HES and STATS19 One-to-one linkage project
Department for Transport · Ministerial Department
Expired The latest version ended on 30 April 2022. The September 2026 register still lists the agreement, but its term has passed.
- Reference
- DARS-NIC-381383-Z9F2P
- Latest version
- v4.4
- Term of latest version
- 31 January 2022 to 30 April 2022
- Start date
- Before 1 December 2017
- Data controller
- Sole Data Controller
- Commercial purposes
- No
- Sublicensing
- No
- Files released to date
- 21
Why the data was released
Objective for processing
This Data Sharing Agreement permits the retention of the data provided under previous iterations of this Agreement for an interim period. This is a pragmatic approach to provide an active Agreement whilst enabling The Department of Transport to complete the necessary actions to enable a subsequent application to extend the Agreement meeting all applicable data sharing standards as published in NHS Digital’s website (see: https://digital.nhs.uk/services/data-access-request-service-dars/dars-guidance).
Road safety is a public health problem. Road collisions cost the country an estimated £36bn a year. With over 500 thousand people estimated as being injured on the roads in Britain a year (source National Travel Survey), about 20 thousand of whom are ‘seriously’ injured (by the definition of the police, source STATS19), and around 5 thousand that have potential ‘life-changing’ injuries (source HES with Maximum Abbreviated Injury Score of 3 or greater (MAIS3+)), there is a clear public health need for the problem to be addressed both in terms of national strategic priorities and through local operational approaches.
The Department for Transport (DfT) has a role in setting the strategic direction of road safety in Great Britain and is the lead agency in producing the evidence-base used by policy makers, operational delivery bodies, research bodies and vehicle manufacturers.
DfT has been collecting and publishing statistics on road accidents and casualties since 1927. The majority of the data used comes from police reports of the accidents. These outputs are National Statistics, and they are critical pieces of evidence which are used to formulate road safety policies at both a national and local level. The police reported data are also extensively used by police forces, researchers, and commercial organisations to make the roads safer.
The police recorded data contains considerable uncertainty in the classification of injuries as being ‘slight’ or ‘serious’. The Hospital Episode Statistics (HES) linkage project started during the mid- to late- 2000s to define injuries using standardised clinical data, rather than guesswork at the side of the road. The first round of matching linked police data to HES data (admitted patient care and accident & emergency records) for each year between 1999 and 2010. This linkage work was carried out by NHS Digital and its predecessor organisations resulting in a report in 2012 highlighting the differences in reported and HES outcomes.
The "Road Safety Statement 2019: A Lifetime of Road Safety" was published in July 2019. This outlines the government’s approach to improving road safety. The document specifically identifies the value of HES data in the road safety context. The government, as a whole, has committed to maximising the use of HES data to improve the evidence used by practitioners.
The objectives for the processing are:
1. To understand the types of injuries sustained by people injured in road traffic accidents.
2. To accurately measure the number of admissions to hospital following a road traffic accident during the COVID-19 lockdown period.
All statistics released by DfT will be aggregated and non-identifiable. Outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guide.
The legal basis for processing falls under GDPR Article 6(1)( e) and GDPR Article 9(2)(j). Processing and dissemination is for research purposes as analysing HES data is necessary for scientific research to understand the type of injuries sustained by people injured in road traffic accidents, and statistical purposes such as determining the number of clinically seriously injured road casualties. The output statistics are released as part of the Reported Road Casualties Great Britain publication each year. Furthermore, it also meets the public interest criteria as defined by the Information Commissioner’s Office (ICO) for good decision-making by public bodies and securing the best use of public resources. The processing of data for this study is a task of public interest since the data will be used in statistics which are used to formulate road safety policies. There are no alternative, less intrusive methods of achieving this purpose.
Stats19 data is collected under section 39 (3) (a) of the Road Traffic Act 1998 by the Department of for Transport to carry out studies into accidents arising out of the use of vehicles. The purpose is to take measures to prevent such accidents. They use the data to formulate road safety policies, monitor trends for road accidents, and give advice relating to the use of roads.
The DfT have requested the retention of HES data for all historical years (1999 to 2020). The reasons for this request are outlined below:
1) The historical pseudonymised HES data enables DfT to examine the number of admitted patient care admissions recorded as being related to a road traffic accident over a long time series.
2) The historical data allows analysis to be completed on new breakdowns of the statistics (e.g. on different road user groups, different types of accidents and roads conditions) over a long period. Road safety is a slowly evolving field, and any change in behaviours needs to be monitored over a long period of time. The different breakdowns required from the time series might change depending on the focus of policy interventions. A new field of interest might be, for example, the type of injuries sustained by car casualties who were not wearing a seatbelt.
3) The historical HES data enables the DfT to investigate any data concerns that can only be detected through analysis of trends over a long period of time. For example, recent analysis of the unmatched HES data has revealed that the matching rates with the International Classification of Diseases (ICD) look-up for injury has worsened over time. Analysis of all years available will enable DfT to understand whether this is due to changes in recording practices for specific injuries.
4) The historical HES data enables the DfT to ground any changes in methodology for future HES/ police recorded reported road casualties (STATS19) data matching exercises in evidence and understand their potential impact. For example, in previous data linkage exercises, matching criteria included a restriction of the geographical scope to only look for potential matches in hospitals located in the same or in adjacent Strategic Health Authorities (SHAs) to where accidents took place. Since SHAs boundaries ceased to exist, recent data linking has instead used a list of the closest hospitals to the accident location, in distance calculated from postcodes/coordinates. For the new methodology, DfT needs to precisely define how many hospitals should be in scope for the matching (e.g. the 10 or 15 closest) and understand how this might affect the matching rate. DfT will use the historical HES data matched to STATS19 to compare the rank in distance of the hospital it was actually matched to with the old methodology. This will allow DfT to refine the range of hospitals ranks and distances that DfT should consider in future for the methodology. Keeping the historical HES data will also enable DfT to compare matching rates, for different matching criteria and groups, between old and new methodologies. Without this analysis of historical HES data, DfT would be at risk of basing their methodology on arbitrary thresholds, which could result in a poor matching rate and/or indications of a large volume of false positive matches (e.g., more use of matching criteria at the less exact end of the scale compared to previous exercises) – making comparison over time unreliable.
5) The historical HES data also gives DfT an external source to evaluate the quality of the STATS19 data against, and this continues to be of prime importance following the recent STATS19 review. The Project Initiation Document from this review (the periodical review of accident data collection) included a strand of work on comparing injuries in HES and STATS19 data. The ability to use the historical HES data to assess the accuracy of reporting on serious injuries within STATS19 data and changes over time is especially important as new injury-based reporting systems (Collision Recording and Sharing CRASH/ Case Overview Preparation Application COPA) are now used by around half of police forces in England. These systems see casualty severity determined based on the type of injury (based on analysis of injuries in HES) rather than based on the police officer’s judgement of seriousness. This has previously caused a major discontinuity in the headline statistics.
The analysis of the matched dataset will help DfT review whether the current list of injuries used in new injury-based reporting systems (CRASH/COPA) is still appropriate, or if it needs to be changed or updated in the future by looking at the injury types and how these have been captured as being serious within STATS19 over time. Analysis of the linked HES data has already informed a recommendation and a decision on what a unified system for recording injury severity should be for all police forces, and this will need to be kept under review in future to ensure it remains appropriate.
6) Historical HES data is likely to be a key evidence source for similar work in future STATS19 reviews, which are regularly carried out (currently every 5 years). Any future requests for HES would require an amendment or renewal to the Data Sharing Agreement.
7) To enable evidence on the number of admissions to hospital following a road accident during the COVID-19 lockdown period. There is a very high public interest in the effects of lockdown on road safety and the data requested from NHS Digital can be used to form the analysis.
As a result, DfT applied to keep the historical HES data to preserve the evidence base and this resulted in the removal of the special condition to destroy any historical HES data.
The DfT are the sole Data Controller and Data Processor under this Agreement. No other organisations have access to the data, or have any role in determining how the data is used.
Processing activities
Under this Agreement, the data may be securely stored but not otherwise processed. No new data will be provided by NHS Digital under this Agreement.
All individuals who have access to the record level data supplied by NHS Digital have it on a need to know basis and are substantive employees of the Department of Transport. All employees are responsible for information training and the information asset owners complete the information asset owner training.
NHS Digital will match STATS19 data (police recorded road accident data collected by DfT) to Hospital Episode Statistics Admitted Patient Care and Hospital Episode Statistics Accident & Emergency data. DfT will supply to NHS Digital the linkage postcode, age, sex, and date of accident from STATS19. NHS Digital will also supply non-matched episodes with matching road accident codes. DfT will receive hospital data that can be compared with police reporting and all accidents not reported to police. All data will be returned to DfT in a pseudonymised format. Re-identification is not permitted and will not be attempted by DfT.
DfT will convert the ICD-10 codes (for causation of the injury, type of injury and severity of injury) into abbreviated injury scale (AIS) scores. DfT will use the information to produce an analysis of the number of people with a MAIS (Maximum Abbreviated Injury Score) of 3 or greater, plus analysis of the types of injuries different road user groups suffer from.
Annual police recorded road traffic accident casualties will be matched with key HES variables (e.g. ICD-10 codes). This has been completed for 2011, 2012, 2013, 2014, 2015 and 2016 data. To be completed for 2017, 2018, 2019, 2020, 2021, and 2022 data.
The output statistics will be released as part of the Reported Road Casualties Great Britain publication in September each year (or alongside other DfT road safety statistical publications).
Outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guide.
All the statistical reports and tables are available to everyone free of charge.
The most recent report is available on the gov.uk website. Previous Annual Reports are also available on The National Archives website.
The retention of HES data has been granted. Reasons for the retention included the following: A lookup file is being developed to determine the Abbreviated Injury Score (AIS) for ICD-10 codes relating to consequences of external causes which enables MAIS3+ estimates to be generated. The lookup file is still under development and will be updated in the future. Losing the data back to 1999 will mean that the Department cannot update its estimates of MAIS3+ casualties back to 1999 when the lookup file is updated. This will represent a loss to the Department’s road safety evidence base.
DfT will hold the pseudonymised data in a secure database with restricted access to specific individuals. The pseudonymised data will be stored separately from the STATS19 data. Most of the analysis will be on the HES data in isolation. However, some analysis will take place linking the HES data to the non-identifiable STATS 19 data consisting of data such as road type, speed limit, conditions etc. The HES data will never be linked back to the postcode and age data held within the STATS19 database.
DfT's SQL database server was moved over to Google Cloud Platform (GCP) in October 2020. The SQL server that hosts the HES data has an audit tool installed that complies with GDPR. This server was built in October 2020 and data (including HES data held by DfT) was transferred to it shortly after. The server’s GCP host is the Europe West One region (Belgium).
All the statistics that will be released by DfT will be aggregated by user group, geography, etc. No identifiable data will be released. Outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guide.
All organisations party to this agreement must comply with the Data Sharing Framework Contract requirements, including those regarding the use (and purposes of that use) by “Personnel” (as defined within the Data Sharing Framework Contract i.e.: employees, agents and contractors of the Data Recipient who may have access to that data).
Expected output
This Agreement permits the secure retention of the data only and no other processing.
No new outputs will be produced under this Data Sharing Agreement.
The expected outputs are:
1. To estimate the number of people who have been injured in road traffic accidents and who were treated at hospital
2. To estimate the number of people who have been seriously injured in road traffic accidents, defined as having a Maximum Abbreviated Injury Score of 3 or higher (MAIS3+)
(3 Serious, 4 Severe, 5 Critical, 6 Unsurvivable taken from https://www.aaam.org/abbreviated-injury-scale-ais/)
The output statistics will be released as part of the Reported Road Casualties Great Britain publication in September each year.
