Accredited official statistics

Reported road casualties in Great Britain involving illegal alcohol levels: 2023

Published 31 July 2025

About this release

This release provides an estimate of the number of personal injury road traffic casualties in Great Britain where at least one driver was over the drink-drive limit in 2023.

These are accredited official statistics and were independently reviewed by the Office for Statistics Regulation in July 2013. For more information, see the about these statistics section.

The recent trends in reported road casualties have been impacted by the national restrictions implemented from March 2020 onwards following the coronavirus (COVID-19) pandemic. More information can be found in our release The impact of lockdown on reported road casualties in Great Britain.

Headline figures

In reported road collisions in Great Britain in 2023 involving a driver or rider who was over the legal alcohol limit, there were:

  • between 230 and 290 fatalities, with a central estimate of 260 fatalities, a decline of 14% from 2022
  • about 1,860 people killed or seriously injured, a decline of 3% from 2022
  • around 6,310 casualties of all severities, a decline of 6% from 2022

While there are different patterns in the numbers of drink-drive fatalities and total casualties in recent years, the percentage of reported road casualties that were drink-drive related, whether fatalities, serious or slight injuries, has increased over the last 10 years (as shown in charts 2, 4 and 6 below). The COVID-19 pandemic may have contributed to this trend.

Table 1: Drink-drive fatalities, killed or seriously injured (KSI) casualties, total casualties and all drink-drive collisions: Great Britain, 2023

Title Drink-drive fatalities Drink-drive KSIs Drink-drive casualties Drink-drive collisions
Number in 2023 260 1,860 6,310 4,390
Change since 2022 -14% -3% -6% -5%
Change since 2013 7% 4% -23% -23%
As a proportion of those in all reported collisions 16% 6% 5% 4%

Things you need to know

Data sources

This release uses data from the STATS19 database and toxicology data for fatalities in road collisions. The STATS19 data provides information on the characteristics of the collisions and driver as well as data on roadside breath tests. The toxicology data is used to identify which deceased drivers has illegal levels of alcohol in their blood at the time of the collision.

The final estimate for 2023 is based on coroners’ and procurators’ fiscal reports for 84% of the drivers or riders who were killed in road traffic collisions in that year in addition to breath tests taken at the scene where the driver was not killed (see definitions).

Data coverage

There is no obligation for people to report all personal injury collisions to the police. These figures, therefore, do not represent the full range of all collisions or casualties in Great Britain. All collisions reported by the police and that occurred on a public highway involving at least one motor vehicle, horse rider or pedal cyclist, and where at least one person was injured, are included in these statistics.

Changes in systems for severity reporting

Non-fatal casualties since 2016 have been affected by a large number of police forces changing their reporting systems which has had a large impact on the classification of injuries recorded. Further details are in the guide to severity adjustments.

All figures in this release are based on adjusted data.

Revisions to statistics

This publication includes revisions to the statistics from 2005 due to a change in methodology. These revisions are minor and don’t affect the overall patterns and trends. More information on the changes can be found in the further information section.

Due to the uncertainty in these estimates, 95% confidence intervals are used for fatality estimates. More details about confidence intervals can be found in the quality and methodology section.

Drink-drive fatalities

The central estimate of the number of deaths in collisions with at least one driver over the alcohol limit for 2023 is 260. This is a decline of 14% on the estimate for 2022, which was around 300. The 95% confidence interval has a lower limit of 230 and an upper limit of 290 fatalities. The central estimate for 2023 suggests a reduction on the previous year, however, when taking into account the confidence interval, we cannot concluded that there is a statistically significant change - this reflects the degree of uncertainty in the estimates.

Chart 1 shows the estimates of fatalities in drink-drive collisions as well as the 95% confidence intervals for the 10 years up to 2023. The number of fatalities increased between 2020 and 2022, following the COVID-19 pandemic. Although there was a decline from 2022 to 2023, the central estimate is higher than for the years before 2020.

