who work fewer hours or not at all. Students who work more than 20 hours per week report more symptoms of daytime sleepiness, including tendencies to arrive at school late because of oversleeping and to have difficulty staying awake in school (Carskadon, 1990). They also report a higher use of stimulants, such as caffeine and cigarettes, possibly as an attempt to increase alertness.
Excessive sleepiness is associated with performance lapses and failures, which can interfere with learning (Carskadon, 1990). It is also possible that lapses in performance because of sleepiness play a role in the injury rates of adolescents. There is evidence that insufficient sleep is associated with moodiness, irritability, and difficulty in modulating impulses and emotions (Carskadon et al., 1989; Dahl, 1996; Pilcher and Huffcutt, 1996). An added risk arises from the tendency for sleepiness to increase the sedative effects of alcohol (Lumley et al., 1987; Roehrs et al., 1986; Zwyghuizen-Doorenbos et al., 1988). This effect may be particularly dangerous for adolescents who are experimenting with alcohol and other risky behaviors, as their sleepiness may increase their vulnerability.
Information about where, how, and under what circumstances young workers are injured or made ill is essential to both developing and evaluating prevention strategies. Surveillance systems can provide this information. Public health surveillance is generally defined as an ongoing systematic collection, analysis, and interpretation of health data that is essential to the planning, implementation, and evaluation of public health practice. Public health surveillance entails not only collecting data, but also using the data to take preventive action. In a surveillance system, the data collectors are responsible for providing information to those who set policy and implement programs and for following up to see how the data have been used.
A 1987 National Research Council report, Counting Injuries and Illnesses in the Workplace, concluded that the surveillance of occupational injuries and illnesses in the United States was inadequate; the report called for a number of changes. Since that time, federal and state agencies have substantially improved data collection and surveillance. Although no comprehensive national surveillance system for work-related injuries has been established and work-