The Reported Road Casualties Great Britain statistical series is the main avenue to get analysis of Hospital Episodes Statistics related to traffic accidents published. HES data might be published in several editions of this series (either on its own publication, or as part of a larger one like the annual report).
HES analysis is also used for internal purposes, for example to contribute to the STATS19 review (the periodical review of accident data collection from the police) which includes a strand of work on comparing injuries in HES and STATS19 data to evaluate severity reporting approaches.
All the statistical reports and tables are available to the public free of charge.
The most recent report is available on the gov.uk website, https://www.gov.uk/government/collections/road-accidents-and-safety-statistics. Previous Annual Reports are also available on The National Archives website, https://webarchive.nationalarchives.gov.uk/20190801212554/https://www.gov.uk/government/collections/road-accidents-and-safety-statistics.
Outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guidance.
Expected measurable benefits
This Agreement permits the secure retention of the data only and no other processing.
DfT currently produces statistics on the number of ‘seriously injured’ casualties in reported road traffic accidents. These figures are based on the judgement of police officers when they attend the scene of an accident. It can be very hard for officers to judge whether a casualty is slightly or seriously injured as they have neither clinical training nor access to diagnostic tools. The current measure also varies between countries, so it is impossible for comparisons to be drawn across Europe, making it harder for countries to learn from one another.
The use of hospital data will achieve the following things :
1. To identify casualties who are truly seriously injured, and in particular, the separation of people with less serious injuries from casualties with life-changing injuries.
2. Identification of all seriously injured casualties in road traffic accidents, including those which are not reported to the police. This will help deal with the underestimation problem with the road traffic accident data (there is no obligation to report accidents to the police, though anyone who is seriously injured will attend hospital and be recorded in HES data).
Each of these factors will allow DfT to produce more accurate statistics on casualties. This, in turn, will help road safety organisations identify patterns of interest and put in place interventions to reduce accidents.
HES linked data from previous years have been used to improve understanding of the types of collisions that result in the most serious injuries, as well as what type of injuries come about for specific road users in certain conditions. Local authorities have made use of this knowledge to better target interventions on the highways, and researchers have used the information to advise manufacturers of how vehicular safety can be improved.
Improving the health of Londoners: transport action plan, published by Transport for London in 2014, is an example of how road safety is integrated in the wider public health context at a local level.
The value of HES data lie in a number of areas:
1. HES data is the only indicator of ‘life-changing’ injuries - The serious severity in police-reported casualty data covers a wide range of injuries, including relatively minor injuries (e.g. ones that casualties will recover from quickly, or injuries for which the casualty is admitted to hospital simply as a precaution). Clinical data from HES allows DfT to understand how many people (and under what conditions and types of road users) suffer from the most serious life-changing injuries. This will help DfT formulate policies to better target / protect at risk road users. There is a public benefit in being able to target different types of interventions at different outcomes. For instance, the type of accident that are more likely to leave a casualty paralysed, or with long-term disabilities required different interventions from those that result in minor fractures. Police-reported data is not and cannot be sophisticated and reliable enough to make this distinction. Information drawn from HES, though, does achieve this.
2. Valuation of prevention of accidents - Most road safety interventions undergo cost-benefit analysis to value the benefits from accidents avoided or reduced in severity. The current valuation models are based around police-reported severities. This can result in a significant variation in the cost of accidents – for instance, a less ‘serious’ injury will have a much lower long-term cost in terms of medical care and national productivity than a more severe ‘serious’ injury. The simplistic police-reported approach makes it impossible to distinguish between these higher and lower costs. HES-derived clinical data should give a more reliable ways of modelling costs for a greater range of injuries.
3. Types of injuries suffered. -The type of injury (e.g. lower limb crushing, head injuries, etc) casualties suffer in specific types of accidents influence what mitigation and development needs to be carried out. With more accurate statistics on casualties this will, in turn, help road safety organisations identify patterns of interest and put in place interventions to reduce accidents. Fewer accidents would result in wider benefits to the health system e.g. a reduced number of visits to A&E and hospital admissions.
None of this information is available in police-records – and if DfT tried to collect it, it would probably be unreliable and misleading. HES data are the only way in which DfT can understand what types of injuries different casualties suffer.
For instance, if vehicle manufacturers and DfT (from a vehicle standards point of view) understand the types and severities of injuries that cyclists and pedestrians suffer in collisions with HGVs, DfT can better target standards and vehicle development to areas that will address these specific injuries.
Without the clinical injury information it is much harder to prioritise research and development in areas that will deliver the highest benefits that will improve the health of road users and reduce the burden on the health system.
The outputs of this work would partly be of direct benefit to DfT and also to health (e.g. by influencing road safety policy, vehicle and road design standards, which would hopefully lead to fewer accidents), but much of it will be reused by external bodies such as vehicle manufacturers and researchers which all cumulatively benefits the road users.
4. Reliability of police-reported data. - DfT has known for some time that police-reported serious injuries are under-reported. However, DfT now have concerns that underreporting is worsening as the police attend fewer accidents and it is harder for members of the public to visit a police station to report an accident. This in itself is a significant public health problem because if underreporting is increasing then the dangers and risks on the road appear to be reducing. Without taking a potential increase in underreporting of accidents into account national and local governments could make a mistake in thinking that the problems from road collisions are decreasing and going away.
Statistics derived from HES provide important evidence in understanding both the scale and the potential changes in scale of underreporting. Seriously injured casualties will attend A&E and be admitted as an in-patient even if an accident is not reported to the police. It is therefore a critical source of information for the wider public health context in both validating and complementing the police-reported data.
Any country (including Great Britain/ United Kingdom) wishing to produce comparable statistics using this standard needs to use clinical data containing ICD codes akin to HES data.
Once statistics are available from a number of countries it will be much easier for public health and road safety leads in Britain to identify areas of good practice from other countries and adopt them here for the health benefit of all road users and to reduce the burden on the NHS from road traffic accidents.
Additionally, as for the current DSA, for this amendment the Department for Transport require the data within the data sharing agreement to investigate road safety issues during the extended COVID time period. There is significant ministerial and public interest in how road usage and road safety has changed during this time period. There have been reports in the media of increases in speeding and other risky behaviours. Several European countries have reported road accidents volumes are not decreasing at the same rate as traffic volumes and a higher proportion of the accidents are severe or fatal.
The data requested from NHS Digital can be used to produce analysis that shows the number of patients admitted to hospital following road accidents and the severity of the accidents in 2020, 2021 and early 2022. Understanding the volume and severity of road accidents occurring during this time period will support policy colleagues working on recovery and restart planning. It will inform the development of interventions, communication and education activities preparing for users to return to the roads. For example, if there is an increase in the severity of accidents occurring during this period, interventions may be targeted towards reducing the severity of accidents. If there is an increase in volume but not severity, reducing the likelihood of accidents occurring may be prioritised. The data requested and the subsequent analysis will enrich the broader evidence base supporting this work.
Incorporate traffic accidents during the extended duration of Covid-19, spanning each of the lockdown periods, will develop understanding of road safety during this time and the severity of accidents occurring. It will support restart and recovery planning and the development of interventions and communications as users once again return to the roads.
Benefits reported so far
HES data has been used in the Reported Road Casualties Great Britain 2016, 2017, 2018 and most recently the 2019 annual report (published September 2020). Analysis has helped assess the trend in serious injuries as a result of road accidents and to understand the context around the severity reporting in STATS19.
HES analysis has been used for internal purposes, for example to contribute to the STATS19 review (the periodical review of accident data collection from the police) which includes a strand of work on comparing injuries in HES and STATS19 data to evaluate severity reporting approaches. This has been useful in identifying the accuracy of reporting on serious injuries within STATS19 data and changes over time. It has been used to assess the accuracy of the severity reporting of road accidents compared with the injury based reporting of road accidents by police forces who use CRASH/COPA. This has been used to produce recommendations to encourage more uptake of the injury-based reporting. This analysis has also been used to further improve and develop definitions of the injury descriptions in STATS19 to allow better recording of injuries by police forces.
Datasets on the latest version
Legal basis for provision: Health and Social Care Act 2012 - s261 - 'Other dissemination of information'
| Dataset | Type of data | Sensitivity | Frequency | Confidential data |
|---|---|---|---|---|
| Hospital Episode Statistics Accident and Emergency (HES A and E) | Anonymised - ICO Code Compliant | Non-Sensitive | One-Off | Section 251 NHS Act 2006 |
| Hospital Episode Statistics Admitted Patient Care (HES APC) | Anonymised - ICO Code Compliant | Non-Sensitive | One-Off | Section 251 NHS Act 2006 |
Files released
Files released counts only files released externally by DARS. Access granted in NHS England's own systems, such as its Secure Data Environment, is not included.
Patient opt-outs were applied to 9 of the 21 files released under this agreement, across every version. About opt-outs
No files recorded as released under the latest version. 21 were released under earlier versions, shown in the version history.
Version history
The register lists each renewal of this agreement as a separate row. This site has 4 versions — earlier versions existed before this site's records begin.
DARS-NIC-381383-Z9F2P-v4.4 31 January 2022 to 30 April 2022
- Title
- HES and STATS19 One-to-one linkage project
- Commercial
- No
- Sublicensing
- No
- Datasets
- 2
- Files released
- 0
Datasets: Hospital Episode Statistics Accident and Emergency (HES A and E); Hospital Episode Statistics Admitted Patient Care (HES APC)
What changed from DARS-NIC-381383-Z9F2P-v3.6
Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.
| Field | Was | Became |
|---|---|---|
| Start date | 2022-01-31 | |
| End date | 2022-04-30 | |
| Hospital Episode Statistics Accident and Emergency (HES A and E): legal basis | Health and Social Care Act 2012 - s261 - 'Other dissemination of information' | |
| Hospital Episode Statistics Admitted Patient Care (HES APC): legal basis | Health and Social Care Act 2012 - s261 - 'Other dissemination of information' |
Objective for processing
The Department for Transport (DfT) has been collecting and publishing statistics on road accidents and casualties since 1927. The majority of this data used comes from police reports of the accidents. These outputs are National Statistics, and they are critical bits of evidence which are used to formulate road safety policies at both a national and local level.
This Data Sharing Agreement permits the retention of the data provided under previous iterations of this Agreement for an interim period. This is a pragmatic approach to provide an active Agreement whilst enabling The Department of Transport to complete the necessary actions to enable a subsequent application to extend the Agreement meeting all applicable data sharing standards as published in NHS Digital’s website (see: https://digital.nhs.uk/services/data-access-request-service-dars/dars-guidance).
The police reported data are also extensively used by police forces, researchers and commercial organisations to make the roads safer.
Road safety is a public health problem. Road collisions cost the country an estimated £36bn a year. With over 500 thousand people estimated as being injured on the roads in Britain a year (source National Travel Survey), about 20 thousand of whom are ‘seriously’ injured (by the definition of the police, source STATS19), and around 5 thousand that have potential ‘life-changing’ injuries (source HES with Maximum Abbreviated Injury Score of 3 or greater (MAIS3+)), there is a clear public health need for the problem to be addressed both in terms of national strategic priorities and through local operational approaches.
