key: cord-0810700-25d779di authors: Xu, L.; Ma, Y.; Yuan, J.; Zhang, Y.; Wang, H.; Zhang, G.; Tu, C.; Lu, X.; Li, J.; Xiong, Y.; Chen, F.; Liu, X.; Xue, Z.; Zhou, M.; Li, W.-Q.; Wu, N.; Chen, H.; Chen, J.; Lu, F.; Su, J.; Qu, J. title: COVID-19 Quarantine Reveals Grade-specific Behavioral Modification of Myopia: One-Million Chinese Schoolchildren Study date: 2020-11-18 journal: nan DOI: 10.1101/2020.11.15.20231936 sha: 5380d28f8e1749fbb79327e22eae9ebde5f40ab7 doc_id: 810700 cord_uid: 25d779di Summary Background High prevalence of myopia of adolescent has been a global public health concern. Their risk factors and effective prevention methods for myopia across schoolchildren developmental stages are critically needed but remain uncertain due to the difficulty in implementing intervention measurements under normal life situation. We aimed to study the impact of the COVID-19 quarantine on myopia development among over one-million schoolchildren. Methods We designed the ongoing longitudinal project of Myopic Epidemiology and Intervention Study (MEIS) to biannually examine myopia among millions of schoolchildren for ten years in Wenzhou City, Zhejiang Province, China. In the present study, we performed three examinations of myopia in 1,305 elementary and high schools for schoolchildren in June 2019, December 2019 and June 2020. We used the normal period (June-December 2019) and COVID-19 quarantine period (January-June 2020) for comparisons. Myopia was defined as an uncorrected visual acuity of 20/25 or less and a spherical equivalent refraction (SER) of -0.5 diopters (D) or less. High myopia was defined as an SER of -6.0 D or less. Findings In June 2019, 1,001,749 students aged 7-18 were eligible for examinations. In the 6-month and 12-month follow-up studies, there were 813,755 eligible students (81.2%) and 768,492 eligible students (76.7%), respectively. Among all students, we found that half-year myopia progression increased approximate 1.5 times from -0.263 D (95% CI, -0.262 to -0.264) during normal period to -0.39 D (95% CI, -0.389 to -0.391) during COVID-19 quarantine (P < 0.001). Multivariate Cox regression analysis identified grade rather than age was significantly associated with myopia (Hazard ratio [HR]: 1.10, 95% CI, 1.08 to 1.13; P < 0.001) and high myopia (HR: 1.40, 95% CI, 1.35 to 1.46; P < 0.001) after adjustment for other factors. The prevalence, progression, and incidence of myopia and high myopia could be categorized into two grade groups: I (grades 1-6) and II (grades 7-12). Specifically, COVID-19 quarantine for 6 months sufficiently increased risk of developing myopia (OR: 1.36, 95% CI, 1.33 to 1.40) or high myopia (OR: 1.30, 95% CI, 1.22 to 1.39) in Grade Group I, but decreased risk of developing myopia (OR: 0.45, 95% CI, 0.43 to 0.48) or high myopia (OR: 0.57, 95% CI, 0.54 to 0.59) in Grade Group II. Interpretation The finding that behavioral modifications for six months during COVID-19 quarantine sufficiently and grade-specifically modify myopia development offers the largest human behavioral intervention data at the one million scale to identify the grade-specific causal factors and effective prevention methods for guiding the formulation of myopia prevention and control policies. Funding Key Program of National Natural Science Foundation of China; the National Natural Science Foundation of China; Scientific Research Foundation for Talents of Wenzhou Medical University; Key Research and Development Program of Zhejiang Province. High prevalence of myopia of adolescent has been a global public health concern. Their risk factors and effective prevention methods for myopia across schoolchildren developmental stages are critically needed but remain uncertain due to the difficulty in implementing intervention measurements under normal life situation. We aimed to study the impact of the COVID-19 quarantine on myopia development among over one-million schoolchildren. We designed the ongoing longitudinal project of Myopic Epidemiology and Intervention Study (MEIS) to biannually examine myopia among millions of schoolchildren for ten years in Wenzhou City, Zhejiang Province, China. In the present study, we performed three examinations of myopia in is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint SER of -6.0 D or less. In June 2019, 1,001,749 students aged 7-18 were eligible for examinations. In the 6-month and 12-month follow-up studies, there were 813,755 eligible students (81.2%) and 768,492 eligible students (76.7%), respectively. Among all students, we found that half-year myopia progression increased approximate 1.5 times from -0.263 D (95% CI, -0.262 to -0.264) during normal period to -0.39 D (95% CI, -0.389 to -0.391) during COVID-19 quarantine (P < 0.001). Multivariate Cox regression analysis identified grade rather than age was significantly associated with myopia (Hazard ratio [HR]: 1.10, 95% CI, 1.08 to 1.13; P < 0.001) and high myopia (HR: 