bs_bs_banner
Clinical and Experimental Ophthalmology 2015; 43: 612–620 doi: 10.1111/ceo.12532
Original Article Study design, rationale and methods for a population-based study of myopia in schoolchildren: the Myopia Investigation study in Taipei Der-Chong Tsai MD PhD,1,2* Li-Ju Lin PhD,3* Nicole Huang PhD,4 Chih-Chien Hsu MD,5,6 Shing-Yi Chen MS,3 Allen Wen-Hsiang Chiu MD PhD1 and Catherine Jui-Ling Liu MD1,6 1
School of Medicine, 4Institute of Hospital and Health Care Administration, 5Institute of Clinical Medicine, National Yang-Ming University, 3Department of Health, Taipei City Government, 6Department of Ophthalmology, Taipei Veterans General Hospital, Taipei, 2 Department of Ophthalmology, National Yang-Ming University Hospital, Yilan, Taiwan
ABSTRACT Background: To describe the study design, rationale and methodology of the Myopia Investigation Study in Taipei (MIT). Design: The MIT was a citywide, population-based cohort study. Participants: Participants were grade 2 students (Fall 2013) of all 153 elementary schools in Taipei City. Methods: The baseline data on the risk factors for myopia development was collected by parentadministered questionnaire surveys covering demographics, medical history, parental myopia, time spent on near work and outdoor activities, reading habits and eye care-seeking behaviour. Ocular examinations focused on the measurement of visual acuity (unaided and best-corrected) and refractive status (before and after cycloplegia), which will be carried out for the eligible schoolchildren biannually for 3 years consecutively. Once myopic children are identified, case manager-led telecoaching for healthcare instructions and reminders will be delivered to parents or caregivers.
Main Outcome Measures: To build a comprehensive database for prevalence, incidence and risk factors of early childhood myopia over a 3-year follow-up period. Results: Of all 19 374 eight-year-old schoolchildren (10 210 [52.7%] boys) eligible for the MIT, 16 486 (85.1%) responded to the questionnaire, 12 019 (62.0%) were examined during the third quarter of 2013 and 11 590 (59.8%) (6267 [52.9%] boys) completed cycloplegic autorefraction on both eyes and were enrolled for further data analysis. There was no significant difference in terms of demographics between the analysed participants and all grade 2 students in Taipei City. Conclusions: Data from the MIT will provide population-based information concerning the prevalence, incidence and risk factors for myopia development among young schoolchildren in a metropolitan area of Taiwan. Key words: cohort study, methodology, myopia, population-based study, schoolchildren.
■ Correspondence: Prof Catherine Jui-Ling Liu, Department of Ophthalmology, Taipei Veterans General Hospital, No. 201, Sec. 2, Shih-Pai Road, Taipei 112, Taiwan. Email: [email protected] Received 1 January 2015; accepted 1 April 2015. *These authors contributed equally to this work. Conflicts of interest: None. Funding sources: The present study was supported by grants from H10237 and P10303 from the Taipei City Government. © 2015 Royal Australian and New Zealand College of Ophthalmologists
Method of myopia investigation in Taipei
INTRODUCTION Myopia is one of the most common eye disorders worldwide, especially in the industrialized regions of East Asia where more than 80% are affected in some populations.1–7 Myopes with an earlier age of onset may suffer from faster myopic progression and eventually be more likely to have high myopia (≤−6.0 diopter [D]).2,8 Although the refractive vision problem is usually correctable with the aid of spectacles, contact lenses or refractive surgeries, the ocular comorbidities associated with high myopia, such as rhegmatogenous retinal detachment, degenerative maculopathy and glaucoma may contribute to an increased risk of irreversible loss of vision and impose heavy socioeconomic burdens to both family and community.2 Several population-based prevalence surveys revealed that the prevalence of myopia has escalated considerably over the past few decades.1,2,4,9 This secular trend is evident in Taiwan, where there was a 3.6-fold increase in myopia prevalence from 1983 (5.8%) to 2000 (21%) among the 7-year-old schoolchild population.4 An epidemiological study recently reported a prevalence of myopia as high as 86.1% among young Taiwanese men aged 18 to 24 years.7 Similarly, the prevalence of myopia in the United States appears to be substantially higher from 1999– 2004 than 30 years earlier.9 Intensive environmental pressures, such as rapid urbanization and a rigorous education system, are considered the driving forces behind the changes in prevalence of myopia over time.10 Given the swift rise in popularity and deep penetration of mobile devices, such as smartphones and tablet computers, over the last few years, the environmental pressures would have become more intensive than ever and this public health issue clearly warrants further epidemiological investigations. Although exact mechanism of myopic onset and progression is still not well understood, pharmacological, environmental and optical interventions have been used for myopia control among schoolchildren.11 Numerous studies have disclosed that atropine is effective in slowing the progression of childhood myopia.12–16 Despite potentially annoying side effects, including photophobia and blurred near vision, there is an increasing trend of atropine eye drop prescription for myopic children in Taiwan.17 Increasing time spent outdoors and reducing peripheral hyperopia with optical devices also have been proven to protect against the development of myopia.18,19 Although the optimal mode of myopia prevention has not yet been made clear, it is desirable to have myopic children and their parents adhere to these interventions until stabilization of refractive status by the late teen years.