The police recorded data contains considerable uncertainty in the classification of injuries as being ‘slight’ or ‘serious’. The Hospital Episode Statistics (HES) linkage project started during the mid- to late- 2000s to define injuries using standardised clinical data, rather than guesswork at the side of the road. The first round of matching linked police data to HES data (admitted patient care and accident & accident and emergency records) for each year between 1999 and 2010. This linkage work was carried out by NHS Digital and its predecessor organisations resulting in a report in 2012 highlighting the differences in reported and HES outcomes.
The Department for Transport (DfT) has a role in setting the strategic direction of road safety in Great Britain and is the lead agency in producing the evidence-base used by policy makers, operational delivery bodies, research bodies and vehicle manufacturers.
DfT has been collecting and publishing statistics on road accidents and casualties since 1927. The majority of the data used comes from police reports of the accidents. These outputs are National Statistics, and they are critical pieces of evidence which are used to formulate road safety policies at both a national and local level. The police reported data are also extensively used by police forces, researchers, and commercial organisations to make the roads safer.
The police recorded data contains considerable uncertainty in the classification of injuries as being ‘slight’ or ‘serious’. The Hospital Episode Statistics (HES) linkage project started during the mid- to late- 2000s to define injuries using standardised clinical data, rather than guesswork at the side of the road. The first round of matching linked police data to HES data (admitted patient care and accident & emergency records) for each year between 1999 and 2010. This linkage work was carried out by NHS Digital and its predecessor organisations resulting in a report in 2012 highlighting the differences in reported and HES outcomes.
The "Road Safety Statement 2019: A Lifetime of Road Safety" was published in July 2019. This outlines the government’s approach to improving road safety. The document specifically identifies the value of HES data in the road safety context. The government, as a whole, has committed to maximising the use of HES data to improve the evidence used by practitioners.
[2 paragraphs unchanged]
2. To accurately measure the number of admissions to hospital following a road traffic accident during the COVID-19 lockdown period.
[3 paragraphs unchanged]
The
Department for Transport (DfT) (Data controller who processes the data) requests
DfT have requested
the retention of HES data for all historical years (1999 to 2020).
Several
The
reasons for this request are outlined below:
1) The historical pseudonymised HES data enables DfT to
determine the longer time series in
examine
the number of admitted patient care admissions
where the admission is
recorded as being related to a road traffic
accident. It is used as an external data source to compare the Stats19 data to, building the understanding and informing the commentary of underlying trends in road safety. For example, the presentation of the latest HES data and its comparison with Stats19 can be found in the latest Reported Road Casualty Great Britain annual report for 2018.
accident over a long time series.
2) The historical data allows analysis to be completed on new breakdowns of the statistics (e.g. on different road user groups, different types of accidents and roads conditions)
on
over
a long
time
period. Road safety is a slowly evolving field, and any change in behaviours needs to be monitored
on
over
a long
time period. What
period of time. The different
breakdowns
are
required from the time series might change depending on the focus of
[13 words unchanged]
of injuries sustained by car casualties who were not wearing a seatbelt.
3) The historical HES data enables the DfT to investigate any data concerns that can only be detected through analysis of trends over a long
time period.
period of time.
For example, recent analysis of the unmatched HES data has revealed that
[27 words unchanged]
whether this is due to changes in recording practices for specific injuries.
4) The historical HES data enables the DfT to ground any changes in methodology for future HES/ police recorded reported road casualties
(Stats19)
(STATS19)
data matching exercises in evidence and understand their potential impact. For example, in
the last
previous
data linkage
exercise,
exercises,
matching criteria included a restriction of the geographical scope to only look
[12 words unchanged]
Strategic Health Authorities (SHAs) to where accidents took place. Since SHAs boundaries
no longer
ceased to
exist,
the current
recent
data
linkage exercise will use
linking has instead used
a list of the closest hospitals to the accident location, in distance calculated from postcodes/coordinates. For the new methodology, DfT
need
needs
to precisely define how many hospitals should be in scope for the
[13 words unchanged]
the matching rate. DfT will use the historical HES data matched to
Stats19 (statistical dataset collected by the Department for Transport from the police on personal-injury road traffic accidents) and calculate
STATS19 to compare
the rank in distance of the hospital it was actually matched to with the old methodology. This will
give
allow
DfT
an idea of
to refine
the range of hospitals ranks and
distance
distances that
DfT should consider in
scope
future
for the
new
methodology. Keeping the historical HES data will also enable DfT to compare matching rates, for different matching criteria and groups, between
the
old and
the
new
methodology.
methodologies.
Without this analysis of historical HES data, DfT would be at risk of basing
the new
their
methodology on
an
arbitrary
threshold,
thresholds,
which could result in a poor matching rate and/or indications of a large volume of false positive matches
(e.g.
(e.g.,
more use of matching criteria at the less exact end of the scale compared to previous exercises) – making comparison over time unreliable.
5) The historical HES data also gives DfT an external source to evaluate the quality of the
Stats19
STATS19
data against, and this
will
continues to
be of prime importance
in
following
the
current Stats19
recent STATS19
review. The Project Initiation Document from
the Stats19
this
review (the periodical review of accident data collection)
includes
included
a strand of work on comparing injuries in HES and
Stats19
STATS19
data. The ability to use the historical HES data to assess the
[5 words unchanged]
injuries within STATS19 data and changes over time is especially important as
there are now
new
injury-based
reporting systems (Collision Recording and Sharing CRASH/ Case Overview Preparation Application COPA)
are now
used by around half of police forces in
England whereby
England. These systems see casualty
severity
is
determined based on the type of injury
from HES
(based on analysis of injuries in HES)
rather than based on the police officer’s judgement of seriousness. This has previously caused a major discontinuity in the headline statistics.
The analysis of the matched dataset will help DfT
determine
review
whether the current list of injuries used in new injury-based reporting systems
[23 words unchanged]
how these have been captured as being serious within STATS19 over time.
The outcome
Analysis
of
this work will be
the linked HES data has already informed
a recommendation and a decision on what a unified system for recording injury severity should be for all police
forces.
forces, and this will need to be kept under review in future to ensure it remains appropriate.
6)
Historical HES data is likely to be a key evidence source for similar work in future
Stats19
STATS19
reviews, which are regularly carried out (currently every 5 years). Any future requests for HES would require an amendment or renewal to the
data sharing agreement.
Data Sharing Agreement.
7) To enable evidence on the number of admissions to hospital following a road accident during the COVID-19 lockdown period. There is a very high public interest in the effects of lockdown on road safety and the data requested from NHS Digital can be used to form the analysis.
[1 paragraph unchanged]
Road safety is a public health problem. Road collisions cost the country an estimated £35bn 36bn a year.
The DfT are the sole Data Controller and Data Processor under this Agreement. No other organisations have access to the data, or have any role in determining how the data is used.
With over 700560 thousand people estimated to being injured on the roads in Britain a year (source National Travel Survey) , 20 thousand of whom being ‘seriously’ injured (by the definition of the police, source STATS19) , and almost 5 thousand having potential ‘life-changing’ injuries (source HES with Maximum Abbreviated Injury Score of 3 or greater (MAIS3+)), there is a clear public health need for the problem to be addressed both in terms and national strategic priorities and through local operational approaches.
DfT has a role in setting the strategic direction of road safety in Great Britain and is the lead agency in producing the evidence-based used by policy-makers, operational delivery bodies, research bodies and vehicle manufacturers. The "Working together to build a safer road system: British road safety statement " was published in December 2015, this outlines the government’s approach to improving road safety. The document specifically identifies the value of HES data in the road safety context and the government, as a whole, has committed to maximising the use of HES data to improve the evidence used by practitioners.
The Department for Transport (DfT) (Data controller who processes the data) requests the retention of HES data for all historical years (1999 to 2020) to:
1) Enable DfT to determine the longer time series in the number of hospital admissions, where the admission is recorded as being related to a road traffic accident
2) Allow analysis to be completed on new breakdowns of the statistics over a long time period
3) Enable DfT to investigate any data concerns that can only be detected through analysis of trends over a long period of time
4) Enable DfT to ground any changes in methodology for further HES/STATS19 data matching exercises in evidence and understand their potential impact
5) Provide DfT with an external source to evaluate the quality of the STATS19 data
6) Used as a key evidence source for future STATS19 reviews
AMENDMENT REQUEST
The amendment request relates to Department for Transport requiring the HES data within the data sharing agreement to investigate road safety issues during the coronavirus disease (COVID-19) time period. There is a very high public interest during lockdown on the effects on road safety and the data request from NHS Digital can be used to form the analysis. The analysis will show the number of patients admitted to hospital following a road accident, in the first few months of the year of 2020. This will support policy colleagues in understanding road safety during the COVID-19 period so far and in preparing for users to return to roads.
As such, the DfT have added the following objective for processing:
1. The number of admissions to hospital following a road traffic accident during the COVID-19 lockdown period
and have added the following reason to those listed 1-6 as to why HES data is needed to be retained for all historical years (1999-2020)
7. To enable evidence on the number of admissions to hospital following a road accident during the COVID-19 lockdown period.
Historical HES data is likely to be a key evidence source for similar work in future Stats19 reviews, which are regularly carried out (currently every 5 years). Any future requests for HES would require an amendment or renewal to the data sharing agreement. Admissions to hospital following a road traffic accident evidence will be used for understanding road safety during the COVID-19 lockdown period which will be useful to inform policy and in preparing users to return to roads.
Processing activities
All individuals who have access to the record level data supplied by NHS Digital have it on a need to know basis and are substantive employees of the Department of Transport . All employees a responsible for information training and the information asset owners complete the information asset owner training.
Under this Agreement, the data may be securely stored but not otherwise processed. No new data will be provided by NHS Digital under this Agreement.
The data disseminated in this agreement will be used only for the purpose outlined in the objectives for processing.
All individuals who have access to the record level data supplied by NHS Digital have it on a need to know basis and are substantive employees of the Department of Transport. All employees are responsible for information training and the information asset owners complete the information asset owner training.
NHS Digital will match STATS19 data (police recorded road accident data collected by
DFT)
DfT)
to Hospital Episode Statistics
admitted patient care
Admitted Patient Care
and
accident
Hospital Episode Statistics Accident
&
emergency
Emergency
data. DfT will supply
for
to NHS Digital the
linkage postcode, age, sex,
and
date of
accident.
accident from STATS19.
NHS Digital will also supply non-matched episodes with matching road accident codes.
[29 words unchanged]
format. Re-identification is not permitted and will not be attempted by DfT.
DfT will hold the pseudonymised data in a secure database with restricted access to specific individuals. The pseudonymised data will be stored separately from the STATS 19 data. Most of the analysis will be on the HES data in isolation. However, some analysis will take place linking the HES data to the non-identifiable STATS 19 data consisting of data such as road type, speed limit, conditions etc. The HES data will never be linked back to the postcode and age data held within the STATS19 database. Data storage and access will only be in DfT’s secure network.
DfT will convert the ICD-10 codes (for causation of the injury, type of injury and severity of injury) into abbreviated injury scale (AIS) scores. DfT will use the information to produce an analysis of the number of people with a MAIS (Maximum Abbreviated Injury Score) of 3 or greater, plus analysis of the types of injuries different road user groups suffer from.
DfT will convert the ICD-10 codes (for causation of the injury, type of injury and severity of injury) into abbreviated injury scale (AIS ) scores. DfT will use the information to produce analysis of the number of people with a MAIS (Maximum Abbreviated Injury Score) of 3 or greater, plus analysis of the types of injuries different road user groups suffer from.
Annual police recorded road traffic accident casualties will be matched with key HES variables (e.g. ICD-10 codes). This has been completed for 2011, 2012, 2013, 2014, 2015 and 2016 data. To be completed for 2017, 2018, 2019, 2020, 2021, and 2022 data.
