BACKGROUND: About 3% of infants are born with a tongue-tie which may lead to breastfeeding problems such as ineffective latch, painful attachment or poor weight gain. The Hazelbaker Assessment Tool for Lingual Frenulum Function (HATLFF) has been developed to give a quantitative assessment of the tongue-tie and recommendation about frenotomy (release of the frenulum). The aim of this study was to assess the inter-rater reliability of the HATLFF. METHODS: Fifty-eight infants referred to the Breastfeeding Education and Support Services (BESS) at The Royal Women's Hospital for assessment of tongue-tie and 25 control infants were assessed by two clinicians independently. RESULTS: The Appearance items received kappas between about 0.4 to 0.6, which represents "moderate" reliability. The first three Function items (lateralization, lift and extension of tongue) had kappa values over 0.65 which indicates "substantial" agreement. The four Function items relating to infant sucking (spread, cupping, peristalsis and snapback) received low kappa values with insignificant p values. There was 96% agreement between the two assessors on the recommendation for frenotomy (kappa 0.92, excellent agreement). The study found that the Function Score can be more simply assessed using only the first three function items (ie not scoring the sucking items), with a cut-off of =4 for recommendation of frenotomy. CONCLUSION: We found that the HATLFF has a high reliability in a study of infants with tongue-tie and control infants.