Diagnostic Accuracy of the LittlEARS Auditory Questionnaire Against Auditory Steady-State Response for Detecting Hearing Loss in Infants Aged 6–24 Months

Abstract

Introduction: Hearing loss is among the most common congenital conditions in childhood, yet most affected infants in Indonesia are identified late because objective audiometry is centralized and preverbal infants show few overt signs. This study aimed to determine the diagnostic accuracy of the parent-completed LittlEARS Auditory Questionnaire against the auditory steady-state response (ASSR) for detecting hearing loss in infants aged 6–24 months.

Methods: In this single-center, cross-sectional diagnostic-accuracy study, 28 children aged 6–24 months (median 9.0 months; 71.4% male) attending a tertiary otorhinolaryngology clinic were enrolled by consecutive sampling. Parents independently completed the Indonesian LittlEARS, and each child underwent ASSR as the reference standard (hearing loss >25 dB HL across 0.5–4 kHz). Diagnostic indices were computed with Wilson 95% confidence intervals (CI); agreement, likelihood ratios, receiver-operating-characteristic analysis, and Firth penalized logistic regression were performed.

Results: ASSR confirmed hearing loss in 19 children (67.9%), of whom 18 had bilateral loss. Referenced to ASSR, LittlEARS yielded a sensitivity of 89.47% (95% CI 68.6–97.1), specificity 77.78% (95% CI 45.3–93.7), positive predictive value 89.47%, negative predictive value 77.78%, and accuracy 85.71% (95% CI 68.5–94.3). Agreement was substantial (Cohen kappa 0.673), the positive likelihood ratio 4.03 and negative likelihood ratio 0.14, and the area under the curve 0.836 (95% CI 0.688–0.985). A suspected-hearing-loss result independently predicted ASSR-confirmed loss (adjusted odds ratio 13.98, 95% CI 1.68–116.25; p=0.015; Nagelkerke R²=0.469).

Conclusion: The LittlEARS questionnaire is an accurate, low-cost screening adjunct for early childhood hearing loss but, given moderate specificity, a positive screen should be confirmed by objective audiometry.