Pedestrians made up nearly one in five serious road casualties in Northern Ireland in 2025

Pedestrians accounted for almost one in five people killed or seriously injured on Northern Ireland’s roads in 2025, despite representing less than one in twelve of all recorded casualties. The difference highlights the importance of examining injury severity as well as total numbers.

The contrast in the official figures

In the 2025 calendar-year road-collision statistics, pedestrians made up 19.0% of killed or seriously injured (KSI) casualties in Northern Ireland. They accounted for 7.7% of all casualties recorded in police-reported injury collisions. Drivers accounted for 41.8% of KSI casualties, meaning drivers and pedestrians together represented 60.8% of the KSI total.

Across all road-user groups, 56 people died and 1,066 were seriously injured during 2025. The combined number of people killed or seriously injured was 1,122, an 11.3% increase on 2024. Overall, 7,879 people were injured in 5,015 recorded injury collisions during the year.

The figures describe the composition of recorded casualties. They do not mean that pedestrians had a 19% chance of being seriously injured, nor do they show the likelihood of a pedestrian being hurt compared with someone travelling in a vehicle. Those questions require information about how many people walk, how far they travel and the environments in which they travel.

Why a casualty share matters

A collision can affect road users differently. People walking have no vehicle structure around them, and the severity of an impact can depend on factors including vehicle speed, the point of impact and the environment in which the collision occurs.

That helps explain why looking only at all-casualty totals can miss important differences in outcomes. The pedestrian share of KSI casualties is substantially larger than the pedestrian share of casualties overall. The statistics establish that contrast, although they do not identify a single reason for it.

It would be incorrect to use the figures to claim that every pedestrian collision is more severe than a vehicle-occupant collision. The aggregate percentages cannot reveal the full distribution of injury severity within each road-user group, and the published figures do not by themselves account for exposure to traffic.

What a deeper investigation should test

PureCar’s collision and casualty archive provides a route to a more detailed question: what recorded circumstances are associated with pedestrian casualties, and do those circumstances differ between slight injuries, serious injuries and fatalities?

The analysis would need to match each casualty record to its collision record, check for missing or inconsistent road-user classifications and define location groups consistently. Where the dataset includes reliable information, analysts could examine whether recorded pedestrian collisions occurred at junctions, on different categories of road or in built-up and rural settings.

Those comparisons must not be turned into an unqualified ranking of “most dangerous” streets. A location with a large number of incidents may also have far more pedestrians and vehicles passing through it. Without a reliable denominator—such as pedestrian movements or comparable exposure data—counts measure recorded volume, not the chance of a collision per journey.

From national proportions to local evidence

National percentages are useful for showing the broad scale of a problem. Local decisions, however, require local evidence. Councils and road authorities need to understand where collisions happen, what the road environment is like, and whether any apparent cluster is supported by sufficient records.

A responsible local study should report the number of cases included, the years covered, how locations were matched and how many records could not be classified. If the location field is missing or uncertain, it should be acknowledged. Small counts should not be presented as proof of an enduring danger pattern.

It is equally important to distinguish fatalities from serious injuries. Combining them into the standard KSI measure is useful for monitoring serious harm, but it should not obscure changes in either outcome where the data supports a separate analysis.

What motorists and pedestrians can do

The statistics cannot prescribe one intervention for every street, but they reinforce the need for care wherever vehicles and pedestrians interact. Drivers should approach junctions and crossings attentively, look carefully for people who may be obscured, and select a speed that is appropriate to visibility and conditions. Pedestrians should use suitable crossing points where available and take care when judging traffic gaps.

These are general safety recommendations, not a claim that behaviour by pedestrians or drivers explains the percentages. Road design, visibility, traffic speed and the circumstances of individual collisions require their own evidence-based assessment.

The finding—and the unanswered question

The clearest finding from the 2025 statistics is the gap between pedestrians’ share of all casualties and their share of serious outcomes. Pedestrians represented 7.7% of casualties overall but 19.0% of people killed or seriously injured.

The next step is not to infer a cause from that difference. It is to use linked casualty and collision data to examine the circumstances and outcomes in more detail, while recognising that exposure data are needed to compare underlying risk fairly.

Why percentages need a denominator

The difference between 7.7% of all casualties and 19.0% of KSI casualties is a comparison between two shares of recorded casualties. It is not a comparison of collision rates. To estimate a pedestrian’s risk relative to that of a vehicle occupant, researchers would need a defensible measure of how much each group is exposed to road traffic. Possible measures include journeys, distance travelled or pedestrian movements, but they are not interchangeable and must match the question being asked.

For example, a town-centre crossing might record several pedestrian collisions because thousands of people cross it each day. A quiet rural stretch might record fewer incidents but offer fewer opportunities for people to walk there. Without a comparable exposure measure, simply sorting locations by casualty count risks confusing the volume of activity with the likelihood of harm.

What should be checked in the underlying records?

A sound PureCar analysis should begin with the casualty records, identify pedestrians using the official road-user classification and link each record to the corresponding collision identifier. It should then check for missing identifiers, duplicate imports and conflicting classifications before summarising severity. If the data contains more than one casualty from a collision, each person should count as a casualty while the incident itself is counted only once in a collision total.

The location analysis should use consistent geographic rules and report how many cases could not be placed reliably. A street name alone may refer to a long road rather than a particular crossing or junction. Where coordinates are imprecise, the result should be grouped more broadly or excluded from a detailed site ranking, with the limitation explained.

From statistics to decisions

Road-safety interventions should respond to confirmed local circumstances rather than national percentages alone. Depending on the site evidence, questions might include whether crossing points are visible, whether vehicle speeds are appropriate, whether parked vehicles block sightlines or whether pedestrian movements interact with turning traffic. These are questions for site-specific assessment, not conclusions established by the annual percentages.

A fair study should look at both the number and severity of incidents and, where available, how the road is used. It should also avoid implying that pedestrians are responsible for their own injuries when the figures do not establish the circumstances of individual events. The purpose is to understand collision conditions and identify potential risk-reduction opportunities, not to assign blame from a broad category.

What readers should watch for in future reporting

When reading claims about the “most dangerous” streets for pedestrians, readers should ask which years are included, whether the count is of collisions or people, how locations are defined and whether traffic or pedestrian exposure has been considered. A high count can be a reason to investigate, but it is not a complete assessment of risk.

PureCar’s value will come from making those distinctions visible. The national figures establish an important disparity in casualty shares. Properly linked records, transparent location methods and suitable exposure data would help explain the circumstances behind it without claiming more than the evidence supports.

Methodology: Percentages are reported from the PSNI/NISRA 2025 calendar-year statistics. The article describes casualty composition, not a per-journey risk rate. PureCar should publish any later location-level conclusions only after checking the underlying record links and available exposure data.