provided in Table 2. The numbers of injury road accidents were first recorded separately in 1966, while the numbers of casualties are available back to 1938. Figures 1 to 7 illustrate the trends in the reported numbers of injury road accidents and casualties including (in some cases) indications of the likely range of random year-to-year variations (see section 1.4). As mentioned in the introduction, the injury road accident statistics are produced from police forces' returns and cannot be comprehensive: there could be many injury accidents not reported by the public to the police, which do not appear in the returns.

The reported numbers of injury road accidents have fallen in most of the past ten years. In 2006, there were 293 fatal accidents, 29 (11%) more than in 2005, the fifth lowest number since the current records began in 1970. The number of serious injury accidents in 2006 (2,247) fell by just 3 to the lowest number since the records of serious accidents began in 1970. The number of "slight injury" accidents (10,569) in 2006 was 351 (3%) fewer than in the previous year. This was also the lowest number since current records began in 1970.

The number of people killed in road accidents in Scotland in 2006 was 314, 28 (10%) more than in 2005. The 2006 figure was the fifth lowest since current records began.

There were 2,625 people recorded as seriously injured in road accidents in 2006, 38 (1%) fewer than in 2005. This was the lowest number since the current records of the numbers of serious injuries began in 1950.

In 2006, 14,328 people were recorded as slightly injured. This is the lowest figure recorded since 1954, and was 601 (4%) fewer than in 2005.

The total number of casualties in 2006 was 17,267. This was 611 (3%) less than in 2005, and was the lowest figure since 1952.

The reductions in the numbers of accidents and casualties in recent years are even more significant given that (for example) in 2006 the number of vehicles licensed in Scotland was about a third higher than in 1996 and that traffic on Scottish roads was estimated to have grown by about a sixth since 1996.

1.2 reported Accidents

In 1966 there were just over 23,200 injury road accidents and the annual total remained around this level until 1973. Numbers then dropped considerably in 1974 and 1975 to about 20,600. This was the time of a fuel crisis when a national speed limit of 50 mph was introduced and the volume of traffic in Great Britain fell by 3% in 1974. Accident numbers increased again in 1976 and reached a peak of nearly 23,100 in 1979.

Figure 2 Scottish fatal road accidents: 1972 on
showing likely range of values (see text) around 5-year moving average

Figure 3 Scottish road accident deaths: 1949 on
showing likely range of values (see text) around 5-year moving average

In the early 1980s numbers began to fall, and did so particularly sharply in 1983 when the total number of injury accidents fell by 7% in a single year to 19,400, serious accidents fell by 13% to just over 6,400, and fatal accidents fell by 11% to 568. The year 1983 was when the 1981 Transport Act came into force and changed the law relating to drink driving, with the introduction of evidential breath testing. Compulsory front seat belt wearing and new procedures for licensing learner motor cyclists were also introduced in 1983. After 1983 the total number of injury accidents increased again to over 20,600 in 1985, and the number of serious accidents rose to just over 6,500 while fatal accidents continued to fall.

By 1987 the total number of injury accidents had fallen to under 18,700, but in 1989 it was up again to just over 20,600. 1989 was the most recent peak in the total number of injury accidents. Since 1989, the total number of injury accidents has fallen in 15 out of 17 years, and in 2006 it was at the lowest level ever recorded. The 2006 figure of 13,109 was 325 less than in 2005.

Since the late 1980s, the number of fatal accidents has fallen considerably from (e.g.) 517 in 1987 to 293 in 2006. For serious accidents, the trend has also been downwards. The number of serious accidents has fallen from (for example) 5,814 in 1989 to 2,247 in 2006, the lowest number ever recorded. The numbers of slight accidents have not changed as much over the years: while sometimes rising and sometimes falling, they remained between 12,000 and 15,000 throughout the period from 1970 to 1998. The most recent "peak" level was 14,443 in 1990. However, they fell below 12,000 in 1999, and the 2006 figure of 10,569 was the lowest since slight accident numbers were first recorded in 1970.

1.3 reported Casualties

As the numbers of accidents have fallen, so have the numbers of casualties. Therefore, this section does not repeat the previous section's detailed analysis of how the numbers have changed.

Numbers killed

The number of people killed in road accidents in Scotland in 2006 was 314, an increase of 10% over 2005. This was the fifth lowest figure recorded. With a few exceptions, there has been a fall in each year since 1978, and for most of that period the figures show a clear, steady long-term downward trend, particularly between 1982 and 1994. Since then, the numbers appear to have been fluctuating around a less pronounced downwards trend. The number in 2006 was 1% below the average for the previous five years (316).

Numbers seriously injured

There were 2,625 people recorded as seriously injured in road accidents in 2006: 38 (1%) fewer than in 2005. The 2006 figure is the lowest number since the current records of the numbers of serious injuries began in 1950. The long term trend shows that the number of serious casualties peaked in the early 1970's at around 10,000 and has generally been falling since the early 1980's. However, there has been some fluctuation around the long-term downwards trend, and there appeared to be a levelling-off when the figures for 1996, 1997 and 1998 were all around 4,050. But the downward trend subsequently resumed: the number of people seriously injured in 2006 was well below that level.

Numbers slightly injured

There were 14,328 people recorded as slightly injured in 2006: 601 (4%) fewer than in 2005, and the lowest number since 1954. Between 1970 and 1990, the figures fluctuated in a range which was broadly 17,000 to 21,000. The fall between 1990 and 1995 in the number of people with slight injuries, followed by an apparent levelling-off at around 17-18,000 in each of the years from 1996 to 1999, could have been a continuation of that pattern. However, the figures for 2000 to 2006 were all below the bottom of that range, with falls each year, suggesting a continuing downward trend.

