Environmental Tobacco Smoke and Respiratory Health in Children: A Critical Review and Analysis of the Literature from 1969 to 1998

Abstract

Listen

Translate article

This analysis of parental/household smoking (a surrogate for environmental tobacco smoke, ETS) and respiratory symptoms and disease in children updates an earlier analysis. Some 94 studies of preschool children and 152 studies of school-age or older children published between 1969 and 1998 were examined. Both analyses have shown an age dependency in the relationship between parental/household smoking and respiratory symptoms and disease in children. A statistically significant, though moderate, relationship between parental/household smoking and respiratory illness was observed in most (86%) of the studies in preschool children. While almost two thirds (98 of 152) of the studies of school-age children showed a statistically significant relationship between parental/household smoking and respiratory symptoms and disease, there was a general lack of consistency of statistical association for specific respiratory endpoints (e.g., asthma, wheeze, bronchitis, and cough). In addition to outcome, specific characteristics of these studies were analyzed for consistency. The commonest index of ETS exposure was a response to a questionnaire regarding adult smoking in the household. Clinical endpoints, usually determined from questionnaire responses, were validated with physical examination and/or medical records in 56% of preschool studies and 30% of school-age studies. The way in which 21 predetermined potential confounding variables were treated in both sets of studies was also examined. The average number considered per study ranged between 7.4 and 8.5 for both sets of studies. In preschool studies the most frequently considered potential confounding variables were socioeconomic status, age, gender, a subject’s health, family health, and family size (60–95% of the studies). In the school-age studies they were socioeconomic status, age, and gender (68–82% of the studies) and less frequently infant feeding, day care, stress, season, quality of housing, and nutrition (less than 20% of the studies). Several variables were identified as potential risk factors on the basis of relatively consistent associations with respiratory endpoints. In preschool children these were family health history, subject’s health history, heating type/presence of air conditioning, young age, maternal smoking during pregnancy, season, low birthweight, and stress. In school-age children, such potential risk factors were family health history, subject’s health history, heating type/presence of air conditioning, active smoking by the subject, and stress.