Review : Parental Smoking and Middle Ear Disease in Children: A Review of the Evidence

Abstract

This report presents a detailed review of the available epidemiological evidence concerning the prevalence of middle ear infections in children in relation to their expo sure to smoking by their parents or other household members and is based on English-language papers cov ering the period 1979-1998. Fifty-eight studies are in cluded in the review, with the most widely used index of exposure being parental smoking. Smoking by all house hold members, by either parent separately and by the mother during pregnancy were also considered in some of the studies. Only 3 studies attempted to quantify expo sure objectively, using measurements of cotinine and nicotine in hair, saliva and pericardial fluid. Tables listing unadjusted and adjusted relative risk estimates were constructed for both postnatal exposure to tobacco smoke and for maternal smoking during pregnancy. The results for postnatal exposure were separated into five disease categories. The overall data suggested some association of postnatal exposure to tobacco smoke with three categories; recurrent otitis media, otitis media with effusion and unspecified middle ear disease, with risk tending to increase with increasing exposure, but no association with acute otitis media or with persistent oti tis media with effusion. However, reported associations were generally quite weak, with relative risks typically below 1.5. Problems in interpreting the data arise from the difficulty of distinguishing possible effects of envi ronmental tobacco smoke and of maternal smoking dur ing pregnancy (for which the data also suggest an asso ciation), the possibility that the underlying cause of mid dle ear disease is in fact an infection, and from the var ious possibilities of bias inherent in epidemiological studies of a weak association, particularly due to inade quate control of confounding. Though it is prudent for parents not to expose their children to high doses of environmental tobacco smoke for long periods, the over all epidemiological evidence does not convincingly dem onstrate that exposure to environmental tobacco smoke increases the incidence of middle ear diseases in chil dren.