Figure 4 Killed and seriously injured casualties
showing likely range of values (see text) around 5-year moving average

Total numbers of casualties

The total number of casualties (of all severities) in 2006 was 17,267, 611 (3%) fewer than in 2005. This represented the lowest number of casualties since 1952. Between about 1970 and 1990, the figures appeared to fluctuate greatly about a general downward trend. Subsequently, the total number of casualties fell markedly from the level of the most recent "short-term" peak (which was over 27,000 in both 1989 and 1990), before appearing to level off: the figures for each of the years from 1993 to 1998 were all within about 600 (3%) of the average of around 22,330 for those six years. However, as the totals for 1999 to 2006 were all under 21,100, with falls each year, it appears that the downward trend has resumed.

Government targets for reductions in the numbers of road accident casualties.

In 1987 the Government adopted a target to reduce road casualties by one third from the 1981-85 annual average by the year 2000. The number of people killed on the roads in Scotland in 2000 was 49% below the 1981-85 average number of fatalities per year, and therefore the target of a one-third reduction by the year 2000 was exceeded for fatalities. For seriously injured casualties, the 2000 figure was 57% below the 1981-85 average, so the target was bettered for seriously injured casualties. However, the figure of 16,618 slight casualties in 2000 was only 9% below the 1981-85 average and so the target of a one-third reduction was not achieved for slight casualties. And, the total number of casualties (of all severities) in 2000 was 24% below the 1981-85 average, and therefore the target of a one-third reduction in the total number of casualties was not met.

In March 2000, the UK Government, the Scottish Executive and the National Assembly for Wales announced a new national road safety strategy and casualty reduction targets for 2010. A separate section on the casualty reduction targets for 2010 (which appears after this Commentary) provides statistics related to these targets, plus a selection of key points. It contains charts and tables for each of the three targets showing the main trends in casualty numbers in comparison to the 1994-98 baseline averages, and to the numbers that might be expected in each year if the targets were to be achieved by means of a constant percentage reduction in each year.

1.4 The likely range of random year-to-year variation in some road accident and casualty numbers for Scotland as a whole(see Figures 2 to 5)

Because road accidents may occur "at random", the numbers of accidents, and the numbers of casualties in those accidents, can fluctuate from year to year. Figures 2 to 5 show, for Scotland as a whole, the numbers of:

• fatal road accidents (1972 to 2006);
• road deaths (1949 to 2006);
• people killed or seriously injured (1950 to 2006);
• children killed or seriously injured (1981 to 2006).

The number of years covered by each chart reflects the availability of the relevant figures. The black dots are the values in each year, and the black lines indicate the year-to-year variation. The grey dashed lines show the likely range of random year-to-year variation in the figures: based on statistical theory, one would expect that only about 5% of years would have figures outwith these ranges. Annex G describes how these ranges were produced: the limits of the likely ranges of values are calculated in a similar way to "95% confidence intervals". It also explains why they cannot be produced for all years.

Figure 5 Children killed or seriously injured
showing likely range of values (see text) around 5-year moving average

Fatal accidents, and deaths in road accidents(see Figures 2 and 3)

Figures 2 and 3 show that the number of fatal accidents is within its likely range of values in every year, and the number of road deaths is within its likely range of values in all but three years. These results are reasonable: one would expect a few years' figures to be outside the likely range of random year-to-year variation, given that there are over 30 years' figures for fatal accidents and over 50 years' figures for road accident deaths. Figures 2 and 3 therefore show that, despite the large percentage changes such as the falls in deaths of 19% between 1998 and 1999, and of 13% between 2001 and 2002, the figures almost always remain within the expected ranges. Hence, one should not put too much weight on a single "large" percentage change.

Children killed or seriously injured(see Figure 5)

Figure 5 shows that the year-to-year fluctuations in the numbers of children killed or seriously injured (for the years for which figures are readily available) are generally within the expected ranges. The exceptions are around 1994, when (we understand) health boards' policies changed, with the result that more child casualties were admitted to hospitals for overnight observation. This changed the classification of many injuries from "slight" to "serious" (since the "Stats 19" definitions state that any non-fatal injury for which a person is detained in hospital as an in-patient should be classed as "serious"). When changes in operational practice or to administrative processes have a marked effect on the statistics, the resulting year-to-year changes can be much greater than would be expected to arise due to normal random year-to-year variation - so it is not surprising that there are figures outwith the expected ranges around 1994.

Killed or seriously injured ( KSI) casualties(see Figure 4)

Figure 4 has many years' figures (around a third) outwith the calculated likely range of values. The reason for this is that statistical variability is not the only reason for year-to-year changes - other factors have contributed to sharp falls and rises in KSI casualty numbers. For example, the sharp fall shown in 1983 may be due in part to seat belt wearing becoming compulsory in that year for drivers and front seat passengers in most cars and light vans. Similarly, the sharp rise in 1994 may be due in part to the change in hospital practices referred to earlier. In effect, such factors change the underlying rate of occurrence of accidents and/or casualties, and therefore, in effect, introduce a break into the series of moving average values. The method used to calculate the likely range of random variation cannot take account of the effect of such changes.

It may be asked why only Figure 4 has many years' figures outwith the calculated interval. The reason is that the likely ranges of random year-to-year variation calculated for "small" numbers are quite "wide" in percentage terms. This is because, for a Poisson process (see Annex G), by definition, the greater the frequency of occurrence of events, the smaller the proportion that the standard deviation of the frequency (which is the square root of that number) represents of that number. For example:

• with 100 cases, the square root is 10 - or 10% of the value;
• with 400 cases, the square root is 20 - 5% of the value;
• with 10,000 cases, the square root is 100 - only 1% of the value.

