House of Commons - Transport

Select Committee on Transport Eleventh Report

2 Progress towards the 2010 targets

7. In 1987 the Government set the first national road safety target: to reduce casualties by one third by 2000. This was seen as groundbreaking. The main target was exceeded and, by 2000, deaths had fallen by 39% and serious injuries by 45%. The overall number of accidents and of slight injuries, however, remained unchanged in the context of a 30% increase in the number of licensed motor vehicles and a 32% increase in vehicle miles travelled.[14]

8. The current targets for road casualty reduction in Great Britain were published in March 2000 in Tomorrow's Roads - safer for everyone.[15] This strategy, co-signed by the Minister for Road Safety in the Department of the Environment, Transport and the Regions and his counterparts in the Scottish Executive and National Assembly for Wales, set the targets to be achieved by the year 2010, taking the average of years 1994-98 as the baseline.[16]

9. The three targets for Great Britain were:

A reduction of 40% in the number of people killed or seriously injured in road accidents;
A reduction of 50% in the number of children killed or seriously injured (children are defined as being those aged under 16); and
A reduction of 10% in the slight casualty rate, expressed as the number of people slightly injured per 100 million vehicle kilometres.

Box 1: Definitions of "killed" and "injured" used in UK road casualty statistics

"Killed: Human casualties who sustain injuries leading to death less than 30 days after the accident. (This is the usual international definition, adopted by the Vienna Convention in 1968.)

"Serious injury: An injury for which a person is detained in hospital as an 'in- patient', injury or any of the following injuries whether or not they are detained in hospital: fractures, concussion, internal injuries, crushings, burns (excluding friction burns), severe cuts, severe general shock requiring medical treatment and injuries causing death 30 or more days after the accident. An injured casualty is recorded as seriously or slightly injured by the police on the basis of information available within a short time of the accident. This generally will not reflect the results of a medical examination, but may be influenced according to whether the casualty is hospitalised or not. Hospitalisation procedures will vary regionally.

"Slight injury: An injury of a minor character such as a sprain (including neck whiplash injury), bruise or cut which are not judged to be severe, or slight shock requiring roadside assistance. This definition includes injuries not requiring medical treatment."

Source: Department for Transport 2008: Road Casualties Great Britain, Main Results 2007 (Definitions).

10. The setting of the targets was supported by considerable statistical and policy evaluation.[17] The potential contributions to casualty reduction, beyond the existing trend lines, were calculated for a range of policy initiatives. These included additional measures to reduce drinking and driving, further work to improve secondary safety in vehicles,[18] new road safety engineering schemes and safety on rural single-carriageways (which have a high rate of fatal accidents). Thus, the targets were not merely extrapolations of trends but ambitious statements of priority and intent.

11. Subsequently, in 2002, the road safety Public Service Agreement of the Department for Transport was enhanced to reflect the higher number of road casualties that occur in disadvantaged areas. It was agreed to achieve a higher level of reduction in disadvantaged areas - defined as the 88 Neighbourhood Renewal Fund areas.

12. Slightly different targets were set for Northern Ireland:[19]

A reduction of one third in the number of people killed or seriously injured in road accidents; and
A reduction of 50% in the number of children killed or seriously injured.

These were to be achieved in the period 2002-2012 and measured against the 1996-2000 average.

13. The Government has continued to recognise improving road safety as one of its core transport objectives. Promoting safety, health and security is one of the five key objectives in Towards a Sustainable Transport System.[20] Whilst this document does not elaborate on how the strategy for sustainable transport links with the road safety strategy, the intent at least is clear.

14. Some organisations have commented positively on how the Government has moved away from dealing with road safety in isolation and is increasingly linking it to other key policy areas, such as social inclusion, young people, and improved health.[21]

15. We commend the Government on having set and maintained ambitious road traffic casualty reduction targets. We also commend it for recognising that road safety needs to be integrated with other important policy objectives such as promoting good health, reducing carbon-dioxide emissions, tackling deprivation and improving quality of life. The Government has not sought to reduce casualties by discouraging vulnerable road users from taking to the streets; but some trends, such as increased traffic, have had this effect. We recommend that in the forthcoming White Paper on sustainable transport, road safety objectives should be integrated with these wider objectives. We also recommend that the road safety strategy for beyond 2010 be explicitly set in the context of wider policy objectives. This should help to ensure that road safety is seen as relevant in other policy areas and that road safety policies do not have unintended consequences on other important objectives, such as improving public health by encouraging walking, cycling and play.

