Objective: To describe the myopia prevalence and spherical equivalent among primary school students in a district of Anhui Province in 2019, to analyze their changes in 2020 and 2024, to obtain baseline data, and to explore the potential impact of changes in screen time on vision, providing scientific evidence for myopia prevention and control strategies. Methods: This cross-sectional study included all the primary school students who underwent vision screening in 2019 (baseline period), 2020 (period of increased screen time), and 2024 (period of slightly resumed screen time). Data on grade, gender, uncorrected distance visual acuity, and refraction were extracted from the screening database. Noncycloplegic refraction was assessed using an FKR-710 autorefractor to measure sphere and cylinder, and uncorrected distance visual acuity was assessed using an K238-AZ LCD visual acuity chart. The spherical equivalent was calculated as sphere + 0.5 × cylinder. Myopia was defined as uncorrected distance visual acuity < 5.0 in either eye with a corresponding spherical equivalent ≤ -0.50 D. Myopia prevalence and spherical equivalent were described using percentage and mean with 95%confidence interval (CI). The Chi-square test and One-way analysis of variance were used to compare myopia prevalence and spherical equivalent across different years, followed by stratified analysis by grade. Results: A total of 183 204 students were screened: 45 294 in 2019, 52 922 in 2020, and 84 988 in 2024. In 2019, the myopia prevalence was 30.8%(95%CI: 30.4%, 31.2%), higher in girls (33.7%) than in boys (28.4%) and increased with grade (P < 0.001), rising from 11.0%in Grade 1 to 35.7%in Grades 2-6. Compared with 2019, the myopia prevalence in 2020 increased by 4.3 (3.7, 4.9) percentage points to 35.1%(P < 0.001). By 2024, it further increased by 6.7 (6.2, 7.2) percentage points from the 2020 level to 41.8%(P < 0.001). In grade-stratified analyses, compared with 2019, Grade 1 students showed the largest increase in 2020; compared with 2020, Grade 1 students showed the smallest subsequent increase in 2024. Quantitative analysis of spherical equivalent showed that the mean spherical equivalent in 2020 was -0.72 D, representing a myopia shift of -0.21 D (-0.23 D, -0.19 D) compared with the 2019 level (P < 0.001); by 2024, the mean spherical equivalent was -0.68 D, indicating a hyperopic shift of 0.04 D (0.02 D, 0.06 D) compared with the 2020 level (P < 0.001). Further grade-stratified analyses indicated that the mean spherical equivalent for Grade 1 students in 2024 (0.08 D) had nearly returned to the 2019 level (0.09 D). Conclusion: The myopia prevalence among primary school students in the study area was 30.8%in 2019, higher than the global level and approximately twice that of the United States and four times that of Germany, indicating that myopia prevention and control among primary school students remains challenging. The myopia prevalence in 2020 and 2024 was higher than in 2019, suggesting that lifestyle changes, such as increased screen time, affect the short-term visual status. Younger students experienced the greatest increase in myopia prevalence in 2020 and the mildest subsequent increased in 2024, and their spherical equivalent in 2024 had almost recovered to the 2019 level, but not for older students. These findings indicate greater ocular plasticity in younger students, i.e. more sensitive to negative exposures and easier to recover after removing such exposures, warranting further investigation on the underlying mechanisms. Additionally, patterns of higher myopia prevalence in girls than in boys and a gradual increase with grade support previous studies, also enhancing the reliability of the relevant results in our study.