Annual police recorded road traffic accident casualties matched with key HES variables (e.g. ICD-10 codes). This has been completed for 2011, 2012, 2013, 2014, 2015 and 2016 data. To be completed for 2017, 2018, 2019 and 2020 data.
The output statistics will be released as part of the Reported Road Casualties Great Britain publication in September each year (or alongside other DfT road safety statistical publications).
The output statistics will be released as part of the Reported Road Casualties Great Britain publication in September each year.
[3 paragraphs unchanged]
The retention of HES data has been granted. Reasons for the retention included the
following.
following:
A lookup file is being developed to determine the Abbreviated Injury Score
[58 words unchanged]
This will represent a loss to the Department’s road safety evidence base.
DfT will hold the pseudonymised data in a secure database with restricted access to specific individuals. The pseudonymised data will be stored separately from the STATS19 data. Most of the analysis will be on the HES data in isolation. However, some analysis will take place linking the HES data to the non-identifiable STATS 19 data consisting of data such as road type, speed limit, conditions etc. The HES data will never be linked back to the postcode and age data held within the STATS19 database.
DfT's SQL database server was moved over to Google Cloud Platform (GCP) in October 2020. The SQL server that hosts the HES data has an audit tool installed that complies with GDPR. This server was built in October 2020 and data (including HES data held by DfT) was transferred to it shortly after. The server’s GCP host is the Europe West One region (Belgium).
[1 paragraph unchanged]
All organisations party to this agreement must comply with the Data Sharing
[11 words unchanged]
that use) by “Personnel” (as defined within the Data Sharing Framework Contract
ie:
i.e.:
employees, agents and contractors of the Data Recipient who may have access to that data).
Expected output
This Agreement permits the secure retention of the data only and no other processing.
No new outputs will be produced under this Data Sharing Agreement.
[8 paragraphs unchanged]
The most recent report is available on
https://www.gov.uk/government/collections/road-accidents-and-safety-statisticsthe
the
gov.uk
website.
website, https://www.gov.uk/government/collections/road-accidents-and-safety-statistics.
Previous Annual Reports are also available on The National Archives
website
website,
https://webarchive.nationalarchives.gov.uk/20190801212554/https://www.gov.uk/government/collections/road-accidents-and-safety-statistics.
[1 paragraph unchanged]
Expected measurable benefits
Summary:
This Agreement permits the secure retention of the data only and no other processing.
[6 paragraphs unchanged]
Improving the health of Londoners: transport action plan, published by Transport for London in 2014, is an example of how road safety is integrated in
the
wider
the
public health context at a local level.
[1 paragraph unchanged]
1. HES data is the only indicator of ‘life-changing’
injuries-
injuries -
The serious severity in police-reported casualty data covers a wide range of
[126 words unchanged]
to make this distinction. Information drawn from HES, though, does achieve this.
2. Valuation of prevention of
accidents-
accidents -
Most road safety interventions undergo cost-benefit analysis to value the benefits from
[76 words unchanged]
more reliable ways of modelling costs for a greater range of injuries.
3. Types of injuries suffered. -The type of injury (e.g. lower limb
[47 words unchanged]
Fewer accidents would result in wider benefits to the health system e.g.
a reduced
number of visits to A&E and hospital admissions.
[4 paragraphs unchanged]
4. Reliability of police-reported data. - DfT has known for some time
[41 words unchanged]
itself is a significant public health problem because if underreporting is increasing
than
then
the dangers and risks on the road appear to be reducing. Without
[19 words unchanged]
thinking that the problems from road collisions are decreasing and going away.
Statistics derived
from
HES
provides
provide
important evidence in understanding both the scale and the potential changes in
[35 words unchanged]
wider public health context in both validating and complementing the police-reported data.
[2 paragraphs unchanged]
AMENDMENT REQUEST:
Additionally, as for the current DSA, for this amendment the Department for Transport require the data within the data sharing agreement to investigate road safety issues during the extended COVID time period. There is significant ministerial and public interest in how road usage and road safety has changed during this time period. There have been reports in the media of increases in speeding and other risky behaviours. Several European countries have reported road accidents volumes are not decreasing at the same rate as traffic volumes and a higher proportion of the accidents are severe or fatal.
The Department for Transport require the data within the data sharing agreement to investigate road safety issues during the COVID time period. There is significant ministerial and public interest in how road usage and road safety has changed during this time period. There have been reports in the media of increases in speeding and other risky behaviours. Several European countries have reported road accidents volumes are not decreasing at the same rate as traffic volumes and a higher proportion of the accidents are severe or fatal.
The data requested from NHS Digital can be used to produce analysis that shows the number of patients admitted to hospital following road accidents and the severity of the accidents in 2020, 2021 and early 2022. Understanding the volume and severity of road accidents occurring during this time period will support policy colleagues working on recovery and restart planning. It will inform the development of interventions, communication and education activities preparing for users to return to the roads. For example, if there is an increase in the severity of accidents occurring during this period, interventions may be targeted towards reducing the severity of accidents. If there is an increase in volume but not severity, reducing the likelihood of accidents occurring may be prioritised. The data requested and the subsequent analysis will enrich the broader evidence base supporting this work.
The data requested from NHS Digital can be used to produce analysis that shows the number of patients admitted to hospital following road accidents and the severity of the accidents, in the first few months of the year on 2020. Understanding the volume and severity of road accidents occurring during this time period will support policy colleagues working on recovery and restart planning. It will inform the development of interventions, communication and education activities preparing for users to return to the roads. For example, if there is an increase in the severity of accidents occurring during this period interventions may be targeted towards reducing the severity of accidents. If there is an increase in volume but not severity, reducing the likelihood of accidents occurring may be prioritised. The data requested and the subsequent analysis will enrich the broader evidence base supporting this work.
Incorporate traffic accidents during the extended duration of Covid-19, spanning each of the lockdown periods, will develop understanding of road safety during this time and the severity of accidents occurring. It will support restart and recovery planning and the development of interventions and communications as users once again return to the roads.
There is an added benefit by amending the purpose to incorporate traffic accidents during the covid-19 lockdown period, building on the points 1 and 2 and above:
3. The data and subsequent analysis on road accidents during the covid-19 period will develop understanding of road safety during this time and the severity of accidents occurring. It will support restart and recovery planning and the development of interventions and communications as users return to the roads.
Benefits reported
HES data has been used in the Reported Road Casualties Great Britain 2016,
2017
2017, 2018
and most recently the
2018
2019
annual report (published September
2019).
2020).
Analysis has helped assess the trend in serious injuries as a result of road accidents and to understand the context around the severity reporting in STATS19.
[1 paragraph unchanged]
DARS-NIC-381383-Z9F2P-v3.6 15 July 2020 to 31 July 2021
- Title
- HES and STATS19 One-to-one linkage project
- Commercial
- No
- Sublicensing
- No
- Datasets
- 2
- Files released
- 9
Datasets: Hospital Episode Statistics Accident and Emergency (HES A and E); Hospital Episode Statistics Admitted Patient Care (HES APC)
What changed from DARS-NIC-381383-Z9F2P-v2.7
Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.
| Field | Was | Became |
|---|---|---|
| Start date | 2020-07-15 | |
| End date | 2021-07-31 |
Objective for processing
The Department for Transport (DfT) has been collecting and publishing statistics on road accidents and casualties since 1927. The majority of
the
this
data used comes from police reports of the accidents. These outputs are
[12 words unchanged]
to formulate road safety policies at both a national and local level.
[1 paragraph unchanged]
The police recorded data contains considerable uncertainty in the classification of injuries as being ‘slight’ or ‘serious’. The
HES
Hospital Episode Statistics (HES)
linkage project started during the mid- to late- 2000s to define injuries
[12 words unchanged]
road. The first round of matching linked police data to HES data
(admitted patient care and accident & accident and emergency records)
for each year between 1999 and 2010. This linkage work was carried
[10 words unchanged]
a report in 2012 highlighting the differences in reported and HES outcomes.
[3 paragraphs unchanged]
The legal basis for processing falls under GDPR Article 6(1)( e) and GDPR Article 9(2)(j). Processing and dissemination is for research
purposes,
purposes as analysing HES data is necessary for scientific research to understand the type of injuries sustained by people injured in road traffic accidents,
and
statistical purposes such as determining the number of clinically seriously injured road casualties. The output statistics are released as part of the Reported Road Casualties Great Britain publication each year. Furthermore, it
also meets the public interest criteria as defined by the
ICO
Information Commissioner’s Office (ICO)
for good decision-making by public bodies and securing the best use of
[30 words unchanged]
policies. There are no alternative, less intrusive methods of achieving this purpose.
Stats19 data is collected under section 39 (3) (a) of the Road Traffic Act 1998 by the Department of
for
Transport to carry out studies into accidents arising out of the use
[23 words unchanged]
for road accidents, and give advice relating to the use of roads.
The Department for Transport (DfT) (Data controller who processes the data) requests the retention of HES data for all historical years (1999 to
2016).
2020).
Several reasons for this request are outlined below:
1) The historical pseudonymised HES data enables DfT to determine the longer
[66 words unchanged]
found in the latest Reported Road Casualty Great Britain annual report for
2017.
2018.
[1 paragraph unchanged]
3) The historical HES data enables the DfT to investigate any data
[20 words unchanged]
the unmatched HES data has revealed that the matching rates with the
ICD
International Classification of Diseases (ICD)
look-up for injury has worsened over time. Analysis of all years available
[5 words unchanged]
whether this is due to changes in recording practices for specific injuries.
4) The historical HES data enables the DfT to ground any changes in methodology for future
HES/Stats19
HES/ police recorded reported road casualties (Stats19)
data matching exercises in evidence and understand their potential impact. For example,
[21 words unchanged]
in hospitals located in the same or in adjacent Strategic Health Authorities
(SHAs)
to where accidents took place. Since SHAs boundaries no longer exist, the
[53 words unchanged]
matching rate. DfT will use the historical HES data matched to Stats19
(statistical dataset collected by the Department for Transport from the police on personal-injury road traffic accidents)
and calculate the rank in distance of the hospital it was actually
[106 words unchanged]
the scale compared to previous exercises) – making comparison over time unreliable.
5) The historical HES data also gives DfT an external source to
[70 words unchanged]
over time is especially important as there are now new reporting systems
(CRASH/COPA)
(Collision Recording and Sharing CRASH/ Case Overview Preparation Application COPA)
used by around half of police forces in England whereby severity is
[18 words unchanged]
seriousness. This has previously caused a major discontinuity in the headline statistics.
[1 paragraph unchanged]
6)
Historical HES data is likely to be a key evidence source for
[19 words unchanged]
HES would require an amendment or renewal to the data sharing agreement.
As a result, DfT
are applying
applied
to keep the historical HES data to preserve the evidence base and this
results
resulted
in the removal of the special condition to destroy any historical HES data.
Road safety is a public health problem. Road collisions cost the country an estimated £35bn
36bn
a year.
With over
700
700560
thousand people estimated to being injured on the roads in Britain a year
(source National Travel Survey)
, 20 thousand of whom being ‘seriously’ injured (by the definition of the
police)
police, source STATS19)
, and almost 5 thousand having potential ‘life-changing’ injuries
,
(source HES with Maximum Abbreviated Injury Score of 3 or greater (MAIS3+)),
there is a clear public health need for the problem to be addressed both in terms and national strategic priorities and through local operational approaches.
