In the recently published study by Kumar and associates, the authors compare the postoperative outcomes of eyes with accidental intraoperative inversion of implantable collamer lens (ICL) with fellow eyes with a normal implantable collamer lens. Two sets of comparisons are presented in the article: (1) within eyes over successive time points and (2) between eyes at specific time points. Regarding the former, the authors applied the correct statistical test to account for the repeated measurements on the same eye at different time points. However, regarding the latter, they did not appreciate the need to account for the clustering of eyes within the same individual. When comparing reinverted and fellow eyes at each time point, they describe using the Mann–Whitney U test, which is not appropriate because it fails to account for the clustering effect. Assuming the first normal and reinverted eye data that are listed in Table 1 correspond to the same individual, we conducted a reanalysis of the study data using the Wilcoxon signed-rank test and provided readers with the correct P values in the Table below. There was no significant difference between the 2 groups with respect to uncorrected visual acuity, central corneal thickness, or anterior chamber depth. Although this does not amend the interpretation, it is still imperative to use appropriate statistical techniques that produce correct P values.