Chart 1: Fatalities in reported drink-drive collisions: Great Britain, 2014 to 2023; error bars show 95% confidence intervals RAS2001

The prevalence of drink-driving in road deaths has fallen over time. In 1979, 26% of road deaths occurred in collisions where at least one driver or rider was over the drink-drive limit. This had fallen to 15% by 1989. Since then, the percentage of road deaths that are drink-drive related has varied between 12% and 18%. In 2023, the proportion was 16%.

Chart 2: Fatalities in reported drink-drive collisions, as a percentage of fatalities in all collisions: Great Britain, 1979 to 2023

Table 2: Fatalities in reported drink-drive collisions, as a percentage of fatalities in all collisions: Great Britain, 1979 to 2023 in 5 year intervals

Year Percentage of fatalities that involved a driver over the drink-drive limit
1979 25.9%
1983 20.3%
1988 15.6%
1993 14.2%
1998 13.4%
2003 16.4%
2008 15.5%
2013 14.0%
2018 13.3%
2023 15.8%

Drink-drive killed or seriously injured (KSI) casualties

The central estimate of the number of killed or seriously injured (KSI) drink-drive casualties in 2023 is 1,860, a decline of 3% from 2022. The overall trend is fairly flat for the last 10 years, other than the years affected by the COVID-19 pandemic where there was a drop in the number of people killed or seriously injured for both drink-drive and all reported road collisions.

Chart 3: Killed or seriously injured casualties in reported drink-drive collisions: Great Britain, 2014 to 2023 RAS2001

The proportion of KSI casualties in drink-drive collisions has varied between 5% and 7% since 2005. In 2023, the proportion was 6%.

Chart 4: Killed or seriously injured casualties in reported drink-drive collisions, as a percentage of KSI casualties in all collisions: Great Britain, 2005 to 2023

Table 3: KSI casualties in reported drink-drive collisions, as a percentage of KSIs in all collisions: Great Britain, 2005 to 2023 in 2 year intervals

Year Percentage of KSI casualties that involved a driver over the drink-drive limit
2005 7.6%
2007 6.8%
2009 6.7%
2011 5.8%
2013 5.5%
2015 5.3%
2017 5.9%
2019 6.4%
2021 6.9%
2023 6.3%

All drink-drive casualties

The central estimate of the number of drink-drive casualties of all severities in 2023 is around 6,310, a decline of 6% from about 6,730 in 2022. The number of drink-drive casualties of all severities has steadily decreased over the last 10 years.

Chart 5: Total casualties in reported drink-drive collisions: Great Britain, 2014 to 2023 RAS2001

It is estimated that around 5% of all casualties in reported road collisions in 2023 were involved in collisions in which at least one driver or rider was over the drink-drive limit.

In 1979, 9% of road casualties occurred in collisions in which at least one driver or rider was over the drink-drive limit. This fell to 5% by 1992 and has mainly varied around 5% since then, which suggests that the reductions in drink-drive collisions over the past decade reflect overall reductions in road casualties.

Chart 6: Casualties in reported drink-drive collisions, as a percentage of casualties in all collisions: Great Britain, 1979 to 2023

Table 4: Casualties in reported drink-drive collisions, as a percentage of casualties in all collisions: Great Britain, 1979 to 2023 in 5 year intervals

Year Percentage of casualties that involved a driver over the drink-drive limit
1979 9.4%
1983 8.6%
1988 7.1%
1993 4.9%
1998 4.8%
2003 6.5%
2008 5.5%
2013 4.4%
2018 5.3%
2023 4.7%

Drink-drive collisions

Trends in drink-drive collisions are similar to those for casualties.

In 1979, around 8% of reported road collisions were drink-drive related. This fell to about 5% by 1990 and has generally been around 5% since then.

It is important to note that the number of overall road traffic collisions has fallen 59% from 1979 to 2023. However, drink-drive collisions have fallen further and are down 77% since 1979. It is therefore likely that some drink-drive initiatives have been effective in reducing the number of drink-drive collisions.