[1 paragraph unchanged]
The Department for Transport (DfT) (Data controller who processes the data) requests the retention of HES data for all historical years (1999 to
2016)
2020)
to:
[6 paragraphs unchanged]
AMENDMENT REQUEST
The amendment request relates to Department for Transport requiring the HES data within the data sharing agreement to investigate road safety issues during the coronavirus disease (COVID-19) time period. There is a very high public interest during lockdown on the effects on road safety and the data request from NHS Digital can be used to form the analysis. The analysis will show the number of patients admitted to hospital following a road accident, in the first few months of the year of 2020. This will support policy colleagues in understanding road safety during the COVID-19 period so far and in preparing for users to return to roads.
As such, the DfT have added the following objective for processing:
1. The number of admissions to hospital following a road traffic accident during the COVID-19 lockdown period
and have added the following reason to those listed 1-6 as to why HES data is needed to be retained for all historical years (1999-2020)
7. To enable evidence on the number of admissions to hospital following a road accident during the COVID-19 lockdown period.
Historical HES data is likely to be a key evidence source for similar work in future Stats19 reviews, which are regularly carried out (currently every 5 years). Any future requests for HES would require an amendment or renewal to the data sharing agreement. Admissions to hospital following a road traffic accident evidence will be used for understanding road safety during the COVID-19 lockdown period which will be useful to inform policy and in preparing users to return to roads.
Processing activities
All individuals who have access to the record level data supplied by
[6 words unchanged]
need to know basis and are substantive employees of the Department of
Transport. The data disseminated in this agreement will be used only
Transport . All employees a responsible
for
information training and
the
purpose outlined in
information asset owners complete
the
objectives for processing.
information asset owner training.
NHS Digital will match STATS19 data (police recorded road accident data collected by DFT) to Hospital Episode Statistics data. DfT will supply for linkage postcode, age, sex, date of accident. NHS Digital will also supply non-matched episodes with matching road accident codes. DfT will receive hospital data that can be compared with police reporting and all accidents not reported to police. All data will be returned to DfT in a pseudonymised format. Re-identification is not permitted.
The data disseminated in this agreement will be used only for the purpose outlined in the objectives for processing.
DfT will hold the pseudonymised data in a secure database with restricted access to specific individuals. The pseudonymised data will be stored separately from the STATS 19 data. Most of the analysis will on the HES data in isolation. However, some analysis will take place linking the HES data to the non-identifiable STATS 19 data consisting of data such as road type, speed limit, conditions etc. The HES data will never be linked back to the postcode and age data held within the STATS19 database. Data storage and access will only be in DfT’s secure network.
NHS Digital will match STATS19 data (police recorded road accident data collected by DFT) to Hospital Episode Statistics admitted patient care and accident & emergency data. DfT will supply for linkage postcode, age, sex, date of accident. NHS Digital will also supply non-matched episodes with matching road accident codes. DfT will receive hospital data that can be compared with police reporting and all accidents not reported to police. All data will be returned to DfT in a pseudonymised format. Re-identification is not permitted and will not be attempted by DfT.
DfT will convert the ICD-10 codes (for causation of the injury, type of injury and severity of injury) into AIS scores. DfT will use the information to produce analysis of the number of people with a MAIS (Maximum Abbreviated Injury Score) of 3 or greater, plus analysis of the types of injuries different road user groups suffer from.
DfT will hold the pseudonymised data in a secure database with restricted access to specific individuals. The pseudonymised data will be stored separately from the STATS 19 data. Most of the analysis will be on the HES data in isolation. However, some analysis will take place linking the HES data to the non-identifiable STATS 19 data consisting of data such as road type, speed limit, conditions etc. The HES data will never be linked back to the postcode and age data held within the STATS19 database. Data storage and access will only be in DfT’s secure network.
Annual police recorded road traffic accident casualties matched with key HES variables (e.g. ICD-10 codes). To be completed for 2011, 2012, 2013, 2014, 2015 and 2016 data.
DfT will convert the ICD-10 codes (for causation of the injury, type of injury and severity of injury) into abbreviated injury scale (AIS ) scores. DfT will use the information to produce analysis of the number of people with a MAIS (Maximum Abbreviated Injury Score) of 3 or greater, plus analysis of the types of injuries different road user groups suffer from.
Annual police recorded road traffic accident casualties matched with key HES variables (e.g. ICD-10 codes). This has been completed for 2011, 2012, 2013, 2014, 2015 and 2016 data. To be completed for 2017, 2018, 2019 and 2020 data.
[4 paragraphs unchanged]
This agreement also requests the
The
retention of HES data
previously supplied
has been granted. Reasons
for the
following reason.
retention included the following.
A lookup file is being developed to determine the Abbreviated Injury Score
[58 words unchanged]
This will represent a loss to the Department’s road safety evidence base.
[2 paragraphs unchanged]
Expected output
[3 paragraphs unchanged]
(3 Serious, 4 Severe, 5 Critical, 6 Unsurvivable taken from
www.aam.org/abbreviated-injury-scale-ais/)
https://www.aaam.org/abbreviated-injury-scale-ais/)
[4 paragraphs unchanged]
The most recent report is available on
the
https://www.gov.uk/government/collections/road-accidents-and-safety-statisticsthe
gov.uk website. Previous Annual Reports are also available on The National Archives
website.
website https://webarchive.nationalarchives.gov.uk/20190801212554/https://www.gov.uk/government/collections/road-accidents-and-safety-statistics.
Outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis
guide.
guidance.
Expected measurable benefits
[2 paragraphs unchanged]
The use of hospital data will achieve
two things:
the following things :
1. To identify casualties who are truly seriously injured, and in particular, the separation of people with less serious injuries from casualties with
live-changing
life-changing
injuries.
[3 paragraphs unchanged]
Improving the health of Londoners: transport action
plan ,
plan,
published by Transport for London in 2014, is an example of how road safety is integrated in wider the public health context at a local level.
[10 paragraphs unchanged]
Any country (including
GB/UK)
Great Britain/ United Kingdom)
wishing to produce comparable statistics using this standard needs to use clinical data containing ICD codes akin to HES data.
[1 paragraph unchanged]
AMENDMENT REQUEST:
The Department for Transport require the data within the data sharing agreement to investigate road safety issues during the COVID time period. There is significant ministerial and public interest in how road usage and road safety has changed during this time period. There have been reports in the media of increases in speeding and other risky behaviours. Several European countries have reported road accidents volumes are not decreasing at the same rate as traffic volumes and a higher proportion of the accidents are severe or fatal.
The data requested from NHS Digital can be used to produce analysis that shows the number of patients admitted to hospital following road accidents and the severity of the accidents, in the first few months of the year on 2020. Understanding the volume and severity of road accidents occurring during this time period will support policy colleagues working on recovery and restart planning. It will inform the development of interventions, communication and education activities preparing for users to return to the roads. For example, if there is an increase in the severity of accidents occurring during this period interventions may be targeted towards reducing the severity of accidents. If there is an increase in volume but not severity, reducing the likelihood of accidents occurring may be prioritised. The data requested and the subsequent analysis will enrich the broader evidence base supporting this work.
There is an added benefit by amending the purpose to incorporate traffic accidents during the covid-19 lockdown period, building on the points 1 and 2 and above:
3. The data and subsequent analysis on road accidents during the covid-19 period will develop understanding of road safety during this time and the severity of accidents occurring. It will support restart and recovery planning and the development of interventions and communications as users return to the roads.
Benefits reported
HES data has been used in the Reported Road Casualties Great Britain
2016
2016, 2017
and most recently the
2017
2018
annual report (published September
2018).
2019).
Analysis has helped assess the trend in serious injuries as a result of road accidents and to understand the context around the severity reporting in STATS19.
HES analysis has been used for internal purposes, for example to contribute to the STATS19 review (the periodical review of accident data collection from the police) which includes a strand of work on comparing injuries in HES and STATS19 data to evaluate severity reporting approaches. This has been useful in identifying the accuracy of reporting on serious injuries within STATS19 data and changes over time. It has been used to assess the accuracy of the severity reporting of road accidents compared with the injury based reporting of road accidents by police forces who use CRASH/COPA. This has been used to produce recommendations to encourage more uptake of the injury-based reporting. This analysis has also been used to further improve and develop definitions of the injury descriptions in STATS19 to allow better recording of injuries by police forces.
Objective for processing
The Department for Transport (DfT) has been collecting and publishing statistics on road accidents and casualties since 1927. The majority of this data used comes from police reports of the accidents. These outputs are National Statistics, and they are critical bits of evidence which are used to formulate road safety policies at both a national and local level.
The police reported data are also extensively used by police forces, researchers and commercial organisations to make the roads safer.
The police recorded data contains considerable uncertainty in the classification of injuries as being ‘slight’ or ‘serious’. The Hospital Episode Statistics (HES) linkage project started during the mid- to late- 2000s to define injuries using standardised clinical data, rather than guesswork at the side of the road. The first round of matching linked police data to HES data (admitted patient care and accident & accident and emergency records) for each year between 1999 and 2010. This linkage work was carried out by NHS Digital and its predecessor organisations resulting in a report in 2012 highlighting the differences in reported and HES outcomes.
The objectives for the processing are:
1. To understand the types of injuries sustained by people injured in road traffic accidents.
All statistics released by DfT will be aggregated and non-identifiable. Outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guide.
The legal basis for processing falls under GDPR Article 6(1)( e) and GDPR Article 9(2)(j). Processing and dissemination is for research purposes as analysing HES data is necessary for scientific research to understand the type of injuries sustained by people injured in road traffic accidents, and statistical purposes such as determining the number of clinically seriously injured road casualties. The output statistics are released as part of the Reported Road Casualties Great Britain publication each year. Furthermore, it also meets the public interest criteria as defined by the Information Commissioner’s Office (ICO) for good decision-making by public bodies and securing the best use of public resources. The processing of data for this study is a task of public interest since the data will be used in statistics which are used to formulate road safety policies. There are no alternative, less intrusive methods of achieving this purpose.
Stats19 data is collected under section 39 (3) (a) of the Road Traffic Act 1998 by the Department of for Transport to carry out studies into accidents arising out of the use of vehicles. The purpose is to take measures to prevent such accidents. They use the data to formulate road safety policies, monitor trends for road accidents, and give advice relating to the use of roads.
The Department for Transport (DfT) (Data controller who processes the data) requests the retention of HES data for all historical years (1999 to 2020). Several reasons for this request are outlined below:
1) The historical pseudonymised HES data enables DfT to determine the longer time series in the number of admitted patient care admissions where the admission is recorded as being related to a road traffic accident. It is used as an external data source to compare the Stats19 data to, building the understanding and informing the commentary of underlying trends in road safety. For example, the presentation of the latest HES data and its comparison with Stats19 can be found in the latest Reported Road Casualty Great Britain annual report for 2018.
2) The historical data allows analysis to be completed on new breakdowns of the statistics (e.g. on different road user groups, different types of accidents and roads conditions) on a long time period. Road safety is a slowly evolving field, and any change in behaviours needs to be monitored on a long time period. What breakdowns are required from the time series might change depending on the focus of policy interventions. A new field of interest might be, for example, the type of injuries sustained by car casualties who were not wearing a seatbelt.
3) The historical HES data enables the DfT to investigate any data concerns that can only be detected through analysis of trends over a long time period. For example, recent analysis of the unmatched HES data has revealed that the matching rates with the International Classification of Diseases (ICD) look-up for injury has worsened over time. Analysis of all years available will enable DfT to understand whether this is due to changes in recording practices for specific injuries.
