OBJECTIVE: The Drive Home Maze Test has been developed as part of a clinic-based driver assessment battery. This article reports normative data and preliminary inter-rater reliability for this test together with predictive validity for a clinical sample. SUBJECTS AND SETTING: 223 individuals contributed to the normative data. Of these, data from 42 clients undergoing driver assessment as part of their rehabilitation or health care plan were used to ascertain the predictive validity of the Drive Home Maze Test for on-road driving performance (Pass/ Fail). MAIN MEASURE: The Drive Home Maze Test was developed for occupational therapy driver assessors to use with all older and/ or functionally impaired drivers prior to on-road testing. The test assesses a person s executive function, attention and visuoconstructive skills. RESULTS: The normative data indicated that time taken and number of errors on the maze increases with age. Performance on this test was strongly indicative of driver assessment outcome using the regression model: ln(Odds of failing on road) = 5.9451 + 2.7047 (log10 time taken to complete the maze). Excellent preliminary inter-rater reliability was also found using Ordinary Least Products linear regression analysis to test for bias between two raters for time taken and number of errors on the maze, with values ranging from r = 0.858 to 0.999. CONCLUSIONS: The Drive Home Maze Test provides useful information concerning a person's cognitive skills prior to undertaking an on-road assessment. Time taken to complete this test is indicative of the outcome of an on-road driver assessment.