Chart 7: Reported collisions and drink-drive collisions: Great Britain, 1979 to 2023 (index 1979 = 100) (RAS2001 for drink-drive casualties, RAS0101 for all casualties)

Where: by region and country

From 5 December 2014, Scotland has a lower drink-drive limit (22 micrograms (mcg) per 100 millilitres (ml) of breath or 50 milligrams (mg) per 100ml of blood) than England and Wales (35 mcg per 100ml of breath or 80 mg per 100ml of blood).

Overall, 5% of casualties in reported road collisions in 2023 occurred in drink-drive collisions.

This varied across the countries of Great Britain. The percentage of casualties which occurred in drink-drive collisions was 5% in England, 7% in Wales and 5% in Scotland. In general, the proportion of collisions involving a drink-driver has been higher in Wales over the past decade.

Chart 8: Percentage of casualties occurring in drink-drive collisions by country, 2014 to 2023 RAS2013

In 2023, the casualty rates within the English regions, varied from about 3% in London to around 6% in the East Midlands. In general, London has a lower proportion of drink-drive collisions than other regions over the past decade, which may reflect the nature of the roads and collisions within the region. Also, the East Midlands has had the highest rate of all English regions since 2017.

Table 5: Percentage of casualties occurring in drink-drive collisions by country and English region, 2023 RAS2013

Region Percentage of casualties occurring in drink drive collisions
England 4.6%
Wales 7.5%
Scotland 5.1%
North East 3.3%
North West 4.8%
Yorkshire and The Humber 4.7%
East Midlands 6.5%
West Midlands 5.0%
East of England 4.7%
South East 5.9%
London 2.5%
South West 5.2%

When: by month and time of day

Month

In 2023, the months with the highest proportion of drink-drive casualties were in summer, with the highest proportion in June. Also, there was an increase in December compared to the other winter months. In contrast, February and March had the smallest proportion. This is broadly in line with other years but patterns for individual years can vary.

Table 6: Percentage of casualties in drink-drive collisions and in all reported collisions that occur in each month in Great Britain, 2023 RAS2011

Month Casualties in drink-drive collisions Casualties in all reported collisions
January 7.6% 8.3%
February 7.2% 7.1%
March 7.2% 7.8%
April 8.0% 7.5%
May 8.2% 8.8%
June 10.4% 9.4%
July 9.2% 9.0%
August 9.2% 8.3%
September 8.1% 8.6%
October 8.0% 8.6%
November 8.0% 8.6%
December 8.9% 8.0%

Hour of day

Overall, in 2023 a greater proportion of drink-drive collisions occurred late at night and in the early hours of the morning, reaching a peak of 8% occurring between 11pm and midnight. The lowest proportion of drink-drive collisions occurred between 11am and midday, with 1%.

Who: drink-drivers by sex and age

Drink-drivers by sex

The majority of drivers involved in road injury collisions overall are male, with an even higher proportion of drink-drivers being male. In 2023, 80.6% of drink-drivers, whose sex is known, were male. However, males made up only 69.3% of drivers (excluding pedal cyclists and horse riders) involved in all collisions where the sex of the driver is known.

Therefore, male drivers are over-represented in drink-drive collisions compared to all reported collisions.

Table 7: Proportion of drink-drivers and drivers in all reported collisions, by sex, in Great Britain, 2023

Type Male Female Total
Drink-drivers involved in reported road collisions 80.6% 19.4% 100%
Drivers involved in all reported road collisions 69.3% 30.7% 100%

Drink-drivers by age

Younger drivers are over-represented in drink-drive collisions compared to all injury collisions.

All age groups up to the age of 49 are over-represented in drink-drivers. However, drivers aged 50 or over are under-represented in drink-drivers.

