Silent reading and secretory otitis media in school children.

In an unselected cohort of 366 8-year-old children, the relationship between secretory otitis media and reading achievement was investigated. The children underwent 10 impedance audiometries and 5 pure tone audiometries during their first year at school. At the beginning of the second grade they all had a Silent Reading Word Test (OS-400). The background parameters were recorded by an interview with one of the parents. There was a significant but small correlation between type B tympanograms in the first grade and silent word reading. No association between silent reading score and otological history or pure tone screening was found. In a stepwise multiple regression model, 37% of the variance could be 'explained' by the included variables. The 'classroom factor' could 'explain' about 17% of the variance, followed by phonology at the start of school (6%), gender (5%), social group of the mother (4%), type B tympanogram (2%), absence from school (2%) and allergy (1%).