Effectiveness of universal newborn hearing screening: A systematic review and meta-analysis

BACKGROUND: Permanent bilateral hearing loss (PBHL) is a serious condition in newborns, with a prevalence of at least one per 1000 live births. However, there has been no recent systematic review and meta-analysis of the effectiveness of universal newborn hearing screening programs (UNHS). METHODS: We registered our study protocol on PROSPERO CRD42020175451. Primary outcomes were any identification of PBHL (ie, PBHL diagnosed at any time), age of identification of PBHL, and neurodevelopment. Two reviewers searched standard databases to March 2022 and extracted data. We used fixed and random effects meta-analysis to pool data and graded the certainty of evidence using standard methods. RESULTS: The search retrieved 2834 records. We identified five studies reporting on the effects of UNHS vs no UNHS in 1 023 610 newborns. The relative risk of being identified with PBHL before nine months in infants with UNHS compared to infants without UNHS was 3.28 (95% confidence interval (95% CI) = 1.84, 5.85, one study, 1 023 497 newborns, low certainty evidence). The mean difference in the age of identification of PBHL in infants with UNHS compared to infants without UNHS was 13.2 months earlier (95% CI = -26.3, -0.01, two studies, 197 newborns, very low certainty evidence). The relative risk of infants eventually being identified with PBHL in infants with UNHS compared to infants without UNHS was 1.01 (95% CI = 0.89, 1.14, three studies, 1 023 497 newborns, low certainty evidence). At the latest follow-up at 3-8 years, the standardised mean difference (SMD) in receptive language development between infants with UNHS compared to infants without UNHS was 0.60 z scores (95% CI = 0.07, 1.13, one study, 101 children, low certainty evidence) and the mean difference in developmental quotients was 7.72 (95% CI = -0.03, 15.47, three studies, 334 children, very low certainty evidence). The SMD in expressive language development was 0.39 z scores (95% CI = -0.20, 0.97, one study, 87 children, low certainty evidence) and the mean difference in developmental quotients was 10.10 scores (95% CI = 1.47, 18.73, 3 studies, 334 children, very low certainty evidence). CONCLUSIONS: UNHS programs result in earlier identification of PBHL and may improve neurodevelopment. UNHS should be implemented across high-, middle-, and low-income countries. REGISTRATION: PROSPERO (CRD42020175451)