Study objective: To test the effects of coffee and napping on nocturnal driving in young and middle-aged participants. Design: A cup of coffee (200 mg of caffeine), a placebo (decaffeinated coffee, 15 mg of caffeine), or a 30-minute nap were tested. Participants drove 125 highway miles between 18:00 and 19:30 and between 02:00 and 03:30 after coffee, placebo, or a nap. Setting: Sleep laboratory and open French highway. Participants: Twelve young (range, 20-25 years) and 12 middle-aged participants (range, 40-50 years). Measurements: Inappropriate line crossings, self-perceived fatigue and sleepiness, and polysomnographic recordings were analyzed. Results: Compared to daytime, after placebo the number of inappropriate line crossings was significantly increased (2 versus 73 for young participants, P < 0.01 and 0 versus 76 for the middle-aged participants, P < 0.05). Both coffee and napping reduced the risk of inappropriate line crossings, compared with placebo, in young participants (respectively, by three-quarters, incidence rate ratios [IRR] = 0.26 95% confidence interval [CI], 0.09-0.74, P < 0.05 and by two thirds, IRR = 0.34 95% CI, 0.20-0.58, P < 0.001) and in middle-aged participants (respectively by nine tenths, IRR = 0.11 95% CI, 0.05-0.21, P < 0.001 and by one fifth, IRR = 0.77 95% CI, 0.63-0.95, P < 0.05). A significant interaction between age and condition (IRR = 2.27 95% CI, 1.28-4.16 P < 0.01) showed that napping led to fewer inappropriate line crossings in younger participants than in middle-aged participants. During napping, young participants slept more (P < 0.01) and had more delta sleep (P < 0.05) than middle-aged participants. Self-perceived sleepiness and fatigue did not differ in both age groups, but coffee improved sleepiness (P < 0.05), whereas napping did not. Conclusions: Coffee significantly improves performance in young and middle-aged participants. Napping is more efficient in younger than in older participants. Countermeasures to sleepiness should be adapted according to the age of drivers.< Réduire