A previous history and earlier onset of low back pain are associated with chronic low back pain in adults, implying that prevention in adolescence may have a positive impact in adulthood. The study objectives were to determine the incidence of low back pain in a cohort of adolescents and to ascertain risk factors. A cohort of 502 high school students in Montreal, Canada, was evaluated during 1995-1996 at three separate times, 6 months apart. The outcome was low back pain occurrence at a frequency of at least once a week in the previous 6 months. Of the 377 adolescents who did not complain of low back pain at the initial evaluation, 65 developed low back pain over the year (cumulative incidence, 17 percent). Risk factors associated with development of low back pain were high growth (odds ratio = 3.09; 95 percent confidence interval (CI): 1.53, 6.01), smoking (odds ratio = 2.20; 95% CI: 1.38, 3.50), tight quadriceps femoris (odds ratio = 1.02; 95% CI: 1.00, 1.05), tight hamstrings (odds ratio = 1.04; 95% CI: 1.01, 1.06), and working during the school year (odds ratio = 1.33; 95% CI: 1.03, 1.71). Modifying such risk factors as smoking and poor leg flexibility may potentially serve to prevent the development of low back pain in adolescents.