Search for a command to run...
To evaluate the contributing role of sleepiness in Italian highway vehicle accidents during the time span 1993–1997. We analyzed separately the hourly distribution of accidents ascribed by police officers univocally to sleepiness and the rest. N/A N/A Using a polynomial regression, we evaluated the relation between accidents (whether sleep-ascribed or not) and sleepiness as derived from a 24-hour sleep propensity curve. The relation between sleep-influenced and non-sleep influenced accidents was analysed using a linear regression The rate of non-sleep ascribed accidents is closely related with sleep propensity and bears a strong similarity with the pattern of sleep-ascribed accidents. A close relationship between the curves of non-sleep ascribed accidents and sleep-ascribed accidents is confirmed. The regression coefficient, which can be seen as the ratio between the quota of accidents that can be considered as sleep affected and those actually ascribed to sleepiness, results in a value of 5.83. Considering that the rate of sleep ascribed accidents is 3.2%, we can calculate the quota of sleep influenced accidents out of those not officially ascribed to sleepiness as 18.7% reaching an estimate of accidents related in some way to sleepiness equal to 21.9%. Our indirect estimate of sleep influenced accidents approaches data reported by other European countries and highlights the importance of sleepiness as a direct and/or contributing factor in vehicle accident rates.