Many patients with obstructive sleep apnea (OSA) or narcolepsy have difficulty driving and increased automobile accidents. Previously we have shown that OSA patients perform poorly on a laboratory-based divided-attention driving test (DADT). Patients with narcolepsy may be as sleepy as OSA patients, so we compared performance on the DADT of OSA patients with that of narcolepsy patients. Twenty-one male OSA patients [age 49.3 +/- 12.7 (SD) years; apnea-hypopnea index (AHI) 73 +/- 29] 21 age- and sex-matched controls, and 16 narcoleptics (12 males, four females; age 39.6 +/- 15.2 years) underwent polysomnography followed by daytime sleep latency testing (MSLT). Following a practice session, all subjects were given the DADT for 20 minute prior to each daytime nap of the MSLT. Narcolepsy patients were younger than OSA or controls and more sleepy than OSA patients. Tracking error was much worse in patients than controls (228 +/- 145 cm for OSA vs. 196 +/- 146 for narcolepsy vs. 71 +/- 31 for controls; p < 0.001), although half of either patient group performed as well as controls. There was only a weak relationship between MSLT and tracking in either patient group. We conclude that impairment in laboratory driving performance skills is seen in both groups of sleepy patients but the degree of impairment is difficult to predict from sleepiness alone.