Passive smoking, salivary cotinine concentrations, and middle ear effusion in 7 year old children.BMJ 1989; 298 doi: https://doi.org/10.1136/bmj.298.6687.1549 (Published 10 June 1989) Cite this as: BMJ 1989;298:1549
- D. P. Strachan,
- M. J. Jarvis,
- C. Feyerabend
OBJECTIVE--To assess the contribution of passive exposure to tobacco smoke to the development of middle ear underpressure and effusion. DESIGN--Cross sectional observational study. SETTING--One third of the primary schools in Edinburgh. SUBJECTS--892 Children aged 6 1/2 to 7 1/2 were examined, and satisfactory tympanograms were obtained in 872. Results of assay of salivary cotinine concentrations were available for 770 children, and satisfactory tympanograms were available for 736 of these. END POINT--Correlation of the prevalence of middle ear underpressure and effusion with concentrations of the marker of nicotine, cotinine, in the saliva of the children. MEASUREMENTS AND MAIN RESULTS--Middle ear pressure and compliance were measured in both ears by impedance tympanometry. Salivary cotinine concentrations were assayed by gas-liquid chromatography. Cotinine concentrations increased with the number of smokers in the household. Girls had higher concentrations than boys, and children living in rented housing had higher concentrations than those living in housing owned by their parents. There was a trend towards more abnormal tympanometric findings with increasing cotinine concentration, the odds ratio for a doubling of the cotinine concentration being 1.14 (95% confidence interval 1.03 to 1.27). After adjustment for the sex of the child and housing tenure the odds ratio for a doubling of the cotinine concentration was 1.13 (1.00 to 1.28). CONCLUSIONS--The results of this study are consistent with those of case-control studies of children attending for an operation to relieve middle ear effusion. They indicate that the disease should be added to the list of recognised hazards associated with passive smoking. About one third of the cases of middle ear effusion in this study were statistically attributable to exposure to tobacco smoke.