Prevalence of visual impairment and refractive errors among different ethnic groups in schoolchildren in Turpan, China

There is currently limited information about ethnic differences in myopia prevalence within mainland China, especially in rural or semi-rural areas. We examined the prevalence of refractive errors, visual impairment and spectacle coverage in school children of varying ethnicity in Turpan, Xinjiang province. Methods: A community eye care service was provided for five schools. Presenting monocular distance and near visual acuity (VA), and ocular alignment were assessed. Retinoscopy and cycloplegic subjective refraction were performed for participants with presenting visual impairment (distance VA worse than 0.3 logMAR; Snellen 6/12 or 20/40) or abnormal binocular vision. Questionnaires administered prior to the eye examinations were used to collect information regarding personal lifestyle and parental myopia. Results: A total of 646 out of 690 (94%) subjects aged four to 19 years (11.9 ± 2.6; mean ± S.D.) completed the eye examination. Three hundred and eighty-two (59%) of participants were of Uyghur ethnicity, followed by Han, 176 (27%) and Hui, 74 (12%). The mean age of Uyghur, Han and Hui students was 12.3 ± 2.7, 11.4 ± 2.6 and 11.4 ± 2.3 years respectively, in which the Uyghur students were significantly older than the Han and Hui students (F_(3,631) = 5.58 p < 0.001). In total, 170 (27%) and 85 (13%) subjects failed the screening examination for one eye or both eyes, respectively. The prevalence of presenting visual impairment was not significantly different among the ethnic groups (p = 0.26). After cycloplegic refraction, most subjects' VA (98%) improved to better than 0.3 logMAR (Snellen 6/12 or 20/40). The prevalence of "clinically-significant myopia" (≤-0.50 dioptres) was 27%, 18% and 13% in Han, Hui and Uyghur children, respectively (p < 0.001). In contrast, Uyghur students had the highest prevalence of astigmatism (Uyghur 12%, Han 5%, Hui 4%). The overall spectacle coverage was 36%, while spectacle coverage among ethnic groups were similar (Han, 41%; Uyghur, 32%; Hui, 41%; χ² = 2.23, df = 2, p = 0.33). Conclusion: The prevalence of clinically significant myopia varied markedly with ethnicity in school children sampled from a semi-rural region of mainland China (Han > Hui > Uyghur). As reported previously, uncorrected/under-corrected refractive error was the main cause of presenting visual impairment.