4) The historical HES data enables the DfT to ground any changes in methodology for future HES/ police recorded reported road casualties (Stats19) data matching exercises in evidence and understand their potential impact. For example, in the last data linkage exercise, matching criteria included a restriction of the geographical scope to only look for potential matches in hospitals located in the same or in adjacent Strategic Health Authorities (SHAs) to where accidents took place. Since SHAs boundaries no longer exist, the current data linkage exercise will use a list of the closest hospitals to the accident location, in distance calculated from postcodes/coordinates. For the new methodology, DfT need to precisely define how many hospitals should be in scope for the matching (e.g. the 10 or 15 closest) and understand how this might affect the matching rate. DfT will use the historical HES data matched to Stats19 (statistical dataset collected by the Department for Transport from the police on personal-injury road traffic accidents) and calculate the rank in distance of the hospital it was actually matched to with the old methodology. This will give DfT an idea of the range of hospitals ranks and distance DfT should consider in scope for the new methodology. Keeping the historical HES data will also enable DfT to compare matching rates, for different matching criteria and groups, between the old and the new methodology. Without this analysis of historical HES data, DfT would be at risk of basing the new methodology on an arbitrary threshold, which could result in a poor matching rate and/or indications of a large volume of false positive matches (e.g. more use of matching criteria at the less exact end of the scale compared to previous exercises) – making comparison over time unreliable.
5) The historical HES data also gives DfT an external source to evaluate the quality of the Stats19 data against, and this will be of prime importance in the current Stats19 review. The Project Initiation Document from the Stats19 review (the periodical review of accident data collection) includes a strand of work on comparing injuries in HES and Stats19 data. The ability to use the historical HES data to assess the accuracy of reporting on serious injuries within STATS19 data and changes over time is especially important as there are now new reporting systems (Collision Recording and Sharing CRASH/ Case Overview Preparation Application COPA) used by around half of police forces in England whereby severity is determined based on the type of injury from HES rather than based on the police officer’s judgement of seriousness. This has previously caused a major discontinuity in the headline statistics.
The analysis of the matched dataset will help DfT determine whether the current list of injuries used in new injury-based reporting systems (CRASH/COPA) is still appropriate, or if it needs to be changed or updated in the future by looking at the injury types and how these have been captured as being serious within STATS19 over time. The outcome of this work will be a recommendation and a decision on what a unified system for recording injury severity should be for all police forces.
Historical HES data is likely to be a key evidence source for similar work in future Stats19 reviews, which are regularly carried out (currently every 5 years). Any future requests for HES would require an amendment or renewal to the data sharing agreement.
As a result, DfT applied to keep the historical HES data to preserve the evidence base and this resulted in the removal of the special condition to destroy any historical HES data.
Road safety is a public health problem. Road collisions cost the country an estimated £35bn 36bn a year.
With over 700560 thousand people estimated to being injured on the roads in Britain a year (source National Travel Survey) , 20 thousand of whom being ‘seriously’ injured (by the definition of the police, source STATS19) , and almost 5 thousand having potential ‘life-changing’ injuries (source HES with Maximum Abbreviated Injury Score of 3 or greater (MAIS3+)), there is a clear public health need for the problem to be addressed both in terms and national strategic priorities and through local operational approaches.
DfT has a role in setting the strategic direction of road safety in Great Britain and is the lead agency in producing the evidence-based used by policy-makers, operational delivery bodies, research bodies and vehicle manufacturers. The "Working together to build a safer road system: British road safety statement " was published in December 2015, this outlines the government’s approach to improving road safety. The document specifically identifies the value of HES data in the road safety context and the government, as a whole, has committed to maximising the use of HES data to improve the evidence used by practitioners.
The Department for Transport (DfT) (Data controller who processes the data) requests the retention of HES data for all historical years (1999 to 2020) to:
1) Enable DfT to determine the longer time series in the number of hospital admissions, where the admission is recorded as being related to a road traffic accident
2) Allow analysis to be completed on new breakdowns of the statistics over a long time period
3) Enable DfT to investigate any data concerns that can only be detected through analysis of trends over a long period of time
4) Enable DfT to ground any changes in methodology for further HES/STATS19 data matching exercises in evidence and understand their potential impact
5) Provide DfT with an external source to evaluate the quality of the STATS19 data
6) Used as a key evidence source for future STATS19 reviews
AMENDMENT REQUEST
The amendment request relates to Department for Transport requiring the HES data within the data sharing agreement to investigate road safety issues during the coronavirus disease (COVID-19) time period. There is a very high public interest during lockdown on the effects on road safety and the data request from NHS Digital can be used to form the analysis. The analysis will show the number of patients admitted to hospital following a road accident, in the first few months of the year of 2020. This will support policy colleagues in understanding road safety during the COVID-19 period so far and in preparing for users to return to roads.
As such, the DfT have added the following objective for processing:
1. The number of admissions to hospital following a road traffic accident during the COVID-19 lockdown period
and have added the following reason to those listed 1-6 as to why HES data is needed to be retained for all historical years (1999-2020)
7. To enable evidence on the number of admissions to hospital following a road accident during the COVID-19 lockdown period.
Historical HES data is likely to be a key evidence source for similar work in future Stats19 reviews, which are regularly carried out (currently every 5 years). Any future requests for HES would require an amendment or renewal to the data sharing agreement. Admissions to hospital following a road traffic accident evidence will be used for understanding road safety during the COVID-19 lockdown period which will be useful to inform policy and in preparing users to return to roads.
Expected output
The expected outputs are:
1. To estimate the number of people who have been injured in road traffic accidents and who were treated at hospital
2. To estimate the number of people who have been seriously injured in road traffic accidents, defined as having a Maximum Abbreviated Injury Score of 3 or higher (MAIS3+)
(3 Serious, 4 Severe, 5 Critical, 6 Unsurvivable taken from https://www.aaam.org/abbreviated-injury-scale-ais/)
The output statistics will be released as part of the Reported Road Casualties Great Britain publication in September each year.
The Reported Road Casualties Great Britain statistical series is the main avenue to get analysis of Hospital Episodes Statistics related to traffic accidents published. HES data might be published in several editions of this series (either on its own publication, or as part of a larger one like the annual report).
HES analysis is also used for internal purposes, for example to contribute to the STATS19 review (the periodical review of accident data collection from the police) which includes a strand of work on comparing injuries in HES and STATS19 data to evaluate severity reporting approaches.
All the statistical reports and tables are available to the public free of charge.
The most recent report is available on https://www.gov.uk/government/collections/road-accidents-and-safety-statisticsthe gov.uk website. Previous Annual Reports are also available on The National Archives website https://webarchive.nationalarchives.gov.uk/20190801212554/https://www.gov.uk/government/collections/road-accidents-and-safety-statistics.
Outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guidance.
Benefits reported
HES data has been used in the Reported Road Casualties Great Britain 2016, 2017 and most recently the 2018 annual report (published September 2019). Analysis has helped assess the trend in serious injuries as a result of road accidents and to understand the context around the severity reporting in STATS19.
HES analysis has been used for internal purposes, for example to contribute to the STATS19 review (the periodical review of accident data collection from the police) which includes a strand of work on comparing injuries in HES and STATS19 data to evaluate severity reporting approaches. This has been useful in identifying the accuracy of reporting on serious injuries within STATS19 data and changes over time. It has been used to assess the accuracy of the severity reporting of road accidents compared with the injury based reporting of road accidents by police forces who use CRASH/COPA. This has been used to produce recommendations to encourage more uptake of the injury-based reporting. This analysis has also been used to further improve and develop definitions of the injury descriptions in STATS19 to allow better recording of injuries by police forces.
DARS-NIC-381383-Z9F2P-v2.7 28 May 2019 to 30 November 2020
- Title
- HES and STATS19 One-to-one linkage project
- Commercial
- No
- Sublicensing
- No
- Datasets
- 2
- Files released
- 0
Datasets: Hospital Episode Statistics Accident and Emergency (HES A and E); Hospital Episode Statistics Admitted Patient Care (HES APC)
What changed from DARS-NIC-381383-Z9F2P-v1.3
Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.
| Field | Was | Became |
|---|---|---|
| Start date | 2019-05-28 | |
| Hospital Episode Statistics Accident and Emergency (HES A and E): common law duty of confidentiality | Section 251 NHS Act 2006 | |
| Hospital Episode Statistics Admitted Patient Care (HES APC): common law duty of confidentiality | Section 251 NHS Act 2006 |
Objective for processing
[2 paragraphs unchanged]
The police recorded data contains considerable uncertainty in the classification of injuries
[65 words unchanged]
a report in 2012 highlighting the differences in reported and HES outcomes.
This request is for the same data from 2011 to 2016.
[1 paragraph unchanged]
1. To
estimate
understand
the
number
types
of
injuries sustained by
people
who have been
injured in road traffic
accidents and who were treated at hospital.
accidents.
2. To understand the types of injuries sustained by people injured in road traffic accidents.
3. To match the hospital records of people injured in road traffic accidents with the police record of the accident / casualty. The matching is carried out by NHS Digital and pseudonymised data returned to Department for Transport (DfT)
4. To estimate the number of people who have been seriously injured in road traffic accidents, defined as having a Maximum Abbreviated Injury Score of 3 or higher (MAIS3+)
(3 Serious, 4 Severe, 5 Critical, 6 Unsurvivable )
(https://www.aaam.org/abbreviated-injury-scale-ais/)
[1 paragraph unchanged]
The legal basis for processing falls under GDPR Article 6(1)( e) and GDPR Article 9(2)(j). Processing and dissemination is for research purposes, and also meets the public interest criteria as defined by the ICO for good decision-making by public bodies and securing the best use of public resources. The processing of data for this study is a task of public interest since the data will be used in statistics which are used to formulate road safety policies. There are no alternative, less intrusive methods of achieving this purpose.
Stats19 data is collected under section 39 (3) (a) of the Road Traffic Act 1998 by the Department of Transport to carry out studies into accidents arising out of the use of vehicles. The purpose is to take measures to prevent such accidents. They use the data to formulate road safety policies, monitor trends for road accidents, and give advice relating to the use of roads.
The Department for Transport (DfT) (Data controller who processes the data) requests the retention of HES data for all historical years (1999 to 2016). Several reasons for this request are outlined below:
1) The historical pseudonymised HES data enables DfT to determine the longer time series in the number of admitted patient care admissions where the admission is recorded as being related to a road traffic accident. It is used as an external data source to compare the Stats19 data to, building the understanding and informing the commentary of underlying trends in road safety. For example, the presentation of the latest HES data and its comparison with Stats19 can be found in the latest Reported Road Casualty Great Britain annual report for 2017.
2) The historical data allows analysis to be completed on new breakdowns of the statistics (e.g. on different road user groups, different types of accidents and roads conditions) on a long time period. Road safety is a slowly evolving field, and any change in behaviours needs to be monitored on a long time period. What breakdowns are required from the time series might change depending on the focus of policy interventions. A new field of interest might be, for example, the type of injuries sustained by car casualties who were not wearing a seatbelt.
3) The historical HES data enables the DfT to investigate any data concerns that can only be detected through analysis of trends over a long time period. For example, recent analysis of the unmatched HES data has revealed that the matching rates with the ICD look-up for injury has worsened over time. Analysis of all years available will enable DfT to understand whether this is due to changes in recording practices for specific injuries.
