OBJECTIVE: To investigate the association of marital status with risk of motor vehicle driver injury. DESIGN: A cohort study with prospective and retrospective outcomes. SETTING: New Zealand. PARTICIPANTS: A total of 10,525 adults (a volunteer sample of a multi-industry workforce, n = 8008; and a random sample of urban electoral rolls, n = 2517). EXPOSURE VARIABLE: Self reported marital status, assessed from a questionnaire administered in 1992-93 (baseline). MAIN OUTCOME MEASURE: Motor vehicle driver injury resulting in admission of the driver to hospital and/or the driver's death, during the period 1988-98; hospitalisation and mortality data were obtained by record linkage to national health databases. RESULTS: During 108 741 person-years of follow up, 139 driver injury cases occurred (85 before baseline, 54 after). After adjustment for age, sex, and study cohort, never married participants had twice the risk of driver injury (hazard ratio [HR] 2.06, 95% confidence interval [CI] 1.35 to 3.16) as married participants (HR 1.00). The relative risk for never married participants was slightly higher (HR 2.29), though less precise (95% CI 1.39 to 3.76), after further adjustment for alcohol intake, driving exposure, area of residence, body mass index, and occupational status. CONCLUSIONS: After taking age, sex, and other variables into account, never married people had a substantially higher risk of driver injury than married people. While requiring corroboration, these findings imply that it may be appropriate for driver injury countermeasures to be targeted to never married people.