The Department for Transport commissioned the Transport Research Laboratory to conduct a literature review to consider the role of infrastructure in relation to the safety of cyclists and their interaction with other road users.
"Overall, it proved problematic to draw firm conclusions from the literature," admitted TRL, but it concluded "Of all interventions to increase cycle safety, the greatest benefits come from reducing motor vehicle speeds."
TRL continued: "Interventions that achieve this are also likely to result in casualty reductions for all classes of road user. This may be achieved by a variety of methods, including physical traffic calming; urban design that changes the appearance and pedestrian use of a street; and, possibly, the wider use of 20 mph speed limits.
"Most cyclist injuries in multi-vehicle collisions take place at junctions. Reducing the speed of traffic through junctions appears to be an effective approach to reducing cycle casualties, and physical calming methods are a reliable means of achieving such a reduction."
TRL was dismissive of a 'segregation-mainly' approach:
"Providing segregated networks may reduce risks to cyclists, although evidence suggests that the points at which segregated networks intersect with highways can be relatively high-risk, sometimes of sufficient magnitude to offset any safety benefits of removing cyclists from the carriageway."
"This may be particularly the case if segregated networks remove cyclists from relatively low-risk links but then increase their exposure at junctions. There is nevertheless a potential application for this approach and it is likely to be attractive to some users. It may be of value in rural settings, where the frequency of junctions is relatively low and where required quality can be achieved and cyclists can be protected at junctions."
TRL says a key priority would be infrastructure improvements at junctions:
"A number of infrastructure interventions that are not widely used in the UK have been implemented on the continent to increase safety at junctions. Particular examples include cycle lane markings continued across junctions, cycle pre-signals and Trixi mirrors (mounted below signal heads to allow drivers of heavy vehicles to see cyclists at their nearside). The literature suggests that, appropriately applied, the former two approaches can have a beneficial effect on cycle casualties."