Reflectometric screening for otitis media: inconsistencies in a sample of Australian aboriginal children

This paper evaluates reflectometry, or acoustic otoscopy, as an objective test for the mobility of intact tympanic membranes in Australian Aboriginal children, who are a population at high risk for otitis media. Reflectometry, pneumotoscopy and tympanometry were compared in 395 ears in 198 Aboriginal children living in remote communities in Northern Australia. A strong interaction was found between pneumotoscopy and tympanometry (P < 0.001), there was some interaction between pneumotoscopy and reflectometry (breakpoint 5.0) (P < 0.05) and there was no significant interaction between reflectometry and tympanometry. Results showed that intra-observer and inter-observer agreement for measures of reflectivity were 0.87 and 0.83, respectively. Unexpected differences between reflectivity measures for right (mean 5.4 units) and left (mean 4.9 units) ears and for different test locations were attributed to procedural artifact or training effects. It was concluded that, despite its practical advantages, reflectometry cannot be recommended as a screening tool for tympanic membrane (TM) immobility in this population.