As a result, if a factor (like the introduction of the compulsory wearing of front seat belts) were to cause the same percentage fall in each of the four types of accident and casualty numbers used in the charts, the following might be observed. The percentage fall could be within the relatively wide percentage range of likely random variation around the smaller numbers, but outwith the relatively narrow percentage range of likely random variation around the larger numbers. The ranges in Figures 2, 3 and 5 appear to be sufficiently wide to encompass the effects of changes such those mentioned above. (That is, the effects of the changes in their first years may fall within the likely range of random variation. Of course, over the longer-term, such changes should make significant contributions to the reductions in casualty numbers and their severity.) However, the intervals in Figure 4 include a much smaller than expected proportion of the figures. This is because the likely range of random variation for KSI casualties represents only a small percentage of the total, and factors like those mentioned above appear to have had a greater percentage effect than that in their first years.

Figure 6 All reported casualties and Slightly injured reported casualties - from 1950

2. Reported Accidents

2.1 Accidents by road type and severity(see Table 4)

Table 4 shows separate figures for trunk roads and for local authority roads. Trunk roads accounted for only small proportions of the total numbers of accidents in 2006: 30% of fatal accidents, 18% of the total of fatal and serious accidents, and 16% of all accidents. The trunk road network's shares of accident numbers in previous years were broadly similar.

When looking at changes over time in the numbers of accidents by type of road, one must remember that the changes for different types of road will be affected by the transfer of traffic away from some roads by the opening of city and town bypasses, and by the construction of new roads with higher average traffic volumes. Therefore, such figures do not provide an accurate measure of the comparative change in the road safety performance of different types of road.

Several changes were made to the trunk road network with effect from 1st April 1996. Annex E refers to them, and explains why the 1994-98 averages for trunk roads and for local authority major roads have been calculated by counting accidents which occurred prior to 1st April 1996 on the basis of whether they occurred on roads which were part of the "post- 1 April 1996" trunk road network.

2.2 Accident rates(see Table 5)

Accident rates showing the number of accidents per 100 million vehicle kilometres are contained in parts (b) and (c) of table 5. These are calculated by dividing the numbers of accidents on each type of road by the estimated volumes of traffic on those roads, which were provided by the Department for Transport, and which are available for all types of road with effect from 1993. The "five year average" accident rates were calculated by dividing the total number of accidents which occurred in each five year period by the total of the estimated volumes of traffic for the same period, rather than by calculating the averages of the individual accident rates for the five years.

Accident rates have fallen markedly since the early 1990s. The overall fatal accident rate has dropped from 0.84 per 100 million vehicle kilometres in 1996 to 0.67 in 2006; the "fatal and serious" accident rate fell from 9.61 to 5.79; and the overall accident rate (all severities) reduced from 42.55 per 100 million vehicle kilometres to 29.87. Motorways had consistently lower accident rates than A roads. Leaving aside the relatively low rate for fatal accidents, minor roads (taken together as a group) tend to have higher accident rates than major roads, and accident rates tend to be higher for "built-up" roads (roads with speed limits of up to 40mph) than for "non built-up" roads (ones with higher speed limits).

Part (c) of the table shows that estimated accident rates vary considerably by police force area. Some of this variation may be attributed to the distribution of traffic by road type within individual areas.

Figure 7 Reported Casualties: 5 year moving average (1947/51 to 2002/06)

2.3 Accidents by month by road type(see Table 6)

The numbers of injury accidents over the years 2002-2006 were fairly evenly spread throughout the year, with minor peaks in August and November, which were 8% and 9% above the average monthly number of accidents respectively. Fatal and serious accidents (taken together) varied more between the months, and their peak, which occurred in September, was 16% above the monthly average. (To allow more equitable comparisons the months are standardised to 30 days.)

On average, there were 23 fatal accidents per month in the years 2002 to 2006. The number did not vary greatly between the months: the lowest average was 18, and the highest was 29.

2.4 Accidents by light condition and road surface condition(see Table 7)

The severity of accidents is associated with the light and road surface conditions, and also with whether the accident occurs on a built-up road or on a non built-up road. Perhaps because of the higher average speeds on non built-up roads, severity rates are higher on non built-up roads than on built-up roads. And, perhaps because of poorer visibility, severity rates are higher in darkness than in daylight. For example, taking the annual averages for 2002-2006, 5% of injury road accidents on non built-up roads in darkness (76 out of 1,508) resulted in one (or more) deaths compared with 1.4% of accidents on built-up roads in darkness (35 out of 2,484) and 3.5% of accidents on non built-up roads in daylight (126 out of 3,592). Similarly, the percentage of accidents classified as either fatal or serious is higher for non built-up roads in darkness than for either built-up roads in darkness or non built-up roads in daylight.

Severity rates did not appear to be higher when the road surface condition was wet, damp or flooded, or affected by snow, frost or ice. For example, taking the annual averages for 2002 to 2006, the percentage of accidents on non built-up roads classified as fatal or serious when the road surface condition was dry was 27.5% (639 out of 2,326) compared with 23.5% (568 out of 2,420) when the surface was wet and 19.1% (63 out of 329) when it was affected by snow, frost or ice.

3. Motorists, breath testing and drink-driving

3.1 Car driver accident rates(see Table 18)

All car drivers involved in injury accidents are included in this table, whether they were injured or not, on the basis of whatever information is known about their ages and their sex. For example, someone whose sex was known, but whose age was not known, will be included in the "all ages" total for the appropriate sex. The grand total includes those for whom neither the age nor the sex was known.

As the car driver accident rates that are shown for each sex and age group are on a "per head of population" basis, rather than being based upon the numbers of driving licence holders or upon the distance driven, they can provide only a general indication of the relative accident rates for each group. The statistics do not provide a measure of the relative risk of each group as car drivers, because they do not take account of the differing levels of car driving by each group.

