One of the challenges of investigating the independent effects of tobacco smoke exposure in childhood on respiratory outcomes is to separate the persisting effects of exposure in utero and early infancy from the effects of current exposure. There is evidence of a strong and consistent relationship between maternal smoking during pregnancy and lung function decrements in infancy15,16 and wheezing in early childhood.17,18 There is also evidence that maternal smoking is associated with an increased risk of asthma in children,19 although this may reflect structural or functional developmental abnormalities of the airway as the association is stronger for non-atopic wheezing illnesses and the effect size diminishes in older children. Therefore, there is still debate about whether ETS exposure of children beyond infancy can initiate asthma or whether it simply acts as a trigger for the exacerbation of symptoms in established asthma.

Many studies have reported robust associations between the exposure of children to tobacco smoke in the home and a variety of lower respiratory symptoms, including wheeze, cough and phlegm,20–23 increased absence from school for respiratory illnesses24 and medical attendances for wheezing illnesses.25 The consistency, robustness to adjustment for confounding factors and evidence of dose response in these relationships led Cook and Strachan to conclude that there was a causal association between tobacco smoke exposure and respiratory symptoms in school-aged children.26 Rushton and colleagues calculated population attributable risks of 10–15% for lower respiratory symptoms caused by ETS exposure in schoolchildren in England and Wales.27 A number of observational studies have reported differential effects of prenatal and postnatal tobacco smoke exposure on respiratory symptoms in childhood. Many of these concluded that there were independent effects, although the strengths of association with different outcomes varied between studies. In general, such studies have tended to support a stronger effect of prenatal exposure to tobacco smoke metabolites than postnatal ETS exposure on lower respiratory illnesses.

Ger associations between respiratory symptoms and maternal compared with paternal smoking can be taken as evidence of a likely prenatal effect due to smoking during pregnancy, although it is also usual for children to spend considerably more time in the company of their mothers during early childhood. Conversely, the presence of a detectable effect of paternal smoking on the respiratory symptoms of children of non-smoking mothers has been used in support of an independent postnatal effect of ETS on children’s respiratory health.

Although there is little conclusive evidence that tobacco smoke exposure during childhood is a strong risk factor for the initiation of asthma, there is good evidence that exposure to ETS exacerbates existing asthma symptoms31–33 and is associated with persistence of wheezing to adolescence 22 and increased risk of asthma in adults.34

In summary, passive exposure to ETS during the school years is associated with an increased risk of respiratory symptoms, including the exacerbation of asthma symptoms in children with established disease. Maternal smoking during pregnancy is associated with an increased incidence of new asthma cases and asthma prevalence in the preschool age group, but the effect is less convincing for later exposure. Exposure to parental smoking is related to the exacerbation of symptoms in children with asthma and may be one of the risk factors for the persistence of asthma in later childhood.