Police accident reports are the main source for official traffic accident data in Norway. However, many accidents often go unreported, especially bicycle and pedestrian accidents. The aim of this study was to look at possible methods to estimate these unreported bicycle and pedestrian accidents in the city of Trondheim, Norway.
This study utilized accident records from the Swedish database STRADA for three Swedish cities Umeå, Uppsala and Norrköping during the four-year period 2015-2018. Two potential methods were studied; (1) estimation using the police reporting grade, and (2) estimation using the size ratio between accidents with and without motorized vehicle involved. Both methods resulted in values used to upscale existing accident values from Trondheim in the same time period. In addition, a survey was conducted to study self-reported accidents among the inhabitants of Trondheim from September 2018 to March 2019.
The estimated values varied, depending on the method and the Swedish city. The results ranged from 82 to 118 for vehicle-pedestrian accidents, from 116 to 154 for vehicle-bicycle accidents, from 568 to 862 for single bicycle accidents and from 1371 to 2133 for single pedestrian accidents. The survey received 336 answers, accumulating a total of 59 accidents where 34 lead to injuries. Only two of the accidents were reported to the police, both severe pedestrian accidents involving motorized vehicles.
This study found that there are a number of uncertainties with the two methods, primarily due to an unknown level of under-reporting in STRADA and lack of reliable source to compare the findings with. Even so, the results showed that there are considerably more bicycle and pedestrian accidents in Trondheim than previously known. Further research is necessary to confirm the validity of the findings.