Table 8: Percentage of drink-drivers and drivers in all collisions by age group in Great Britain, 2023

Age Drink-drivers All drivers
Under 20 6.1% 5.8%
20 to 24 13.8% 10.8%
25 to 29 14.8% 11.1%
30 to 34 14.8% 12.4%
35 to 39 13.1% 10.7%
40 to 49 18.9% 17.7%
50 to 59 11.7% 15.2%
60 or over 6.8% 16.3%

Age of driver by hour of day

A higher proportion of collisions which involve a drink-driver who is under the age of 25 occur late at night and in the early hours of the morning compared to those involving older drivers. 56% of collisions involving drink-drivers, aged 24 or under, occurred between 10pm and 4am, compared to 38% of all drink-drive collisions that occurred between those hours.

On the other hand, 48% of collisions involving drink-drivers who were aged 60 or over occurred between 2pm and 7pm whereas the same was true for only 23% of all drink-drive collisions.

Chart 9: Percentage of collisions involving drink-drivers of different ages that occurred at different times of the day in Great Britain, 2023

Who: casualties by sex and age

Casualties by sex

In 2023, 68% of casualties in drink-drive collisions where the sex of the casualty was known were male. In comparison, 61% of casualties in all reported road collisions are male, meaning males are over-represented in drink-drive casualties.

Table 9: Proportion of casualties in drink drive collisions and all reported collisions where the casualty sex is known, by sex, in Great Britain, 2023 RAS2031

Type Male Female Total
Casualties in drink-drive collisions 68.0% 32.0% 100%
Casualties in all reported road collisions 61.4% 38.6% 100%

Casualties by age

A higher proportion of casualties in drink-drive collisions were aged between 16 and 59 than in all reported collisions in 2023. On the other hand, children (under 16) and older people (aged 60 or over) represented a lower proportion in drink-drive collisions than collisions overall.

The percentages in the table below do not add up to 100 as the proportions casualties with unknown ages are not shown.

Table 10: Percentage of casualties by age group in all collisions and in drink-drive collisions in Great Britain, 2023 RAS2031

Age Drink-drive All collisions
0 to 15 5.1% 8.8%
16 to 24 22.7% 18.9%
25 to 59 63.0% 55.2%
60 or over 8.4% 14.9%

Further information

Tables published with this release

With this release, the headline table on drink-driving, RAS2001, has been updated with 2023 figures. It shows collisions and casualties by severity since 1979.

The tables on drink-driving are organised by topic:

Digital breath tests

Statistics on the results of roadside breath alcohol screening tests, administered by police forces in England and Wales in 2022, using digital breath testing devices is also published.

The devices are able to record exact breath alcohol readings and the result of individual tests, as well as reason for test, and age and sex profiles of those tested. The results are downloaded to data systems on a monthly basis and provided to the Department for Transport. These data are not provided by all police forces so are incomplete and therefore do not cover England and Wales as a whole.

Crime Survey for England and Wales

The Crime Survey for England and Wales can be used to look at trends in self-reported drink driving by age, sex and frequency of alcohol consumption. Within the survey a driver is considered to be anyone who stated they had driven at least once in the past 12 months.

Data is also provided for self-reporting of the following:

  • driving after taking medication with advice not to drive
  • driving under or over the drink-drive limit
  • drug driving
  • diving after taking legal highs
  • mobile phone use when driving

Home Office breath test data

Analysis on the number of roadside tests carried out by police in England and Wales is produced by the Home Office. The figures show geographic patterns and seasonal variation. Commentary is also provided to aid the interpretation of the trends.

Drink-drive convictions data

Below is a summary of the number of defendants prosecuted, convicted and sentenced at criminal courts in England and Wales for drink-drive offences under Section 5 of the Road Traffic Act, relating to driving or being in charge of a motor vehicle with alcohol concentration above prescribed limit.

Note that these values are for those whose principal offence was “Driving or attempting to drive a motor vehicle while having a breath, urine or blood alcohol concentration in excess of the prescribed limit” and that was the offence for which the heaviest penalty was imposed. Further data on driving convictions is available from the Ministry of Justice.

Type 2017 2018 2019 2020 2021 2022 2023 2024
Proceeded against 34,102 34,049 35,006 28,163 32,046 32,997 30,531 28,028
Convicted 33,416 33,255 34,215 27,513 31,169 31,990 29,529 27,208
Sentenced 33,435 33,254 34,198 27,504 31,178 31,985 29,528 27,219

Drug-driving statistics

Updated statistics on drug-driving to include 2023 data are now available.

