Full report (.pdf format)

Transport accidents and injuries present a cost to the European Union estimated at around 2% of GDP - around twice its entire annual budget for all activity.

Transport safety is a key priority in the Common Transport Policy and a high level of protection for EU citizens in all harmonisation activity is required by the Treaty.

Policymakers and managers aiming for a higher level of safety need to take an interest in as many of the factors influencing safety as possible and, at least, those factors they are able to affect or control.

Safety performance indicators provide a means by which policymakers can ensure that their actions are as effective as possible and represent the best use of public resource.

These are defined as any measurement that is causally related to crashes or injuries, used in addition to a count of crashes or injuries, in order to indicate safety performance or understand the process that leads to accidents.

A large number of potential safety performance indicators exist. Not all of them are equally important. In general, the importance of a safety performance indicator can be assessed in terms of the strength of its relationship with accident or injury occurrence, if it makes a major contribution to accidents and if it can be influenced by road safety measures or programmes.

Transport safety can be assessed in terms of the social cost of accidents and injuries. Yet, it is clear that simply counting crashes or injuries is often an imperfect indicator of the level of transport safety. There are several reasons for this:

• The number of transport crashes or injuries is subject to random fluctuations, meaning that a short-term change in the recorded number does not necessarily reflect a change in the underlying, long-term expected number.
• Reporting of crashes and injuries in official road accident statistics is incomplete. This means that an observed change in the number of crashes could merely be a change in the propensity to report crashes to the police.
• A count of crashes says nothing about the processes that produce crashes. It is, to some extent, a matter of chance whether a hazardous situation or a near miss results in a crash or not. It is, therefore, entirely possible that conditions are very risky, but that, luckily, no accidents have occurred.
• In order to develop effective measures to reduce the number of accidents or the number of killed or injured people, it is necessary to understand the process that leads to accidents. Safety performance indicators can serve this end.

Such indicators can give a more complete picture of the level of transport safety and can point to the emergence of new problems at an early stage, before these problems show up in the form of accidents. A regular monitoring of safety performance indicators improves the understanding of road accident trends. Since these monitoring results can become available far more quickly than registered accidents, they are particularly useful for policymakers.

There are fundamental differences between road transport and other modes of transport with respect to the regulatory and operational regimes to which individual operators in the system are subjected. While there is a longer tradition of use of performance indicators in the non-road transport modes, and particularly in aviation, it is not so easy to correlate them with crash data as is the case in road transport. Therefore, no attempt has been made in this report to make a case for the standardisation of safety performance indicators across transport modes. The appropriate set of safety performance indicators will vary from one mode of transport to the next.

Road transport represents by far the greatest transport safety problem in all European countries with around 90% of all transport accident fatalities occurring in road transport. The main emphasis in this report has, therefore, been put on road transport, although reference has also been made to the other transport modes.

This report considers best practice in the use of road safety performance indicators in transport and presents recommendations for the use of such indicators as an integral part of safety programmes both at EU and national level.

This model illustrates the role of safety performance indicators in a wider context and serves as a device for identifying important safety performance indicators.

(Targeted) safety programmes

â

Safety measures implemented

â

Operational conditions of transport production (performance indicators)

â

Consequences of operational conditions (accidents) social costs

á

Safety targets (policy intentions)

Targeted safety programmes produce a set of safety measures to be implemented. These measures result in certain operational conditions of transport production, which in addition to the safety measures, are also influenced by a broad set of environmental and societal factors. The term 'operational conditions' comprises measures of operator behaviour and the technical condition and quality of the infrastructure and the vehicles used. The operational conditions of transport production result in a certain accident rate and number of casualties, usually taking into account the differing severity of accidents. These numbers are compared to safety targets, in order to monitor progress in achieving them.

Among the road safety performance indicators most commonly used are those that relate to behavioural characteristics such as speed levels, the rate of drink driving and the use of seat belts. In addition, a number of infrastructure, vehicle or trauma-related indicators are relevant. These provide a more straightforward means of monitoring the impact of a measure or programme and enable early, target-oriented adjustments of specific interventions. In addition, they allow for a more detailed understanding of the reasons for safety problems than is possible by looking at crash frequency alone.

Some EU Member States - usually those that have performed best in reducing casualties - have shown how safety performance indicators can be used efficiently in targeted safety programmes. In fact, some countries have specified certain behavioural characteristics as concrete targets in their national safety programmes in addition to pure casualty reduction targets. Working with performance indicators results in an increase of the understanding of policymaking authorities in the effects of their policies. Experience in some EU Member States show that the authorities using performance indicators are more engaged with their policies if performance indicator data are reported to them regularly.

Safety in the maritime, rail and aviation sectors has a long history and strong tradition based on regulation (initially only national, now increasingly pan European and international) and inspection to assess compliance with regulations. The underlying idea being that the greater the compliance the better the safety performance. Safety is an important design criterion for these transport modes. However, these sectors are organised in such a way that the interaction of many independent actors, both public and private sector, play important roles in achieving high levels of safety.