Car driver accident rates per head of population vary markedly by age and sex. In 2006, the overall rate was 3.9 per thousand population aged 17+. The peak occurs for males in the 17-22 age group, with a rate of 9.2 per thousand population in 2006. This rate is almost double those of females of the same age (4.8 per thousand in 2006), and males aged 30-59 (4.9 per thousand in 2006).

The overall male car driver accident rate in 2006 (4.9 per thousand) was less than in the previous year, and this was the case for each of the age groups. The overall female car driver accident rate in 2006 (2.7 per thousand) was the same as the previous year. The rates for the age groups also remained the same apart from those aged 17-22 which rose from 4.2 to 4.8 per thousand.

Between 1996 and 2006, the male car driver accident rate fell from 6.8 to 4.9 per thousand population, whereas the female car driver accident rate tended to remain around 3.0 per thousand population (with some year-to-year fluctuations) for most of the period - but it was only 2.7 per thousand in both 2005 and 2006. As a result, the overall, ratio of male to female car driver accident rates has fallen from 2.2 : 1 for 1996 to 1.8 : 1 in 2006.

3.2 Breath testing of drivers(see Tables 19, 20 and 21)

These tables cover all motorists who were known to be involved in injury road accidents (the figures do not include, for example, those involved in "hit and run" accidents who were not traced). For these tables, a motorist is defined as the driver or the rider of a motor vehicle, including (e.g.) a motorcyclist.

In 2006, 59% of motorists involved in injury accidents were asked for a breath test (the percentage varied among the police forces, from about 50% to around 82%). The breath test proved positive (or the motorist refused to take the test) for 4.1% of those drivers breathalysed. This represented 2.4% of the total number of motorists involved (including those who were not asked for a breath test). These percentages have not changed much in the past five years, although the latter two appeared to rise in 2006.

Tables 20 and 21 show the figures for each time of day for different days of the week ( Table 20) and for a number of years ( Table 21). Table 21 shows that, in 2006, 43% of the "positive / refused" cases occurred between 9 p.m. and 3 a.m.: 76 between 9 p.m. and midnight, plus 142 between midnight and 3 a.m., out of a total of 509. Table 20 shows that, using 2002 to 2006 averages, the number of "positive / refused" cases, expressed as a percentage of motorists involved in accidents, was highest (at around 16 - 17%) between midnight and 6 a.m., but varied depending upon the day of the week, from 11.1% (the average for 3 a.m. to 6 a.m. for Mondays to Thursdays) to 21.6% (3 a.m. to 6 a.m. on Sundays). Table 20 shows that although the period from 9 p.m. to midnight had the second highest number of "positive / refused" cases, the equivalent percentages were not as high, because between 9 p.m. and midnight there were many more motorists involved in accidents than between midnight and 3 a.m.

3.3 Drink-drive accidents and casualties ( see Table 22)

Table 22 shows the estimates (made by the Department for Transport) of the numbers of injury road accidents involving illegal alcohol levels. They are higher than the number of drivers with positive breath test results (or who refused to take the breath test) because they include allowances for the numbers of cases where drivers were not breath tested because of the severity of their injuries, or because they left the scene of the accident. Information about the blood alcohol levels of road users who died within 12 hours of being injured in a road accident is supplied by the Procurators Fiscal.

The estimates show that the numbers of drink-drive accidents fell by 16% and the number of casualties by 18% between 1995 and 2005 (the latest year for which estimates are available): from about 790 to roughly 660 (accidents) and from around 1,210 to some 990 (casualties). While fluctuating from year to year, the number of people killed as a result of drink-drive accidents is estimated to have fallen slightly, from about 50 in 1995 to around 30 in 2005. The number of serious casualties is estimated to have dropped by around a half (from roughly 310 in 1995 to some 170 in 2005).

4. Reported Casualties

4.1 Casualties by type of road ( see Table 23)

In 2006, "non built-up" roads accounted for two-fifths of the total number of casualties (42%: 7,264 out of 17,267). However, perhaps because average speeds are higher on non built-up roads than elsewhere, they accounted for almost three quarters of those killed (73%: 230 out of 314) and for over half of the total number of killed and seriously injured combined (52%: 1,519 out of 2,939).

Compared with 1996, the fall in the total number of casualties has been greater for built-up roads (24%) than elsewhere (16%). The difference in the numbers killed on built up roads is also greater than those on non built-up ones (down by 30% for built-up roads compared with 3% elsewhere). However, over the years, some traffic will have been transferred away from built-up roads by the opening of city and town bypasses, and by the construction of non built-up roads with higher average traffic volumes. Therefore, these figures do not provide an accurate measure of the comparative change in the road safety performance of "built-up" and "non built-up" roads.

4.2 Casualties by mode of transport(see Table 23)

A total of 10,704 car users were injured in road accidents in 2006, representing 62% of all casualties. Of these car users, 175 died. There were 2,851 pedestrian casualties (17% of the total), of whom 61 died, 781 pedal cycle casualties (5% of the total), of whom 10 died, and 1,068 motorcycle casualties (6% of the total), of whom 58 died. Because of the numbers of car user, pedestrian, pedal cyclist and motorcyclist casualties, the figures for each of these four groups of road users are the subject of separate sections, which follow this one, and are followed by a section on child casualties, which gives details of their modes of transport. Together, all the modes of transport other than the four mentioned above accounted for 1,863 casualties in 2006 (11% of the total), and for smaller percentages of the numbers of killed and seriously injured. These included 763 bus and coach users injured in 2006, of whom 57 suffered serious injuries (none died) - these low proportions presumably being due to the greater protection of their passengers by buses and coaches. There were also 393 casualties who were travelling in light goods vehicles, 191 people in heavy goods vehicles, 248 users of taxis, 94 users of minibuses and 174 people with another means of transport.

