Anda di halaman 1dari 8

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/325019625

The associations between near visual activity and incident myopia in children:
A nationwide 4-year follow-up study

Article  in  Ophthalmology · May 2018

CITATIONS READS

0 147

8 authors, including:

Po-Wen Ku Hsiao-Yun Hu
National Changhua University of Education National Yang Ming University
98 PUBLICATIONS   844 CITATIONS    74 PUBLICATIONS   550 CITATIONS   

SEE PROFILE SEE PROFILE

Yung Liao Li-Jung Chen


National Taiwan Normal University National Taiwan University of Sport, Taiwan
51 PUBLICATIONS   168 CITATIONS    76 PUBLICATIONS   728 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Personal Health Index View project

Prospective study of subjective and objective measures of late-life sedentarty behaviors and mental well-being View project

All content following this page was uploaded by Yung Liao on 20 June 2018.

The user has requested enhancement of the downloaded file.


The Associations between Near Visual
Activity and Incident Myopia in Children
A Nationwide 4-Year Follow-up Study
Po-Wen Ku, PhD,1,2 Andrew Steptoe, DPhil, DSc,3 Yun-Ju Lai, MD, PhD,4,5,6 Hsiao-Yun Hu, PhD,7,8
Dachen Chu, MD, PhD,6,9,10 Yung-Feng Yen, MD, PhD,6,10,11 Yung Liao, PhD,12 Li-Jung Chen, PhD2,5

Objective: This nationwide population-based study aimed to examine the prospective association between
near visual activities and incident myopia in Taiwanese children 7 to 12 years old over a 4-year follow-up period.
Design: Prospective cohort design.
Participants: There were 1958 children aged 7 to 12 years from the Taiwan 2009 National Health Interview
Survey who were linked to the 2009 through 2013 claims data from the National Health Insurance system.
Methods: Multivariable Cox proportional hazard models were used to estimate the associations between 3
types of near visual activities in sedentary posture, namely reading (< 0.5, 0.5-0.9, 1.0 hours per day [h/d]), use
of computer, Internet, and games (<0.5, 0.5-0.9, 1.0 h/d), and “cram school” attendance (<0.5, 0.5-1.9, 2.0 h/
d), and incident myopia.
Main Outcome Measures: Prevalent myopia was defined as those who had 2 ambulatory care claims
(International Classification of Diseases code 367.1) in 2008-2009. Incident myopia was defined by those who had
at least 2 ambulatory care claims (International Classification of Diseases code 367.1) during the 4-year follow-up
period (2010-2013) after excluding prevalent cases.
Results: Overall, 26.8% of children had myopia at baseline, and 27.7% of those without myopia at baseline
developed incident myopia between 2010 and 2013. On average, they spent 0.680.86 h/d on computer/Internet
use, 0.630.67 h/d on reading, and 2.783.53 h/d on cram school. The results showed that children attending
cram schools 2 h/d (hazard ratio, 1.31; 95% confidence interval, 1.03-1.68) had a higher risk of incident myopia.
The effects of these activities remained similar in sensitivity analyses.
Conclusions: Cram school attendance for 2 h/d may increase the risk of children’s incident myopia. This
effect may be due to increased near visual activity or reduced time outdoors. Ophthalmology 2018;-:1e7 ª 2018
by the American Academy of Ophthalmology

Supplemental material available at www.aaojournal.org.

