We have recently completed a short review of the Department for Transport’s (DfT) Reported road casualties in Great Britain statistics.

This review was requested by DfT specifically to examine new adjusted estimates of injury severity which have been developed to provide comparative data over time and across police forces. Since 2018, DfT has published these adjusted severity data as ‘official statistics in development’ while it developed and evaluated its new methodology.

Our review considers this enhanced reporting to be of sufficient quality and value to be regarded as accredited official statistics alongside the other data in the statistical release.

New methodology and adjusted estimates

DfT worked in collaboration with ONS’s Methodology Advisory Service, to develop a methodology to quantify the effect of changing police force recording systems and to estimate the level of ‘slight’ and ‘serious’ injuries as if all police forces were using newly introduced and more-accurate injury-based recording systems. DfT’s approach to development and working with methods experts aligns with the standards of innovation and improvement of the Code and DfT was recognised for its work at the Government Statistical Service Awards in 2019.

DfT has provided users of the statistics with accessible and clear information relating to the severity adjustment. It offers users various supporting information to aid interpretation of the statistics and help their understanding of the new reporting method. The severity adjustment is prominently flagged to users in the main bulletin, with further information provided in an accessible guidance document – including when it considers adjusted and unadjusted figures are appropriate to use. More-detailed information is also available on the methodology, published in technical reports for interested users.

DfT plans to continue to publish both adjusted and unadjusted estimates at least until all police forces are using new injury-based recording systems. This will allow users to both select the appropriate estimate for their needs and understand the scale of the historical adjustment. DfT will continue to support users to use the most appropriate data for their needs.

Meeting users’ needs

As part of this review, we spoke to some users of the underlying vehicle collision data. Users told us of the positive engagement they have had with DfT, either informally or through more formal routes.

Users recognised the thorough approach DfT had taken in developing this new method and the importance of the system recording changes and the benefits that brings within the data and statistics.

Reported road casualties in Great Britain statistics

While this review was focussed on the injury-based reporting change we took the opportunity to consider the range of other data and supporting documentation in DfT’s Reported road casualties statistical release. Of note was the substantial amount of good quality supporting material made available to users which both enhance the usability of the statistics and improve trustworthiness by being transparent about your approach to the statistics and communicating the strengths and weaknesses of the data.

We also recognise that your team seems very responsive to user needs, introducing new analysis or standing down publications in response to areas of user interest, for example the series of factsheets on topics of interest such as Deprivation and road casualties or e-Scooters. This allows the team to focus resource on where the greatest value can be added.

One aspect that I particularly want to highlight relating to these statistics and more generally across DfT relates to pre-release access. The list of individuals with pre-release access was highlighted as an issue when we last reviewed these statistics in 2019. We note that the list has increased since then. From speaking to the team this has occurred even though they are following processes and guidance on pre-release access within DfT.

We recommend you review the process and rules around how these lists are maintained to ensure that only those who are essential to commenting on or taking action on the statistics at or just prior to the release of the statistics are included, and that suitable challenge is embedded within the process.

We also note that statistics for Wales and Scotland are published by the Welsh Government and Transport Scotland based on the same data. However, there are some differences in approaches. Your team includes some material within the supporting documentation that touches upon these differences, but we consider that expanding this where appropriate may be beneficial for users. We understand that the Welsh Government and Transport Scotland are looking at the quality of different elements of the underlying data; DfT should take account of any outcomes of this, consider how these may impact on the wider GB statistics and update its quality information accordingly.

We appreciate the open engagement and clear enthusiasm and commitment to the development of these statistics that your team showed. We will continue to engage with you on the outcomes of this short review and in particular on pre-release access processes.

I am copying this letter to Melanie Brown, Welsh Government and Andrew Paterson, Transport Scotland.

Yours sincerely

Mark Pont

Assessment Programme Lead