4.3 Car user casualties

A total of 10,704 car users were injured in road accidents in 2006, representing 62% of all casualties. Of these people, a total of 1,430 were either killed or seriously injured, 175 of whom died. Non built-up roads accounted for a little over half of all car user casualties (55%: 5,856 out of 10,704). Perhaps because average speeds are higher on non-built up roads, they accounted for much higher percentages of the total numbers of car users who were killed (90%: 157 out of 175) or were killed or seriously injured (75%: 1,068 out of 1,430). (see Table 23)

The number of car users killed in 2006 was 14% more than the 2005 figure. However, the number who were killed or seriously injured fell by 2% and the total number of casualties of all severities was down by 3%. Since 1996, the number killed has dropped by 5%, and there have been falls of 38% in the number who were killed or seriously injured and of 18% in the total number of car user casualties. (see Table 23)

Looking at annual averages over the years 2002-2006, the killed and seriously injured casualty rate for 16-22 year old car users was 0.87 per thousand population. This was much higher than the rate for car users in the older age groups, which varied from 0.23 to 0.51 per thousand population. (see Table 32)

On average, over the years 2002-2006, 74% of car user fatalities occurred on roads with a speed limit of 60mph. Such roads accounted for 63% of those car users who were killed or seriously injured, but for only 43% of the total number of car user casualties (of all severities). (see Table 33)

Adult car users

On weekdays, the peak time for adult car user casualties was from 4pm to 6pm. The 5pm to 6pm average of 668 (the annual average for the years 2002-2006) was 22% higher than the average of 549 in the morning 8am to 9am peak. (see Table 28)

Adult car user casualties varied by month, with fewer in the months of March and April and more between October and December. The peak month was December, which had 33% more adult car user casualties than the lowest month, March (annual averages over the years 2002-2006; months standardised to 30 days). (see Table 29)

Friday had the peak numbers of adult car user casualties over the years 2002-2006 with 15% more than the average daily number of adult car user casualties. (see Table 30)

4.4 Pedestrian casualties

There were 2,851 pedestrian casualties in 2006: 17% of all casualties. Of these, 745 were killed or seriously injured (61 died). Presumably because of the greater vulnerability of pedestrians, a high proportion (25%) of the total number of people who were killed or seriously injured were pedestrians In addition, 26% of pedestrian casualties were killed or seriously injured (745 out of 2,851) compared with 17% of all casualties (2,939 out of 17,267). About 95% of pedestrian casualties occurred on built-up roads (2,717 out of 2,851). Perhaps because of higher average speeds on non built-up roads, 50% of the pedestrian casualties on such roads were killed or seriously injured (67 out of 134) compared with 25% on built-up roads (678 out of 2,717). (see Table 23)

The number of pedestrians killed and seriously injured in 2006 was 1% more than 2005 and the overall number of pedestrian casualties was 6% less. Since 1996, the number of pedestrians killed has fallen by 42%, the number who were killed or seriously injured has dropped by 42%, and there has been a 34% reduction in the total number of pedestrian casualties. Looking at the annual average for the period 2002 to 2006, the pedestrian fatality rate was higher for those aged 70+ (0.04 per thousand population) than for any other age-group. However, the 12-15 age-group had the highest 'killed and serious' and 'all severities' pedestrian casualty rates (0.42 and 1.88 per thousand population, respectively). The corresponding casualty rates for the 5-11 age-group were only slightly lower. (see Tables 23 & 32)

The overall pedestrian 'all severities' casualty rate for males was 0.75 per thousand population, compared with 0.46 per thousand for females, using the averages for the period 2002 to 2006. (see Table 34)

Adult pedestrian casualties

On average, in the period 2002 to 2006, on weekdays, the peak time for adult pedestrian casualties was from 4pm to 6pm; at weekends it was from midnight to 2am. (see Table 28)

November and December were the peak months for adult pedestrian casualties, with each having 20-21% more than the monthly average. Adult pedestrian casualties in the four "winter" months, November to February, were 16% more than the monthly average (annual averages over the years 2002-2006; months standardised to 30 days). (see Table 29)

Friday and Saturday have the highest numbers of adult pedestrian casualties; respectively 21% and 18% more than the daily average over the period 2002 to 2006. (see Table 30)

4.5 Pedal Cycle Casualties

There were 781 pedal cycle casualties in 2006, the same as the previous year. The combined total of killed and seriously injured pedal cycle casualties in 2006 was 141, 7% more than in 2005. There were 10 pedal cycle fatalities in 2006, 6 fewer than in 2005. Since 1996 there has been a 40% reduction in all pedal cycle casualties, the number who were killed or seriously injured has fallen by 35%, and the number of fatalities has fluctuated between 7 and 16. In 2006, 89% of pedal cycle casualties were on built-up roads. (see Table 23)

In terms of the averages for the period 2002 to 2006, the pedal cycle casualty rate per head of population was highest for those aged 12-15 (0.38 per thousand population) and 5-11 (0.35 per thousand). The other age groups with above-average casualty rates were: 16-22, 23-29 and 30-39. Of course, it must be remembered that, as noted earlier, "per capita" casualty rates do not provide a measure of the relative risk, because they do not take account of the levels of usage of (in this case) pedal cycles. (see Table 32)

Adult pedal cycle casualties

Using the averages for the period 2002 to 2006, on weekdays, the peak numbers of adult pedal cycle casualties were from 4pm to 6pm and from 8 am to 9 am. At weekends the numbers were smaller, and there was no clear peak. (see Table 28)

The peak month of the year for adult pedal cycle casualties was August, which was 30% more than the monthly average (2002-2006 annual averages standardised to 30 days). (see Table 29)