Myopia is one of the common disorders of the eye and the spent outdoors could be a simple behavioral change strategy
development of a high prevalence of serious myopia in- for reducing the risk of developing myopia in childhood,5-7
creases the risk of later health problems, such as glaucoma, but it remains unclear whether other modifiable behavioral
retinal detachment, and cataract.1 A recent systematic risk factors are associated with incident myopia.
review and meta-analysis based on the pooled data from Children spend large amounts of time playing games on
145 studies covering 2.1 million participants showed that smart phones, tablet computers and PCs; reading and cram
the global prevalence of myopia was 28.3% in 2010. This schools (also known as tutorial or coaching schools);8 all of
figure is projected to increase rapidly from 34.0% in 2020 to which involves close visual work in a sedentary posture.9
39.9% in 2030 and reach 49.8% in 2050.2 Although the Whether near work is a risk factor for the development of
increasing prevalence of myopia has already been myopia in children has received attention in recent years. A
observed in some countries, regional differences were systematic review and meta-analysis documented inconsis-
evident. In 2010, the prevalence in East Asia (47.0%) was tent conclusions and heterogeneity across 14 studies (11 cross-
significantly higher than those in North America (34.5%) sectional and 3 cohort studies), suggesting more prospective
and Western Europe (28.5%). Countries in East and cohort studies are needed to verify the relationships between
Southeast Asia such as China, Japan, South Korea, near visual activities and incident myopia.10 To date, among the
Singapore, and Taiwan have been gripped by an 7 prospective cohort studies, only 2 reported that more time
unprecedented increase in myopia.3,4 Another systematic spent in near visual work could increase the risk of myopia
review and meta-analysis indicated that increasing time incidence11,12; others revealed no association.13-16 Besides,

ª 2018 by the American Academy of Ophthalmology https://doi.org/10.1016/j.ophtha.2018.05.010 1


Published by Elsevier Inc. ISSN 0161-6420/18
Ophthalmology Volume -, Number -, Month 2018