Progress since 2000

16. The Government is on track to meet all its road safety targets by 2010.[22] By 2007:

the number of people killed or seriously injured was 36 per cent below the 1994-98 average;
the number of children killed or seriously injured was 55 per cent below the 1994-98 average;
provisional estimates show the slight casualty rate was 30 per cent below the 1994-98 average; and
the additional target for reducing casualties in areas of deprivation was met in 2005.[23]

17. Not surprisingly, the Government has tended to report this as "good progress".[24] In his evidence to us, the Minister, Jim Fitzpatrick MP, was more circumspect. He told us

In general terms, casualty reduction has been good but not good enough, particularly in terms of deaths. Our target was set in terms of the combined figure for deaths and serious injuries. The trend in these has diverged unexpectedly […][25]

18. Those who have looked more closely at the figures have raised concerns. Two official reviews have been undertaken, in 2004[26] and 2007.[27] The more recent review noted:

good overall progress against the targets but serious injuries falling much more rapidly than deaths;
little progress in reducing car user deaths;
a significant rise in motorcyclist deaths;
particular concerns about male drivers, younger drivers and rural roads; and
possible changes in the level of accident reporting to the police, based on comparison of hospital and police data.[28]

19. A further report concluded that the casualty reductions anticipated in 2000 would not be achieved on current trends by 2010. Three policy measures had exceeded expectations: road safety engineering measures, improved secondary safety in cars and additional measures for speed reduction. However, in terms of casualty reductions from certain other major policy measures, little or no progress was likely to be achieved by the end of the target period due to a lack of initiatives.[29] These were notably:

improving safety on rural single-carriageway roads;
reducing casualties in drink-drive accidents; and
reducing the accident involvement of novice drivers.

20. The road safety target for deprivation was met in 2005. However, significant disparities remain in casualty rates according to levels of income. Child pedestrians from the lowest socio-economic groups are 21 times more likely to be killed in a traffic accident than those from the top socio-economic group.[30] This inequality is not restricted to child pedestrians: less affluent car users are also at greater risk of death than the more affluent.[31] Dr Christie felt that the target had not been sufficiently stretching and that, having achieved it, there was a danger that the problems would be overlooked.[32] We urge the Government to renew its focus on tackling the appalling level of child road traffic deaths associated with deprivation.

Reliability of data

21. The Government's assessment of "good progress" towards its main target (reducing the number of people killed or seriously injured by 40%) relies largely on the reduction in serious injuries, rather than deaths. There are approximately ten serious injuries reported for each death, so trends in serious injuries dominate the target. Whereas deaths have declined by 18% since the baseline, serious injuries have declined by more than twice as much (37%). This was not anticipated when the target was set.

22. The Government's monitoring data are based on police records of road accidents and casualties on a form known as "STATS19". The forms are collated by the local highway authorities and forwarded to the Department for Transport. It has long been known that not all accidents are recorded on the STATS19 system[33] and that some accident types, such as pedal cyclist accidents, are disproportionately under-recorded (see Box 2). It has been assumed, however, that these data would give a robust picture of the trends (if not the absolute numbers), just as a sample survey should do. This assumption, sometimes referred to as the "trend defence", rests on the underlying principle that, even if not all accidents are reported, the proportion which go unreported is likely to remain pretty constant. There is, however, increasing evidence that reporting and recording has changed over the period and that the STATS19 data do not give a reliable picture, particularly of the trends in serious injuries. Indeed, the trend defence is now challenged by some of the statisticians responsible for compiling the data.[34]