Background information

A methodology note is available and describes how the estimates are produced, including how numbers are scaled to address missing data.

Drink-drive legislation

The Road Safety Act 1967 introduced the first drink-driving limit in the UK, set at a maximum Blood Alcohol Content (BAC) of 80mg of alcohol per 100ml of blood (or the equivalent 107mg of alcohol per 100ml of urine). It became an offence to drive, attempt to drive or be in charge of a motor vehicle on a road or other public place with a BAC that exceeded the maximum prescribed legal limit. The 1967 Act also made it an offence to fail to provide a specimen for a laboratory test without reasonable excuse.

The Transport Act 1981 introduced evidential breath testing and established a maximum breath alcohol concentration of 35 micrograms of alcohol in 100ml of breath. This was implemented in 1983. Today, people are given a preliminary breath test at the roadside and then taken back to the police station for an evidential breath test.

Drink-driving legislation has been strengthened over the years, including tougher penalties for offenders which can include potentially unlimited fines, disqualification from driving, or facing imprisonment for the most serious offences.

On 5 December 2014, Scotland reduced the legal BAC limit for all drivers from 80mg per 100ml of blood to 50mg per 100ml. The drink-drive limit introduced by the 1967 Act remains in place for England and Wales today.

Definitions used throughout this publication

Drink-drive collision: A reported incident on a public road in which someone is killed or injured, where at least one of the motor vehicle drivers or riders involved met one of these criteria:

  • failed a roadside breath test by registering above 35 micrograms of alcohol per 100ml of breath (in England and Wales) or 22 micrograms (in Scotland)

  • refused to give a breath test specimen when requested by the police (other than when incapable of doing so for medical reasons)

  • had a blood sample taken within 12 hours of the collisions and was subsequently found to have more than 80 milligrams of alcohol per 100ml of blood (in England and Wales) or 50 milligrams (in Scotland)

Drink-drive casualties: All road users killed or injured in drink-drive collisions.

Drink-driver: A driver or rider of a motor vehicle who was detected to have been over the drink-drive limit. This includes those who failed a roadside breath test or refused to give a breath sample when requested by the police (other than when incapable of doing so for medical reasons) as well as deceased drivers or riders who had a blood sample taken within 12 hours of the collision in which alcohol levels above the legal limit were detected.

A full list of the casualty definitions used in this release is available.

Revisions to statistics

A change was made to the methodology for estimating the number of drink-drive collisions and casualties. To prevent double-counting collisions which involved a driver who failed a breath test and a driver who died and alcohol was detected in their blood, the collision is treated the same as collisions which had a deceased drink-driver and no drivers who failed a breath test.

To account for missing data, the number of collisions is scaled up to estimate the actual number of drink-drive collisions. The number of casualties is also scaled up. Previously, the number of casualties was scaled based on the severity of the casualty but they are now scaled based on the severity of the collision the casualty was in.

These changes only affect statistics from 2005 onwards and only cause minor changes to the statistics. This means that the main trends shown by the statistics are unaffected by these changes.

Detailed information about the methodology used to produce these statistics can be found in the methodology note.

Impact of COVID-19

It is important to note that the number of reported drink-drive collisions and casualties involved in them is likely to have been impacted by the COVID-19 pandemic and the national restrictions implemented during periods of 2020 and 2021, which led to a reduction in traffic and collisions, but may also have affected driver behaviour. Since then, there has been an increase in the overall number of collisions reported with numbers broadly showing a return to levels seen before the COVID-19 pandemic.

Quality and methodology

Data sources

Data on the characteristics of the collisions are taken from the STATS19 database and more information on STATS19 and how road traffic casualty data is collected from police forces is available.

