Inter-observer agreement of soft contact lens fit assessment

Purpose: Soft contact lens fit assessment and recording varies among clinicians. The aim of this study is to compare the inter-student variation in contact lens grading using two scales, a traditional scale (using mm or % estimation) and a three point grading scale. Methods: Forty-four second year optometry students with 10 weeks of contact lens experience participated in this study evaluating 6 different lens fits of daily and reusable commercially available soft contact lenses. Students evaluated up-gaze blink, push-up (displacement) and horizontal lag and were randomly assigned to use the traditional or three point grading scale. Students were briefed prior to the study to ensure scale interpretation consistency. The study was time-controlled and after a 3-minute slit-lamp set-up, students received 1-minute for the measurement of each aspect of the fit. Results: Inter-observer agreement was assessed using: a) concordance and 95% confidence intervals of paired observations and b) intra-class correlation (ICC) statistic. The three-point scale showed a higher concordance than the traditional scale for all but the horizontal lag measurement (up-gaze blink 32% vs 11%, push-up 43% vs 11% and horizontal lag 36% vs 43%). The 95% confidence intervals for the traditional scale were narrower than that of the three-point scale (up-gaze blink 0.4mm vs 1.8, push-up 43% vs 1.8 and horizontal lag 0.3mm vs 1.9). The ICC was higher for the traditional scale (horizontal lag 0.42 vs 0.28 and push-up 0.49 vs 0.20) but slightly lower than the three-point scale for the blink in up-gaze (0.77 vs 0.88). Conclusions: Overall, the three point scale showed higher concordance indicating higher agreement between randomly paired student observations. However, the 95% confidence interval was wider as compared to the traditional scale reducing the ability to detect clinical change. Thus, our results suggest that the traditional scale is preferable during Optometrist’s training.