Assessment of Visual Function and Performance in Paralympic Judo for Athletes with Vision Impairment.

SIGNIFICANCE: Paralympic judo currently requires all athletes to compete against each other in one class irrespective of their level of vision impairment (VI). Recent evidence suggests that multiple classes are required to enhance fairness, yet it remains unclear how many classes are necessary and what vision tests should be used to define those classes. PURPOSE: The aim of this study was to quantify the relationship between vision and performance in judo for individuals with VI. The results were expected to inform the development of evidence-based criteria to structure Paralympic judo competition. METHODS: The visual function of 53 elite VI judokas was assessed using a test battery that included tests of visual acuity (VA), contrast sensitivity, light sensitivity, depth perception, motion perception, visual search, and central visual field. Performance was assessed by measuring the ratio of fights won across all competitions the participants took part in in the 2 years before and after vision testing. Pearson correlation coefficients and decision tree analyses were used to determine the relationship between vision and performance. Partial correlations were also conducted to determine the unique ability of each measure of visual function to predict judo performance. RESULTS: Visual acuity was the best predictor of judo performance and remained the only visual function related to performance when controlling for correlations between VA and other visual functions. Decision tree analyses suggested to split athletes into two groups for more legitimate competition, using a cutoff point around 2.6 logMAR. Within each of the two resulting subgroups, no correlations remained between any of the visual functions and performance. CONCLUSIONS: The results of this study suggest that VI judo competition should be split into separate categories for partially sighted and functionally blind athletes. The inclusion of visual functions in addition to VA does not improve the ability to predict performance in VI judo.