Cycling Fatalities 2015: STATS19 Discrepancies

Almost inevitably, due to the way the information is obtained, the Beyond The Kerb fatality lists are not comprehensive. However, neither are the official STATS19 data. This page looks at the discrepancies between the two.

Total fatalities

The two lists counted a different number of fatalities.

  • Beyond The Kerb listed 108 as collision fatalities, including single party collisions, and 7 as medical episodes (note that it is fairly likely that not all medical episode fatalities were listed). Five now-known collision fatalities were omitted.
  • STATS19 listed 100 fatalities.

However, it is worth noting that the criteria for recording incidents are a little different, meaning that not all fatalities listed in Beyond The Kerb would be expected to be recorded as cycling fatalities in STATS19 (specific cases are discussed below).

  • Beyond The Kerb lists all cycling deaths that occur on what might normally be considered UK roads in lay terms, which excludes off-road rights of way but includes car parks and private roads. Northern Ireland is included (though no fatalities were recorded there in 2015). Where there is doubt as to whether a fatality should be included, it is included, although in some cases it is later removed or reclassified. Deaths due to medical episodes (mostly heart attacks) are included, partly in case inquests later reveal different information.
  • STATS19 has tighter criteria, which are defined in STATS20. It excludes deaths which occur 30 or more days after an incident, incidents which occur on private roads or in car parks, and incidents where the sole cause of death is medical episode (note that if a non-fatal medical episode occurs and causes loss of control, and a resulting collision causes fatal injuries, it is recorded). STATS19 does not cover Northern Ireland, which has its own recording system.

Beyond The Kerb omissions: 5

There are four fatalities which were recorded STATS19 data but which were not in the Beyond The Kerb list.

Online reports have thus far been found for only the second of these incidents.

There is a fifth fatality which was recorded neither by Beyond The Kerb nor in STATS19.

Classification discrepancies: 5

There are twelve fatalities (not including medical episode deaths) which were recorded in the Beyond The Kerb list but which do not appear as fatalities in STATS19.

Five of these are classification discrepancies, in that they met the criteria for inclusion in the Beyond The Kerb list but not those for inclusion in STATS19.

Two are due to deaths occurring 30 days or more from the date of the incident, and are thus recorded as serious injuries in STATS19:

One is due to the victim being on foot at the time of the collision (the exact pattern of events is unknown, but shortly before the collision he was repairing a puncture at the roadside) and thus being recorded as a pedestrian fatality:

The remaining two are omitted from STATS19 as they were victims of the Shoreham air crash:

STATS19 omissions: 7

The cause of one fatality is, as far as can be discerned, unexplained and therefore it may or may not be a justifiable omission:

However, the remaining six are seemingly erroneous omissions:

From the information available at the time of writing, there seem to be no complications with these incidents which would obstruct the recording of them. Indeed, a number of them have resulted in the charge, prosecution and even conviction of the driver involved.

One particularly remarkable point is that Lancashire’s entire set of fatalities have been omitted from STATS19: zero deaths were recorded for that area, yet there were three collision fatalities, as well as a fourth which was recorded as a medical episode by Beyond The Kerb.


Aside from the handful of fatalities which did not meet STATS19 criteria, it would appear that the official figure for cycling fatalities should be at least 107—potentially even a little higher. An under-reporting rate of nearly 7% seems surprisingly high for fatality cases. Of particular concern is Lancashire’s apparent failure to record any of its fatalities.


It transpires that all four of Lancashire’s fatalities were recorded as medical deaths. An article about these incidents may be forthcoming once more information has come to light.