Our mission at THINK! is to change behaviour to reduce road accidents. We need to measure how well we’re doing at this to make sure we’re spending money effectively. Statistics are one way of measuring behaviour change on our roads. Here are some of the statistics we use to help measure the success of our campaigns.
Most of the statistics are based on road accidents reported to the police (STATS19 system). These provide detailed statistics about the circumstances of personal injury road accidents, including the types of vehicles involved and the consequent casualties.
Other sources directly related to road safety are also used, including hospital admissions, death registrations, coroners’ reports, national travel survey, crime survey from England and Wales, and statistics on breath tests and motoring offences from the Home Office and Ministry of Justice.
An index sheet of all road accidents and safety statistical tables is available.
- road safety research reports
- road accident statistics (STATS19) research datasets available from the Economic and Social Data Service
- international comparisons: International Traffic Safety Data and Analysis group (IRTAD)
- all statistics published by DfT
The following road safety statistical releases are published during the year:
- Reported road casualties in Great Britain: main results (June) – first release of key statistics on casualties and accidents reported to police
- Reported road casualties in Great Britain: quarterly provisional estimates (August, November, February) – main quarterly statistics, with no release for final quarter of the year as this is covered by the annual results
- Reported road casualties in Great Britain: annual reports (September) – detailed data and analyses of road casualties, with articles presenting further analysis on specific road safety topics. On rare occasions the figures that were first published in June may be revised at this point. This only happens when police forces identify significant changes after June.
- Reported road casualties in Great Britain: estimates for accidents involving illegal alcohol levels (July) – an annual statistical release providing provisional estimates on accidents involving drinking and driving in Great Britain
Other ad-hoc surveys or factsheets are also published from time to time. These are available at:
Detailed guidance about STATS19 data and other road safety data is available at:
Background information on road safety statistics
Statistics on road safety in Great Britain are mostly based on accidents reported to the police through the STATS19 system. This system allows police forces to report all personal-injury accidents to the department. It does not collect any information about damage-only accidents. Information on STATS19 can be found in the report form and the guidance document used by the police when completing the form.
Comparisons with death registration statistics show that very few, if any, road accident fatalities are not reported to the police. However, it has long been known that a considerable proportion of non-fatal casualties are not known to the police, as hospital, survey and compensation claims data all indicate a higher number of casualties than are reported.
The department produces an annual ‘best estimate’ of the total number of road casualties in Great Britain each year, including those not reported to police. This is derived primarily from National Travel Survey (NTS) data. The latest such estimates, along with a description of how they have been derived and their limitations, are set out in an annual article published in the ‘Reported road casualties Great Britain: annual report’.
The STATS19 data are therefore not a complete record of all injury accidents and this should be kept in mind when using and analysing the data. However, they remain the most detailed, complete and reliable single source of information on road casualties covering the whole of Great Britain, in particular for monitoring trends over time.
Some of the statistics published in the annual report are based on linking STATS19 data and hospital admissions data (HES).
Personal and sensitive information
The road accident data provided to the department by the police do not include any names, addresses or telephone numbers.
It is not unusual for claims management companies to call members of the public claiming that they got their name and telephone number from the national road accident database or from a governmental agency. This is not correct: no government agency or department holds the names or addresses of people who have been in road traffic accidents. Even if a body did hold this information, they would not release it as it is protected under the data protection laws.
If you have been contacted in such a way, it is most likely to be as a result of cold calling or random number dialing. If you are able to get a contact name and number for the company you can report them to the Claims Management Regulator.
However, the STATS19 data do include variables which we regard as sensitive. These include: drivers’ and casualties’ home postcodes, breath test results, and contributory factors. These variables are all withheld from the public downloads of the raw data so they are not disclosed. The sensitive variables are only released in very limited circumstances
Researchers from recognised research institutes can apply for access to the sensitive variables. Any application needs to include a short description of the research proposal and how the sensitive variables will be used. Research work can be carried out under an end-user licence which stipulates that researchers cannot release the raw data, attempt to match to other data to identify individuals, or disclose any information about individuals.
We supply the postcode information to NHS Digital. They use the information to match police-reported casualty data to hospital data (HES). We then receive anonymised clinical data back. NHS Digital do not provide any names, address or other identifiable data to the department and they destroy the data once the match has been completed. We only use the resulting matched data set to produce aggregated statistics.