The day of the week with the peak numbers of adult pedal cycle casualties was Wednesday, 29% higher than the daily average, over the years 2002-2006. There were substantially fewer adult pedal cycle casualties on Saturday and Sunday, with 42% and 29% less than the daily average respectively. (see Table 30)

4.6 Motorcyclist casualties

A total of 1,068 motorcyclists were injured in road accidents in 2006, representing 6% of all casualties. Of these, 408 were either killed or seriously injured, of whom 58 died. Just under half of all motorcyclist casualties occurred on non built-up roads but (perhaps because of their higher average speeds) such roads accounted for about three-fifths of those killed or seriously injured, and four-fifths of those killed. (see Table 23)

The number of motorcyclist casualties in 2006 was 1% fewer than in the previous year. The number killed rose by 24 and the total who were killed or seriously injured increased by 1%. Over the period since 1996, the total number of motorcyclist casualties rose in each of the next five years to a peak in 2001; since then, it has tended to decline. As a result, the figure for all casualties in 2006 was 26% higher than in 1996. Twenty nine more motorcyclists died in 2006 than in 1996. (see Table 23)

On average, over the years 2002 to 2006, the motorcyclist casualty rate was highest for the 16-22 and 30-39 year old age groups (0.47 and 0.44 per thousand population respectively), followed by the 23-29 and 40-49 age groups (0.35 and 0.32 per thousand population); other age-groups had much smaller casualty rates. (see Table 32)

Looking at the averages for the period 2002 to 2006, the peak time of day for adult motorcyclist casualties was 4 p.m. to 6 p.m. on weekdays (see Table 28), the peak months of the year were June and August, with relatively high numbers in the other months from May to September (see Table 29) and there were more casualties on Sundays than on any of the other days (see Table 30).

4.7 Child casualties

There were 2,021 child casualties in 2006, representing 12% of the total number of casualties of all ages. Of the child casualties, 373 were killed or seriously injured, of whom 25 died (see Table 24).

There were 14 more children killed in 2006 than in 2005 and a rise of 1% in the number of children killed or seriously injured. The total number of child casualties fell by 7%. Since 2002, the number of children killed has risen by 11, there has been a reduction of 29% in child killed and seriously injured casualties, and a 26% fall in the total number of child casualties. (see Table 25)

In terms of the averages for the period 2002 to 2006, on weekdays, the peak time for child casualties was from 3 p.m. to 5 p.m., with 28% of all weekday casualties in those two hours. A further 28% occurred in the three hours between 5 p.m. and 8 p.m. There was a smaller peak in the morning, between 8 a.m. and 9 a.m. There was no real clear peak at weekends: the numbers of casualties were very broadly the same each hour from 1 p.m. to 6 p.m. (see Table 27)

August was the peak month for child casualties, with 26% more than in an average month. June and September both had 12-13% more than an average month. (2002-2006 annual averages standardised to 30 days). (see Table 29)

Using the averages for 2002 to 2006, Friday was the peak day of the week for child casualties, with 21% more than an average day. Sunday, on the other hand, had 20% less than an average day. (see Table 30)

Child casualties by mode of transport

In 2006, there were 992 child pedestrian casualties. They accounted for 35% of all pedestrian casualties of all ages (992 out of 2,851). Of the child pedestrian casualties, 247 were killed or seriously injured (9 died). (see Table 24)

There were 209 child pedal cycle casualties in 2006 (27% of the total of 781 pedal cycle casualties of all ages). The child pedal cycle casualties included 40 who were killed or seriously injured, 5 of whom died. (see Table 24)

In 2006, there were 657 child casualties in cars, 6% of the total number of car user casualties of all ages (657 out of 10,704). Of the child casualties in cars, 70 were killed or seriously injured (10 died). (see Tables 23 and 25)

Child casualty rates (per head of population)

Children's casualty rates (per head of population) increase with age: using the averages for the years 2002-2006 taken together, for children aged 0-4 the rate was 1.15 per thousand population, whereas it was 2.69 per thousand for those aged 5-11 and for the 12-15 age group it was 3.67 per thousand. The pedestrian casualty rate for younger children (0-4 years) was less than a third of those for 5-11 and just over a fifth of the 12-15 year old rate. (see Table 32)

The pedestrian casualty rate for boys in the 5-11 age group was almost twice that for girls. The difference between the sexes was even more pronounced in the case of the "driver or rider" casualty rates, particularly for the 12-15 age group. (see Table 34)

The overall child pedestrian casualty rates for killed and seriously injured and for all severities, at 0.29 and 1.23 per thousand child population respectively, were roughly double the corresponding rates for pedestrian casualties of all ages. (see Table 32)

4.8 Casualty rates for local authority roads by local authority area, and the likely range of random year-to-year variation in these figures(see Annex H)

There can be some large percentage year-to-year fluctuations in the numbers of some types of casualty for local authority areas. In order to illustrate this, the table and charts in Annex H were prepared in 2006 and published in "Road Accidents Scotland 2005". They provide the following overall casualty rates (calculated per 100 million vehicle kilometres) for local authority roads in each local authority area for 2003:

• child killed and seriously injured casualty rate;
• (all ages) killed and seriously injured casualty rate;
• slight casualty rate.

These figures were calculated (or taken) from the data in two of the tables in "Road Accidents Scotland 2005":

• the numbers of children killed and seriously injured, and the total number of people killed and seriously injured - Table 40; and
• the number of slight casualties, the estimated volume of traffic (in millions of vehicle kilometres) and the resulting slight casualty rate - Table 41.

