The presence of excessive amounts of alcohol in the blood of the driver plays a substantial role in the production of accidents that cause fatal or serious injuries to vehicle occupants and pedestrians, since, in addition to the affected driver, innocent people who have not been drinking (passengers, pedestrians, and people in other cars) are often victims. A person who is severely affected by alcohol may be physically incapable of driving a motor vehicle; one who is less affected but who is still obviously under the influence of alcohol constitutes a danger as a driver. Further, in subclinical intoxication, in which a person has consumed alcohol but shows no clinical signs whatever, his judgement may be significantly impaired.
Alcohol is implicated in a very large number of road accidents because it leads to slowed reflexes, problems with vision and a loss of self control. At 0.05 g/dL (50mg/100ml), the risk of accident is multiplied by 2. At 0.08 g/dL it is multiplied by 10. Overall, alcohol is implicated in less than 10% of accidents which are not serious, 25 % of accidents causing injury to the driver and/or passengers, 50 % of fatal collisions and 65 % of accidents where a single driver is involved. Young drivers are particularly affected by the problem of drink driving because of their lack of experience and maturity as drivers.
In Canada, studies indicate than 50% of all drivers and pedestrians fatally injured were drinking shortly before the crash; 78% of those drivers had blood alcohol concentrations of 0.10% by weight or greater. Compared to the nondrinking driver, a driver with a concentration of 0.10% is 8 times more likely to be involved in a fatal crash. At the 0.15% level, the risk of crash involvement increases by a factor of 25 or more. In the USA, from one-third to two-thirds of drivers involved in fatal crashes have a record of previous arrests for public drunkenness, for drunken driving, or have registered as having a drinking problem. Eastern Europe has an increasing number of drivers but has yet to implement effective driver education; in Poland in 1995 there were 5 times more accidents caused by drunken driving than in the UK.
In the same way that drinking and driving do not mix, neither do drinking and bicycle riding. A study published in 2001 reported that bike riders with a blood alcohol level of 0.08 g/dL or greater were 20 times more likely to be seriously injured or killed in a bicycle accident.
The drink-driving problem is exacerbated by the increased use of drugs and prescribed medicines. The Belgian government screened 2,500 involved in accidents to gauge the extent of drug use. The tests found that 28 percent were over the 50mg/100ml limit and that 19 percent had taken illegal drugs or medicines likely to impair their judgement.