Comparison between estimated and measured myopia progression in Hong Kong Children without myopia intervention.

To the Editor, We read with interest the original article “Comparison between estimated and measured myopia progression in Hong Kong children without myopia control intervention” by Yang et al. The authors compared myopia progression estimates for children without myopia management with the BHVI myopia calculator (bhvi.org/myopiacalculatorresources/) to rates measured in approximately 216 Hong Kong children wearing single vision spectacles (using spherical equivalent refraction (SER) data). They graphically estimated the 95% confidence limits to ascertain if the measured values fell within the calculator's range. They concluded that: (a) their measured cycloplegic SER was in close agreement to the estimated SER from the calculator, but (b) the measured individual progression fell within the 95% confidence intervals (CI) for only 32– 38% of children, and the remainder fell equally above and below the limits. They concluded that that the data should be interpreted with caution and that inclusion of additional data and larger sample sizes should be considered. We disagree with their method of applying confidence intervals to estimate individual progression and hence their conclusions for the reasons explained below. Childhood myopia is progressive, and there is substantial evidence on the benefits of reducing the risk of the eye reaching higher levels of myopia. Many of the current myopia control strategies slow but do not arrest myopia. The BHVI myopia calculator was designed to be a simple tool to illustrate the progressive nature of myopia, and to help practitioners demonstrate to patients and parents the benefits of myopia control treatments. The calculator was not designed to monitor the progression of myopia for each eye of an individual. As the authors rightly suggest, such evaluation of progression may require more than one parameter. The calculator's “without management” estimates for the Asian population was derived using data from 4504 myopic eyes, and provides the mean refractive error with increasing age (each year) and the 95% CI for the mean refractive error. Thus, the CI is the range within which the “mean refractive error” for a given age will lie. The fact that the authors report that the mean refractive error values are closely aligned with the “without management” values proves that the calculator provides a close approximation of the mean refractive error at that age, and indicates that the calculator is functioning as intended. In this regard, increasing the sample size as the authors suggest will likely narrow and not expand this CI. For diverse populations, Published online: 7 January 2022