Material and method
The study is based on retrospective data from the registries of the six police districts and eleven hospitals with an accident and emergency function in Northern Norway, including Svalbard, for the years 2013 and 2014.
The sample includes children and adults in Northern Norway who died or were hospitalised in connection with the use of ATVs or snowmobiles during this period. From the police we collected information on fatalities, and from the hospitals on both fatalities and injuries (hospitalisations). Injured persons who were treated by the primary health services or in outpatient departments were not included. Incidents that occurred near the borders with Sweden and Finland were registered in the county where the injured person was hospitalised.
Police officers retrieved data on fatalities in a manual review of their records, and the trauma registrars in the hospitals retrieved data on fatalities and injuries from all trauma registries and patient records. The trauma reporting forms were reviewed manually, and patient records were consulted for further information as required. Some hospitals also had overviews of hospitalisations associated with the use of ATVs and snowmobiles in which the injured person had not been admitted as a trauma patient. These were included. To avoid double entries, personal ID numbers were used to collate data, in accordance with the approvals for the study, while later analyses were undertaken on a de-identified data file. In addition, we used Google to search for media reports on deaths linked to the use of ATVs and snowmobiles, with the search terms ATV, firhjuling [four-wheeler], snøscooter, snøskuter [snowmobile], dødsulykke [fatal accident], dødsfall [fatality], døde [died], omkom, omkommet [killed]. To exclude double entries, we collated the media information on vehicles, dates, gender and accident location with the data from the police and hospitals. Fatalities discovered by internet searches were not reported among the other fatalities in our study, but were included as a separate group.
The number of injuries and fatalities in snowmobile-related incidents in 2013 and 2014 were seen in relation to the number of registered snowmobiles in these years (11). We thereby found an average number of injuries/1 000 snowmobiles/year, that could be compared to the figures from a previous study conducted in Western Finnmark (5).
In compliance with the approvals for the study, the following variables were included: fatalities; personal injury with hospitalisation; ATV or snowmobile; gender; age; county; year; month; and day of the week. The study had no approval for collection of information on degree of injury, helmet use, whether the injured person was the driver, passenger etc., weather, light conditions, time of day, driving on- or off-piste, driving on a public road or off-road, driving under the influence of alcohol, hypothermia, drowning or avalanches.
The material is presented descriptively. SPSS version 23 was used for data processing. The study obtained prior approval by the Norwegian Centre for Research Data (reference no. 49099/4/AMS), the Regional Committee for Medical and Health Research Ethics (reference no. 2016/709/REK Nord), the Council for Confidentiality and Research and the Police Directorate (reference no. 201601958–8–641). The request for data was also given prior approval by the personal data officials in the various health enterprises concerned.