Search for a command to run...
Abstract Topic This systematic review examines variability regarding measuring visual acuity (VA) and refractive error (RE). Clinical Relevance VA and RE are key factors that determine visual function, and although accurately measuring VA and RE is fundamental to clinical practice, no scientific consensus has been reached with respect to variability. This lack of consensus affects our ability to establish a limit of agreement (LoA) for clinically accepted variability and may subsequently affect the clinical decision-making process, regulations, and research in the field of ophthalmology. Methods This systematic review was performed in accordance with PRISMA guidelines and was registered at PROSPERO (ID: CRD42024554663). We searched the Medline, PubMed, and Embase databases. The Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool was then used to assess the risk of bias (RoB). We included studies that directly compared the accuracy of VA or RE and reported the LoA and mean difference expressed in logarithm of the minimum angle of resolution (LogMAR) for VA and diopters (D) for refraction. We then report the summarized mean LoA values with 95% confidence intervals (CIs). Results After applying the inclusion and exclusion criteria, our analysis included 13 and 6 studies that reported VA and RE measurements, respectively. From these 19 studies, only 6 (32%) scored a low RoB in at least three out of four QUADAS-2 domains. In 27 out of 38 subgroups (71%), LoA exceeded the generally accepted ranges for VA or RE (namely, ±0.15 LogMAR and ±0.5D, respectively). Based on our analysis, the summarized mean LoA values were ±0.20 LogMAR (95% CI: 0.17, 0.23) and ±0.70D (95% CI: 0.50, 0.89) for VA and RE measurements, respectively. Conclusions We found that both VA and RE measurements have relatively high variability, exceeding currently accepted limits. Thus, the currently accepted limits of this variability are either unrealistic or reflect studies with methodological weaknesses. Rigorous studies are therefore needed in order to more accurately estimate the acceptable limits of variability for both VA and RE measurements. For now, we propose using the mean LoA values obtained in our study as a provisional frame of reference. In addition, we discuss several aspects of understanding the variability that exists when measuring VA and RE, particularly with respect to comparing studies.