Box 2: Sources of under-reporting and under-recording of road accidents

Not all road accidents are "reportable": for example, if no injury occurs. The requirements to stop, provide information and report a road traffic accident are set out in the Road Traffic Act 1988 (section 170), as amended by the Road Traffic Act 1991 (Schedule 4);
There is no legal obligation for drivers to report road accidents to the police, provided the parties concerned exchange personal details at the scene;[35]
Some accidents that should be reported by drivers to the police are not reported. This may be because the driver is ignorant of the legal requirements or is reluctant to do so, for example, if the driver has been drinking or is uninsured;
The police do not record all accidents reported to them. Up to one fifth of casualties reported to the police are not recorded in the STATS19 system;[36] and
It is often difficult for a police officer to judge whether a casualty should be classified as having a serious or slight injury (see Box 1). For example, the full severity of the injury may not be apparent until some time after the collision when the police officer is no longer present. Research has found that the police tend to underestimate the severity of the injury.[37]

23. There are a number of reasons to believe that the actual decline in serious injuries is not as great as that recorded in STATS19 statistics:

Hospital data suggest that serious injuries are not falling.[38] There are differences in the criteria and hospital data are not necessarily 'correct' but this divergence has been confirmed by a number of in-depth studies.[39] Personal injury insurance claims are rising[40] and collisions attended by the Fire and Rescue Service have not shown a decline.[41]
Over long periods the ratio between deaths and serious injuries has been 11-13 serious injuries for each death. Yet since 2000 the ratio has declined almost every year, without explanation, so that now there are only 9 serious injuries reported for each death.[42]
Improvements in medical care can be expected to have saved the lives of some road accident victims who would previously have died. The number, however, is disputed. The College of Emergency Medicine says that advances in medical care have had a considerable impact in recent years[43] whilst the Minister told us that this has made little difference to the casualty figures.[44] Life-saving medical care would tend to reduce the figure for deaths but not for serious injuries.
Witnesses representing the police pointed out that the police tended to report only those accidents that they attended and that a reduction in the number of roads police officers had led to a reduction in accident reporting.[45]
Analysis of serious injury records shows that it is the less severe 'serious injuries' which are declining more rapidly than the more severe 'serious injuries'.[46] These are also the types of injury less likely to be reported to the police.

24. We asked many of our witnesses why serious injuries appeared to have declined so rapidly whilst deaths had not. Mr David Lynam of the Transport Research Laboratory (TRL) thought that there had been some changes in policy and in vehicle design that might explain a small increase in the chance of fatal outcomes but "that is explaining only a proportion of the difference we see."[47] No-one was able to offer a more convincing explanation, other than the possibility that reporting rates had changed. The Transport Research Laboratory investigated possible explanations, particularly seat-belt wearing rates and an increase in 'unsurvivable' accident types, but found that these did not explain the divergence.[48] An analysis of Scottish casualty data found that, relative to hospital records, police reporting rates fell substantially between 1997 and 2005.[49]

25. The Department for Transport has acknowledged weaknesses in the police data and publishes caveats with the relevant statistics. For example:

"[…] research has shown that an appreciable proportion of non-fatal injury accidents are not reported to the police. […]. up to a fifth of casualties reported to the police are not included in the statistical return. Moreover, studies also show that the police tend to underestimate severity of injury because of the difficulty in distinguishing severity at the scene of the accident."[50]

26. Differences between police and hospital records are not peculiar to road-traffic casualties: there is a similar discrepancy in the recorded numbers of injuries due to violence.[51] The Department has commissioned research to further compare police and hospital casualty data. This shows a decline in the rate of reporting of serious injuries.[52]

27. The reporting of accidents involving only slight injuries has always been acknowledged to be less robust than that for more serious injuries. Furthermore, the reporting of slight injuries also appears to have declined.[53] Although slight injuries are still an important issue, particularly for vulnerable road users,[54] our witnesses generally attached little credence to the slight injury target.

28. The Department for Transport seems reluctant to contemplate that all these findings might imply some fundamental problems. "We do not see it as a major problem, but there is a discrepancy. […] We are very happy with the STATS19 figures we get from the police and have no reason to doubt that they are not in about the right place […]."[55]

29. Up to this point we have accepted the assurances of the Government that its casualty data were robust and that good progress was being made on bringing down the number of people killed or seriously injured. Given the significant yet unexplained divergence in the trends for deaths and serious injuries, and given the growing body of evidence of changes in the reporting rates, we can no longer conclude that good progress is being made on casualty reduction. Indeed, we are worried that Ministers are not challenging their officials sufficiently and that policy-makers and practitioners are being lulled into a false sense of security.

