Fewer Road Deaths, More Serious Injuries: Northern Ireland's 2025 Safety Divide
Road deaths fell by 19% in 2025, but the number of people seriously injured rose by 14%. A closer look shows why the headline improvement should not be mistaken for a broad safety breakthrough.
Fifty-six people died on Northern Ireland's roads in 2025, compared with 69 the previous year. On its own, that fall of 13 deaths — approximately 19% — looks encouraging. But another figure moved in the opposite direction: 1,066 people were seriously injured, compared with 939 in 2024, an increase of around 14%.
The combined number killed or seriously injured reached 1,122 in 2025, up from 1,008 in 2024. In other words, although fewer lives were lost, the total number of people suffering the most severe recorded casualty outcomes went up by 114, or 11.3%.
The wider picture also deteriorated. Police recorded 5,015 injury collisions, resulting in 7,879 casualties across all severity levels. The total number of casualties was 5.1% higher than the previous year. Of those injured in 2025, 6,757 were recorded as slightly injured. These are police-recorded injury incidents and exclude damage-only crashes.
One way to understand the paradox is to look at the different parts of the population affected. Young people aged 16 to 24 accounted for 227 killed or seriously injured casualties, or 20.2% of the total. Among children under 16, 115 were killed or seriously injured, 22 more than in 2024. People aged 65 and over accounted for 201 KSI casualties, compared with 170 in 2024.
Road-user type also matters. Drivers accounted for 41.8% of KSI casualties, while pedestrians accounted for 19.0%. Pedestrians represented a much smaller share of all casualties, 7.7%, illustrating their disproportionate presence among more severe outcomes. A national strategy that centres only on car occupants would miss an important part of the casualty burden.
The apparent contradiction is not a statistical error. Fatalities and serious injuries are related but different outcomes. The number killed can fall while the number of severe nonfatal injuries rises. The figures alone cannot explain why those two measures diverged in 2025, nor can they prove that emergency response, vehicle safety or a particular infrastructure scheme caused the change.
The road safety strategy comparison adds weight to the concern. DfI reported that the 2025 number of serious injuries was 42% above the strategy baseline, while fatalities were 17% below it. This is a meaningful difference in performance against long-term goals.
A stronger investigation would examine which districts and road-user groups accounted for the largest increases in serious injuries, then compare changes in speed limits, enforcement, traffic volumes and road design. The official figures establish the central issue. Understanding the reasons requires another layer of evidence.
A dashboard cannot call this a simple success
| Indicator | 2024 | 2025 | Direction | |---|---:|---:|---| | Deaths | 69 | 56 | Down 13 | | Serious injuries | 939 | 1,066 | Up 127 | | Killed or seriously injured | 1,008 | 1,122 | Up 114 |
The combined harm total follows directly: 127 additional serious injuries minus 13 fewer deaths = 114 additional people killed or seriously injured. The official strategy uses a multi-year baseline as well, which is worth comparing with the one-year trend. None of this proves a causal explanation for the different directions of deaths and injuries.
Why averages and denominators matter
Fatalities can vary markedly from year to year, especially with relatively small national counts. A fair long-term evaluation should use multi-year trends and measures tied to traffic exposure where available. The fact that deaths fell does not prove a particular policy succeeded; the fact that serious injuries rose does not prove one failed. Neither conclusion follows from the totals alone.
Some readers will wonder whether improved emergency care could explain fewer deaths alongside more surviving injuries. That is an unverified hypothesis, not a conclusion supported by these records. Changes in collision mix, journey volumes, road-user vulnerability or random year-to-year variation are other possibilities. Any explanation requires evidence, not editorial guesswork.
The exposure-adjusted comparison missing from most headlines
DfI's 2025 strategy statistics provide a much sharper risk comparison: 4 KSI casualties per 100 million kilometres travelled by car or van, compared with 344 per 100 million motorcycle kilometres, 63 per 100 million kilometres cycled, and 41 per 100 million kilometres walked. These are different road-user exposure rates, not the proportion of all casualties. On the published point estimates, the motorcycle rate is 86 times the car/van rate; that is an arithmetic comparison of the reported rates, not a prediction of any individual's journey risk. Differences in journey type, speed, infrastructure and exposure are not controlled for by this simple ratio. DfI 2025 Road Safety Strategy figures (https://www.infrastructure-ni.gov.uk/news/northern-ireland-road-safety-strategy-nirss-2030-annual-statistical-report-2025).
This is precisely why totals alone are insufficient. A mode used for fewer kilometres can carry a disproportionate injury burden per kilometre. Reporting the rates alongside counts exposes the road-user groups most in need of evidence-led protection without implying that a particular individual is to blame.
