Cycling in the dark: how dangerous is it? Analysis of bicycle crashes in relation with light conditions
There has been no previous research in the Netherlands into how dangerous it is for cyclists to ride their bikes in the twilight and by night. This is one of the reasons why it cannot yet be determined how much bicycle lighting generally contributes to road safety, in other words, how many casualties it saves among cyclists. Partly in answer to a question that was asked by the Dutch sector organization RAI Association, SWOV has investigated the risks of cycling in several light conditions. This was done by making use of the data files on crashes, casualties and mobility that are available to SWOV, supplemented with the data for each day about sunrise and sunset. Dividing the number of casualties by the distance travelled by bicycle (in millions of kilometres) gives the casualty rate for each light condition: the higher the rate, the more casualties per kilometre travelled. When the rate is higher in certain light conditions, it may be the case that it is more dangerous to cycle in those light conditions. A different explanation, however, may be that those specific light conditions are always used by a category of cyclists who always have a higher casualty rate, irrespective of light conditions. The casualty rates have therefore been calculated for specific categories of cyclists, for specific age groups in particular. The casualty rates for the days of the week and the four seasons have also been determined.
The analyses did not include all injured cyclists, but only the serious (road) bicycle injuries. A serious road injury is a casualty who has been admitted to hospital with a minimum injury severity of 2 expressed in MAIS, the Maximum Abbreviated Injury Score. The MAIS is an international measure to indicate the injury severity. As is the case here, casualties among cyclists are often distinguished in casualties in crashes involving a motorized vehicle and casualties in crashes with no involvement of a motorized vehicle. Most seriously injured cyclists were involved in daylight crashes, both those involving and those not involving a motorized vehicle. This is an expected result, as more cycling takes place during daytime. But in the dark there are relatively more casualties among cyclists than in daylight hours. During the period 1993-2008, the percentage of cyclists who were seriously injured in motor vehicle crashes in the dark fluctuated between 14% and 17%. For seriously injured cyclists in crashes not involving motorized vehicles the percentage increased from 13% in 1993 to 23% in 2008. However, only 10% of the distance cycled is travelled in the dark, which allows the conclusion
that casualty rate for cyclists is higher in the dark than in daylight. In the dark, the casualty rate is usually higher in the morning (between midnight and the beginning of dawn) than it is in the evening (between the end of dusk and midnight).For seriously injured cyclists in crashes involving motorized vehicles the casualty rate in the dark during the morning hours is roughly twice as high as in other light conditions. Until 2001, these rates declined for the different light conditions, afterwards they have been fluctuating. In recent years the casualty rates in the dark seems to be 6 SWOV-rapport R-2010-32 Stichting Wetenschappelijk Onderzoek Verkeersveiligheid SWOV - Leidschendam increasing. For seriously injured cyclists in crashes without motorized vehicles being involved, the early morning casualty rate in the dark in 1993 was a factor of 4 higher than in other light conditions, and in 2008 this was even a factor of 10. This casualty rate shows an increase by a factor of 2.6 for the period 1993-2008.
When we look at the casualty rates per age category and light condition, we see that for all age categories the casualty rate is highest in the dark. From this we may conclude that the casualty rate is higher in the dark than it is in daylight because cycling in the dark is indeed more dangerous and not because those cyclists travelling in the dark have an increased casualty rate in all conditions. Especially for those in the age group 18-29 year-old in particular it is dangerous to cycle in the dark (compared to their overall casualty rate). There are, however, clear indications that it is not only the dark, but also the use of alcohol plays a role in the high casualty rate for this group. In 1993, 24% of the 18-24 year-old cyclists who were seriously injured on a weekend night in a crash without motorized vehicles being involved had used alcohol (according to information in the hospital registration); this percentage increased to 58% in 2008. Also among the 25-59 year-olds alcohol use is relatively high and increasing: 21% in 1993 and 44% in 2008.
Days are shorter in winter and relatively more cycling takes place in the dark: in winter 36% of bicycle mobility takes place during dusk and dark as opposed to some 5% in spring and summer and about 19% in autumn. When it is also taken into account that it can be slippery in winter, the casualty rate for cyclists could be expected to be highest in winter. This is indeed the case for cyclists in motorized vehicle crashes, although the rate is not much higher in winter than it is in other seasons. The casualty rate for cyclists in non-motorized vehicle crashes, on the other hand, is highest in summer.
The present study did not systematically investigate the possible explanations for the higher rates during dusk and dark and this subject is recommended for further study. Some kind of in-depth study and behavioural study provide more clarity about possible explanations. It can already be recommended to discourage 'cycling under the influence'.
The relationship between using bicycle lights (and their quality) and the casualty rate when cycling in dark en dusk has not been investigated in this study. It is also difficult to investigate, because it is not known whether or not cyclists who were involved in a crash had working lights on their bicycle. It is also interesting to look for an explanation of the fact that the casualty rate for cyclists in crashes not involving motorized vehicles is higher than the casualty rate for cyclists in crashes in which a motor vehicle was involved.