The NHMRC Road Accident Research Unit (RARU) was funded by the Motor Accident Commission in 1997 to investigate the role of roadside hazards in road accidents resulting in death or serious injury to a car occupant in South Australia. The main aim of this project was to document the extent to which roadside hazards contribute to severe and fatal car crashes in South Australia and to comment on the opportunities that exist to make our roads ides safer. A secondary aim of the project was to conduct investigations in such a way as to provide for the training of engineers from Transport SA in the recognition of hazardous roadside features and an appreciation of their importance in road safety. The study was based on information contained in the Traffic Accident Reporting System data base on road accidents reported to or by the Police and on information in Coronial records of fatal crashes. Some roadside hazard crashes were also investigated at the scene. We found that: Roadside hazards were the immediate cause of at least one death in 40 per cent of all crashes in which a car occupant was fatally injured in South Australia from 1985 to 1996. Collisions with roadside hazards were the immediate cause of 39 per cent of all car occupant deaths during those years. Roadside hazards also played a role in 38 per cent of all car crashes in which an occupant was admitted to hospital in South Australia from 1994 to 1996. Countermeasures aimed at reducing travelling speed, drink driving, and driver fatigue are likely to decrease the frequency of roadside hazard crashes, probably to a greater degree than crashes in general. However, reliance on attempts to change driver behaviour alone will not be an adequate response to the dangers presented by roadside hazards. There is much that can be done to further improve the safety of our roads and roads ides. Cars that ran onto the unsealed shoulder of the road were 20 per cent more likely to be involved in a fatal collision with a roadside hazard than were cars involved in other types of crash in which an occupant was fatally injured. Sealing part of the shoulder of the road and providing edge lining reduces the risk of a driver running off the road, particularly on curves. The potential benefits of clear zones alongside roads are evident in this study. More than half of the fatal hazards in the crashes in this study were within 3 meters of the edge of the road, partly reflecting the number of crashes involving poles and trees in urban areas. Overall, 90 per cent of the roadside hazards which were struck by a car, fatally injuring at least one occupant, were located less than 9 meters from the edge of the roadway. Trees were by far the most common roadside hazard in South Australia, accounting for 23 per cent of all deaths to car occupants and being involved in 17 per cent of all serious injury crashes. Some of the fatal crashes were with trees which had been planted by the then Highways Department in what are should be clear zones. Collisions with Stobie poles accounted for an average of about 10 deaths and 70 hospital admissions per year in South Australia..Sixty five car occupants were fatally injured when their car either rolled down a steep embankment, ran into a drain, or struck the face of a cutting. Many of these deaths could have been prevented by either changes to the earthworks alongside the road or the provision of guard rails. Although cases in which a guard rail performed as intended did not appear in the fatal accident data files, there were 13 car occupants who were fatally injured when their car hit a guard rail. These fatal cases show that further investigation and refinement of the design and installation of guard rails is warranted. Six car occupants who died in collisions with fences on the boundaries of private property were impaled by the top rail of the fence. Many similar fences which were erected by the Highways Department adjacent to signalised pedestrian crossings in the Adelaide area have yet to be replaced, even though at least one has been a direct cause of fatal injury to a car occupant. This study has shown that the investigation and monitoring of the causes and consequences of road crashes in South Australia is hampered by the inadequacies of the main source of data, the Traffic Accident Reporting System. RECOMMENDATIONS Current levels of enforcement of legislation relating to speed and drink driving be maintained or increased. Consideration be given to the elimination of 110 km/h zones in rural areas of South Australia and to a reduction in the speed limit to 50 km/h on all roads in urban areas. The horizontal alignment of the road, the condition of the road surface, and the provision of sealed shoulders and edge lines be subject to timely review, initially on all State and National Highways in South Australia, and a case be made for the provision of funds to permit early rectification of such deficiencies as may be identified. As soon as practicable, shoulders be sealed to a width of at least half a metre on all highways and major roads in South Australia, commencing with curved sections of road, and despite standard traffic engineering practice, edge lining be provided on highways and major roads in South Australia regardless of the width of the road. The roadside tree planting policies of Transport SA and local government authorities be reviewed taking into account current best road safety practice to prevent new planting in hazardous locations and to rectify the dangers posed by trees which have been planted within 9 meters of rural roads. In urban areas a clear zone should be maintained within a minimum of 1 metre from the edge of the road and preferably 3 meters or more wherever practicable. A standing committee similar to the NSW Hazardous Pole Program Committee be convened to identify and implement all practicable ways of reducing the deaths and injuries resulting from collisions with Stobie poles. Casualty crashes with guard rails be the subject of detailed investigation, supplemented by the investigation of cases that can be identified in which a guard rail has performed as intended. A review be conducted of recent developments in the design of guard rails, including a French design which is constructed from softwood logs and is widely used on minor rural roads in that country. The system of reporting on road accidents in South Australia be reviewed with reference to the systems operating elsewhere, including the United States. (A)
Abstract