The initial study population meeting the inclusion criterion consisted of 2,510 patients. After exclusion criteria were applied, a total of 2,064 patients were included for analysis. The mean age of the study population was 9.63 ± 3.68 years in 2019, 9.93 ± 3.68 years in 2020, and 10.47 ± 3.74 years in 2021.
As shown in Fig. 1, The overall cohort (n = 2,064) had a mean SE of 0.12 ± 3.70 D in 2019, -0.07 ± 3.95 D in 2020, and − 0.49 ± 3.85 D in 2021. A greater change in mean SE was observed between 2020 and 2021 compared to between 2019 and 2020. When grouped by sex, males (n = 977) had a mean SE of -0.069 ± 3.705 in 2019, 0.087 ± 3.973 in 2020, and − 0.288 ± 3.799 in 2021. Females (n = 1,087) had a mean SE of 0.280 ± 3.696 in 2019, -0.200 ± 3.924 in 2020, and − 0.677 ± 3.898 in 2021. There was no significant difference between males and females in 2019 (p = 0.781) or 2020 (p = 0.556), or 2021 (p = 0.491).
A subset analysis was conducted for patients who had a cycloplegic refraction in all three years of the study 2019, 2020, and 2021 (Fig. 1). 109 patients (mean age 10.0 ± 3.51 years in 2021) returned in the same months for all three years. The mean SE decreased from 0.66 ± 3.40 D in 2019 to 0.51 ± 3.58 D in 2020 and 0.22 ± 3.72 D in 2021. On repeated measures ANOVA, there was a significant difference in myopic shift between years (F-ratio = 14.4, p < 0.00001), and post-hoc testing with Bonferroni correction demonstrated a significant change from 2020 to 2021 (p = 0.00008) but not from 2019 to 2020.
These three-year return patients were further grouped by patients who were initially myopic (SE ≤ -0.5 D) in 2019 (Fig. 1). 41 myopic patients (mean age 11.3 ± 3.58 years in 2021) were followed for all three visits. Mean SE decreased from − 2.58 ± 2.12 D in 2019 to -2.82 ± 2.36 D in 2020 and − 3.25 ± 2.51 D in 2021. On repeated measures ANOVA, there was a significant difference in myopic shift between years (F-ratio = 7.49, p = 0.001), and post-hoc testing with Bonferroni correction demonstrated a significant change from 2020 to 2021 (p = 0.0037) but not from 2019 to 2020.
Analysis by refractive categories (moderate to high myopia, low myopia, emmetropia, hyperopia) was conducted for the three-year return patients cohort (Table 1, Fig. 2). All refractive categories demonstrated a lowering of mean SE between 2020 and 2021. A greater change was observed between 2020 and 2021 compared to between 2019 and 2020 in all refractive categories except for in the emmetropic patients. The greatest myopic shift occurred in the low myopic patients, who displayed an increased rate of change in mean SE of 0.337 D in 2020 to 2021 compared to baseline.
Table 1
Mean spherical equivalent (SE) of all patients grouped by refractive error categories in the years 2019–2021.
Category | n | 2019 | 2020 | 2021 |
Hyperopia | 34 | 4.61 ± 1.65 | 4.59 ± 1.86 | 4.30 ± 2.08 |
Emmetropia | 34 | 0.61 ± 0.68 | 0.45 ± 0.93 | 0.34 ± 1.10 |
Low myopia | 28 | -1.49 ± 0.76 | -1.63 ± 1.42 | -2.10 ± 1.64 |
Moderate + High myopia | 13 | -4.94 ± 2.20 | -5.39 ± 1.87 | -5.71 ± 2.32 |
Refractive error categories: high myopia (≤-6.00 D); moderate myopia (-6.00 D to -3.00 D]; low myopia (-3.00 D to -0.50 D]; emmetropia (-0.50 D to + 2.00 D); hyperopia ( ≥ + 2.00 D). |
Further analysis was conducted according to zip codes of residence, with patients grouped by those in the CPS (urban) district (n = 779) and those in non-CPS (suburban) districts (n = 1285). The urban cohort consisted of 382 males and 397 females, while the suburban cohort consisted of 595 males and 690 females. The mean age of the urban cohort was 9.64 ± 3.72 in 2019, 10.04 ± 3.58 in 2020, 10.34 ± 3.72 in 2021, while the mean age of the suburban cohort was 9.63 ± 3.66 in 2019, 9.86 ± 3.74 in 2020, and 10.56 ± 3.75 in 2021. There was no significant differences in sex and age between the urban cohort and the suburban cohort. The mean SE of the urban cohort was − 0.23 ± 3.38 in 2019, -0.27 ± 3.63 in 2020, and − 0.56 ± 3.47 in 2021. For the suburban cohort, the mean SE was 0.35 ± 3.88 in 2019, 0.04 ± 4.12 in 2020, and − 0.45 ± 4.08 in 2021. There was a significant difference (p = 0.039) between the two cohorts in 2019, but not in 2020 and 2021.
Myopia Prevalence
The prevalence of myopia for every year in the overall cohort was recorded and stratified by all ages 2–17 as displayed in Table 2. The greatest increases in prevalence from 2020 to 2021 were observed in ages 8 and 17, increasing by 25% in both age groups. From 2019 to 2020, prevalence decreased from 28–27% in 8 year olds, and from 79–55% in 17 year olds.
Table 2
Prevalence of myopia (SE≤-0.50 D) in the overall cohort (n = 2,064) of all patients aged 2 to 17 years from 2019 to 2021.
Age (years) | n | 2019 | 2020 | 2021 |
2 | 1 | 0.00 | - | - |
3 | 17 | 0.27 | 0.50 | - |
4 | 62 | 0.30 | 0.28 | 0.33 |
5 | 293 | 0.29 | 0.21 | 0.29 |
6 | 295 | 0.24 | 0.26 | 0.31 |
7 | 290 | 0.36 | 0.26 | 0.37 |
8 | 242 | 0.28 | 0.27 | 0.52 |
9 | 212 | 0.27 | 0.45 | 0.38 |
10 | 196 | 0.49 | 0.47 | 0.47 |
11 | 209 | 0.53 | 0.52 | 0.53 |
12 | 182 | 0.51 | 0.65 | 0.55 |
13 | 151 | 0.63 | 0.56 | 0.59 |
14 | 141 | 0.50 | 0.64 | 0.62 |
15 | 120 | 0.711 | 0.68 | 0.74 |
16 | 90 | 0.63 | 0.75 | 0.72 |
17 | 86 | 0.79 | 0.55 | 0.80 |
Patients were further grouped into education levels (Table 3, Fig. 3): preschool (2–5 years), elementary school (6–9 years), middle school (10–13 years), and high school (14–17 years). All education levels demonstrated an increase in myopia prevalence from 2020 to 2021 except for middle school patients, which remained the same at 54%. The greatest increase in prevalence was observed in elementary school patients (10%).
Table 3
Prevalence of myopia (SE≤-0.50 D) in all patients grouped by school grade in the years 2019–2021.
School Grade (years) | n | 2019 | 2020 | 2021 |
Preschool (2–5) | 373 | 0.28 | 0.23 | 0.29 |
Elementary school (6–9) | 1039 | 0.29 | 0.29 | 0.39 |
Middle school (10–13) | 738 | 0.523 | 0.54 | 0.54 |
High school (14–17) | 437 | 0.63 | 0.65 | 0.71 |