these cohort studies may have suffered from methodologic failed the vision screening were required by the school to take
limitations such as a short follow-up period,16,17 high rates of their child to a local eye clinic for regular eye examinations. It is
participant attrition,11,13,14 low precision in determining the compulsory that documentation of the examination signed by an
onset time of myopia,13 and small16 or nonrepresentative ophthalmologist is submitted to the school health center. The
nationwide completion rate for receiving clinical evaluation was
samples.12,15
approximately 95% in the school year 2003.21
To address these issues, strengthen the evidence base, Exposure variables: Near visual activities. Time spent in 3
and inform school policy, the present investigation extends types of near visual activities during both weekdays and weekend days
these findings by prospectively studying a national sample were assessed, including reading (including reading a book, magazine,
of young children 7 to 12 years old at baseline and including comic, or story); use of computer, Internet, and games; and cram
assessments of a range of near visual activities, over 4 years school attendance (including doing homework, assignments, and ex-
of follow-up. Thus, this nationwide population-based study aminations). Cram schools are a common way of enhancing academic
aimed to examine the prospective association between near learning in Taiwan and involve attendance at private classes outside
visual activities and incident myopia in Taiwanese children. the regular school system in the evening or on the weekends. Each type
We also included potentially confounding factors (i.e., of activity was assessed using the following questions: “On average,
during a weekday (Monday to Friday), how many hours/minutes in a
outdoor leisure activities) in multivariable analyses and
day do you engage in this activity?” and “On average, during a
conducted sensitivity analyses to test for the robustness of weekend (Saturday and Sunday), how many hours/minutes in a day do
the findings. you engage in this activity?” The average time engaging in each type
of activity per day was computed as follows: (51 weekdays þ 21
Methods weekend day)/7, which were categorized into 3 groups (reading, <0.5,
0.5-0.9, 1.0 hours/day [h/d]; use of computer, Internet, and games,
<0.5, 0.5-0.9, 1.0 h/d; and cram school attendance: <0.5, 0.5-1.9,
Study Design and Sample 2.0 h/d), based on previous studies.22,23 The content validity of all of
This prospective cohort study included participants aged 7 to 12 the items in the National Health Interview Survey was reviewed by
years in the Taiwan 2009 National Health Interview Survey who expert panels. The whole questionnaire was then examined twice in
were linked to the 2008 through 2013 claims data from the Na- pilot testing.18 The research team undertook a substudy to further
tional Health Insurance system. The 2009 National Health Inter- assess the reliability and validity of the measure among 49
view Survey included 25 636 participants (<12 years, n ¼ 3531; elementary school students (male/female, 29/20; mean age,
12 to 64 years, n ¼ 19 201; 65 years, n ¼ 2904; overall response 10.240.43 years) and their caregivers in Taiwan. The results of 3-
rate, 84.0%), and was conducted by the National Health Research day testeretest reliability and concurrent validity are shown in
Institutes and Bureau of Health Promotion Administration, Min- Table S1 (available at www.aaojournal.org), demonstrating the
istry of Health and Welfare in Taiwan. Participants were selected psychometric properties of these measures are adequate. Testeretest
using multistage stratified systematic sampling design to select a reliability of the time spent in near visual activities with a 3-day in-
nationally representative sample.18 terval was examined using Pearson’s correlation ranged between 0.76
In 2009, 2085 children aged 7 to 12 years participated in the and 0.95 (P < 0.001) and paired-samples t tests (all P > 0.05).
National Health Interview Survey. Information about the children Concurrent validity showed that Pearson’s correlation coefficients
was obtained using face-to-face interviews with an adult proxy between time spent in near visual activities self-reported by children
respondent (mother, 70.3%; father, 21.7%; others, 8%) familiar and their parents/caregivers demonstrating a high correlation (0.72-
with the child’s health. Among them, 1958 respondents (93.9%) 0.85) between the 2 groups. This provides some evidence of reliability
provided the consent to link the children’s claims data from the and validity of these measures.
National Health Insurance program. The National Health Insurance Covariates. Based on previous literature,4,10,24 the following
program is compulsory for all citizens starting from birth (coverage potentially relevant factors were included as covariates: (i) socio-
rate of 99.6% of the total population 23.1 million in 2009).19 demographic factors: sex, age (4-6, 7-8, 9-10, 11-12 years old),
Ethical approval for this study was obtained from Taipei City parents’ education level (junior high school or below, senior high
Hospital Institutional Review Board, Taiwan (reference number school, and college or higher), household monthly income (US$;
TCHIRB-10404118-W). To ensure adequate data protection, all <1000, 1000-1665, 1666-2332, and 2333), and urbanization
data access and statistical analyses were conducted in the Health (urban, suburban, and rural); (ii) outdoor leisure (sessions per
and Welfare Data Science Center, Ministry of Health and Welfare, week; <1, 1-2, 3-5, and 6); (iii) TV viewing (<2.0, 2.0-2.9, 3.0
Taiwan. h/d); and (iv) number of other eye disorders (0 vs 1), including
strabismus, amblyopia, color blindness, astigmatism, retinal dis-
Measures eases, and hypermetropia, which were measured by the items of the
National Health Interview Survey.
Outcome variable: Myopia. Based on claims data from the Na-
tional Health Insurance Research Database from 2008 through
2013, prevalent myopia was defined as those who had 2 ambu- Data Analysis
latory care claims (International Classification of Diseases, Ninth
Revision, Clinical Modification, [ICD-9-CM] as ICD 367.1) in Descriptive statistics for prevalent and incident myopia were
2008 or 2009. Incident myopia was defined by those who had 2 calculated first to characterize the sample structure. We used c2
ambulatory care claims during the 4-year follow-up period (2010- tests to test for group differences in percentage of each group of
2013) after excluding prevalent cases. According to the Imple- near visual activities and covariates. Previous research indicated
mentation Regulations Governing Health Examinations to that the use of conventional level (P ¼ 0.05) may fail to identify
Elementary and Junior High School Students in Taiwan,20 at the variables known to be important. Covariates with P < 0.25 were
beginning of each academic year (from September of 1 year to included in the subsequent regression models for adjustment.25
June of the next year), the annual vision screening was Mean values and standard deviations for time spent in different
conducted in the school. Parents whose child (aged 7 years) types of near visual activities were calculated separately for age

2
Ku et al 
Visual Activity and Myopia in Children

Table 1. Characteristics of Prevalent and Incident Myopia in Taiwanese Children at Baseline (Taiwan, 2009-2013)

Prevalence (2009) Incidence (2010-2013)


