Assessing the construct validity of the Farra Eye Model for capsulorhexis training in cataract surgery: a pilot study

Capsulorhexis is a critical and technically demanding step in cataract surgery where slight errors can cause serious complications, making repeated simulation-based training essential. Limited access to hands-on surgical training has increased reliance on simulators, but existing models are often costly and limited. To address this, the Farra Eye Model was developed as a low-cost, reusable 3D-printed simulator offering realistic eye simulation. This study aims to assess its construct validity in distinguishing between different levels of surgical expertise.

A cross-sectional pilot study was conducted with 33 participants (11 novices, 11 intermediates, and 11 experts) grouped by cataract surgery experience. Each performed three capsulorhexis attempts on the Farra Eye Model, with the second and third attempts recorded. Two blinded consultants independently assessed the masked videos using the capsulorhexis indices in the ICO-OSCAR: Phaco assessment tool, along with forceps grabs and procedure duration. Construct validity was tested using one-way ANOVA or Kruskal-Wallis tests, and inter-rater reliability was measured using intraclass correlation coefficients (ICC), with values > 0.7 considered satisfactory.

Intermediate and expert groups outperformed novices in capsulorhexis, with higher ICO-OSCAR scores, fewer forceps grabs, and shorter procedure durations (p < 0.001). No statistically significant differences were found between the intermediate and expert groups (p > 0.05). Inter-rater reliability was excellent for all parameters (ICC > 0.7).

The Farra Eye Model can distinguish between novice and experienced surgeons (intermediate and expert groups), demonstrating construct validity. However, the inability to distinguish between intermediate and expert performance suggests limited sensitivity to advanced skill levels, highlighting the need for further refinements.