Safety Data Value and Use

Prev Next
This content is currently unavailable in English (UK). You are viewing the default (English) version.

Implications for Delivery

Data is an integral part of implementing road safety interventions aligned to the Safe System. There is a wide range of relevant data available so practitioners need to:

  • Ensure that a variety of casualty data sources are used beyond just STATS19

  • Ensure that data used is supporting a Safe System approach to road safety

  • Always use exposure data as well as casualty data to calculate risk

  • Use and monitor road risk data from multiple sources as part of a more proactive approach to road safety. This should include data relating to all the Safe System components.

  • Investigate new data sources that can add value to analysis

Background

Reliable collection and use of safety data is crucial for delivering Transport Scotland’s Road Safety Framework. Casualty data is essential for measuring progress towards the long-term goal of zero fatalities, the interim targets to 2030 and the intermediate outcome targets.

Data enables evidence-led decisions about road safety interventions. This helps to address issues on the road network before crashes happen, supporting a proactive Safe System approach to road safety. Monitoring our Key Performance Indicators (KPIs) relating to road safety behaviours also helps us to be more proactive.

The data collection and analysis processes are equally important. Reliable and fit-for-purpose data leads to useful and actionable insight. Analysts can then work with this data to support policy makers with decision making and prioritisation. A variety of data sources should be used to understand risk on the network and help with targeting funds.

Data Sources

Effective road safety management needs a range of data sources. To deliver Scotland’s ambitious targets, data needs to be collected reliably and used effectively to support decision-makers.

Safety data should be used relating to:

  • Safety outcomes – for example casualties and collisions

  • Exposure – for example traffic levels

  • Road risk – for example relating to road condition and speeds travelled

  • Safety performance indicators (SPIs)

  • Performance of road safety interventions

Data for all these areas is crucial for understanding network performance, taking a proactive approach and monitoring progress towards targets.

Collision data gives a comprehensive and detailed understanding of collisions occurring across the country. This should include details of vehicles and casualties involved as well as causation factors. STATS19 is Great Britain’s main dataset of collisions and includes several important variables, for example:

  • Date and time

  • Location (GPS co-ordinates)

  • Road characteristics – road type, junction type, speed limit

  • Number and type of vehicles involved

  • Number and basic demographics of casualties involved

STATS19 is a useful and comprehensive data source, however it is important to note that only police reported collisions are included. Therefore, the number of collisions is underreported, particularly for less severe collisions.

It is important to use other data sources in addition to STATS19 to get a broader understanding. These include:

  • In-depth collision investigation data to understand the root causes of collisions. The number of collisions investigated is smaller than for STATS19, however the insights are much more detailed.

  • Fire and ambulance service incident data which may have more detailed information on casualty condition and post-crash care.

  • Hospital data for understanding casualty outcomes and post-crash care, however this data can be more challenging to obtain and match to collision data.

  • Operator data for understanding collisions within different vehicle types or geographical areas. This could include data from e-scooter or e-bike operators, for example.

  • Vehicle data such as e-call data which can provide precise vehicle location information on collisions attended by emergency services.

To accurately assess road risk, exposure data should be combined with casualty data. Typically, this is information on distances travelled from traffic counts or travel surveys. Road risk is then estimated by dividing the number of crashes or casualties by the distance.

Travel data can be difficult to obtain, however. If this is not available, proxy measures can be used as an alternative. These could include:

  • Road length

  • Driving population - for example from driver licensing data

  • Vehicle fleet - for example from vehicle registration data

Ideally exposure data should be available for different road user types to enable comparisons.

Road risk data supports a proactive approach so that issues can be addressed before collisions occur. Several data sources can be utilised to understand risk on the network:

  • Speed data can be obtained from traditional approaches such as spot speed surveys. With an increasing number of connected vehicles, more extensive speed data is now available from commercial suppliers such as TomTom.

  • Vehicle data from Euro NCAP is useful for understanding the fleet and the safety rating of vehicles involved in collisions.

  • Harsh braking and other forms of near miss data from commercial suppliers can help to identify high-risk areas on the network.

  • Road infrastructure asset data can be used in assessment tools such as iRAP. This gives an understanding of risk based on the features of the road and roadside. Roads-based interventions can then be selected based on identified risk for different road user types.

  • Public survey data can help to understand the prevalence of certain behaviours. These could include seat belt wearing, mobile phone use and speeding.

Using Data

A useful way of quantifying risk is through Safety Performance indicators (SPIs). SPIs are indicators with strong causal links to road safety outcomes.

It is important to understand what works and what doesn’t on the road network. Evaluating the effectiveness of interventions on our roads helps us to achieve this. Fit-for-purpose data should be collected and analysed to support this aim, helping to inform future interventions. Insight should be obtained from observational studies and before-after studies to assess the effectiveness of schemes. Public survey data also helps to understand risk and perception of risk relating to an intervention.

In addition to data already available, new data sources can further help to deliver a Safe System approach. There should be a proactive approach to identifying new data as digital technology advances. Combining different data sources helps to give a better understanding of risk on Scotland’s roads.

Sources

Raffo, V. I., Feypell, V., Martensen, H., Duchamp, G., Burlacu, F. A., Turner, B. M., & Paala, M. (2021). Guidelines for conducting road safety data reviews. Mobility and Transport Connectivity (MTC) series. Washington, DC: World Bank Group. http://documents.worldbank.org/curated/en/099140001132222667

World Health Organization. (2010). Data systems: A road safety manual for decision-makers and practitioners. World Health Organization. https://iris.who.int/handle/10665/44256