Ohlendorf Arne, Leube Alexander, Wahl Siegfried
Institute for Ophthalmic Research, University of Tuebingen, Geschwister-Scholl-Platz, 72074 Tübingen, Germany.
Carl Zeiss Vision International GmbH, 73430 Aalen, Germany.
Healthcare (Basel). 2016 Jul 13;4(3):41. doi: 10.3390/healthcare4030041.
To investigate the inter-device agreement and mean differences between a newly developed digital phoropter and the two standard methods (trial frame and manual phoropter).
Refractive errors of two groups of participants were measured by two examiners (examiner 1 (E1): 36 subjects; examiner 2 (E2): 38 subjects). Refractive errors were assessed using a trial frame, a manual phoropter and a digital phoropter. Inter-device agreement regarding the measurement of refractive errors was analyzed for differences in terms of the power vector components (spherical equivalent (SE) and the cylindrical power vector components J0 and J45) between the used methods. Intraclass correlation coefficients (ICC's) were calculated to evaluate correlations between the used methods.
Analyzing the variances between the three methods for SE, J0 and J45 using a two-way ANOVA showed no significant differences between the methods (SE: p = 0.13, J0: p = 0.58 and J45: p = 0.96) for examiner 1 and for examiner 2 (SE: p = 0.88, J0: p = 0.95 and J45: p = 1). Mean differences and ±95% Limits of Agreement for each pair of inter-device agreement regarding the SE for both examiners were as follows: Trial frame vs. digital phoropter: +0.10 D ± 0.56 D (E1) and +0.19 D ± 0.60 D (E2), manual phoropter vs. trial frame: -0.04 D ± 0.59 D (E1) and -0.12 D ± 0.49 D (E2) and for manual vs. digital phoropter: +0.06 D ± 0.65 D (E1) and +0.08 D ± 0.45 D (E2). ICCs revealed high correlations between all methods for both examiner (p < 0.001). The time to assess the subjective refraction was significantly smaller with the digital phoropter (examiner 1: p < 0.001; examiner 2: p < 0.001).
"All used subjective methods show a good agreement between each other terms of ICC (>0.9). Assessing refractive errors using different subjective methods, results in similar mean differences and 95% limits of agreement, when compared to those reported in studies comparing subjective refraction non-cylcoplegic retinoscopy or autorefraction".
研究一种新开发的数字验光仪与两种标准方法（试验框架和手动验光仪）之间的设备间一致性及平均差异。
两组参与者的屈光不正由两名检查者测量（检查者1（E1）：36名受试者；检查者2（E2）：38名受试者）。使用试验框架、手动验光仪和数字验光仪评估屈光不正。分析所用方法之间在屈光不正测量方面关于屈光力矢量分量（球镜等效度（SE）以及柱镜屈光力矢量分量J0和J45）差异的设备间一致性。计算组内相关系数（ICC）以评估所用方法之间的相关性。
使用双向方差分析对SE、J0和J45的三种方法之间的方差进行分析，结果显示检查者1和检查者2的各方法之间均无显著差异（SE：p = 0.13，J0：p = 0.58，J45：p = 0.96）（检查者2：SE：p = 0.88，J0：p = 0.95，J45：p = 1）。两位检查者关于SE的每对设备间一致性的平均差异和±95%一致性界限如下：试验框架与数字验光仪：+0.10 D ± 0.56 D（E1）和+0.19 D ± 0.60 D（E2），手动验光仪与试验框架：-0.04 D ± 0.59 D（E1）和-0.12 D ± 0.49 D（E2），手动验光仪与数字验光仪：+0.06 D ± 0.65 D（E1）和+0.08 D ± 0.45 D（E2）。ICC显示两位检查者的所有方法之间均具有高度相关性（p < 0.001）。使用数字验光仪评估主观验光的时间显著更短（检查者1：p < 0.001；检查者2：p < 0.001）。
“所有使用的主观方法在ICC方面（>0.9）彼此之间显示出良好的一致性。与比较主观验光、非睫状肌麻痹检影验光或自动验光的研究报告相比，使用不同主观方法评估屈光不正会产生相似的平均差异和95%一致性界限”。
