STUDY OBJECTIVE: To investigate the association between motor vehicle driver injury and socioeconomic status. DESIGN: Cohort study with prospective and retrospective outcomes. SETTING: New Zealand. PARTICIPANTS: 10 525 adults (volunteer sample of a multi-industry workforce, n=8008; and a random sample of urban electoral rolls, n=2517). OUTCOME MEASURE: Motor vehicle driver injury resulting in admission of the driver to hospital or the driver's death, or both, during the period 1988-98; hospitalisation and mortality data were obtained by record linkage to national health databases. MAIN RESULTS: After adjustment for age and sex, driver injury risk was inversely associated with both occupational status (p for linear trend <0.0001) and educational level (p for linear trend =0.007). Participants in the lowest approximate quartile of occupational status were four times as likely (HR 4.17, 95% CI 2.31 to 7.55) to have experienced a driver injury during follow up as participants in the highest approximate quartile. Participants who had been to secondary school for less than two years were twice as likely (HR 2.26, 95% CI 1.34 to 3.81) to have experienced a driver injury as those who had been to university or polytechnic. There was little evidence that driver injury risk was associated with neighbourhood income (p for linear trend =0.12) CONCLUSIONS: Occupational status and educational level seem to be important determinants of driver injury risk. Driver injury countermeasures should be targeted to people in low status occupations, as well as to people with comparatively little formal education.