Drink-drive collisions are identified using data on roadside breath tests and toxicology data for deceased drivers. The breath test data is included in the STATS19 database. The toxicology data is collected from coroners’ and procurators’ fiscal reports for drivers or riders who were killed in road traffic collisions. These two data sources are independent and the trends in each are not necessarily related.

Completeness and timeliness

The road safety statistics team is working to allow the data collection from coroners to start earlier and more regularly based on STATS19 data. The aim is to improve the overall response rate and allow the publication of drink-drive estimates to be released less than 20 months after the end of the year. This approach is currently being tested and, although initial findings were encouraging, the COVID pandemic has impacted on data supply timescales to date.

The return rates for the last 4 years are as follows:

  • 2020: 77%
  • 2021: 78%
  • 2022: 83%
  • 2023: 84%

The figures for 2020 and 2021 are lower than previous years due to the COVID-19 pandemic. Return rates have increased for 2022 and 2023.

Sampling uncertainty

Toxicology data are not available for all killed drivers or riders recorded in STATS19 and are typically available for around 60% to 70% of relevant cases.

To account for the killed drivers without a known BAC, the casualties from the known cases are scaled up. The estimates are therefore based on a sample, rather than a complete count, which introduces an element of uncertainty.

Due to the nature of the data used to create these estimates, there is considerably more uncertainty in the number of fatalities and fatal collisions than any other severity level. The reason for this is that 65% of fatalities in drink-drive collisions in 2023 were drink-drivers themselves.

The estimates in this release are based only on those road collisions which are reported to the police. Comparisons of road accident reports with death registrations 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. The data used as the basis for these statistics are therefore not a complete record of all personal injury road accidents, and this should be borne in mind when using and analysing the figures.

Communicating uncertainty

To communicate the uncertainty in these estimates, numbers of casualties and collisions have been rounded to the nearest 10. Also, 95% confidence intervals are used when referencing estimates for the number of fatalities throughout the publication.

The 95% confidence level is the standard against which statistics are typically tested. The bars on chart 1 are ranges of values for an estimate which we are 95% confident that the ‘true’ value falls in. Technically, it indicates that in 100 years with the same risk of fatalities (or injury), 95 of those years will result in a number of fatalities (or injuries) between a given range.

When we compare figures for the latest year with the previous year, we say that a change is ‘statistically significant’ if we are sufficiently confident (at the chosen level – here, 95%), based on the data available, that the difference between the two figures is the result of a genuine change, rather than being a product of chance - broadly, this is when the two sets of bars on chart 1 don’t overlap. Otherwise, we say the change is ‘not statistically significant’ – this does not necessarily mean that there is not a genuine change, but that we are not able to determine one exists from the data available, at the chosen level of confidence.

User engagement

Following user engagement, we produce drink-drive estimates annually in July and then revise the estimate in the following year. In the longer term we aim to reduce the delay from the end of the year making the release more timely.

About these statistics

These official statistics were independently reviewed by the Office for Statistics Regulation in July 2013. They comply with the standards of trustworthiness, quality and value in the Code of Practice for Statistics and should be labelled ‘accredited official statistics’.

They undergo regular quality assurance reviews to ensure they meet customer needs. More information can be found on our accredited official statistics status webpage.

Accredited official statistics are called National Statistics in the Statistics and Registration Service Act 2007. Accredited Official Statistics Policy – Office for Statistics Regulation

Details of ministers and officials who receive pre-release access to these statistics up to 24 hours before release are available.

Instructions for printing and saving

Depending on which browser you use and the type of device you use (such as a mobile or laptop) these instructions may vary.

You will find your print and save options in your browser’s menu. You may also have other options available on your device. Tablets and mobile device instructions will be specific to the make and model of the device.

Select Ctrl and F on a Windows laptop or Command and F on a Mac

This will open a search box in the top right-hand corner of the page. Type the word you are looking for in the search bar and press enter. Your browser will highlight the word, usually in yellow, wherever it appears on the page. Press enter to move to the next place it appears.

Contact details

Road safety statistics

To hear more about DfT statistical publications as they are released, follow us on X at DfTstats.