Guidance

STATS19 collection and use: road safety factor recording

Updated 28 May 2026

1. About this research

This page summarises findings from research commissioned by the Department for Transport (DfT) on the introduction of road safety factors (RSFs) in STATS19, which was carried out by Agilysis working on behalf of Road Safety Great Britain.

The research examined how RSFs are recorded in practice and how the transition from contributory factors (CFs) to RSFs affects the interpretation of STATS19 data. Further details of the transition are available in the guidance on RSF and CF recording.

2. Key messages

This research suggests:

• the transition from contributory factors (CFs) to road safety factors (RSFs) has led to clear changes in how factors are distributed in STATS19 data
• these changes occur across police forces and reflect differences in classification and recording, as well as possible real-world trends
• speed-related factors decrease as a proportion, while non-motorised road user factors increase
• recording is influenced by system design, operational pressures and training
• care is needed when interpreting trends during the transition period, particularly when comparing across years

3. Background

STATS19 is the system used to record detailed information about personal injury road traffic collisions reported to the police. Contributory factors have historically been used to record factors which may have contributed to a collision.

Road safety factors (RSFs) were introduced from 2023 in some forces to replace CFs, implementing a recommendation from the most recent STATS19 review. Implementation has taken place at different times across police forces, and during 2024 some forces recorded both CFs and RSFs.

This means that data from recent years reflects a transition period where differences in recording practice may affect interpretation.

For information on how STATS19 data is collected and used, see the STATS19 collection and use guidance.

4. Approach

The research combined quantitative analysis of STATS19 data with qualitative insight from police forces that had transitioned to RSFs.

This included:

• analysis of national and police force-level data comparing 2022 CF data with 2024 RSF data, where available
• detailed examination of changes in factor distributions
• in-depth interviews with officers and analysts

Interviews were carried out with representatives from 5 police force areas, including front-line officers, supervisors and data analysts who worked with the data collected.

5. Patterns in road safety factor recording

5.1 Overall distribution

At national level, behaviour-related factors continue to account for the largest proportion of recorded factors.

However, comparisons between 2022 CF data and 2024 RSF data show changes in distribution between categories:

• the proportion of speed-related factors decreases
• the proportion of non-motorised road user factors increases
• smaller increases are observed in distraction and vehicle-related factors
• behaviour-related factors remain broadly stable overall

5.2 Changes within categories

Within categories, there are also notable changes in how individual factors are recorded.

For example, within speed-related factors:

• the proportion attributed to aggressive or reckless driving decreases
• the proportion attributed to exceeding the speed limit increases

Within non-motorised road user factors:

• increases are observed in factors relating to visibility and use of crossing facilities
• decreases are observed in some categories such as pedestrian carelessness and non-motorised vehicles entering the road from pavements

Within behaviour-related factors:

• the proportion attributed to ineffective observation decreases
• more specific driver behaviours, such as passing too close or manoeuvring errors, increase

5.3 Variation between police forces

The patterns outlined above were broadly similar across police forces, in that several forces showed the same patterns. For this project, the 10 forces with the highest number of RSFs recorded in 2024 were studied, and comparing the RSFs data for 2024 with the CFs data for 2022 for these forces showed that:

• speed-related factors decreased across all forces examined
• non-motorised road user factors increased across all forces, although to different extents
• changes within behaviour-related factors are consistent in direction but vary in magnitude

The means of recording of CFs or RSFs varies across forces, which may account for differences, but similar patterns were shown for forces recording the data in different ways (for example, using desktop-based recording or mobile devices).

6. Factors influencing RSF recording

The research identified a number of factors that influence how RSFs are recorded in practice. These findings are based on interviews with officers and analysts from 5 police force areas. With a relatively small number of interviews, definitive conclusions cannot be drawn, instead this works illustrates that it is likely that there are a number of factors influencing patterns.

6.1 Implementation of the transition

The transition from CFs to RSFs was often experienced as a system change rather than a conceptual change. In many cases, recording systems were updated to replace CFs with RSFs. However, in some forces, the change to RSFs coincided with a change or reporting method (for example, from paper forms to digital collection via mobile devices).

Some participants reported limited communication or formal training around the change, although this was not necessarily expected as the change was seen as minimal from the recording perspective: “it just happened… you tick the boxes on there instead”

DfT response: DfT will consider whether guidance to better explain the purpose of RSFs and support consistent understanding across forces would be worthwhile.

6.2 Operational context

Workload and operational pressures were identified as influencing recording behaviour. For example: “if they’re snowed under, they’ll use the quickest method”.

Selecting certain factors may also lead to additional investigative work. However, this is not something that has notably changed in the transition from CFs to RSFs.

DfT response: DfT will consider how guidance can emphasise the importance of accurate recording of STATS19 data alongside operational pressures.

6.3 System and interface design

System design influences factor selection. For example, participants noted: “you do miss things… if you haven’t got that far down the list” .

This suggests that ordering and visibility of factors can affect recording, and this varies both between CFs and RSFs (with a longer list for CFs) and in the different systems used (for example the factors are ordered differently on desktop and mobile devices).

6.4 Data structure and usability

The RSF framework contains fewer and broader categories than the previous CF system: “there were too many choices before” .

This may improve usability (for officers) but reduce analytical detail (for data analysts) illustrating a trade-off in the data collection.

6.5 Summary of findings

The qualitative analysis shows that the recording of the main RSF categories remain broadly consistent with the previous CF framework. However, there are clear changes in how factors are distributed both between and within categories.

Based on the qualitative work, these changes are likely to reflect a combination of:

• the operational context affecting how data is recorded in practice (such as police officer workload)
• the design and structure of the RSF framework (different grouping and ordering of factors)
• differences in classification of RSFs compared with CFs (for example, a shorter list of factors)

7. Implications for users of RSF data

The transition from CFs to RSFs represents a structural change in how collision factors are recorded.

• there is a discontinuity between CF and RSF data
• changes in distributions are systematic rather than random
• both framework design and implementation influence how data is recorded

Interpreting changes in road safety factors (RSFs)

When using RSF data, changes over time should be interpreted with care.

• the transition from CFs to RSFs has not occurred at the same time across all police forces
• some forces recorded both CFs and RSFs during 2024
• changes may reflect differences in classification and recording practice, as well as real-world trends
• system design and operational pressures can influence factor selection
• proportional comparisons are more reliable than absolute counts during the transition period

These factors should be considered when comparing data across years or between areas.

8. Overall conclusions

This research has highlighted how observed changes in RSF data may be influenced by a combination of:

• the structure of the RSF framework compared to that for CFs
• differences in implementation across forces
• practical factors affecting data entry, including officer recording behaviours

However, at this point it is not possible to draw definitive conclusions as to the relative importance of different factors in explaining the changes in the proportion of collisions with different factors recorded. Further analysis of the data will be required as more RSF data becomes available.

DfT response: DfT will continue to include a comparison of the differences between CFs and RSFs .

9. Contact

If you have feedback or questions about STATS19 data collection, please contact the Department for Transport road safety statistics team.

Road safety statistics

Email roadacc.stats@dft.gov.uk