1.40, 95% CI, 1.35 to 1.46; P < 0.001) after adjustment for other factors. The prevalence, progression, and incidence of myopia and high myopia could be categorized into two grade groups: I (grades 1-6) and II (grades 7-12). Specifically, COVID-19 quarantine for 6 months sufficiently increased risk of developing myopia The finding that behavioral modifications for six months during COVID-19 quarantine sufficiently and grade-specifically modify myopia development offers the largest human behavioral intervention data at the one million scale to identify the grade-specific causal factors and effective prevention methods for guiding the formulation of myopia prevention and control policies. Key Program of National Natural Science Foundation of China; the National Natural Science is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020. 11.15.20231936 doi: medRxiv preprint neovascularization. In light of the increasing prevalence of myopia and high myopia has been a global public health concern, the impact of COVID-19 lockdown on myopia development has gained substantial attention. We searched PubMed, Google Scholar, and MEDLINE databases for original articles reported between database inception and November 10, 2020, using the following search terms: (coronavirus OR COVID* OR SARS-COV-2 OR lockdown OR quarantine) AND (myopia OR short-sightedness OR refractive error). To date, there was no original study reported to uncover the influence of COVID-19 quarantine on myopia progression. This study provides the largest longitudinal intervention data on myopia progression in Chinese schoolchildren covering all grades of schoolchildren at one-million scale. COVID-19 quarantine model uncovers that behavioral modifications for six months may lead to significant increase of overall prevalence of myopia associated with their increased screen times and decreased outdoor activity times. Importantly, their effects on developing myopia or high myopia of students are . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint Myopia is the most-common cause of visual impairment worldwide 1 . Many studies have reported the prevalence of myopia among schoolchildren, which ranges from 20% in early school-aged children to as high as 80% among 17-year-olds, in China 2-6 and other East and Southeast Asia countries such as Singapore and Japan 7-9 . The incidence of myopia also tends to be gradually increasing in populations of European and Middle Eastern ancestry 10, 11 . The latest World Health Organization (WHO) World Report on Vision (issued Oct 8, 2019) 12 indicated there were at least 312 million people <19 years of age with myopia in 2015, and myopia was estimated to affect 2.6 billion people of all ages worldwide by 2020, and predicted to affect approximately 5 billion people by 2050, including 938 million people with high myopia 1 . In addition, myopia with younger onset may progress to high myopia 13 , which are at . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint a substantially increased risk of sight-threatening complications such as myopic maculopathy, glaucoma, and retinal detachment. The increasing prevalence of myopia is a global public health concern 14, 15 , promoting many countries including China 16 to implement nationwide policies for preventing myopia development. The progression of myopia can be influenced by genetic and environmental components 17 . Many observational studies have reported numerous environment factors involved in myopia development, including education level, near work, urbanization, season of birth, and outdoor activities [18] [19] [20] [21] . Excessive near work including study, reading, and screen use were associated with having or developing myopia 22 . Of these, the influence of electronic products use time remains controversial. While some studies reported a significant association between increased computer and smart phone use time and myopia development 23, 24 , a meta-analysis reported s no association of screen time with the prevalence or incidence of myopia 25 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint schoolchildren in Wenzhou City with semiannual examination for ten years, taking into consideration of gender, educational level, educational system, and region of habitation. Until now, we have completed three test points of myopia: June 2019, December 2019, and June 2020, which provide a large-scale longitudinal study of 12-month follow-up, including the precious data for 6-month before and after the coronavirus disease (COVID-19) epidemic. Since the outbreak of COVID-19 was first reported in late December 2019 in Wuhan, Hubei province, China 28 , it has quickly caused an unprecedented global pandemic 27 . Governments from many countries have imposed various strict containment measures on citizens including case isolation, limited outdoor activities, and school closures to prevent COVID-19 virus spread 29 . According to the