Development of a low cost screen to identify hearing loss in young children and appropriate services for deaf children in Binga, Zimbabwe

There is a high prevalence of hearing loss estimated between 8 and 16% in young children in rural areas in Zimbabwe. Deaf children are usually identified late and do not benefit from early interventions. This study was conducted to evaluate a questionnaire screen in identifying permanent hearing loss in excess of 50dBHL of the better ear, compared with the pure-tone audiometric screen, in children aged 36-72 months living in Binga district, Zimbabwe. Subjects were recruited into the study by using two questions which identified 417 at-risk children who were registered as "Failing" children. In addition, 417 children were recruited as controls who were matched by age and sex and were registered as "Following" children (n=834). The "Questionnaire" screen used for this study had two parts; "Part I" had 8 general questions for every child, while "Part 2" had 3 age specific sections (A, B and C) with a set of 10 questions in each section and took between 20-25 minutes to administer for each child. The screen was administered on 747 (90%) children of the original sample (n=834) by four trained Tonga fluent interviewers, 87 children (10%) having dropped out. An experienced audiologist administered the gold standard pure-tone audiometry screen on the same 747 (90%) children who were "Questionnaire" screened. There were four test sound frequencies used: 0.5k, lk, 2k and 4k set at a flat cut-off point of 50dBHL generated by a calibrated Kamplex screening audiometer. The pure-tone screen administration required 10 - 15 minutes per child. Pass or fail results were plotted on the audiogram. For reliability testing of both the "Questionnaire" and the pure-tone screens, repeats were administered on 131 and 110 children respectively who were randomly selected from 747 children. Children with marked physical malforination, neurological problems and those for whom Tonga was not their first language were excluded from the study. The results revealed that the "Questionnaire" screen had a sensitivity of 79% and specificity of 96%. It was inter-and intra-user reliable (r---0.89) in identifying permanent hearing loss in children aged 36-72 months (p>0.05). The "Questionnaire" was easy to use and found to be a low-cost screen that can be appropriately adapted and used in service delivery or research programmes in different cultural settings in developing countries