Variables in 2009 n (%) P for c2 n % P for c2
Sociodemographic
Sex 0.325 0.457
Male 970 27.8 700 28.6
Female 986 25.9 731 26.8
Age (y) <0.001 <0.001
7-8 (grades 1-2) 567 12.0 499 34.5
9-10 (grades 3-4) 666 29.1 472 28.6
11-12 (grades 5-6) 723 36.4 460 19.4
Father’s educational level <0.001 <0.001
Junior high school or below 504 18.7 410 22.2
Senior high school 825 26.8 604 26.3
College or higher 600 33.7 398 35.4
Mother’s educational level <0.001 <0.001
Junior high school or below 393 18.3 321 19.9
Senior high school 974 27.4 707 26.3
College or higher 548 32.5 370 36.2
Household monthly income (US$) <0.001 <0.001
<1000 439 17.5 362 21.3
1000-1665 546 24.7 411 25.8
1666-2332 413 32.0 281 33.5
2333 481 33.5 320 33.8
Urbanization <0.001 0.325
Urban 309 32.0 210 29.1
Suburban 1364 27.4 990 28.3
Rural 255 18.0 209 23.4
Near visual work
Computer/Internet/video game (h/d) 0.664 0.919
<0.5 996 26.2 735 28.2
0.5-0.9 416 27.6 301 26.9
1 513 28.3 368 27.7
Reading (h/d) <0.001 <0.001
<0.5 905 23.1 696 23.7
0.5-0.9 528 31.3 363 34.7
1 490 30.0 343 28.6
Cram school attendance (h/d) <0.001 0.029
<0.5 675 15.9 568 23.8
0.5-1.9 259 27.8 187 30.5
2.0 960 34.5 629 30.2
Lifestyle behaviors
TV watching (h/d) <0.001 0.473
<2 625 35.5 403 27.8
2.0-2.9 540 28.9 384 30.0
3 757 18.9 614 26.4
Frequency of outdoor leisure (session/wk) <0.001 0.735
<1 506 26.7 371 29.1
1-2 735 31.7 502 28.5
3-5 269 24.2 204 25.5
6 443 20.8 351 26.5
Health status
No. of other eye disorders <0.001 0.029
0 1481 20.6 1176 28.9
1 473 46.5 253 22.1

h/d ¼ hours per day.

groups. One-way analysis of variance was used to explore age To assess the potential problems of confounding and under-
group differences. diagnosed myopia, 2 sensitivity analyses were conducted. First,
A multivariable logistic regression model was computed to given that participants with more eye disorders might be at high
assess the associations between near visual activities and prevalent risk of myopia and might be less likely to engage in outdoor
myopia (model 1). Multivariable Cox proportional hazard models physical activities and near visual work, the inclusion of these
were analyzed to address the prospective associations between near subjects may influence the near visual activityeincident myopia
visual activities at baseline and incident myopia (model 2). relationships. The Cox regression was, therefore, repeated after

3
Ophthalmology Volume -, Number -, Month 2018

Table 2. Average Activity Hours per Day (Mean  Standard have fathers and mothers with a college or higher degree, a
Deviation) by Age Groups (Taiwan, 2009) higher household income, and to live in urban areas. They were
also more likely to have more reading time and more time
Age Groups (y) Computer/Internet Reading Cram school attending cram schools. They were more likely to watch TV
relatively little (<2 h/d), engaged in less outdoor leisure activity
7-8 0.410.64 0.650.71 2.293.49
and were more likely to have other eye disorders. Participants
9-10 0.700.85 0.640.65 2.843.54
11-12 0.860.96 0.600.66 3.103.50
who had incident myopia in 2010 through 2013 tended to be
Total sample 0.680.86 0.630.67 2.783.53 younger, lived in urban areas, and have fathers or mothers with
P value* <0.001 0.240 0.001 college or higher degrees, and enjoy a higher household monthly
income. They spent more time in reading and attending cram
schools.
Post-hoc tests with a significant difference: computer/internet 11-12 > 9-10
> 7-8; Cram school: 11-12 > 7-8; 9-10 > 7-8.
*Tests for 1-way ANOVA.
Patterns of Different Types of Near Visual
Activities by Age Groups
Patterns of different types of near visual activities stratified by age
excluding those who had any other eye disorders (model 3). Sec- groups are shown in Table 2. Among the 3 types of close visual
ond, previous research based on Taiwan insurance claim data work, school children spent the largest amount of time in
found that some physicians may categorize myopia diagnosis using attending cram school, followed by computer/Internet use, and
ICD 367 instead ICD 367.1. Therefore, restricting the diagnoses to reading. Apart from reading, it seems that time spent in cram
367.1 was likely to underestimate a number of children with school and computer/Internet use increase with age among
myopia.21 To mitigate the potential bias of underdiagnosis, the Cox children in elementary school.
regression was repeated based on the study sample with the
selection criterion of ICD 367 (model 4). Associations Between Near Visual Activity and
All analyses were conducted using SAS 9.4 software (SAS Prevalent Myopia
Institute, Cary, NC) and P < 0.05 was considered significant.
After adjusting for potential confounders, multivariable logistic
regression analyses (model 1 in Table 3) showed that children who
Results read more (0.5-0.9 h/d: odds ratio, 1.28 [P ¼ 0.006]; >1.0 h/d:
odds ratio, 1.43 [P ¼ 0.020]), and spent 2 h/d in cram school
Characteristics of Prevalent and Incident Myopia (odds ratio, 1.65; P < 0.001), were more likely to have prevalent
myopia.
Table 1 shows characteristics of prevalent (2009) and incident
(2010-2013) myopia in Taiwanese children at baseline. Overall, Associations between Near Visual Activities and
26.8% of children had myopia at baseline, and 27.7% of those Incident Myopia
without myopia at baseline developed incident myopia between
2010 and 2013. At baseline, c2 tests show that children who had Model 2 in Table 3 and Table S2 (Cox regression analyses) showed
myopia in 2009 were more likely to be 11 or 12 years of age, that, after covariates had been considered, children who attended