data from UNESCO, more than 87% of the world's student population from > 160 countries are affected by these lockdown measures 30 . COVID-19 quarantine has led to substantial changes in daily life among school-going children, including sharply dropped schooling times, but the increased online digital screen times and decreased outdoor activity times 31 . This provides unique and unprecedented opportunity to identify and evaluate the effect of behavioral modifications reeling from COVID-19 quarantine on the incidence and/or the worsening of myopia among schoolchildren 32 . Wenzhou is a coastal city located in Zhejiang province in China with 9.25 million local residents, of which many are usually businessmen by profession and frequently work and live in different cities including Wuhan. There were approximately 180,000 Wenzhou people working in Wuhan and 33,000 of them returned from Wuhan after COVID-19 outbreak, which led to Wenzhou was the worst affected city outside Hubei province in China in early February 2020. Therefore, schoolchildren in Wenzhou were strictly isolated to study at home from January to May 2020. It is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint provided a practical experimental model to identify the critical risk factors for behavioral modification of developing myopia. To the best of our knowledge, no comprehensive evaluation of the influence of COVID-19 quarantine on myopia development was reported. In the current longitudinal study across over one-million schoolchildren, we aimed to determine the prevalence of myopia and assess the impact of COVID-19 quarantine on myopia progression and incidence. The one-million mega dataset of myopia general survey allows to investigate the prevalence and to identify risk factors of myopia at the grade (even at birth month) scale. Specifically, we sought to address two critical questions: (1) Was six-month behavioral modifications during COVID-19 quarantine sufficient to change myopia progression course? (2) What are key risk factors for the occurrence and progression of myopia especially high myopia? -The findings from our study not only provides the first human behavioral intervention data at the one million-scale to identify the grade-specific causal factors, but also offer valuable information for the development of evidence-based myopia prevention and control strategies for children by school grade level. As part of our ongoing efforts within the MEIS project, the current survey of refraction-confirmed myopia was conducted among more than one-million schoolchildren in is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint Information Management System (AMIMS), which involves manual inspection of automated refractometry data, automatic data import, and collaboration between clinicians and a statistician (Supplemental Figure S1 ). In June of 2019, a census of 1,305 elementary and high schools encompassing different educational systems (key school and non-key school [i.e., public, sport, and martial arts]) in 11 districts of Wenzhou was conducted. There were 1,060,925 students aged 7 to 18 years recruited at the baseline, including 612,648 elementary school students (grades 1-6 in Chinese education system), 264,661 junior high school students (grades 7-9), and 183,616 senior high school students (grades [10] [11] [12] . All students identified in the census were invited to complete a self-administered is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint participate. The present study was approved by the Ethics Committee of the Wenzhou Medical University Affiliated Eye Hospital (approval numbers Wmu191204 and Wmu191205). All procedures were conducted in accordance with institutional/national research committee ethical standards as well as the 1964 Helsinki declaration. Participant completion of the self-administered questionnaire was considered informed consent. Due to the large sample size, performing the criterion standard of cycloplegic refraction to diagnose myopia was not applicable. Thus, the current survey adopted an alternative method of VA and autorefraction testing to identify myopia. VA was evaluated using an ''E-type'' standard logarithmic visual chart at a distance of 5 m. All students underwent noncycloplegic refraction testing using an automated refractometer, followed by complement subjective refraction testing for validation using an ''E-type'' standard logarithmic visual chart. Students who required corrective lenses were examined with and without eyeglasses, and naked eye refraction data were used to calculate the spherical equivalent. Myopia was identified as an uncorrected VA of less than 20/25, and spherical equivalent refraction (SER) was equaled to (sphere + [cylinder/2]) of -0.5 diopters (D) or less 33 . Since analysis for right and left eyes yielded a satisfactory correlation (Pearson correlation coefficient = 0.954, Supplemental Figure S2 ), SER data from the right eye were arbitrarily used for assessing myopia development, as reported in previous studies 34, 35 . An SER of -6.0 D or less indicated as high myopia, and an SER between -0.5 D and -6.0 D defined as non-high myopia 36 . From June 2019 to December 2019, the first 6-month interval indicated as "Normal" period . