The primary aim has been to study whether the standard values for road links and intersections - number of accidents, risks, consequences, number of injured etc. - according to the Swedish Road Administration's The effects of roadworks 2000: New construction and improvements can still be regarded valid or whether modification is needed. Accident data have been extracted both from OLY/VITS for the period 2000-2002 and from STRADA for the period 2003-2005. One further objective has been to find whether data from the above accident data systems can be processed and treated in a similar way. The road network studied comprised all roads assigned numbers in VDB as per 31.12.2005. Information has been collected for a total of 55,591 injury accidents in OLY/VITS, and for 55,026 injury accidents in STRADA. Of these, 58,707 or 53 per cent can be referred to the VDB network, and 15 per cent have been classified as node accidents and 85 per cent as link accidents. The study primarily concentrated on motor vehicle accidents outside urban areas. The foremost reason is that here we have a comprehensive accident material and no interference from municipal streets which are not described in VDB.