30. The reality is that STATS19 is a system for recording accidents reported to the police, in order to assist with road safety measures. It is well established that some common categories of injury-accident are disproportionately under-recorded. It was never designed to be a scientific method for recording overall trends in accidents or casualties. It is neither a census of all accidents, nor a properly structured sample.

31. The Government should establish a British Road Safety Survey to track overall casualty and safety trends. This would be a structured survey, gathered from a statistically significant sample of households, similar to the National Travel Survey. It would, therefore, not rely on levels of reporting by road users or police. It would be akin to the British Crime Survey which is seen as a more reliable long-term monitor of crime than the police crime statistics. This would involve original survey work, and might also draw on existing data sources, including police, hospital and insurance company data, to obtain a more rounded picture. A survey would have the additional benefit of being able to monitor attitudes to road safety including, for example, the fears of vulnerable road-users.

32. Chief Constable Steve Green of the Association of Chief Police Officers emphasised that the police were not deliberately misreporting or under-recording road casualty data.

One thing I would say absolutely categorically is there is no organised conspiracy to under-record. There is no incentive to do so because there is no result at the end of it, partly because so little priority is given to road safety in the Home Office list of priorities.[56]

33. Other witnesses emphasised the value and quality of STATS19 data for a range of practical purposes.[57] We make no criticism of the police with regard to STATS19 reporting. The police have no mandate to seek out unreported accidents, nor the time or qualifications to make complex assessments of the severity of injuries. Equally, we accept that STATS19 data provides some valuable information.

34. There is a significant body of evidence to suggest that the current methods for recording road-traffic injuries are flawed. We recommend that the Government commissions an independent review of the STATS19 system in order to establish its strengths and weaknesses, bearing in mind our recommendation above for a British Road Safety Survey. The review should also examine ways in which the system could be simplified, with a view to promoting greater consistency, and consider ways of routinely linking police and hospital data.

Reductions in road deaths

35. Whereas we have reservations about the accuracy of the serious injury data, there seems to be agreement that few, if any, deaths go unrecorded. These give a less controversial account of the Government's success with reducing casualties. The reviews of 2004 and 2007 noted the disappointing progress in reducing deaths.

36. The casualty figures for Great Britain for 2007 were significantly different to previous years in that the number of deaths fell sharply. Compared with 2006:

overall deaths fell by 7%, contrasting with only 11% in the previous ten years;

the number of child deaths fell by 28% to its lowest ever total; and

deaths among car users declined by 11%.

The 2007 figures are extremely welcome.

37. The Minister was understandably wary about placing too much emphasis on a single year and there has not been time for the data to be fully analysed to see if there are particular reasons for them.[58] It remains to be seen whether this represents the result of particular events, a random fluctuation, or the start of a new trend.[59]

38.  The changes in deaths over the whole period, compared to the 1994-98 average, are shown in Table 1. This shows that 632 fewer people (18%) died on the roads in 2007. Table 1: Change in deaths by road user group, 1994-98 average to 2007

Road user group 1994-98 Average 2007 Change (%)

Pedestrians 1008 646 -362  (-36%)

Cyclists 186 136 -50  (-27%)

Motorcyclists   467 588 +121  (+26%)

Car users 1762 1432   -330  (-19%)

Bus & Coach   20 12 -8  (-40%)

LGV & HGV   118   110 -8   (-7%)

All users   3578 2946 -632  (-18%)

Source: Department for Transport, Road Casualties Great Britain 2007: Annual Report, September 2008.

39. Excluding bus and coach users where the numbers are very small, the biggest absolute and percentage fall is in pedestrian deaths, which have declined steadily, by 36% over the period. Cyclist deaths have also declined substantially, by 27%. The distances walked and cycled per person have stayed fairly constant over the period but total distances travelled by these modes have increased due to growth in the population.[60] The percentage reductions in child pedestrian and cyclist deaths are even greater (Table 2). There is, however, some evidence that part of the reduction is due to increased restrictions on the independent mobility of younger children,[61] something that the National Travel Survey is not designed to monitor.

40. Car user deaths, particularly car drivers, had reduced by only 9% prior to 2007. As this is the largest fatality group, it is particularly worrying. In 2007 there was a sharp fall giving a 19% reduction over the whole period.

