Purpose:
The SureSightTM Autorefractor (Welch Allyn, Skaneateles Falls, NY) uses wavefront analysis to measure refractive error. The manufacturer recommends using the instrument in "adult" mode for testing of adults and for testing of children under cycloplegia. "Child" mode, which adds a accommodation correction factor (value not specified by the manufacturer), is recommended for testing of children 10 years or younger who are not cyclopleged and who are known not to be myopic. The present study examines whether the correction factor added in "child" mode is consistent for different magnitudes of sphere when refractive error is measured under standardized conditions (model eye).
Methods:
The SureSightTM (Software Version 2.20) was used to make 20 repeated readings of a model eye (+20 D lens power) in 6 simulated refactive error settings. At each setting of the model eye, 20 readings were obtained first in adult mode, and then in child mode. Sphere values were analyzed, and a small astigmatism arising from tilt of the model eye lens was ignored. The means and standard deviations were calculated for each group. A linear model was then fit to the data to determine the least-squares offset between child and adult modes across all measured refractions.
Results:
A linear model estimate of a single child mode offset across all values of simulated refractive error was 1.96 D (95% CI 1.89 to 2.02, p< 0.001).
Conclusions:
These findings are not consistent with literature reports that a consistent correction factor of +2.50 D is added to measurements of sphere obtained using the "child" mode. Our data suggest that this rule is applied to hyperopic measurements, but for myopic measurements, a lesser correction factor is applied.
Keywords: astigmatism • refraction • imaging/image analysis: clinical