Why This Article Matters
Summary
This analysis investigated whether previous myopia progression can reliably predict future progression in children. The study included 136 children aged 5-12 years with myopia between -1.00 and -6.00 D who had participated in a 24-month trial comparing 0.01% atropine eye drops with placebo. The authors evaluated whether changes in spherical equivalent refractive error (SER) and axial length (AL) during the first 12 months were associated with changes observed during the following 12 months. Multivariable linear regression models were used while controlling for age and refractive status or axial length at 12 months. The study found that prior progression had limited predictive value. A myopic progression of at least 0.50 D during the first year predicted similar progression in the second year with a positive predictive value of only 42%. Likewise, greater SER or AL change in the first year showed only weak associations with subsequent progression, with minimal improvement in prediction accuracy when prior progression was included in the models. Similar findings were observed in analyses restricted to the placebo group. Overall, the results suggest that previous myopia progression is a poor predictor of future change. The findings challenge the common assumption that identifying “fast progressors” is useful for guiding treatment decisions or enriching clinical trial populations in myopia control research.
Commentary
Limitations
The post hoc nature of the analysis is the main limitation. Although the parent study was randomized, this specific question was not the primary objective of the original trial. In addition, only 136 of 187 randomized children had complete 12- and 24-month data, largely due to missing measurements during the COVID-19 period. The cohort was also drawn from U.S. clinical centers and included a majority of White participants, which may limit generalizability to populations with higher myopia prevalence and faster progression, particularly East Asian cohorts. Finally, the analysis focused on a 12-month prior progression window; shorter or longer observation periods, or more complex multivariable prediction models, may yield different results.
Clinical relevance
The study challenges the common assumption that children who progressed rapidly over the previous year will necessarily remain “fast progressors” in the following year. Prior SER and AL change showed only weak predictive value, and adding prior progression produced minimal improvement in prediction accuracy. This suggests that treatment decisions should not rely too heavily on previous-year progression alone. Instead, clinicians should continue to consider age, baseline refractive error, axial length, family history, ethnicity, lifestyle factors, and overall risk profile.
Comparison with existing evidence
These findings are consistent with previous evidence reported by Mutti et al. (Optom Vis Sci, 2022), who also found limited value of prior progression in predicting future myopia progression. They further align with recent International Myopia Institute (IMI) recommendations emphasizing multifactorial risk assessment rather than reliance on a single progression metric (IMI Digest, 2025; Wolffsohn et al., IOVS, 2019). Similarly, Brennan et al. (Optom Vis Sci, 2024) highlighted the importance of age and race in axial elongation patterns, supporting the idea that progression risk cannot be adequately captured by prior change alone. The present findings are also relevant in the context of the PEDIG atropine trial reported by Repka et al. (JAMA Ophthalmology, 2023), in which 0.01% atropine did not significantly reduce myopia progression or axial elongation in U.S. children.
Unanswered questions
Future studies should determine whether prediction improves when prior progression is combined with age, ethnicity, parental myopia, baseline AL, choroidal or biometric markers, environmental exposure, and genetic risk. It also remains unclear whether predictive models perform differently across ethnic groups, treatment modalities, and younger children at the earliest stages of myopia onset.
Key Take-Home Messages
- Previous-year myopia progression alone is a poor predictor of future progression in children. - Children identified as “fast progressors” may not continue to progress at the same rate over time. - Prior changes in spherical equivalent refraction and axial length added minimal improvement to prediction models. - Clinical decisions in myopia management should rely on a broader risk assessment, including age, baseline refractive error, axial length, family history, and environmental factors. - These findings question the usefulness of selecting participants for myopia control trials based solely on prior progression rate.
Conflict of Interest Statement
The author declares no conflict of interest related to this article.
Strengths
This study uses data from a well-characterized, multicenter randomized controlled trial, with standardized measurements of both spherical equivalent refractive error (SER) and axial length (AL). The inclusion of AL is particularly valuable, as axial elongation is increasingly regarded as a key outcome in myopia control studies. The authors also applied clinically meaningful thresholds and complemented categorical analyses with multivariable regression models and cross-validation, making the assessment of predictive performance more robust. Sensitivity analyses restricted to the placebo group further strengthen the interpretation that the findings were not driven by atropine exposure.