Validation of a Digital GOSLON Yardstick for Outcomes Assessments: A Comparison of Novice and Expert Raters

ObjectiveTo evaluate the validity, reliability, and usability of a newly digitized GOSLON Yardstick model set for unilateral cleft lip and palate (UCLP) outcome assessment. It was hypothesized that digital ratings would demonstrate reliability comparable to traditional plaster models and support consistent outcome comparisons across centers.DesignRetrospective secondary analysis of dental casts. Raters were blinded to patient and institutional identity. Calibration sessions were held prior to assessment. Reliability, workload, and usability were evaluated using validated instruments.SettingOne craniofacial center.ParticipantsThirty-eight Caucasian UCLP patients in mixed dentition, with primary surgical care at a single institution. Seven orthodontic residents (novices) and 5 craniofacial orthodontists (experts) completed all ratings.InterventionsPlaster casts were digitized using a 3Shape TRIOS scanner and uploaded to Sketchfab. Novices rated both formats across 2 sessions post-calibration. Experts completed 2 digital-only sessions after virtual calibration.Main Outcome MeasuresIntra- and inter-rater reliability (Cohen's kappa), GOSLON score distribution, task workload (NASA-TLX), and system usability (SUS).ResultsDigital GOSLON ratings showed moderate-to-high reliability (expert kappa: 0.82-0.86; novice: 0.75-0.81). No significant differences were found between plaster and digital scores for novices. Experts assigned more GOSLON 3 scores; novices assigned more GOSLON 4 (P < .001). Experts preferred digital models; novices preferred plaster. Workload and usability scores were acceptable across both groups.ConclusionsEven though calibration remains critical for novice raters, the Digital GOSLON is a reliable tool for outcomes assessment and may facilitate inter-center comparisons.