Measurement of hearing impairment among Greenlandic school-children: association between self-reported data and clinical examinations

Abstract

Abstract Background Multiple ear infections is causing hearing impairment among children all over the world and the health and social consequences track into early adolescence and later in life, if not treated. The monitoring of prevalence in a population is important to assess the need for interventions in a population. Methods One hundred eighty five children from 5 to 10th grade from Sisimiut town and the nearby settlements participated in a clinical examination to have ear-examination and pure tone audiometry. Participants filled out a questionnaire at home with their parents before the clinical examination, and hearing impairment was collected as individual self-reports and as audiometric measurements. Results A total of 185 children between 9 and 15 years of age (median: 11 years, IQR: 10–13) were included, 60% (n = 111) were girls. 247 (70%) of the 355 available otoscopies were clinically assessed as normal. Cohen’s Kappa coefficient was 0.31. Eighteen children (10%) were found to have hearing impairment. None of the children had hearing aids. Test performance for self-reports were that sensitivity was 56% and specificity was 87%. The predictive value of a positive test was 31%, and the predictive value of a negative test was 95%. 32 children (17%) reported hearing impairment to the extent that they were not able to keep up in school, of which half reported that it had lasted for more than one year. 7 of the 32 children reporting hearing impairment (22%) reported that the extent of their hearing impairment was affecting their classroom experience so they were not able to follow. Conclusion Self-reported and clinically screening for hearing impairment are two different concepts. Even though the two concepts are statistically correlated, the correlation coefficients are low. The test performance indicated that self-reported data might be measuring hearing as an experience in a social environment and not directly comparable to pure tone audiometry which examines hearing in controlled testing conditions. Since both measure hearing impairment, they supplement each other in research on impaired hearing, and the choice of measure should relate to the purpose and method of the investigation.

Keywords

• Hearing impairment
• Prevalence
• Measurement properties
• Methodology
• School-children
• Greenland