Related Experiment Video
Updated: Jul 21, 2025

Comparison of Three Clinical Stereoscopic Methods for Measuring Binocular Visual Function During Amblyopic Treatment in Unilateral Amblyopia
Published on: September 27, 2024
Systematic review of simulation-based education in strabismus assessment and management
Niveditha Pattathil1, Christine C Moon1, Zahra Haq1
1School of Medicine, Faculty of Health Sciences, Queen's University, Kingston, Canada.
Background:
Strabismus is a dynamic condition for which simulation-based training is valuable, given the variable complexity and relatively reduced exposure compared with other ophthalmic presentations. This study assessed the performance of simulation models available for medical training in the assessment and management of strabismus.
Methods:
A systematic review of relevant peer-reviewed academic databases was conducted, without publication date restrictions. English-language publications evaluating the performance of simulation models for education on strabismus were included. Risk of bias was assessed using the Cochrane RoB-2 tool and CLARITY Risk of Bias Instrument for Cross-Sectional Surveys of Attitudes and Practices. Validity of evidence was evaluated using the Kirkpatrick framework.
Results:
Of the total 3,298 citations exported for title and abstract screening, 54 advanced to full-text screening, and 7 were included in final review. Model types were either dry (2), wet (4), or virtual reality (1). All models were deemed to be successful, but few standardized parameters were specified. Costs of models ranged from a few dollars (ball and wood), to moderate (non-cadaveric), to costly (virtual reality). All studies scored a moderate or high risk of bias, and the majority (4/7) of studies scored level 1 on the Kirkpatrick scale.
Conclusions:
Research on simulation for strabismus assessment and management is limited and varied for model fidelity and testing audiences. All models were deemed individually successful compared to non-simulation-based teaching methodologies, although no direct comparisons were made. The limited evidence available suggests that low-fidelity and low-cost models can be used for trainees without sacrificing educational quality.