41. The figure that most stands out is the 26% increase in motorcyclist deaths. The amount of motorcycling has increased over the period but so too have the amounts of driving, walking and cycling whilst the numbers of deaths for these groups has reduced.

42. Accidents involving young drivers, particularly young males, are also a major concern. We have previously drawn attention to it in our Report Novice Drivers which showed that:

27% of 17-19 year-old males are involved in a road collision as a driver in their first year of driving; and
one in eight driving licence holders is aged under 25, yet one in three drivers who die in a collision is under 25, and almost one in two drivers killed at night is under 25.

Despite the improvements in 2007, drivers aged between 16 and 29 years still make up 42% of all driver deaths. We highlight this issue again in paragraphs 67-77 below.Table 2: Change in child deaths, 1994-98 average to 2007

Road user group 1994-98 Average 2007 Change (%)

Pedestrians 133 57 -76  (-57%)

Cyclists 43 13 -30  (-70%)

Car users 77 45 -32  (-42%)

Others   82 51 -31  (-38%)

All users   260 121 -139  (-53%)

Source: Department for Transport, Road Casualties Great Britain 2007: Annual Report, September 2008.

43. Safety - as opposed to the mere absence of accidents - can be measured by casualty rates based on distance travelled. The fatality rates for different users are shown in Table 3 below. These show that over the period 1997-2006 the fatality rate has fallen for each of the main user groups, but more steeply for pedestrians and pedal cyclists. Table 3: Deaths per billion passenger kilometres

Road user group 1997 2006

Pedestrians 57 36

Cyclists 45 31

Motorcyclists 119 107

Car users 3 2.5

Source: Road Casualties Great Britain 2007, Table 52

International comparisons

44. Compared to many other countries, the UK has a low number of road deaths relative to its population. Its position, however, has been slipping. In 2001 it was second behind Malta in a table of 29 nations compiled by the European Transport Safety Council. By 2006 it was down to sixth, behind Malta, the Netherlands, Sweden, Switzerland and Norway.[62] Of the top ten nations, all had made bigger percentage reductions in deaths than the UK.[63]

45. Mr Fred Wegman, Managing Director of the Netherlands road safety institute SWOV, told us that

"Until 2000 we were always looking to the United Kingdom when it came to road safety. You were the inventors of many good activities and policies. All of a sudden, somewhere in 2000, you stopped doing things and we [the Netherlands] continued with our efforts."

46. His view is echoed by others with an international perspective.[64] Many of our witnesses suggested that the UK could learn most about road safety from the examples of the Netherlands and Sweden, and in different ways, France and Germany.[65]

47. Along with the evidence provided by our witnesses, we have used the above tables on road deaths to identify priorities for the future. We set these out below.

14 Department for Transport, Transport Statistics Great Britain 2007, November 2007, Tables 8.1 and 9.1 Back

15 Department for Environment, Transport and the Regions, Scottish Executive and National Assembly for Wales, Tomorrow's Roads - safer for everyone, March 2000 Back

16 Major aspects of road safety have been devolved to the Scottish and Welsh administrations and road safety as a whole is devolved to the Northern Ireland Assembly. The Department for Transport has provided a detailed statement of the legislative and executive competencies of the devolved administrations with regard to road safety in Ev 357

Back

17 Broughton, J et al, The Numerical context for setting national casualty reduction targets. TRL Report 382, 2000, Table 6 Back

18 Primary safety measures reduce the likelihood of a collision occurring; secondary safety measures prevent or reduce the severity of an injury in a collision. Back

19 Department of the Environment Northern Ireland, Northern Ireland Road Safety Strategy 2002-2012, 2002 Back

20 Department for Transport, Towards a Sustainable Transport System, Cm 7226, October 2007 Back

21 Ev 187 Back

22 Department for Transport, Road Casualties Great Britain 2007: Annual Report, September 2008, p 21 Back

23 Department for Transport, Second Review of the Government's Road Safety Strategy, February 2007. Published jointly with the Scottish Executive and the Welsh Assembly Government. Back

24 Department for Transport, Annual Report 2008, Cm 7395, May 2008, p 125 Back