Effects of outdoor activities on myopia among rural school children in Taiwan
Wu PC, Tsai CL, Hu CH, Yang YH · 2010
grade Dlongitudinalindependentunreplicated
Sample
145 students aged 7-12 from two schools on Chimei Island
Population
Rural elementary school children, Chimei Island, Taiwan
Design
Cross-sectional survey with parent questionnaire, cycloplegic autorefraction and axial length. No intervention, no control group, no temporal ordering. Exposure (outdoor activity) was parent-reported.
Key findings
Myopia prevalence 31%. In multiple logistic regression, myopia was significantly associated with school year and with having a myopic parent; outdoor activity was inversely associated with myopia (OR 0.3, 95% CI 0.1-0.9, P = 0.025). Television watching was not significant after adjustment. Notably the study's own strongest predictor is parental myopia.
Genetic confound
High and self-evident: the second-strongest predictor in the model is myopic parent. Families in which parents are myopic - and therefore in which children carry myopia risk alleles - may also differ in how much time children spend outdoors, so the outdoor odds ratio is not separable from heredity in a cross-section of 145 children.
Effects
| Outcome | Metric | Value | Measure | Timing | Vs | Horizon | Class |
|---|---|---|---|---|---|---|---|
| Myopia, association with parent-reported outdoor activity | adjusted odds ratio | OR 0.3 (95% CI 0.1-0.9, P = 0.025) | standardized | cross-sectional | none | not-applicable | health |
| Myopia, association with having a myopic parent | adjusted logistic regression | significant; among the model's strongest predictors alongside school year | standardized | cross-sectional | none | not-applicable | health |
| Myopia prevalence | prevalence | 31% of 145 children aged 7-12 | standardized | cross-sectional | none | not-applicable | health |