Table 4 provides a summary of the evidence that specific medications adversely affect the safety of the older driver. The preponderance of evidence suggests that benzodiazepines adversely affect the safety of the older driver, particularly for high doses and long half-life compounds. This conclusion is based upon the very consistent psychomotor function data showing pronounced dose-related impairment, the more limited epidemiologic data on crash involvement, epidemiologic data associating benzodiazepines with other types of injuries, and the fact that the reasons for most benzodiazepine use are not plausible confounders. This conclusion thus reinforces the need to prescribe benzodiazepines cautiously, including assessment of nonpharmacologic alternatives, use of the lowest possible dose for the shortest possible time, and avoidance of the very long half-life compounds. As more new nonbenzodiazepine anxiolytics and hypnotics become available, their effects on the safety of the elderly driver need to be determined. There is some evidence that cyclic antidepressants, currently the mainstay for treatment of depression in the elderly population, adversely affect driving safety; however, because of the paucity of experimental and epidemiologic data concerning the effects of depression per se on driving, further research is needed. Nevertheless, the existing data reinforce the need for careful prescribing of antidepressants, particularly avoidance of agents with high side-effect profiles (such as amitriptyline and imipramine) in the older driver. For hypoglycemics, although there is sufficient evidence of driving impairment to create a basis for concern, there are many unresolved questions. Currently, diabetic patients should be advised concerning the risk and management of hypoglycemia. For other sedating drugs, it always is prudent to advise patients concerning potential effects on driving. Unlike younger drivers, the typical older driver is a medication-taker. There now is a substantial body of evidence that commonly used medications can interfere with driving safety. Because many questions remain unanswered, there is a pressing need for further research that more fully elucidates how patient characteristics, disease, and drugs interact to affect driving safety; however, sufficient data are available to reinforce an underlying theme in geriatric medicine that is not yet fully implemented in practice: the need for caution in pharmacotherapy, with selection of a drug, dose, and regimen suitable for the unique characteristics of this population.