Table 3. Multivariable Logistic and Cox Regression Analyses for the Effects of Different Types of Close Work on Prevalent and Incident
Myopia (Taiwan, 2009-2013)

Model 1 (n [ 1956) Model 2 (n [ 1431) Model 3 (n [ 1176) Model 4 (n [ 1077)


2009 Prevalent Myopia 2010-2013 Incident Myopia 2010-2013 Incident Myopia 2010-2013 Incident Myopia
Types of Close Work ORs 95% CI P HR 95% CI P HR 95% CI P HR 95% CI P
Computer/Internet/video game (h/d)
<0.5 1.00 1.00 1.00 1.00
0.5-0.9 0.79 0.58-1.06 0.120 1.00 0.76-1.31 0.981 1.04 0.77-1.40 0.807 1.06 0.81-1.39 0.661
1 1.05 0.78-1.40 0.761 1.14 0.89-1.48 0.306 1.24 0.95-1.63 0.114 1.12 0.87-1.43 0.381
Reading (h/d)
<0.5 1.00 1.00 1.00 1.00
0.5-0.9 1.50 1.13-1.99 0.006 1.30 1.01-1.67 0.045 1.32 1.01-1.74 0.045 1.28 1.00-1.63 0.052
1 1.43 1.07-1.91 0.020 1.07 0.82-1.40 0.637 1.05 0.79-1.41 0.734 1.10 0.85-1.42 0.480
Attending cram schools (h/d)
<0.5 1.00 1.00 1.00 1.00
0.5-1.9 1.28 0.87-1.90 0.214 1.19 0.85-1.67 0.302 1.24 0.86-1.78 0.248 1.26 0.91-1.75 0.166
2.0 1.65 1.24-2.19 <0.001 1.31 1.03-1.68 0.030 1.46 1.12-1.89 0.005 1.35 1.07-1.71 0.013

CI ¼ confidence interval; h/d ¼ hours per day; HR ¼ hazard ratio; OR ¼ odds ratio.
Boldface indicate that P < 0.05.
For model 1, Covariates are age, parental educational levels, household income, urbanization, TV watching and outdoor leisure activities, and eye disorders
(variables with P < 0.25 for incident myopia in Table 1).
For model 2, covariates are age, parental educational levels, household income, and eye disorders (variables with P < 0.25 for incident myopia in Table 1).
For model 3, sensitivity analyses excluding eye diseases.
For model 4, sensitivity analyses excluding eye diseases and using ICD367.