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint without the influence of COVID-19 quarantine, and the second 6-month interval from January 2020 to June 2020 indicated as "COVID-19" period with the influence of COVID-19 quarantine. Since COVID-19 quarantine had a great influence on alterations of schoolchildren lifestyle and study behaviors, it provided a practical intervention model for comparing the differences of risk factors related to myopia between normal period and COVID-19 period. Earlier studies [37] [38] [39] have used the 6-month interval to measure myopia progression for exploring the intervention strategies. In addition, during August 2020, we further used a new-designed questionnaire to survey 12,013 students, which were randomly selected from all grades. The questionnaire consisted of 17 questions, including the basic characteristics of schoolchildren, myopia information of their parents and siblings, and students' lifestyle or learning approaches such as outdoor activities times and online course times during normal period and COVID-19 period. The prevalence of myopia and high myopia were calculated by all eligible participants in three examinations. For prevalence, 95% confidence intervals (CIs) were calculated according to the procedure reported by Newcombe 40 . To examine factors associated with the incidence of myopia, the Cox proportional hazard regression analysis was performed on students who attended all three examinations and did not have myopia at the baseline. Similarly, Cox proportional hazard regression analysis was conducted to reveal risk factors associated with the incidence of high myopia on students who attended all three examinations and did not have high myopia at the baseline. These multivariate Cox regression analyses were adjusted for grade, age, gender, birth month, educational system (key/non-key school), and habitation (urban/rural). The survival package in R was applied to is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint perform multivariate Cox regression analyses. The significance of differences between categorical variables was assessed using the Fisher's test, and that of differences between continuous variables was assessed using Student's t test. Two-sided P values were used in all statistical analyses. The Pearson correlation analysis was used to calculate the correlation of SER between right-and left-eye. The R software (ver. 3.6.1) was used for all statistical analyses (http://www.r-project.org). The funders had no role in the study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data and had final responsibility for the decision to submit for publication. We enrolled and collected data from a total of 1,060,925 students at the baseline (June 2019). After stringent sample filtering, 1,001,749 students (550,756 males and 450,993 females) were included in the current study (Figure 1 and Supplemental Table S1 ). There were 580,177 students in elementary schools, 250,839 in junior high schools, and 170,733 in senior high schools, covering almost all schools (1,305/1,336 = 97.8%) in 11 districts of Wenzhou City (Supplemental Figure S3 ). Of the 1,001,749 students, 813,755 (81.2%) were reexamined after the first six months (Normal period) and 768,492 (76.7%) after the second six months (COVID-19 period) (Figure 1 ). At the baseline, the myopia prevalence was 52.89% (95% CI, 52.79% to 52.99%), and the high myopia is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint prevalence was 4.11% (95% CI, 4.07% to 4.15%). Table 1 shows the prevalence of myopia and high myopia among schoolchildren across three examinations categorized by gender, educational level, educational system, and habitation. The high prevalence of myopia among schoolchildren is similar with other cities of China and East Asia countries (Supplemental Table S2 ). At the baseline, cross-sectionally, the prevalence of myopia increased with grade level increasing (Table 1) , and was 35.41% (95% CI, 35.29% to 35.53%) among elementary school (grades 1-6), 74.01% (95% CI, 73.84% to 74.18%) among junior high school (grades 7-9), and 81.26% (95% CI, 81.07% to 81.44%) among senior high school (grades 10-12). The high myopia prevalence was significantly higher among senior high school (12.03%; 95% CI, 11.87% to 12.18%) than that among elementary school (0.82%; 95% CI, 0.8% to 0.85%). A grade-stratified analysis indicated that myopia development could be categorized into two grade groups: Grade Group I (grades [1] [2] [3] [4] [5] [6] representing myopia sensitive stage and Grade Group II (grades 7-12) representing high myopia sensitive stage (Supplemental Figure S4-S6) . A high increase in myopia prevalence of 9.9% per advance in grade level was observed in Grade Group I, whereas the rate increase in myopia prevalence slowed to 2.4% per advance in grade level in Grade Group II (Supplemental Tables S3-S5 ). The prevalence of high myopia increased from 4.46% (95% CI: 4.32% to 4.6%) to 13.25% (95% CI: 12.95% to 13.55%) during advancement through Grade Group II (Supplemental Table S3 ). Interestingly, the prevalence of myopia increased in a stepwise manner by grade, as determined by sequential analyses stratified by the students' birth month (Figure 2A , Supplemental Figure S7A -B, and Supplemental Table S6 ). Notably, an average increase in myopia prevalence of 8.54% (95% CI, 6.77% to 9.93%; P < 0.001) in the Grade Group I and 4.32% (95% CI, 0.69% to 7.42%; P < 0.001) in is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint the Grade Group II was observed in students of the same age born in the August of the year compared with those born in the September (Supplemental Figures S8-S9 and Supplemental Table S6 ), consistent with policies regarding the age of school enrollment in China that students born in August went to school a year earlier. It provides a direct clue that grade increasing of education burden could play more effects on myopia occurrence rather than age. Consistently, by performing multivariate Cox regression analyses with adjustment, we found grade was statistically significantly associated with myopia (HR: 1.1, 95% CI, 1.08 to 1.13, P < 0.001) and high myopia (HR: 1.4, 95% CI, 1.35 to 1.46, P < 0.001), whereas age and birth month showed non-significant association with myopia and high myopia ( Figure 2B-C) . Furthermore, we found gender, habitation, and educational system showed a statistically significant association with myopia ( Figure 2B ). Compared with males, non-key school, and rural habitation separately, factors of females (HR: 1.2, 95% CI, 1.19 to 1.22, P < 0.001), key school (HR: 1.04, 95% CI, 1.03 to 1.06, P < 0.001), and urban habitation (HR: 1.11, 95% CI, 1.09 to 1.13, P < 0.001) had a higher risk to develop myopia. During the COVID-19 pandemic, we found students' online time (> 2h) substantially increased is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint times from 40.1% in normal period to 23.5% in COVID-19 period ( Figure 3B ). These increased online time and decreased outdoor time among school-aged children may lead to occur or worsen myopia. To assess the impact of COVID-19-induced lifestyle changes on myopia development, we found myopia progression speed among all schoolchildren increased of approximate 1.5 times from -0.263 D (95% CI, -0.262 to -0.264) during normal period to -0.39 D (95% CI, -0.389 to -0.391) during COVID-19 period (P < 0.001, Supplemental Table S7 ). Stratified by gender, educational level, educational system, and habitation, we also observed a significant exacerbation of myopia progression during COVID-19 period (P < 0.001, Supplemental Table S7 ). Notably, compared with normal period, we found that the faster myopia progression of students from Grade Group I than ones from Grade Group II during COVID-19 period ( Figure 3C and Supplemental Table S8 ). For all schoolchildren, we found that the half-year incidence rate of myopia increased from 8.5% in normal period to 9.87% in COVID-19 period (OR: 1.18, 95% CI, 1.15 to 1.20; Figure 4A ). Stratified by female/male, key/non-key school, and urban habitation, schoolchildren had a significantly higher half-year incidence rate during COVID-19 period than that during normal period (P < 0.001, Figure 4A ). COVID-19 quarantine showed no significant impact on half-year incidence rate among students from rural habitation (P = 0.62). With regard to educational level, students in Grade Group I showed a remarkably higher risk of developing myopia during COVID-19 period than normal period (OR: 1.36, 95% CI, 1.33 to 1.40; P < 0.001, Figure 4A ), whereas students in Grade Group II had a prominently lower risk of incident myopia during COVID-19 period than normal is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint period (OR: 0.45, 95% CI, 0.43 to 0.48; P < 0.001, Figure 4A ). Interestingly, we found a distinct pattern that the half-year incidence rate of myopia during COVID-19 period consistently increased with grade increasing in Grade Group I, and the myopia half-year incidence rate during COVID-19 period gradually decreased with grade increasing in Grade Group II ( Figure 4B and Supplemental Table S9 ). These results suggested COVID-19 was a risk factor for developing myopia among students in Grade Group I, whereas it was a protective factor for developing myopia among students in Grade Group II. As for high myopia, the half-year incidence rate among all schoolchildren decreased from 1.31% during normal period to 0.096% during COVID-19 period (OR: 0.73, 95% CI, 0.70 to 0.75; Figure 4C ). Among