Investigators surveyed a large sample of students and reported that rope skipping performance exhibited a nonlinear relationship with refractive outcomes in younger children. The study used cycloplegic autorefraction to derive an average spherical equivalent (SEmean) across both eyes and classified myopia as SEmean ≤ −0.50 D. The cohort of 6–12‑year‑olds showed a reference threshold at 66 skips per minute identified within the nonlinear model. Below that threshold, the study team observed that each additional 10 skips per minute was associated with lower odds of myopia and a small shift of SEmean toward the hyperopic direction. The source provided point estimates and confidence intervals for these associations; those numerical details are reported in the original preprint.
The investigators adjusted models for multiple covariates, and school clustering was handled via robust standard errors. The source did not report additional mechanistic measures linking rope‑skipping to ocular growth (for example, no direct measures of accommodation or axial length were described in the text). Therefore the observed association should be viewed as an epidemiologic signal rather than proof of physiological mediation.
Among older adolescents, the relationship between endurance performance and refractive outcomes differed. The authors reported no clear overall nonlinearity across the full timescale of the endurance measure; however, they identified an inflection beyond 307 seconds where each additional 10 seconds of endurance time was associated with increased odds of myopia and a shift of SEmean toward myopia. The study team presented these as exploratory thresholds.
The source did not specify certain procedural details commonly used to interpret field fitness tests—such as test‑retest reliability, warm‑up protocols, or whether the endurance measure reflected time to exhaustion on a standard course—beyond the summary statistics and modeled associations. Readers should therefore avoid assuming uniform testing conditions across sites without consulting the original dataset or protocol documentation.
The paper describes a cross‑sectional survey performed in December 2023 that enrolled 17,235 students aged 6–18 years from 35 schools across seven counties or cities in Hainan Province. Cycloplegic autorefraction formed the basis for refractive classification. The investigators adjusted their models for age, sex, ethnicity, urban–rural residence, near‑work duration, outdoor activity time, parental history of myopia, body‑mass index and a centered BMI‑squared term; they also performed sensitivity analyses using waist‑to‑height ratio. School clustering was addressed using robust standard errors clustered by school.
The analysis aimed to detect nonlinear associations and derive reference thresholds for performance measures in relation to refractive outcomes. The source states that the identified thresholds—66 skips/min for rope‑skipping in younger children and 307 s for endurance in older adolescents—are exploratory values intended for potential school‑level risk stratification rather than being presented as causal cutoffs or diagnostic criteria. Precise details about the functional form of the nonlinear models (for example spline specifications) are provided in the preprint; readers should refer to the source for the exact modeling approach and full numerical results.
The authors noted several constraints that limit causal interpretation. The study's cross‑sectional design prevents inference about temporality or directionality between fitness performance and refractive development. The source explicitly acknowledged missing direct body‑composition measures; therefore independent effects of adiposity or lean mass could not be disentangled from the reported BMI and waist‑to‑height ratio adjustments. The thresholds reported are exploratory; the study team cautioned against using them as clinical diagnostic cutoffs.
Data governance restrictions also limit public access to the raw individual‑level dataset; the preprint indicates that qualified researchers may request de‑identified aggregate data and analysis code subject to approvals by the data custodian and relevant ethics authorities. The authors declared no competing interests and reported ethics approval from the Ethics Review Committee of Hainan Normal University.
Open questions remain. The analysis did not report axial‑length measurements or other ocular biometric data that would strengthen causal inference on refractive development. Likewise, protocol details for the fitness tests were not expanded in the source beyond the performance metrics used in modeling; this constrains comparison with other cohorts or replication studies. Given these gaps, the reported associations should be treated as hypothesis‑generating signals that require longitudinal follow‑up and external validation before informing screening or intervention policies.