Research has been conducted into three subjects that relate to the safety of roundabouts (R-95-58). The first subject concerns the development of safety in the long term. The results of a previous study gave the impression that, after some time, the number of road accident victims on roundabouts was again increasing. This new study, in which the accident history of about 200 roundabouts was expanded by the years 1992 to 1994 inclusive, shows that safety is certainly not declining in the long term.
The expansion of the previously performed accident study again shows that replacement of an intersection by a roundabout reduces both the number (by 50%) and the severity of accidents. Also, the replacement of an intersection controlled by traffic lights by a roundabout leads to positive results. The study demonstrates, even more clearly than the previous study did, that - certainly on the somewhat busier roundabouts - a separate cycle path is almost always a safer solution than a cycle lane on the carriageway or the absence of a cycle facility. Replacement of intersections by roundabouts with separate cycle paths (cyclists do not have right of way) has led to an average decrease of the number of casualties by 90%.
The second subject concerns the regulation of priority for cycle traffic on separate cycle paths around roundabouts. In follow up to a study conducted in 1994, the road hazard on 17 roundabouts where cyclists have priority was compared with the road hazard on 62 roundabouts where cyclists do not have right of way. Again, it was determined that more accidents and road accident victims were registered, on average, on those roundabouts where cyclists had right of way. More than 90% of these victims related to cyclists and moped riders. The large variation in the number of road accident victims on these roundabouts represents an indication that the design and organisation of the roundabout could have a major influence on safety where this priority rule applies.
Apart from considering safety, this study also devotes attention to how the priority rule for cyclists influences the capacity of the roundabout for motorised traffic. Cyclists who have priority indeed exert a deleterious effect on that capacity, but this is only of significance with very busy roundabouts that carry a large volume of cycle traffic.
The third subject relates to the change in the priority rule for existing roundabouts that are older and generally somewhat larger. Accident studies on roundabouts where the priority rule was amended, whether or not in combination with (limited) reconstruction, offer a conflicting impression. The number of accidents continues to fall, while the number of road accident victims sometimes increases - and this generally relates to cyclists and moped riders.
Experts that have been consulted on this matter tend to be unanimous in their opinion that the priority rule should be linked to a reconstruction of the roundabout in the sense of radially-oriented approach roads, tighter curves and often narrowing of the carriageways.
An analysis of the problem, which also involves the desired uniformity, led to the recommendation to impose priority for traffic on the roundabouts for all roundabouts in the short term, provided that a safe solution for cycle traffic is ensured at the same time. Depending on traffic intensities and the local possibilities, these will be in the form of separate cycle paths, flyover solutions or alternative cycle roads.
A full reconstruction of the roundabout is recommended, but in order to prevent a delay in the priority rule amendment, this could also be done at a later stage.