Otitis Media and Compliance

To the Editor.— The recent study of otitis media by Schwartz et al,1 has basic methodologic problems that call into question the validity of the conclusions. The first relates to the definition of otitis media with effusion (OME). Criteria in the study have not been validated, and (more importantly) only 60% of the diagnoses of OME were confirmed by tympanometry. Did the other 40% really have OME? Would analysis using only children whose diagnosis was confirmed by tympanometry have led to the same results?