Compare road outcomes against the amount and type of travel.
The executive takeaway
Per 100,000 motor vehicle miles · the same 190 eligible authorities in Great Britain each year.
20160.04286
20250.03038
injury collisions per 100,000 miles
This fixed panel is a comparison of eligible areas, not the whole Great Britain total. Changes are descriptive; they do not establish cause or insured claim cost.
The rate is the sum of injury collisions divided by the sum of motor vehicle miles in the same eligible areas. Areas with incomplete or flagged evidence are excluded.
| Year | Rate per 100,000 miles | Collisions | Motor vehicle miles |
|---|---|---|---|
| 2016 | 0.04286 | 124,267 | 289,941,000,000 |
| 2017 | 0.0405 | 119,059 | 293,991,100,000 |
| 2018 | 0.03801 | 112,309 | 295,449,700,000 |
| 2019 | 0.03601 | 107,763 | 299,246,700,000 |
| 2020 | 0.03558 | 83,751 | 235,369,100,000 |
| 2021 | 0.03537 | 93,105 | 263,254,200,000 |
| 2022 | 0.03413 | 97,758 | 286,397,700,000 |
| 2023 | 0.03271 | 95,724 | 292,640,700,000 |
| 2024 | 0.03086 | 91,727 | 297,202,300,000 |
| 2025 | 0.03038 | 92,007 | 302,823,900,000 |
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
Keep in mindReported injury collisions do not measure all claims, fault or financial losses.
Approved evidence published 2 Oct 2026This topic, its downloads and linked Data Lab views use the same selected release.
Inspect the sources and their datesA busy road may have more collisions simply because more vehicles use it. Compare exposure, injury outcomes and persistent patterns before drawing an insurance conclusion.
Established: Police injury records and traffic data support descriptive comparisons.
Developing: Vehicle, road, time and weather breakdowns help form testable questions.
Still uncertain: Reported injury collisions do not measure all claims, fault or financial losses.
Recorded injury collisions: 2021: 101,087; 2022: 106,004; 2023: 104,258; 2024: 100,927; 2025: 101,525. These are counts, not rates.
STATS19 collision records, latest five final years
Evidence period: 2021–2025Checked in this release: 25 Sept 2026
STATS19 vehicle records, latest five final years
Evidence period: 2021–2025Checked in this release: 25 Sept 2026
STATS19 casualty records, latest five final years
Evidence period: 2021–2025Checked in this release: 25 Sept 2026
Rates, injury severity and persistent differences need separate interpretation.
Current response: Test
Watch next: Reassess when exposure-controlled patterns also hold in developed insurer claims.
Reported injury collisions do not measure all claims, fault or financial losses.
Evidence period: 2021–2025Checked in this release: 25 Sept 2026
Evidence period: 2021–2025Checked in this release: 25 Sept 2026
Evidence period: 2021–2025Checked in this release: 25 Sept 2026
Evidence period: 2021–2025Checked in this release: 25 Sept 2026
Evidence period: 2021–2025Checked in this release: 25 Sept 2026
Evidence period: 2021–2025Checked in this release: 25 Sept 2026
Evidence period: 2021–2025Checked in this release: 25 Sept 2026
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
Evidence period: 2016–2025Checked in this release: 23 Sept 2026
THE ACTUARY LENS · ROAD RISK
Interpret the evidence on injury-collision occurrence, severity and exposure. The detailed analysis lives in the STATS19 Data Lab.
INSURANCE IMPLICATIONS
What would change this view? Reassess when exposure-controlled patterns also hold in developed insurer claims.
Calculated from this release using the exact same selectors as Data Lab. Eligible-authority totals are not whole-GB totals. A repeated ordering is descriptive evidence, not a causal result or an insurer pricing validation.
In 2025, 127,883 casualties were recorded. 34.8% were in collisions injuring more than one person. The Lab separates casualty role, age, road conditions and adjusted KSI, with every person counted once per breakdown.
Multi-claimant exposure can matter for bodily-injury accumulation, but police injury counts do not measure compensation or developed insurer losses.
Inspect injury burden and multiplicity →The Lab standardises conditional collision severity across shared vehicle-age, driver-age, road, urban/rural and light strata, separating cars from taxis. It reports how much of each fuel population remains in common support. The frozen 2025 file has no vehicle-age values, so that year’s adjusted comparison is unavailable.
These controls improve descriptive comparability. They do not establish accident frequency per mile, fault or a causal powertrain effect. Sparse taxi support and unmeasured vehicle/driver differences remain important.
Inspect comparable EV / ICE outcomes →Recorded home-area agreement and published home-to-collision distance bands now sit alongside explicit missingness in the Lab. Collision geography describes where events occurred; pricing geography usually describes where a policyholder lives. Residential mileage or policy exposures are needed before estimating a postcode relativity.
Inspect residence and collision evidence →In 2025, the national matched-road comparison records 0.0296 injury collisions per 100,000 motor vehicle miles. Road categories use the corresponding official annual mileage, and vehicle comparisons count involvements against same-class mileage.
Annual road and vehicle rates are available. Additional regional and time-profile inputs require approval in the shared data release.
Compare frequency, mileage and uncertainty →The 2025 broad car-and-taxi rate is 0.0483 involvements per 100,000 matching miles. It includes all fuels and ages.
