Purpose To conduct a meta-analysis on the effects of atropine in slowing myopia progression and to compare Asian and white children and randomized controlled trials (RCTs) and observational studies. Methods Randomized controlled trials and observational studies that assessed the effects of all concentrations of atropine in slowing myopia progression in children were searched from MEDLINE, EMBASE, and the Cochrane Library up to April 2013. Jadad scoring was used to evaluate the quality of RCTs, and the Newcastle-Ottawa Scale was used for observational studies. Results Four RCTs and seven cohort studies (a kind of observational study) with 1815 children aged 5 to 15 years were included. The children had a baseline refraction of -0.50 to -9.75 diopters (D) and were followed up for 22.0 months (range, 12.0 to 36.5 months). The weighted mean differences in myopia progression in RCTs and cohort studies of Asian children were 0.55 D per year (p < 0.01) and 0.54 D per year (p < 0.001), respectively, and 0.35 D per year (p = 0.01) in cohort studies of white children. Compared with placebo, the risk of fast myopia progression (>1.0 D per year) using atropine was significantly decreased in both RCTs (odds ratio [OR], 0.14; p < 0.01) and cohort studies (OR, 0.08; p < 0.01), and the benefit of slow myopia progression (<0.50 D per year) using atropine was significantly increased in both RCTs (OR, 6.73; p < 0.01) and cohort studies (OR, 22.10; p < 0.01). Conclusions Atropine could significantly slow myopia progression in children, with greater effects in Asian than in white children. Randomized controlled trials and cohort studies provided comparable effects.