Inspection and investigation reports by government appointed agencies could be considered as the backbone of safety in these transport modes. In such reports, accidents and incidents play a dominant role. A growing interest is observed in all these transport modes to enrich the existing procedures to improve safety. Safety performance indicators are a promising development in this respect. In the maritime, aviation and rail sectors, examples are evident of initial steps to develop safety performance indicators, although these attempts cover only parts of the whole sector. Working with safety performance indicators allows for comparisons both within and between sectors. This is a useful aid to achieving a higher level of safety through a better understanding of the causes of accidents and by more transparent and rational decision-making.

For decades, the concept of utilising indicators for the continuous monitoring and analysis of processes has been standard practice in industrial quality management. The safety community should exploit this simple and robust concept. Once introduced and established for all the transport modes, the application of safety performance indicators will further stimulate safety work and thus reduce crash rates across Europe.

The European Commission should encourage Member States to agree upon and regularly collect a scientifically established set of safety indicators for all transport modes. At the same time, recommendations on harmonised sampling and application methodologies should be given. It is recommended that this implementation of safety performance indicators be built systematically over time and reviewed regularly.

In road transport, a natural starting point would be the main behavioural indicators - speed levels, drink driving rates and seat belt and helmet use - the comparability of which over time and between countries is of utmost importance. Thereafter, quality indicators such as road networks, vehicle fleets and emergency services should be added. There will be considerable benefit from the harmonisation of collection methods and the extension of the list of regular indicators. Initially, this will be easiest to accomplish among those Member States that currently do not collect safety indicators, and within the applicant countries.

In view of the above, ETSC recommends to the EU:

• To set up and specify a set of relevant safety performance indicators (SPI) to be used at European and national level as a means of assessing the trends in safety levels and the success of casualty reduction programmes;
• To convene, for this purpose, a working group of the High Level Group on Road Safety, comprising road safety experts and representatives of the Member States to consider the list of 'best practice SPI' as suggested in section 5 of this report, i.e. behavioural indicators on speed, alcohol, restraint systems and safety devices as well as SPI in the areas of quality management of road networks, vehicle fleets and emergency services;
• To support feasibility and pilot studies on SPI collection and application at Member State level;
• To require, on the basis of the above studies, the Member States to collect SPI in a regular and harmonised way and to encourage them to make continuous safety monitoring an integral part of their national safety strategies;
• To give guidance to the Member States as regards data collection and observation methodologies (such as sampling frequencies, sample size and time, measuring protocols) as well as on data administration and fusion, in order to produce harmonised sets of SPI that are comparable over time and between countries at a European scale; to recommend that independent bodies collect data on SPI;
• To help disseminate current best practice in collection and use of safety performance indicators across the Member States and make annual results and trends available in statistical publications as well as through the electronic media such as the Internet and the Commission's future Road Safety Information System;
• To require the Member States to deliver regularly an agreed set of national SPI data to the Commission for international communication, taking the Commission's "Road Safety Quick Indicator" and the CARE database as an example;
• To negotiate, during the consultations on the coming action programme, a common set of relevant SPI with the Member States and to implement quantified targets for SPI into the programme;
• To include a task in the 6th R&D framework programme of carrying out further research into running and benchmarking integrated safety management programmes, with a view to the causal relationships between safety performance indicators and crashes or injuries; and
• For the other transport modes, there is a need to develop a comprehensive set of performance indicators covering the whole sector and to use that information to improve safety and for more transparent and rational decision making. This requires a general accepted model of why accidents occur and how to prevent them effectively and efficiently. It is recommended to develop such a model and to derive from that model performance indicators.

The added value for the European Union lies in a harmonised and comprehensive system of transport safety performance indicators allowing the EU to use these data for its own policies and to provide a facility for Member States to reflect on their own performance.

ETSC gratefully acknowledges the contributions of members of ETSC's Transport Safety Performance Indicator Working Party to this review:

Mr. Fred WEGMAN (Chairman)
Ms. Martha BROUWER
Mr. Jean-Pierre CAUZARD
Dr. Rune ELVIK
Mr. Klaus MACHATA
Dr. Tapani MÄKINEN
Mr. Lars STENBORG
Prof. George YANNIS

Also Mr. Cees GLANSDORP (ETSC Maritime and Inland Waterway Safety Working Party), Messrs. Veli-Pekka KALLBERG and Bertil HYLEN (ETSC Rail Safety Working Party) and Mr. Alfred ROELEN (NLR).

ETSC Working Party Secretary: Ms. Pam LEWIS

ETSC is grateful for the financial support provided by Directorate of Energy and Transport of the European Commission and the Finnish Ministry of Transport and Communications. ETSC also acknowledges the contribution towards the printing and dissemination costs of this review provided by BP, Ford of Europe, KeyMed, Rail Safety and Railtrack Group plc, Scania, and Shell International. The contents of this review are the sole responsibility of ETSC and do not necessarily reflect the view of sponsors nor organisations to which research staff participating in the Working Party belong.