students in Grade Group I, the half-year incidence rate of high myopia during COVID-19 period was significantly higher than that during normal period (OR: 1.30, 95% CI, 1.22 to 1.39, Figure 4C) , whereas there was a lower half-year incidence rate of students in Grade Group II during COVID-19 period than that during normal period (OR: 0.57, 95% CI, 0.54 to 0.59, Figure 4C ). Consistent with the identified COVID-19-influenced pattern of myopia, half-year incidence rate of high myopia during COVID-19 period gradually increased with grade increasing in Grade Group I, but decreased with grade increasing in Grade Group II ( Figure 4D ). In contrast, the half-year incidence rate of high myopia during normal period tended to persistently increase with grade increasing from 0.063% to 3.87% (Supplemental Table S10 ). Our finding consistently indicated COVID-19 contributed risk to incident high myopia among students in Grade Group I and had a protective effect on developing high myopia among students in Grade Group II. . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint difficulty in implementing behavioral modification at large scale. As the result, the sufficiency of behavioral modification pattern and time remain unclear. To the best of our knowledge, the EMIS study provides the first human intervention evidence at one-million scale to demonstrate that behavioral modification for six months during COVID-19 quarantine was sufficient to alter myopia progression. This sufficiency time for behavioral modification to be effective is essential in guiding the formation of myopia development and control among schoolchildren. Additional studies are needed to evaluate and demonstrate whether the COVID-19 pandemic exert lasting effects on the progression of myopia or high myopia (i.e. this specific pattern would persist during or after the COVID-19 quarantine) and to identify the shortest time for behavioral modification (such as three months) that is effective in influencing myopia development. Furthermore, previous studies have demonstrated that myopia onset and peak progression are rapid in childhood, particularly around puberty 43 . Compared with normal period, we found the COVID-19 quarantine had grade-specific effects on worsening the progression of myopia among schoolchildren, particularly among students in elementary school (Grade Group I), suggesting that elementary children were more vulnerable to be influenced by altered external environment resulting from COVID-19 quarantine. Interestingly, we uncovered a distinct pattern that COVID-19 quarantine was a risk factor of developing myopia or high myopia among students from Grade Group I, whereas COVID-19 quarantine was a protective factor for the development of myopia or high myopia among students from Grade Group II. Although COVID-19 increased the online time and decreased the outdoor activity time for all schoolchildren, COVID-19 still decreased the incidence rate of myopia and high myopia among students in Grade Group II, which may be due to the largely reduced study pressure. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org /10.1101 /10. /2020 Third, genetic factors play an important role in development of high myopia 22, 44 . As result the extent of high myopia can be modified by environmental factors such as behavioral modification in COVID-19 quarantine is largely unknown. Interestingly, we found that behavioral modification increased the risk of developing high myopia in Grade I group and decreased the risk of incident high myopia in Grade II group. This study provide the clearest data yet in support of the environmental factors contribution to high myopia, which is collaborated by previous studies 22, 45 . The EMIS study provide the first human behavioral intervention data at the one million-scale to identify the grade-specific causal factors for myopia prevention and control. Within our EMIS study population, the overall prevalence of myopia and high myopia were 52.89% and 4.11%, respectively. The prevalence of myopia among students of senior high school were 81.26%, which was similar to that reported among students of high school in 2010 in Shandong Province, China (84.6%) 46 While several studies have evaluated and showed positive correlation between age and grade level with myopia development 18, 52, 53 . However, these studies are limited by sample size and lack of the detailed analysis of these risk factors, rendering the importance of these risk factors uncertain. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org /10.1101 /10. /2020 In our survey of over one-million students, over 5,000 participants for each birth month, enabling stratification of the results by birth month. In students of the same age, differences in myopia were observed between those born in August and September, indicating that educational grade is more closely associated with development of myopia than age. Multivariate Cox regression analyses further validated the significant association between grade and myopia as well as high myopia after adjusting for other risk factors. Thus, our MEIS study with one-million participants establishes that educational burden could play key roles in myopia and high myopia. Based on our multivariate Cox models, we found that female students had a higher risk of developing myopia or high myopia than male students, which was consistent with previous studies reporting a positive correlation between female sex and increased myopia risk in China 47, 54, 55 . This difference could have a biological explanation related to hormonal differences, as postulated in a review of myopia in opposite-sex twins 56 . In addition, traditional cultural factors might influence the behaviors of Chinese girls, such as Chinese girls prefer to white skin and avoid sun exposure, and are encouraged to be gentle and quiet, which may predispose girls to stay indoors more 57 . We also observed that students from key school or urban habitation yielded a significantly higher risk of developing myopia or high myopia than those from non-key school or rural habitation. Consistently, Jan et al. 55 reported the prevalence of visual impairment was higher among Chinese students living in urban than ones living rural areas. More economically developed regions in China, such as Zhejiang, Jiangsu, and Shanghai, had a higher risk of visual impairment than less developed regions regardless of survey year 55 . The limitations of this study should be noted. First, the study participants were schoolchildren and did not include adults at universities and middle-aged or quinquagenarian individuals; hence our . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint findings are limited to younger population. Furthermore, student's online time and outdoor activity time changes during COVID-19 period were based on self-reported, thus these measurements might exist recall bias. Second, only Chinese schoolchildren were participants. In view of the ongoing COVID-19 spread worldwide, more studies are warranted to explore the influence of COVID-19 on myopia progression of schoolchildren based on more different ethnicities. In summary, the present study is the largest longitudinal survey to assess the impact of COVID-19 quarantine-related lifestyle changes on myopia development among schoolchildren. We found an interesting COVID-19-induced pattern of developing myopia among children and adolescents, indicating different intervention strategies should be applied to control myopia among elementary and high schools. Our results should help inform educational and health policy decision making and the development of programs aimed at the prevention and control of myopia. Over the next decade, the MEIS will continue to support surveys of myopia in students twice annually, with a focus on discerning the effects of genetics, nutrition, and learning environment using questionnaire surveys and medical intervention studies to promote nationwide strategies for the prevention and control of myopia. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint All authors declare no competing interests. The authors are open to sharing statistical codes and providing descriptive data in table form. Requests should be made to Jianzhong Su (sujz@wmu.edu.cn) This work was supported by the Key Program of National Natural Science Foundation of China is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint multivariate Cox model. In both Cox models, adjusted predictors included grade, age, gender, birth month, educational system, and habitation. Forest plot indicates the differences of myopia incidence between normal and COVID-19 period stratified by gender, educational level, educational system, and habitation. B) The differences of grade-specific half-year myopia incidence between normal and COVID-19 period. C) Forest plot indicates the differences of high myopia incidence between normal and COVID-19 period stratified by gender, educational level, educational system, and habitation. B) The differences of grade-specific half-year high myopia incidence between normal and COVID-19 period. . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint Global prevalence of myopia and high myopia and temporal trends from Prevalence of myopia in school children in greater Beijing: the Beijing Childhood Eye Study Associations between anthropometric indicators and both refraction and ocular biometrics in a cross-sectional study of Chinese schoolchildren Refractive error and visual impairment in school children in rural southern China Age-specific prevalence of visual impairment and refractive error in children aged 3-10 years in Refractive status and prevalence of refractive errors in suburban school-age children Component dependent risk factors for ocular parameters in Singapore Chinese children Current Prevalence of Myopia and Association of Myopia