Background <p>To investigate the impact of spectacle correction on visual development and axial length (AL) in children with moderate to high astigmatism.</p> Methods <p>A nested cohort study analyzed 82 first-grade students (age: 7.13 ± 0.36 years) with cylindrical power ≥ 1.25 diopters (D) and best corrected visual acuity (BCVA) ≤ 0.1 (logMAR) from the Anyang Childhood Eye Study. Uncorrected visual acuity (UCVA), BCVA, cycloplegic autorefraction and AL were measured at baseline, 1-year and 2-year follow-up. Based on whether parents provided spectacle correction for their children, children were categorized into the spectacles group (<i>n</i> = 17) and non-spectacles group (<i>n</i> = 65), which was stratified into tertiles by cylindrical power (C): 1.25 D ≤ C &lt; 1.50 D (<i>n</i> = 18), 1.50 D ≤ C &lt; 1.75 D (<i>n</i> = 21), C ≥ 1.75 D (<i>n</i> = 26).</p> Results <p>Compared to the non-spectacles group, the spectacles group exhibited a significantly greater myopic shift in spherical equivalent refraction (<i>P</i> = 0.006) and longer AL (<i>P</i> &lt; 0.05) by the 2-year follow-up. Conversely, the non-spectacles group demonstrated a significant reduction in cylindrical power (<i>P</i> &lt; 0.001) and improvement in UCVA (<i>P</i> = 0.013). Longitudinal analysis of the non-spectacles subgroups revealed a significant emmetropization in SER and axial elongation over the 2-year period (all <i>P</i> &lt; 0.05), with no significant inter-group deviations among the three subgroups (<i>P</i> &gt; 0.05). However, higher baseline astigmatism was significantly associated with worse BCVA at the 2-year follow-up (<i>P</i> = 0.014).</p> Conclusions <p>For school-aged children with moderate to high astigmatism and normal baseline BCVA (≤ 0.1 logMAR), the absence of spectacle correction may not impede visual development or accelerate axial elongation. Clinical decision-making should prioritize individualized assessment of visual demands and symptoms, rather than relying solely on cylindrical power.</p>