Introduction Published studies on road traffic injuries in Kenya are all descriptive. There is a dearth of data from controlled studies. The purpose of this study was to determine the contribution of driver risk factors for motor vehicle crashes.

Methods This was a case–control study undertaken in Eldoret, Kenya. Cases were drivers of motor vehicles that had crashed and resulted in hospitalisation of the injured; controls were non-accident involved drivers randomly selected through a roadside survey from the driving population.

Results 150 case and 327 control drivers were interviewed. They were mostly young (mean age 36.8 years), male (96.7%), married (86%) with post-primary education (60%). The demographic characteristics of cases and controls were similar. The most important personal risk factors were age, not having a driving licence, alcohol use, overspeeding and use of mobile phones. Younger drivers (18–44 years) were twice as likely as older drivers to be involved in a crash [OR=1.84]; unlicensed drivers were 5 times more likely than those licensed to be involved in a crash; drivers who had consumed alcohol were three times more likely to crash [OR=2.94]; and those driving at a speed greater than the posted speed limit were 10 times more likely to crash [OR=10.378].

Conclusions Modifiable driver factors such as licensing, speed, alcohol and use of mobile phones are important risk factors for cashes on urban roads. Enhanced awareness and enforcement of legislation on these factors should be emphasised as part of broader road safety interventions for all road users.