4
Ku et al 
Visual Activity and Myopia in Children

cram schools 2 h/d were at increased risk of incident myopia strategy requires further investigation through intervention
(hazard ratio, 1.31; P ¼ 0.030). Similar results were observed in studies.
sensitivity analyses in models 3 and 4 after excluding A major strength of our study is that it is conducted on a
participants with other eye disorders at baseline and using nationally representative sample with a prospective design,
ICD367 for myopia, respectively. However, the relationship
thus strengthening the evidence base with in greater
between incident myopia and reading was weaker than those
with cram school attendance. The relationship between incident generalizability. Second, the data from large-scale health
myopia and reading (0.5-0.9 h/d) was a borderline significance surveys was cross-linked with a national health insurance
across models 2, 3 and 4 (P ¼ 0.045-0.052). In contrast, use of database with >99% coverage nationwide. Because the
computer/Internet was not related to incident myopia in any model. participation rate was high in the Taiwan National Health
Interview Survey (approximately 84% in 2009), the possi-
bility of selection bias is small. Finally, based on school
Discussion health regulations of Taiwan, eye examinations are required
for every school child in elementary and junior high schools
This study fills a gap in the literature by using a nationally in each semester. Thus, the outcome variabledincident
representative sample with a longer follow-up period than myopia based on medical recordsdis less likely to have
previous research, having low rates of loss to follow-up been underestimated or diagnosed after a delay. However,
because of the use of ambulatory care claims data from the this approach may not work in other countries.
National Health Insurance Research Database. This nation- Despite these strengths, several limitations of the present
wide 4-year follow-up study revealed that the association of study should be also considered. First, the measures of close
different types of near visual activities on incident myopia visual work, including reading; use of computer, Internet,
may be distinct. Among 3 types of close visual work in and games; and cram school attendance, were self-reported,
sedentary posture, school children attending cram school 2 and thus could be subject to bias. In addition, these items
h/d may have a higher risk of incident myopia. In contrast, have not been validated against external objective measures.
the association strength between time spent reading and Moreover, using baseline questionnaire information alone
incident myopia is weak, and only marginally significant. Use but not annual information for close visual work might also
of computer, Internet, and games is not related to incident result in bias. Second, this study used claim data as a
myopia. The prospective results were verified when adjusting measure of myopia, which cannot be used to assess the
for underlying covariates, including baseline sociodemo- severity of myopia. Future studies should consider exam-
graphic factors and outdoor leisure activities. Sensitivity an- ining the associations between close visual work and the
alyses also supported the stability of these findings. incidence of high myopia. Nevertheless, the systematic
This study demonstrates that close visual work, especially regime of school-based monitoring of vision problems in
cram school attendance 2 h/d, is positively associated with Taiwan may ensure the validity and minimize the problem
higher risks of developing myopia in children, supporting the of using claim data for myopia. Finally, a potential con-
findings of previous cohort studies.10 This finding is founderdoutdoor activitydwas assessed using sessions but
consistent with earlier findings of a cohort study17 Notably, not exact duration. Furthermore, outdoor brightness was not
time spent in cram school (2.78 h/d) is much higher than considered in the present study. This limitation may partly
those spent on reading (0.63 h/d) and use of computer/ explain why outdoor leisure activity was not a significant
Internet and games (0.68 h/d). This factor may partially predictor of incident myopia in the current study.
account for why the association of cram school attendance In conclusion, this study expands our understanding of
is stronger than those with reading and use of computer/ how time spent in different types of near visual activity is
Internet and games among school children in Taiwan. related to incident myopia in school children. It presents
However, the underlying mechanism linking cram evidence that cram school attendance for 2 h/d may in-
schooling and incident myopia remains unclear. crease the risk of incident myopia in children. Notably, time
Like many newly industrialized countries in Asia, most spent in cram school increases with age among elementary
families in Taiwan are nuclear with few children. Given school children. Greater time spent in cram schooling may
busy work schedules, most dual-earner parents are not able increase the amount of near visual activity and at the same
to pick up children from school on time. Moreover, although time deprive children of time outdoors. These issues warrant
parents sometimes do not provide help with children’s further study.
homework after work, they expect their children to be
academically successful. Therefore, enhancement of aca-
demic performance, especially in mathematics, science, and References
English courses, could potentially be another key reason
why cram schools are prevalent in Taiwan and other Asian
countries.26,27 It is possible that prolonged time spent in near 1. World Health Organization (WHO). Impact of increasing
prevalence of myopia and high myopia: report of the Joint
visual work for homework, assignments, and examinations
World Health Organization. Brien Holden Vision Institute
at cram school may bring heavy pressure and hinder outdoor Global Scientific Meeting on Myopia. Sydney, Australia:
physical play, which has been recognized as an underlying University of New South Wales; 2015.
factor of myopia. It may be important for cram schools to 2. Holden BA, Fricke TR, Wilson DA, et al. Global prevalence of
decrease the time spent in near visual activities and increase myopia and high myopia and temporal trends from 2000
outdoor activities to help prevent incident myopia. This through 2050. Ophthalmology. 2016;123:1036e1042.