They have not been updated to take account of any subsequent revisions to the underlying data. The same applies to all the figures which are quoted in this section. The table in Annex H also shows the likely upper and lower limits of the ranges within which these casualty rates would be expected to fall, given the likely random statistical variation that might affect the number of casualties in that year. Based on statistical theory, one would expect that the actual figures would be outwith these ranges in only about 5% of cases. The text in Annex H describes how the ranges were calculated, using the annual averages for 2001 to 2005, as that is the five year period centred on 2003 (the year to which the casualty rates relate). That is why the table and charts are not for 2005: the calculation of ranges for 2005 would require the annual averages for 2003 to 2007. When the table and charts were prepared, 2005 was the latest year for which data were available.

The charts which accompany the Annex H table show the actual casualty rates for 2003, casualty rates based upon the 2001-2005 annual averages, and the likely ranges of values within which the 2003 rates might fall, given the likely levels of random statistical variation in that year (calculated from the 2001-2005 annual averages). The "2003" rates are identified by black diamonds, the rates based upon the 2001-2005 annual averages by small circles, and the likely ranges of values by the thin bars which extend to either side of the small circles. (In any case where the 2001-2005 annual average is zero, there is no likely range of values as, by definition, the value for 2003 could only be zero.) For example, the "slight casualty rate" chart shows that (for local authority roads in 2003):

• Shetland had the lowest slight casualty rate (22 per 100 million vehicle-kilometres) and Glasgow the highest (99 per 100 million vehicle kilometres), as can be seen from the table;
• in both these cases, as it happens, the actual figure for 2003 is near to the rate which is based on the 2001-2005 annual average value (Table 41 of "Road Accidents Scotland 2005" showed that, in 2003, Shetland had more slight casualties than their 2001-2005 annual average numbers, whereas Glasgow had less than their 2001-2005 annual average numbers);
• Orkney and Eilean Siar had the widest likely ranges of values. This is due to their having relatively few slight casualties (2001-2005 annual averages of 42 and 54, respectively). The smaller the casualty numbers are, the greater in percentage terms the potential random year-to-year variation (this is discussed in Section 1.4 and Annex G). Edinburgh and Glasgow have much narrower likely ranges of values, because their numbers of slight casualties on local authority roads are much larger (2001-2005 annual averages of 1,575 and 2,088 respectively). The "Scotland" figure (at the foot of the chart) has a very narrow likely range of values, because it is based on an annual average of 12,927 in 2001-05.
• Few local authorities had slight casualty rates that were markedly outwith the likely range of values;
• Inverclyde was an exception: it had a slight casualty rate (48 per 100 million vehicle-kilometres) which was noticeably above the upper limit (of 45 per 100 million vehicle-kilometres) of the estimated likely range of values - in other words, the slight casualty rate that year was unusually high, compared with what would have been expected on the basis of the casualty numbers for the five-year period. Table 41 of "Road Accidents Scotland 2005" showed that its number of slight casualties in 2003 was 211, compared with the annual average of 176 for the years 2001 to 2005.

5. Comparisons of Scottish figures against those of other countries

5.1 Casualty rates: against England & Wales ( see Tables A to D on the pages which follow)

Historically, the "killed" and "killed and seriously injured" casualty rates per head of population in Scotland have been above those for England & Wales, whereas the "all severities" casualty rate has been lower in Scotland than in England & Wales. In 2006, Scotland's casualty rates were 15% higher (killed), 7% higher (killed and serious) and 25% lower (all severities). In all three cases, this represented an improvement in the position in Scotland relative to that in England & Wales (compared with the 1994-98 average).

For years, the Scottish child casualty rates per head of population have been higher than those of England & Wales for "killed and seriously injured" and slightly lower for "all severities". In 2006, the Scottish rates were 93% higher (killed) than those in England and Wales, 42% higher (killed and seriously injured) and 5% lower (all severities). Apart from those killed, this represented an improvement in Scotland's figures relative to England & Wales (compared with the 1994-98 average).

It should be noted that the ratio of the fatality rates for Scotland and for England and Wales can fluctuate markedly from year to year, particularly for the child fatality rates due to the relatively small numbers in Scotland, (which may be subject to year-to-year changes which are large in percentage terms). Therefore, subsequent paragraphs do not refer to the fatality rates for children using different modes of transport. In addition, it should be remembered that the rates for some other sub-groups may be affected by year-to-year fluctuations: for example, the numbers are relatively small for most categories of child "killed and seriously injured" casualties in Scotland.

The casualty rates of car users in Scotland have for many years been substantially higher than those of England & Wales for "killed" and "killed and seriously injured" casualties, while for "all severities" the rate has been much lower. In 2006, Scotland's car user fatality rate was 28% higher than that of England & Wales, the "killed and seriously injured" rate was 17% higher, while the "all severity" car user rate was 30% lower. For child car users, the "killed and seriously injured" rate was 47% higher in Scotland and the "all severities" rate was 23% less than that of England and Wales.

In 2006, the pedestrian fatality rate per capita was 4% higher in Scotland than England & Wales, the "killed and seriously injured" rate in Scotland was 24% higher and the "all severities" rate was 6% higher. The child pedestrian casualty rates in Scotland were 54% higher ("killed and seriously injured") and 20% higher (all severities) than those for England & Wales.

Pedal cyclists casualty rates (all ages) in Scotland were substantially lower than in England & Wales in 2006 for "killed and seriously injured" (36% lower) and for "all severities" (47% lower). The child pedal cycle casualty "all severities" rate was also lower in Scotland than in England & Wales. These differences may reflect the fact that, according to the National Travel Survey, on average, people in Scotland do not travel as far by bicycle as people in England and Wales.

Further information about the numbers of casualties in England and Wales, and for Great Britain as a whole, can be found in "Road Casualties Great Britain 2006", which is published by the Department for Transport.

