The post-winter prevalence rate of middle ear effusion in four-year-old children, judged by tympanometry

Abstract Tympanometry and otologic examinations were performed in 373 four-year-old children randomly selected from the Central National Register at the end of February, 1979. A close interaural correlation of the middle ear pressures of a child was found and therefore the tympanometric results represent related as well as independent data. Consequently, the post-winter prevalence rates of type B tympanograms were calculated according to different sampling units: per all ears 13.7%; per right ears 12.9%; per left ears 14.5%; per ears selected at random 13.9%. Additionally, the prevalence rate was estimated per child at 20.4% type B in at least one ear and at 7% type B in both ears. Based upon the literature the validity of the tympanometric method in detecting middle ear effusion is described. The predictive values of positive and negative tests mostly depend on the degree of the cutoff point (pass—fail point) which has been chosen. In this study the cutoff point was the presence of a type B tympanogram. The advantage and disadvantage of applying ‘one ear’ or ‘one child’ as the sampling unit is discussed from a statistical as well as a clinical point of view. When calculating either per ear or per child no sex differences in the prevalence rates could be proved.