Authors' response

Abstract

loss, the other 3 reasons were directly implicated with the thickness of the retainers. We assumed that this concern was because retainer loss may confound the comparison of failure of retainers with different thicknesses between the 2 groups. However, retainer loss is indeed a reason for failure, and we could not exclude it. To eliminate its potential confounding effect, we compared not only failure of retainers, but also each independent reason (fracture, retainer loss, no longer fitting, and serious abrasion) between the 2 groups. Last, Dr Goje was concerned about the strength of the statistical significance indicated with a P value of 0.028 for retainer fracture and suggested that we should provide the confidence interval for that index. Our results showed that the 95% confidence interval of the odds ratio for retainer fracture ranged from 1.057 to 26.607. This confidence interval is clearly above 1 and indicates a significant difference between the 2 groups, consistent with the P value of 0.028. Thus, we can confirm that fracture rates were significantly higher in the 0.75-mm group than in the 1.00-mm group, not by chance. We appreciate the feedback on our research and hope our response has fully addressed the concerns.