5.2 Road deaths: international comparison 2005 ( see Tables E and F)

This section compares Scotland's road death rates in 2005 with the fatality rates of some countries in Western Europe and some developed countries world-wide. The comparisons involve a total of up to 33 countries (including Scotland, and counting each of the UK, Great Britain, England, Wales and Northern Ireland as an individual country). Some of the figures are not available for some of the countries. The fatality rates were calculated on a "per capita" basis (the statistics given are rates per million population), and the countries were then listed in order of their fatality rates in Table E sections (a), (b) and (c). In cases where two countries appear to have the same rate, the order takes account of decimal places which are not shown in the tables. Section (d) of the table ranks countries by a set of car user fatality rates which were calculated on a "per motor vehicle" basis (the statistics given are rates per million motor vehicles).

Tables E and F were provided by the Department for Transport, which obtained the figures for foreign countries from the International Road Traffic and Accident Database ( IRTAD) Web site, the address of which is: http://www.cemt.org/irtad/IRTADPUBLIC/ .

In accordance with the commonly agreed international definition, most countries define a fatality as being due to a road accident if death occurs within 30 days of the accident. However, the official road accident statistics of some countries limit the fatalities to those occurring within shorter periods after the accident. The numbers of deaths, and the death rates, which appear in the IRTAD tables take account of the adjustment factors used by the Economic Commission for Europe and the European Conference of Ministers of Transport to represent standardised 30-day numbers of deaths.

In 2005, Scotland's overall road death rate of 56 per million population was the sixth lowest of the 31 countries surveyed (counting each of Scotland, England, Wales and Northern Ireland as a separate country, but not counting the overall " GB" and " UK" figures).

However, Scotland's overall road safety position does not appear as good when the fatality rates of pedestrians are considered separately. In 2005, Scotland's pedestrian fatality rate was 13 per million population. Scotland ranked seventeenth of the 29 countries for which figures are available (again counting Scotland, England, Wales and Northern Ireland separately, and again not counting the " GB" and " UK" figures).

When the car user fatality rate is calculated on a per capita basis, Scotland has a low car user fatality rate (30 per million population: the eighth lowest of 29 countries, again not counting the " GB" and " UK" figures. However, it may be argued that the car user fatality rate should be calculated on "per motor vehicle" basis, in order to try to approximate better the differing levels of car use in different countries, and hence reflect differences between countries in car drivers' "exposure to risk". (Rates based on the amount of car traffic in each country would be even better, but the data required to calculate them are not available for some countries.) When car user fatality rates are calculated on a "per motor vehicle" basis, Scotland's car user fatality rate of 61 per million motor vehicles was again the eighth best out of the 28 countries for which figures are available (again, not counting the " GB" and " UK" figures).

The fatality rates per head of population for 29 countries (including Scotland, England, Wales and Northern Ireland as separate countries, but not counting the overall " GB" and " UK" figures) are shown, for each of four broad age-groups, in Table F. Again, the ordering takes account of decimal places not shown in the table. In most cases, Scotland has one of the lowest rates per capita. The Scottish rate is ninth lowest for child casualties aged 0-14, the fifth lowest for those aged 15-24, the sixth lowest for those aged 25-64 and the fifth lowest for those aged 65+ (in each case, not counting the overall " GB" and " UK" figures).

International comparisons of road safety are based on road death rates, because that is the only basis for which there is an international standard definition. As indicated above, the OECDIRTAD tables provide comparable figures for each country, after making adjustments to the data for countries which do not collect their figures on the standard basis. One should not try to compare different countries' overall road accident casualty rates (i.e. the total numbers killed or injured, relative to the population of each country) because there is no internationally-adopted standard definition of a injury road accident. There are considerable differences between countries in the coverage of their injury road accident statistics. For example, many countries count only accidents which result in someone being admitted to hospital - so their figures would not include the kinds of accident which, in Britain, are classified as causing only "slight" injuries or certain types of serious injury. Because many countries' definitions of injury road accidents are much "narrower" than the definition used in the UK, their reported numbers of injury road accidents will appear low relative to ours - so comparing the reported numbers of people injured in road accidents may provide a misleading impression of different countries' road safety records.

Table A Casualties in Scotland, England & Wales by severity
Number of casualties : All ages and child casualties

Table B Casualties in Scotland, England & Wales by severity
Rates per 1,000 population : All ages and child casualties

⁽¹⁾ Child 0-15 years

Table C Casualties in Scotland, England & Wales by mode of transport and severity, 2006
Number of casualties : All ages and child casualties

Table D Casualties in Scotland, England & Wales by mode of transport and severity, 2006
Rate per 1,000 population : All ages and child casualties

⁽¹⁾ Child 0-15 years

Table E International Comparisons
Fatality rates per capita, for (a) all road users, (b) pedestrians, (c) car users; and (d) per motor vehicle for car users: ranked by respective rates - 2005 (as recorded in IRTAD⁽¹⁾)

(1) Source: International Road Traffic and Accident Database ( OECD). The basis of the numbers is described in the text.
Some of the countries may have updated their figures since they provided the data to OECD.

Table E (continued) International Comparisons
Fatality rates per capita, for (a) all road users, (b) pedestrians, (c) car users; and (d) per motor vehicle for car users: ranked by respective rates - 2005 (as recorded in IRTAD⁽¹⁾)

(1) Source: International Road Traffic and Accident Database ( OECD). The basis of the numbers is described in the text.
Some of the countries may have updated their figures since they provided the data to OECD.

Table F International Comparisons ⁽¹⁾
Road accident fatality rates per capita, by age group, ranked by respective rates - 2005

(1) Source: International Road Traffic and Accident Database ( OECD). The basis of the numbers is described in the text.

« Previous | Contents | Next »