Baseline data obtained from a study of 276 children with asthma from 259 low income families were analyzed to test the hypothesis that passive smoking is associated with frequency of emergency room (ER) visits, hospitalizations, and impaired pulmonary function. The data were analyzed using multiple regression techniques. We controlled for other variables that might affect the frequency of ER visits, including smoking by the children themselves and the presence of other irritants or allergens in the child's home. Passive smoking was positively associated with ER visits (p less than 0.01), but not with hospitalizations or abnormalities in pulmonary function. The frequency of days with symptoms of asthma per month was also directly associated with ER visits (p less than 0.02). The estimated mean annual increase in ER visits attributable to the presence of one or more smokers in the household was 1.34 +/- 0.50, an increase of 63% over nonsmoking households. The estimated annual health care cost for emergency care of children with asthma that can be attributed to passive smoking is 92 dollars (95% confidence interval from 24 to 160 dollars) for families with 1 or more smokers.