5
Ophthalmology Volume -, Number -, Month 2018

3. Ding B-Y, Shih Y-F, Lin LL, et al. Myopia among school- 16. Wu PC, Tsai CL, Wu HL, et al. Outdoor activity during class
children in East Asia and Singapore. Surv Ophthalmol. recess reduces myopia onset and progression in school chil-
2017;62:677e697. dren. Ophthalmology. 2013;120:1080e1085.
4. Rudnicka AR, Kapetanakis VV, Wathern AK, et al. Global 17. Tsai DC, Fang SY, Huang N, et al. Myopia development
variations and time trends in the prevalence of childhood among young schoolchildren: the myopia investigation study
myopia, a systematic review and quantitative meta-analysis: in Taipei. Invest Ophthalmol Vis Sci. 2016;57:6852e6860.
implications for aetiology and early prevention. Br J Oph- 18. Shih Y-T, Hung Y-T, Chang H-Y, et al. The design, contents,
thalmol. 2016;100:882e890. operation and the characteristics of the respondents of the 2001
5. French AN, Ashby RS, Morgan IG, et al. Time outdoors and National Health Interview Survey in Taiwan. Taiwan J Public
the prevention of myopia. Exp Eye Res. 2013;114:58e68. Health. 2003;22:419e430.
6. Rose KA, Morgan IG, Ip J, et al. Outdoor activity reduces the 19. Jiang YD, Chang CH, Tai TY, et al. Incidence and prevalence
prevalence of myopia in children. Ophthalmology. 2008;115: rates of diabetes mellitus in Taiwan: analysis of the 2000-2009
1279e1285. Nationwide Health Insurance database. J Formos Med Assoc.
7. Xiong S, Sankaridurg P, Naduvilath T, et al. Time spent in 2012;111:599e604.
outdoor activities in relation to myopia prevention and control: 20. Taiwan Ministry of Education. Implementation Regulations
a meta-analysis and systematic review. Acta Ophthalmol. Governing Health Examinations to Elementary and Junior High
2017;95:551e566. School Students. Ministry of Education, Taipei, Taiwan 2003.
8. Morgan IG, Rose KA. Myopia and international educational 21. Fang YT, Chou YJ, Pu C, et al. Prescription of atropine eye
performance. Ophthalmic Physiol Opt. 2013;33:329e338. drops among children diagnosed with myopia in Taiwan from
9. Saunders TJ, Vallance JK. Screen time and health indicators 2000 to 2007: a nationwide study. Eye (Lond). 2013;27:
among children and youth: current evidence, limitations and 418e424.
future directions. Appl Health Econ Health Policy. 2016;15: 22. Chang T-W, Lin Y-H, Tsai H-F, et al. Preliminary exploration
323e331. of sedentary activities and daily behaviors in Taiwanese chil-
10. Huang HM, Chang DS, Wu PC. The Association between near dren: results of the 2005 National Health Interview Survey.
work activities and myopia in children-a systematic review and Taiwan National Health Interview Survey Research Brief.
meta-analysis. PLoS One. 2015;10: e0140419. 2006;5:1e12.
11. French AN, Morgan IG, Mitchell P, et al. Risk factors for inci- 23. Lo K-H, Pan L-YP, Lin M-C, et al. The analysis of special
dent myopia in Australian schoolchildren: the Sydney adolescent physical inactivity higher energy intakes and body mass index
vascular and eye study. Ophthalmology. 2013;120:2100e2108. of children in Taiwan: results of 2001 National Health Inter-
12. Jones-Jordan LA, Mitchell GL, Cotter SA, et al. Visual activity view Survey. Taiwan J Public Health. 2003;22:474e482.
before and after the onset of juvenile myopia. Invest Oph- 24. Foster PJ, Jiang Y. Epidemiology of myopia. Eye (Lond).
thalmol Vis Sci. 2011;52:1841e1850. 2014;28:202e208.
13. Guggenheim JA, Northstone K, McMahon G, et al. Time 25. Hosmer DW, Lemeshow S, Sturdivant RX. Applied logistic
outdoors and physical activity as predictors of incident myopia regression (3rd ed.). New York: John Wiley & Sons; 2013.
in childhood: a prospective cohort study. Invest Ophthalmol 26. Chiu S-C, Chung Y-C, Pai L-W, et al. Using a qualitative
Vis Sci. 2012;53:2856e2865. triangulation method to explore the challenge of myopia pre-
14. Jones LA, Sinnott LT, Mutti DO, et al. Parental history of vention in school children for the role of school health nursing.
myopia, sports and outdoor activities, and future myopia. Int J Nurs. 2016;3:55e64.
Invest Ophthalmol Vis Sci. 2007;48:3524e3532. 27. Tsai DC, Fang SY, Huang N, et al. Myopia development
15. Saw SM, Shankar A, Tan SB, et al. A cohort study of incident among young schoolchildren: the Myopia Investigation
myopia in Singaporean children. Invest Ophthalmol Vis Sci. Study in Taipei. Invest Ophthalmol Vis Sci. 2016;57:
2006;47:1839e1844. 6852e6860.

