Kashiura and colleagues also voiced concern about possible confounding effects of active smoking by children. Tobacco dependence usually starts in adolescence, in utero smoke exposure increases risk for later development of tobacco dependence, and active smoking would affect a child’s respiratory symptoms. Our data set did not assess children’s active smoking. As the onset of tobacco use is not common before adolescence, if this were an important confounder of the association between maternal smoking and childhood asthma, I would expect to see a significant interaction of maternal smoking and age 12 to 18 years. On reanalysis of the data set, this interaction was not significant (P > .3) in any of the models. It is possible, however, that maternal smoking is a proxy measure for adolescents’ smoking; sorting out the independent effects of each would not be possible in our data set.