Passive exposure to tobacco smoke was assessed in children with asthma (age 3-15) and in referents. There was statistically significantly (P less than 0.0005) higher excretion of the nicotine metabolite, cotinine, in the urine of 49 children with asthma (geometric mean 10 ng/ml) compared with 77 referents (4.8 ng/ml). Maternal smoking was statistically significantly more prevalent among the asthmatics than among the referents (relative risk = RR = 2.6, 95% C1 = 1.2-5.3). In conclusion, the exposure to environmental tobacco smoke in asthmatic children was higher than among healthy children, indicating that passive smoking may be a predisposing and/or aggravating factor for childhood asthma.