4) The historical HES data enables the DfT to ground any changes in methodology for future HES/Stats19 data matching exercises in evidence and understand their potential impact. For example, in the last data linkage exercise, matching criteria included a restriction of the geographical scope to only look for potential matches in hospitals located in the same or in adjacent Strategic Health Authorities to where accidents took place. Since SHAs boundaries no longer exist, the current data linkage exercise will use a list of the closest hospitals to the accident location, in distance calculated from postcodes/coordinates. For the new methodology, DfT need to precisely define how many hospitals should be in scope for the matching (e.g. the 10 or 15 closest) and understand how this might affect the matching rate. DfT will use the historical HES data matched to Stats19 and calculate the rank in distance of the hospital it was actually matched to with the old methodology. This will give DfT an idea of the range of hospitals ranks and distance DfT should consider in scope for the new methodology. Keeping the historical HES data will also enable DfT to compare matching rates, for different matching criteria and groups, between the old and the new methodology. Without this analysis of historical HES data, DfT would be at risk of basing the new methodology on an arbitrary threshold, which could result in a poor matching rate and/or indications of a large volume of false positive matches (e.g. more use of matching criteria at the less exact end of the scale compared to previous exercises) – making comparison over time unreliable.
5) The historical HES data also gives DfT an external source to evaluate the quality of the Stats19 data against, and this will be of prime importance in the current Stats19 review. The Project Initiation Document from the Stats19 review (the periodical review of accident data collection) includes a strand of work on comparing injuries in HES and Stats19 data. The ability to use the historical HES data to assess the accuracy of reporting on serious injuries within STATS19 data and changes over time is especially important as there are now new reporting systems (CRASH/COPA) used by around half of police forces in England whereby severity is determined based on the type of injury from HES rather than based on the police officer’s judgement of seriousness. This has previously caused a major discontinuity in the headline statistics.
The analysis of the matched dataset will help DfT determine whether the current list of injuries used in new injury-based reporting systems (CRASH/COPA) is still appropriate, or if it needs to be changed or updated in the future by looking at the injury types and how these have been captured as being serious within STATS19 over time. The outcome of this work will be a recommendation and a decision on what a unified system for recording injury severity should be for all police forces.
6) Historical HES data is likely to be a key evidence source for similar work in future Stats19 reviews, which are regularly carried out (currently every 5 years). Any future requests for HES would require an amendment or renewal to the data sharing agreement.
As a result, DfT are applying to keep the historical HES data to preserve the evidence base and this results in the removal of the special condition to destroy any historical HES data.
Road safety is a public health problem. Road collisions cost the country an estimated £35bn a year.
With over 700 thousand people estimated to being injured on the roads in Britain a year , 20 thousand of whom being ‘seriously’ injured (by the definition of the police) , and almost 5 thousand having potential ‘life-changing’ injuries , there is a clear public health need for the problem to be addressed both in terms and national strategic priorities and through local operational approaches.
DfT has a role in setting the strategic direction of road safety in Great Britain and is the lead agency in producing the evidence-based used by policy-makers, operational delivery bodies, research bodies and vehicle manufacturers. The "Working together to build a safer road system: British road safety statement " was published in December 2015, this outlines the government’s approach to improving road safety. The document specifically identifies the value of HES data in the road safety context and the government, as a whole, has committed to maximising the use of HES data to improve the evidence used by practitioners.
The Department for Transport (DfT) (Data controller who processes the data) requests the retention of HES data for all historical years (1999 to 2016) to:
1) Enable DfT to determine the longer time series in the number of hospital admissions, where the admission is recorded as being related to a road traffic accident
2) Allow analysis to be completed on new breakdowns of the statistics over a long time period
3) Enable DfT to investigate any data concerns that can only be detected through analysis of trends over a long period of time
4) Enable DfT to ground any changes in methodology for further HES/STATS19 data matching exercises in evidence and understand their potential impact
5) Provide DfT with an external source to evaluate the quality of the STATS19 data
6) Used as a key evidence source for future STATS19 reviews
Processing activities
All individuals who have access to the record level data supplied by
[7 words unchanged]
to know basis and are substantive employees of the Department of Transport.
They
The data disseminated in this agreement
will
use the data
be used
only for the
purposes described
purpose outlined
in
this document
the objectives for processing.
NHS Digital will match STATS19 data (police recorded road accident
data)
data collected by DFT)
to Hospital Episode Statistics data. DfT will supply for linkage postcode, age, sex, date of accident. NHS Digital will also supply non-matched episodes with matching road accident codes.
This means that
DfT will receive hospital data that can be compared with police reporting
[6 words unchanged]
police. All data will be returned to DfT in a pseudonymised format.
Re-identification is not permitted.
[1 paragraph unchanged]
DfT will convert the ICD-10 codes (for causation of the injury, type
[13 words unchanged]
information to produce analysis of the number of people with a MAIS
score
(Maximum Abbreviated Injury Score)
of 3 or greater, plus analysis of the types of injuries different road user groups suffer from.
Annual police recorded road traffic accident casualties matched with key HES variables (e.g. ICD-10 codes). To be completed for 2011, 2012, 2013, 2014, 2015 and 2016
data, and repeated annually.
data.
[3 paragraphs unchanged]
The most recent report is available
at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/9279/rrcgb2011-06.pdf.
on the gov.uk website. Previous Annual Reports are also available on The National Archives website.
Previous Annual Reports are also available on The National Archives website.
This agreement also requests the retention of HES data previously supplied for the following reason. A lookup file is being developed to determine the Abbreviated Injury Score (AIS) for ICD-10 codes relating to consequences of external causes which enables MAIS3+ estimates to be generated. The lookup file is still under development and will be updated in the future. Losing the data back to 1999 will mean that the Department cannot update its estimates of MAIS3+ casualties back to 1999 when the lookup file is updated. This will represent a loss to the Department’s road safety evidence base.
This application also requests the retention of HES data previously supplied for the following reason. A lookup file is being developed to determine the Abbreviated Injury Score (AIS) for ICD-10 codes relating to consequences of external causes which enables MAIS3+ estimates to be generated. The lookup file is still under development and will be updated in the future. Losing the data back to 1999 will mean that the Department cannot update its estimates of MAIS3+ casualties back to 1999 when the lookup file is updated. This will represent a loss to the Department’s road safety evidence base. Further information can be found in the report published at: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/556648/rrcgb2015-03.pdf
[2 paragraphs unchanged]
Expected output
The expected outputs are:
1. To estimate the number of people who have been injured in road traffic accidents and who were treated at hospital
2. To estimate the number of people who have been seriously injured in road traffic accidents, defined as having a Maximum Abbreviated Injury Score of 3 or higher (MAIS3+)
(3 Serious, 4 Severe, 5 Critical, 6 Unsurvivable taken from www.aam.org/abbreviated-injury-scale-ais/)
[1 paragraph unchanged]
The Reported Road Casualties Great Britain statistical series is the main avenue to get analysis of Hospital Episodes Statistics related to traffic accidents published. HES data might be published in several editions of this series (either on its own publication, or as part of a larger one like the annual report).
HES analysis is also used for internal purposes, for example to contribute to the STATS19 review (the periodical review of accident data collection from the police) which includes a strand of work on comparing injuries in HES and STATS19 data to evaluate severity reporting approaches.
All the statistical reports and tables are available to the public free of charge.
The most recent report is available on the gov.uk website. Previous Annual Reports are also available on The National Archives website.
[1 paragraph unchanged]
All the statistical reports and tables are available to everyone free of charge.
The most recent report is available at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/9279/rrcgb2011-06.pdf.
Previous Annual Reports are also available on The National Archives website.
Expected measurable benefits
[2 paragraphs unchanged]
The use of hospital data will achieve
three
two
things:
1.
The identification of
To identify
casualties who are truly seriously injured, and in particular, the separation of people with less serious injuries from casualties with live-changing injuries.
[3 paragraphs unchanged]
Further details and examples:
Improving the health of Londoners: transport action plan , published by Transport for London in 2014, is an example of how road safety is integrated in wider the public health context at a local level.
Road safety-related benefits of using the HES data.
Road safety is a public health problem. Road collisions cost the country an estimated £35bn a year . With over 700 thousand people estimated to being injured on the roads in Britain a year , 20 thousand of whom being ‘seriously’ injured (by the definition of the police) , and almost 5 thousand having potential ‘life-changing’ injuries , there is a clear public health need for the problem to be addressed both in terms and national strategic priorities and through local operational approaches.
DfT has a role in setting the strategic direction of road safety in Great Britain and is the lead agency in producing the evidence-based used by policy-makers, operational delivery bodies, research bodies and vehicle manufacturers. "Working together to build a safer road system: British road safety statement " (,https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/487704/british_road_safety_statement_print.pdf) published in December 2015, outlines the government’s approach to improving road safety. The document specifically identifies the value of HES data in the road safety context and the government, as a whole, has committed to maximising the use of HES data to improve the evidence used by practitioners.
Improving the health of Londoners: transport action plan , (http://content.tfl.gov.uk/improving-the-health-of-londoners-transport-action-plan.pdf ) published by Transport for London in 2014, is an example of how road safety is integrated in wider the public health context at a local level.
[3 paragraphs unchanged]
3. Types of injuries suffered. -The type of injury (e.g. lower limb
[9 words unchanged]
of accidents influence what mitigation and development needs to be carried out.
With more accurate statistics on casualties this will, in turn, help road safety organisations identify patterns of interest and put in place interventions to reduce accidents. Fewer accidents would result in wider benefits to the health system e.g. number of visits to A&E and hospital admissions.
[3 paragraphs unchanged]
The outputs of this work would partly be of direct benefit to DfT
and also to health
(e.g. by influencing road safety policy, vehicle and road design
standards),
standards, which would hopefully lead to fewer accidents),
but much of it will be reused by external bodies such as vehicle manufacturers and researchers which all cumulatively benefits the road users.
4. Reliability of police-reported data. - DfT has known for some time
[35 words unchanged]
to report an accident. This in itself is a significant public health
problem:
problem because
if underreporting is increasing than the dangers and risks on the road
[24 words unchanged]
thinking that the problems from road collisions are decreasing and going away.
[1 paragraph unchanged]
5. Benchmarking national performance and learning lessons from elsewhere. Road safety has long been a topic for considerable international cooperation and learning. Although Britain is regarded as one of the leading countries in some aspects of road safety, there are a number of areas which we have worse performance than elsewhere (see, for instance, the PACTS / TRL report Understanding the strengths and weaknesses of Britain’s road safety performance ).
Historically international benchmarking has only been possible using information on fatalities. This is because police-reported severity definitions are both unreliable and vary across the world. One of the work packages of the SafetyCube project specifically address standardisation of clinically defined serious injuries. Although not yet published, the standard that the project has adopted is based on the Abbreviated Injury Scale (AIS – as created by the AAAM ) and identifies all casualties with an injury scoring 3 or more on the scale (referred to as MAIS3+) as being ‘seriously’ injured. This definition has also been adopted by the European Commission and is, therefore, regarded as the de facto standard throughout Europe.
[2 paragraphs unchanged]
Benefits reported
None, we are currently working on the data matching exercise.
HES data has been used in the Reported Road Casualties Great Britain 2016 and most recently the 2017 annual report (published September 2018). Analysis has helped assess the trend in serious injuries as a result of road accidents and to understand the context around the severity reporting in STATS19.
Objective for processing
The Department for Transport (DfT) has been collecting and publishing statistics on road accidents and casualties since 1927. The majority of the data used comes from police reports of the accidents. These outputs are National Statistics, and they are critical bits of evidence which are used to formulate road safety policies at both a national and local level.
The police reported data are also extensively used by police forces, researchers and commercial organisations to make the roads safer.