Footnotes and Financial Disclosures


8
Originally received: November 11, 2017. Institute of Public Health, National Yang-Ming University, Taipei,
Final revision: May 5, 2018. Taiwan.
Accepted: May 8, 2018. 9
Department of Neurosurgery, Taipei City Hospital, Taipei, Taiwan.
Available online: ---. Manuscript no. 2017-2592.
10
1 Department of Health Care Management, National Taipei University of
Graduate Institute of Sports and Health, National Changhua University of
Nursing and Health Sciences, Taipei, Taiwan.
Education, Taiwan.
11
2 Section of Infectious Diseases, Taipei City Hospital, Taipei City Gov-
Department of Epidemiology and Public Health, University College
ernment, Taipei, Taiwan.
London, UK.
12
3
Department of Behavioural Science and Health, University College Department of Health Promotion and Health Education, National Taiwan
London, UK. Normal University, Taichung, Taiwan.
4 Financial Disclosures:
Division of Endocrinology and Metabolism, Department of Internal
The authors have no proprietary or commercial interest in any materials
Medicine, Puli Branch of Taichung Veterans General Hospital, Nantou,
discussed in this article.
Taiwan.
5
HUMAN SUBJECTS: This study includes human subject/tissues. Study
Department of Exercise Health Science, National Taiwan University of protocol was approved by the Taipei City Hospital Institutional Review
Sport, Taiwan. Board, Taiwan (reference number: TCHIRB-10404118-W). All tenets of
6
School of Medicine, National Yang-Ming University, Taipei, Taiwan. the Declaration of Helsinki were followed. Informed consent was obtained
7
Department of Education and Research, Taipei City Hospital, Taipei, from all human subjects.
Taiwan. No animal subjects were used in this study.

6
Ku et al 
Visual Activity and Myopia in Children
Author Contributions: Correspondence:
Conception and design: Ku, Lai, Hu, Chu, Yen, Chen Li-Jung Chen, PhD, Department of Exercise Health Science, National Taiwan
Analysis and interpretation: Ku, Steptoe, Lai, Liao, Chen University of Sport, No. 16, Section 1, Shuang-Shih Road, Taichung, 404
Taiwan. E-mail: ljchen@gm.ntupes.edu.tw; and Yung Liao, PhD,
Data collection: Ku, Steptoe, Lai, Hu, Chu, Yen, Liao, Chen
Department of Health Promotion and Health Education, National Taiwan
Obtained funding: Not applicable Normal University, 162, Section 1, Heping E. Road, Taipei City 106,
Overall responsibility: Ku, Steptoe, Lai, Hu, Chu, Yen, Liao, Chen Taiwan. E-mail: liaoyung@ntnu.edu.tw.
Abbreviations and Acronyms:
ICD ¼ International Classification of Diseases.

View publication stats

Anda mungkin juga menyukai