Significance: Autorefractors are commonly used by eye care practitioners worldwide as a starting point for clinical prescribing and by researchers as an instrument to study development of refractive errors and accommodation. This study demonstrates that the Nidek ARK-1 provides a reasonable and repeatable estimate of refractive error.
Purpose: The purposes of this study were (a) to compare refraction measurements of the Nidek ARK-1 (2016 release) autorefractor with that of subjective refraction and retinoscopy performed by an experienced clinician and (b) to determine the intratest and intertest variability of autorefraction measures taken using the ARK-1 autorefractor.
Methods: Sixty-seven adult patients aged 18 to 69 years underwent retinoscopy, subjective refraction, and ARK-1 autorefraction on a same day by a single clinician. A separate subset of 14 participants was invited for the repeatability and reproducibility study. Both eyes of each participant were included in the analysis.
Results: A statistically significant (but not clinically significant) positive spherical difference was observed between the ARK-1 and subjective refraction (P = .003). Spherical equivalent refractive errors were statistically similar between the ARK-1 and subjective refraction (P = .20). A statistically and clinically significant difference was observed in the cylindrical component between the ARK and subjective refraction (P < .01). No statistically significant difference was observed between the ARK and subjective refraction in both the horizontal (J0; P = .08) and oblique cylindrical vector (J45; P = .96). Bland-Altman analysis revealed that the 95% limits of agreement were widest between the ARK and subjective refraction in all of the refractive components (-0.60 to 0.89 diopter for spherical component, -0.80 to 0.69 diopter for spherical equivalent, and -0.98 to 0.30 diopter for cylindrical component). The intertest and intratest variability of the ARK-1 was small.
Conclusions: The Nidek ARK-1 autorefractor is a useful clinical tool that provides a reasonable and repeatable estimation of refractive error in adults.