Fuel-specific mileage and age-adjusted rates remain withheld: the MOT sample includes non-car Class 4 vehicles, grouped use does not identify calendar mileage, and the fleet expansion is incomplete. The Lab reports the observed MOT age profiles, missing intervals, calibration gaps and required sensitivities. Stock-based indicators remain separate.
Inspect vehicle mileage and fuel exposureThe local-weather pilot is awaiting a reviewed data release.
The Lab maps station and counter coverage, separates observed from imputed traffic and compares alternative gauges and prior traffic on common days. Temperature and ice effects remain unavailable. This sampled network does not justify a causal or insurer pricing multiplier.
Inspect local rain, measured exposure and sensitivityThe NTS driver exposure extension is awaiting a reviewed data release.
Age exposure profiles include survey intervals. Age-specific rates, commercial business comparisons and joint age-purpose rates remain withheld until population and driving scopes can be reconciled.
Inspect driver age, journey purpose and exposureThe road-mix extension is awaiting approval in the shared release.
Only 2025 authority road-mix estimates pass the exposure checks. Earlier authority comparisons and M25 section frequencies remain withheld. Broad regional persistence and two later years do not establish a causal or insurer pricing effect.
Inspect road mix, frozen groups and route matching →The comparable panel moves from 0.04 to 0.03 injury collisions per 100,000 motor vehicle miles between 2016 and 2025. Changes in how much, when and where vehicles travel can affect the result.
For an insurer, this supports testing road exposure alongside annual mileage. It does not establish a change in insured claim frequency: damage-only events and many unreported injuries are outside this dataset.
Inspect annual rates and coverage →The highest-risk discovery group remains above the lowest in 4 of 4 eligible later years. Earlier-versus-later area rank correlation is 0.954. These comparisons retain the same areas and freeze the original groups.
Stable road mix, exposure patterns and reporting differences can also generate persistence. A pricing study should test decomposed geographic features alongside the existing postcode factor, using features available at quote and later claims held out from model construction.
Compare geographic persistence and exceptions →Adjusted serious-injury measures help compare road outcomes through changes in police reporting. They do not estimate compensation, repair costs, credit hire or large-loss development. The Lab separates casualties per mile from the proportion of recorded injury collisions that are fatal or serious.
For UK motor insurance, these are candidate indicators for bodily-injury exposure. Financial severity requires an insurer’s developed claims experience.
The detailed 2021–2025 records support road, time and vehicle comparisons. Licensed-car stock joins provide a fuel-cohort involvement indicator. Raw stock comparisons do not control for vehicle age, mileage, driver selection or road use. The separate matched analysis reports its own controls and remaining limitations.
The separate annual context view joins country weather to a fixed geographic road panel. It describes association across years; it does not estimate the effect of rain or ice on an individual journey.
A road-exposure or telematics analysis should use measured driving exposure in each road and time environment before translating these patterns into pricing factors.
UPDATES
Observation periods and publication dates appear with the supporting sources in Data & sources. A source check records an inspection; it does not make the evidence newer.
The assessment changes when the evidence justifies a different insurance conclusion. Its change date is recorded separately from source checks and the date this release was published.
Trace the assessment to its publications, datasets and observation periods.
Approved evidence published 2 Oct 2026This topic, its downloads and linked Data Lab views use the same selected release.
Inspect the sources and their datesEvidence limit: Reported injury collisions do not measure all claims, fault or financial losses.
Dates belong to the selected release. A source check is an inspection, not new evidence. External source links open in a new tab.
Source 1 · Department for Transport
Qualification: Original bytes archived losslessly as gzip; sha256 applies to decompressed bytes. Initial reproducible extraction, no human source review asserted.
Used for: Department for Transport
Source 2 · Department for Transport
Qualification: Original bytes archived losslessly as gzip; sha256 applies to decompressed bytes. Initial reproducible extraction, no human source review asserted.
Used for: Department for Transport
Source 3 · Department for Transport
Qualification: Original bytes archived losslessly as gzip; sha256 applies to decompressed bytes. Initial reproducible extraction, no human source review asserted.
Used for: Department for Transport
Source 4 · Department for Transport
Qualification: Original bytes archived losslessly as gzip; sha256 applies to decompressed bytes. Initial reproducible extraction, no human source review asserted.
Source 5 · Department for Transport
Qualification: Original bytes archived losslessly as gzip; sha256 applies to decompressed bytes. Initial reproducible extraction, no human source review asserted.
Source 6 · Department for Transport
Qualification: Original bytes archived losslessly as gzip; sha256 applies to decompressed bytes. Initial reproducible extraction, no human source review asserted.
Source 7 · DfT / DVLA
Qualification: Original bytes archived losslessly as gzip; sha256 applies to decompressed bytes. Initial reproducible extraction, no human source review asserted.
Source 8 · Department for Transport
Qualification: All roads; latest geography in the workbook. Severity-adjusted FSC and KSI. Reporting warnings for Avon and Somerset and Staffordshire in 2022–2023 are excluded from persistence tests.
Source 9 · Department for Transport
Qualification: Motor vehicle miles; pedal cycles excluded. Exact authority code and year join. Independent rate reconciliation against RAS0403; mismatches are suppressed.