The police recorded data contains considerable uncertainty in the classification of injuries as being ‘slight’ or ‘serious’. The HES linkage project started during the mid- to late- 2000s to define injuries using standardised clinical data, rather than guesswork at the side of the road. The first round of matching linked police data to HES data for each year between 1999 and 2010. This linkage work was carried out by NHS Digital and its predecessor organisations resulting in a report in 2012 highlighting the differences in reported and HES outcomes.
The objectives for the processing are:
1. To understand the types of injuries sustained by people injured in road traffic accidents.
All statistics released by DfT will be aggregated and non-identifiable. Outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guide.
The legal basis for processing falls under GDPR Article 6(1)( e) and GDPR Article 9(2)(j). Processing and dissemination is for research purposes, and also meets the public interest criteria as defined by the ICO for good decision-making by public bodies and securing the best use of public resources. The processing of data for this study is a task of public interest since the data will be used in statistics which are used to formulate road safety policies. There are no alternative, less intrusive methods of achieving this purpose.
Stats19 data is collected under section 39 (3) (a) of the Road Traffic Act 1998 by the Department of Transport to carry out studies into accidents arising out of the use of vehicles. The purpose is to take measures to prevent such accidents. They use the data to formulate road safety policies, monitor trends for road accidents, and give advice relating to the use of roads.
The Department for Transport (DfT) (Data controller who processes the data) requests the retention of HES data for all historical years (1999 to 2016). Several reasons for this request are outlined below:
1) The historical pseudonymised HES data enables DfT to determine the longer time series in the number of admitted patient care admissions where the admission is recorded as being related to a road traffic accident. It is used as an external data source to compare the Stats19 data to, building the understanding and informing the commentary of underlying trends in road safety. For example, the presentation of the latest HES data and its comparison with Stats19 can be found in the latest Reported Road Casualty Great Britain annual report for 2017.
2) The historical data allows analysis to be completed on new breakdowns of the statistics (e.g. on different road user groups, different types of accidents and roads conditions) on a long time period. Road safety is a slowly evolving field, and any change in behaviours needs to be monitored on a long time period. What breakdowns are required from the time series might change depending on the focus of policy interventions. A new field of interest might be, for example, the type of injuries sustained by car casualties who were not wearing a seatbelt.
3) The historical HES data enables the DfT to investigate any data concerns that can only be detected through analysis of trends over a long time period. For example, recent analysis of the unmatched HES data has revealed that the matching rates with the ICD look-up for injury has worsened over time. Analysis of all years available will enable DfT to understand whether this is due to changes in recording practices for specific injuries.
4) The historical HES data enables the DfT to ground any changes in methodology for future HES/Stats19 data matching exercises in evidence and understand their potential impact. For example, in the last data linkage exercise, matching criteria included a restriction of the geographical scope to only look for potential matches in hospitals located in the same or in adjacent Strategic Health Authorities to where accidents took place. Since SHAs boundaries no longer exist, the current data linkage exercise will use a list of the closest hospitals to the accident location, in distance calculated from postcodes/coordinates. For the new methodology, DfT need to precisely define how many hospitals should be in scope for the matching (e.g. the 10 or 15 closest) and understand how this might affect the matching rate. DfT will use the historical HES data matched to Stats19 and calculate the rank in distance of the hospital it was actually matched to with the old methodology. This will give DfT an idea of the range of hospitals ranks and distance DfT should consider in scope for the new methodology. Keeping the historical HES data will also enable DfT to compare matching rates, for different matching criteria and groups, between the old and the new methodology. Without this analysis of historical HES data, DfT would be at risk of basing the new methodology on an arbitrary threshold, which could result in a poor matching rate and/or indications of a large volume of false positive matches (e.g. more use of matching criteria at the less exact end of the scale compared to previous exercises) – making comparison over time unreliable.
5) The historical HES data also gives DfT an external source to evaluate the quality of the Stats19 data against, and this will be of prime importance in the current Stats19 review. The Project Initiation Document from the Stats19 review (the periodical review of accident data collection) includes a strand of work on comparing injuries in HES and Stats19 data. The ability to use the historical HES data to assess the accuracy of reporting on serious injuries within STATS19 data and changes over time is especially important as there are now new reporting systems (CRASH/COPA) used by around half of police forces in England whereby severity is determined based on the type of injury from HES rather than based on the police officer’s judgement of seriousness. This has previously caused a major discontinuity in the headline statistics.
The analysis of the matched dataset will help DfT determine whether the current list of injuries used in new injury-based reporting systems (CRASH/COPA) is still appropriate, or if it needs to be changed or updated in the future by looking at the injury types and how these have been captured as being serious within STATS19 over time. The outcome of this work will be a recommendation and a decision on what a unified system for recording injury severity should be for all police forces.
6) Historical HES data is likely to be a key evidence source for similar work in future Stats19 reviews, which are regularly carried out (currently every 5 years). Any future requests for HES would require an amendment or renewal to the data sharing agreement.
As a result, DfT are applying to keep the historical HES data to preserve the evidence base and this results in the removal of the special condition to destroy any historical HES data.
Road safety is a public health problem. Road collisions cost the country an estimated £35bn a year.
With over 700 thousand people estimated to being injured on the roads in Britain a year , 20 thousand of whom being ‘seriously’ injured (by the definition of the police) , and almost 5 thousand having potential ‘life-changing’ injuries , there is a clear public health need for the problem to be addressed both in terms and national strategic priorities and through local operational approaches.
DfT has a role in setting the strategic direction of road safety in Great Britain and is the lead agency in producing the evidence-based used by policy-makers, operational delivery bodies, research bodies and vehicle manufacturers. The "Working together to build a safer road system: British road safety statement " was published in December 2015, this outlines the government’s approach to improving road safety. The document specifically identifies the value of HES data in the road safety context and the government, as a whole, has committed to maximising the use of HES data to improve the evidence used by practitioners.
The Department for Transport (DfT) (Data controller who processes the data) requests the retention of HES data for all historical years (1999 to 2016) to:
1) Enable DfT to determine the longer time series in the number of hospital admissions, where the admission is recorded as being related to a road traffic accident
2) Allow analysis to be completed on new breakdowns of the statistics over a long time period
3) Enable DfT to investigate any data concerns that can only be detected through analysis of trends over a long period of time
4) Enable DfT to ground any changes in methodology for further HES/STATS19 data matching exercises in evidence and understand their potential impact
5) Provide DfT with an external source to evaluate the quality of the STATS19 data
6) Used as a key evidence source for future STATS19 reviews
Expected output
The expected outputs are:
1. To estimate the number of people who have been injured in road traffic accidents and who were treated at hospital
2. To estimate the number of people who have been seriously injured in road traffic accidents, defined as having a Maximum Abbreviated Injury Score of 3 or higher (MAIS3+)
(3 Serious, 4 Severe, 5 Critical, 6 Unsurvivable taken from www.aam.org/abbreviated-injury-scale-ais/)
The output statistics will be released as part of the Reported Road Casualties Great Britain publication in September each year.
The Reported Road Casualties Great Britain statistical series is the main avenue to get analysis of Hospital Episodes Statistics related to traffic accidents published. HES data might be published in several editions of this series (either on its own publication, or as part of a larger one like the annual report).
HES analysis is also used for internal purposes, for example to contribute to the STATS19 review (the periodical review of accident data collection from the police) which includes a strand of work on comparing injuries in HES and STATS19 data to evaluate severity reporting approaches.
All the statistical reports and tables are available to the public free of charge.
The most recent report is available on the gov.uk website. Previous Annual Reports are also available on The National Archives website.
Outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guide.
Benefits reported
HES data has been used in the Reported Road Casualties Great Britain 2016 and most recently the 2017 annual report (published September 2018). Analysis has helped assess the trend in serious injuries as a result of road accidents and to understand the context around the severity reporting in STATS19.
DARS-NIC-381383-Z9F2P-v1.3 1 December 2017 to 30 November 2020
- Title
- HES and STATS19 One-to-one linkage project
- Commercial
- No
- Sublicensing
- No
- Datasets
- 2
- Files released
- 12
Datasets: Hospital Episode Statistics Accident and Emergency (HES A and E); Hospital Episode Statistics Admitted Patient Care (HES APC)
Objective for processing
The Department for Transport (DfT) has been collecting and publishing statistics on road accidents and casualties since 1927. The majority of the data used comes from police reports of the accidents. These outputs are National Statistics, and they are critical bits of evidence which are used to formulate road safety policies at both a national and local level.
The police reported data are also extensively used by police forces, researchers and commercial organisations to make the roads safer.
The police recorded data contains considerable uncertainty in the classification of injuries as being ‘slight’ or ‘serious’. The HES linkage project started during the mid- to late- 2000s to define injuries using standardised clinical data, rather than guesswork at the side of the road. The first round of matching linked police data to HES data for each year between 1999 and 2010. This linkage work was carried out by NHS Digital and its predecessor organisations resulting in a report in 2012 highlighting the differences in reported and HES outcomes. This request is for the same data from 2011 to 2016.
The objectives for the processing are:
1. To estimate the number of people who have been injured in road traffic accidents and who were treated at hospital.
2. To understand the types of injuries sustained by people injured in road traffic accidents.
3. To match the hospital records of people injured in road traffic accidents with the police record of the accident / casualty. The matching is carried out by NHS Digital and pseudonymised data returned to Department for Transport (DfT)
4. To estimate the number of people who have been seriously injured in road traffic accidents, defined as having a Maximum Abbreviated Injury Score of 3 or higher (MAIS3+)
(3 Serious, 4 Severe, 5 Critical, 6 Unsurvivable )
(https://www.aaam.org/abbreviated-injury-scale-ais/)
All statistics released by DfT will be aggregated and non-identifiable. Outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guide.
Expected output
The output statistics will be released as part of the Reported Road Casualties Great Britain publication in September each year.
Outputs will contain only aggregate level data with small numbers suppressed in line with HES analysis guide.
All the statistical reports and tables are available to everyone free of charge.
The most recent report is available at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/9279/rrcgb2011-06.pdf.
Previous Annual Reports are also available on The National Archives website.
Benefits reported
None, we are currently working on the data matching exercise.
Register history
When this agreement appeared in, or was edited in, each monthly edition of the register. Built by comparing every edition this site holds, the earliest of which is July 2021.
-
July 2021 —
already listed in the earliest edition this site holds, so it may be older. 3 versions: DARS-NIC-381383-Z9F2P-v1.3, DARS-NIC-381383-Z9F2P-v2.7, DARS-NIC-381383-Z9F2P-v3.6
-
March 2022
1 version added: DARS-NIC-381383-Z9F2P-v4.4
-
December 2022
Register-wide edit DARS-NIC-381383-Z9F2P-v1.3, DARS-NIC-381383-Z9F2P-v2.7, DARS-NIC-381383-Z9F2P-v3.6 — Datasets: legal basis: “
s261(1) and” taken out. Made to 639 agreements in this edition, so it is reported once, on the changes page, and not counted as an amendment of this agreement.
Cite this page
NHS England (2026) Data Uses Register, September 2026 edition, agreement DARS-NIC-381383-Z9F2P, “HES and STATS19 One-to-one linkage project”. Read via NHS Data Access Explorer (unofficial), https://healthdatauses.uk/agreements/dars-nic-381383-z9f2p/ (accessed [date]).
This address stays the same, but the page is rebuilt with each monthly edition, so the citation names the edition it shows. Every edition's data is kept in the facts store.
Source: datausesregister_september2026.xlsx, September 2026 edition of the NHS England Data Uses Register. Search that workbook for DARS-NIC-381383-Z9F2P to see the original rows.