We have continued to analyze national data sets to determine the prevalence of hearing impairment in children and associations with various risk factors. Our main source of data for school age children (6 to 19 years old) is the Third National Health and Nutrition Examination Survey (NHANES III), 1988-1994. In this study, air-conduction audiometric thresholds were obtained at .5, 1, 2, 3, 4, 6, and 8 kHz frequencies for 5,915 children in randomly-selected communities across the United States. African-American and Hispanic (Mexican-American) children were oversampled to insure sufficient numbers for race and/or ethnic-specific prevalence rates. Pure-tone averages (PTA) based on thresholds at .5, 1, and 2 kHz (lower frequencies), and at 3, 4, and 6 kHz (higher frequencies) were computed for the better and worse ear responses. Hearing outcomes appropriate for children were defined as: a) PTA less than 15 dB (normal range), b) PTA equal to or greater than 15 dB (slight/mild or greater HL), and c) PTA equal to or greater than 25 dB (moderate or severe HL). Also, multivariately adjusted odds ratio (OR) estimates of relative risk and 95% confidence intervals (CI) were calculated using logistic regression models. The prevalence of hearing loss (HL) equal to or greater than 15 dB in the better ear in the low frequencies was 2.1%. The prevalence of HL for Mexican- American children was significantly higher. Also, younger children aged 6-8 years had an increased prevalence of HL. The prevalence of HL equal to or greater than 25 dB in the low frequencies was 2.9 per 1,000 children. For children with fair/poor general health, this prevalence estimate was significantly increased to 14.6 per 1,000 (OR=5.2; 95% CI, 1.5-18.5). Overall, we found the prevalence of bilateral (better ear) low frequency hearing impairment increased with young age, low family income, Mexican-American ethnicity, and fair/poor childrens health status. These findings illustrate the need to identify HL early and develop effective interventions to insure the development of communication skills needed in school and later in life. Additional studies using NHANES III have examined the reliability of the hearing threshold determinations and, also, measures obtained to assess the middle ear status (tympanograms and acoustic reflex). We have recently begun analyses of hearing levels for children with normal (type A) tympanograms in both ears versus those with left-shifted tympanograms (type C) and/or flat tympanograms (type B) in one or both ears. As predicted, children with normal tympanograms in both ears have lower mean thresholds in both the low and high frequency ranges. Using a tympanogram width criterion, we have shown that children with flat tympanograms have higher hearing thresholds, with a median level between 10 to 15 dB and a 95th percentile of 35 to 45 dB. These children most likely have otitis media with effusion (OME) and may have suffered from a persistent conductive hearing loss for a long time. These analyses suggest that OME may account for a considerable portion of the differences in hearing levels by age, race/ethnicity, and other factors. We are also collaborating with NIEHS and NICHD on a study of in-utero exposure to organochlorine compounds in relation to health effects in children, utilizing the US Collaborative Perinatal Projects stored blood specimens and health data, including hearing thresholds obtained on the children at age 8. We also collaborated with the NICHD Newborn Network on the analysis of childrens hearing status at 2 years of age, following treatment in the NINOS clinical trial at birth with inhaled nitric oxide to improve oxygenation and reduce the need for ECMO in near-term hypoxic neonates. A paper describing results of this study has beensubmitted for publication.
