Background : School-based hearing screening is likely to be the first opportunity to identify childhood hearing loss in South Africa. Criteria for school-based hearing screening requires balancing the targeted degree of hearing loss while ensuring that referral rates are sufficiently low for a cost-effective and sustainable programme. The study aim was to investigate the effect of screening intensity (loudness) levels on the referral rate and to establish the effect of an immediate rescreen in reducing the referral rate. Methods : A within-subject study was conducted in two phases. Phase 1: compared the referral rate in a counterbalanced sequence at screening levels of 20 dB HL, 25 dB HL and 30 dB HL across 1, 2 and 4 kHz for 135 children. Phase 2: determined the effect of an immediate rescreen on referral rate for 337 children screened at 25 dB HL. If a further referral was obtained on rescreen, diagnostic audiometry was subsequently conducted. Results : Referral rate was reduced to 6.7% from 17% when using 25 dB HL as opposed to 20 dB HL as screening intensity. Referral rate was reduced to 4.4% when employing 30 dB HL as screening intensity. An immediate rescreen reduced the overall referral rate by more than one-third. Diagnostic audiometry confirmed that almost half (47%) of the referred children had a hearing loss. Conclusion : A screening intensity of 25 dB HL andimmediate rescreen reduces the referral rate significantly and will limit the burden of the screening programme on health care resources.