This issue of Annals of Internal Medicine has the most relevant medical study I’ve read in a long time. We’ve all been in the scary situation of driving late at night while sleepy and struggling to stay awake. This study asked the important question: Does drinking coffee or taking a nap improve driving late at night?

The study design involved having volunteer drivers each take 90 minute drives on different weeks under four different conditions. One drive was during the day, as a control. The other three drives were from 2:00 a.m. to 3:30 a.m. One night time drive was 30 minutes after drinking 125 ml (about a half a cup) of coffee containing 200 mg of caffeine. Another drive was 30 minutes after drinking the same amount of decaffeinated coffee (containing 15 mg of caffeine). Another drive was 30 minutes after awakening from a 30 minute nap which was taken at a highway rest area with the participant sleeping in the car with the driver seat reclined. The participants didn’t know if they were drinking coffee or decaf.

Driving quality was measured by videotaping the road from the car and counting inappropriate crossings of the painted lines that separate lanes. A driving instructor accompanied the driver and was prepared to take over control of the car. (So don’t try this at home!)

The results were interesting. The drivers drinking decaf did much worse (i.e. had more line crossings) than they did in their daytime drives. Both after napping and after drinking real coffee drivers did better on average than after drinking decaf, but not as well as during the daytime drive. The improvements due to a nap and due to coffee were about the same.

The authors conclude

Of interest, some participants respond very well to caffeine but do not improve greatly after a nap, while others benefit more from a short sleep than from caffeine. Both [coffee and napping] should be proposed and promoted to decrease the risk for accidents at night. Both countermeasures are about equally efficient in reducing line crossings, and we have no preference of one over the other. Participants should choose according to their own physiologic response.

The moral of this study is to try to avoid driving late at night. If you have to drive when sleepy, it’s better to first take a nap and drink some coffee, but that still won’t make you as safe as you are when driving during the day.