This paper presents a new method for assessing the risk of accidents associated with darkness. The method estimates the risk of accident associatedwith darkness in terms of an odds ratio, which is defined as follows: [(number of accidents in darkness in a given hour of the day)/(number of accidents in daylight in the same hour of the day)]/[(Number of accidents in agiven comparison hour when the case hour is dark)/(Number of accidents ina given comparison hour when the case hour is in daylight)]. This estimate of the risk of accident associated with darkness does not require data on exposure, but relies on the count of accidents in the same pair of hoursthroughout the year. One of the hours is dark part of the year, but has daylight the rest of the year. The comparison hour, which has daylight the whole year, is used to control for seasonal variations. The aim of relyingon the same pair of hours throughout the year is to minimise the influence of potentially confounding factors. Estimates of the risk of injury accidentsassociated with darkness are developed on the basis of accident datafor Norway, Sweden and the Netherlands. It is found that the risk of an injury accident increases by nearly 30% in darkness in urban areas, by nearly 50%in rural areas, and by about 40% for urban and rural areas combined(adjusted estimate). (A) Reprinted with permission from Elsevier.
Abstract