With Environmental Factors Among Schoolchildren in Japan Prevalence of myopia and its association with body stature and educational level in 19-year-old male conscripts in seoul Increased prevalence of myopia in the United States between 1971-1972 and 1999-2004 The changing prevalence of myopia in young adults: a 13-year series of population-based prevalence surveys Mechanism of neutralization of influenza virus infectivity by antibodies Factors associated with high myopia after 7 years of follow-up in the Correction of Myopia Evaluation Trial (COMET) Cohort Global magnitude of visual impairment caused by uncorrected refractive errors in 2004 Myopia-a 21st century public health issue A polymorphism that affects OCT-1 binding to the TNF promoter region is associated with severe malaria Genetic and environmental effects on myopia development and progression Myopia among schoolchildren in East Asia and Singapore Effect of time spent outdoors at school on the development of myopia among children in China: a randomized clinical trial Outdoor activity reduces the prevalence of myopia in children Worldwide prevalence and risk factors for myopia. Ophthalmic & physiological optics : the journal of the British College of Ophthalmic Opticians The impact of computer use on myopia development in childhood: The Generation R study Smartphone use as a possible risk factor for myopia Ophthalmic & physiological optics : the journal of the British College of Ophthalmic Opticians (Optometrists) 2020 Parental history of myopia, sports and outdoor activities, and future myopia World Health Organization declares global emergency: A review of the 2019 novel coronavirus (COVID-19) A Novel Coronavirus from Patients with Pneumonia in China Covid-19 -The Law and Limits of Quarantine May home confinement during the COVID-19 outbreak worsen the global burden of myopia? Graefe's archive for clinical and experimental ophthalmology = Albrecht von Graefes Archiv fur klinische und experimentelle Impact of COVID-19 pandemic on children and adolescents' lifestyle behavior larger than expected Digital Screen Time During COVID-19 Pandemic: Risk for a Further Myopia Boom? American journal of ophthalmology IMI-Defining and classifying myopia: a proposed set of standards for clinical and epidemiologic studies Two-year changes in refractive error and related biometric factors in an adult Chinese population Effect of Time Spent Outdoors at School on the Development of Myopia Among Children in China: A Randomized Clinical Trial KW-6002 (Kyowa Hakko Kogyo) Effect of outdoor activity on myopia onset and progression in school-aged children in northeast China: the Sujiatun Eye Care Study Protective behaviours of near work and time outdoors in myopia prevalence and progression in myopic children: a 2-year prospective population study Posterior Chamber Phakic Intraocular Lens Implantation for the Correction of Myopia and Myopic Astigmatism: A Retrospective 10-Year Follow-up Study Two-sided confidence intervals for the single proportion: comparison of seven methods Impact of the COVID-19 Epidemic on Lifestyle Behaviors and Their Association With Subjective Well-Being Among the General Population in Mainland China: Cross-Sectional Study Impact of the COVID-19 virus outbreak on movement and play behaviours of Canadian children and youth: a national survey. The international journal of behavioral nutrition and physical activity 2020 The relationship between growth spurts and myopia in Singapore children The heritability of high myopia: a reanalysis of Goldschmidt's data Incidence of and Factors Associated With Myopia and High Myopia in Chinese Children, Based on Refraction Without Cycloplegia Refractive error, visual acuity and causes of vision loss in children in Shandong, China. The Shandong Children Eye Study Prevalence and associated factors of myopia in high-school students in Beijing Prevalence of refractive errors in teenage high school students in Singapore Refractive errors in koreans: the Korea national health and nutrition examination survey Changes of the ocular refraction among freshmen in National Taiwan University between Prevalence of myopia in Taiwanese schoolchildren The epidemics of myopia: Aetiology and prevention Increasing Prevalence of Myopia in Europe and the Impact of Education Prevalence of myopia in schoolchildren in Ejina: the Gobi Desert children eye study Association of Visual Impairment With Economic Development Among Chinese Schoolchildren Optometry and vision science: official publication of the American Academy of Associations between near work, outdoor activity, and myopia among adolescent students in rural China: the Xichang Pediatric Refractive Error Study report no It is made available under a perpetuity.is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint It is made available under a perpetuity.is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.15.20231936 doi: medRxiv preprint