613 For early identification of myopes and retardation of myopic progression among the youth population, the Myopia Investigation study in Taipei (MIT) was designed to provide a citywide eye examination programme for grade 2 students of all 153 elementary schools in Taipei. Once myopic schoolchildren are identified, telecoaching service will be provided to promote their adherence to myopic intervention. Furthermore, the MIT will follow-up the refractive status of this school cohort biannually for 3 years to bridge the knowledge gap concerning the myopia incidence and risk factors of progression in Taiwan. Compared with numerous cross-sectional studies of school myopia prevalence, there is relatively fewer data on myopia incidence from longitudinal cohort studies.1,2,20,21 The purpose of this paper is to summarize the study design and rationale of the MIT and to describe the procedures used to carry out this citywide programme.
METHODS Study design The MIT is a population-based cohort study initiated and funded through a programme grant from the Taipei City Government. It was set up in 2013 and designed to investigate prospectively the development of myopia among the 8-year-old schoolchildren (born mainly between September 2005 and August 2006) in Taipei City, which is one of the areas with the highest prevalence of myopia in the world.2–4 A previous survey in Taiwan revealed that the mean refractive status becomes myopic at the age of 9 years.3 For early identification and intervention of myopia in young children, we chose grade 2 students and will continue to follow their refractive changes for 3 years. Written informed consent was obtained from either parent of each child and the Declaration of Helsinki was adhered to throughout. The protocol and consent procedure were approved by the Institutional Review Board of the Taipei City Hospital (TCHIRB-1020501).
Specific aims In addition to the baseline examination in 2013, there will be up to five episodes of citywide refractive status surveys performed biannually (i.e. once per semester) in the following 3 years. The specific aims of the MIT include the following: 1
2
© 2015 Royal Australian and New Zealand College of Ophthalmologists
Determining the baseline prevalence of myopia in the population of aforementioned young schoolchildren (primary outcome); Determining the 1-year to 3-year incidence of myopia among those non-myopic participants at baseline (primary outcome);
614 3
4
5
6
Tsai et al. Determining the degree of myopia progression during the up to 3-year follow-ups among those myopic participants at baseline (secondary outcome); Examining the associations between both modifiable (environmental) and non-modifiable (hereditary) risk factors and myopia development (secondary outcome); Determining the prevalence and associated factors of previously undiagnosed myopia (secondary outcome). Determining the prevalence and associated factors of anisometropia and astigmatism (secondary outcome).
Study area Taipei City, the capital of Taiwan situated in the northern part of the island, has a population of approximately 2.6 million and an area of 271.7 km2.22 It is the political, economic and cultural center of Taiwan. There are 12 administrative districts in Taipei City. Of those, seven districts are defined as urban and five districts as suburban (Supporting Information Table S1), according to the criteria of urbanization classification based mainly on population density, medical resources, age and residents’ education levels.23 There are 153 elementary schools across the city.24 Due to the well-executed compulsory education system, the student attendance rate at elementary schools is greater than 99% in Taipei City and there were 19 374 schoolchildren in grade 2 during the 2013 Fall semester.24 The distribution of these 19 374 students by administrative districts is provided in Supporting Information Table S1. In addition, the nationalized health-care system is also well established in Taiwan. A total of 24 hospitals and 123 private clinics provide eye health-care services in Taipei City.
Organizational structure National Yang-Ming University has a mandate from the Taipei City Government to help organize and monitor the MIT. A total of 20 hospitals and 54 private clinics across the 12 districts have joined the MIT, providing hospital-based or clinic-based eye examinations for eligible schoolchildren. To meet a consistent standard operating procedure, the MIT organizing committee formulated a standardized protocol for staff training and held a series of coordination courses for participating ophthalmologists and other registered medical practitioners. The principal investigators of the MIT also invited 11 senior ophthalmologists and four epidemiologists to build a committee to monitor the quality of the study execution. They regularly visited the MIT-associated
hospitals/clinics to evaluate eye examination procedures during the survey campaign. It was requested that all examinations be carried out according to the standardized protocol. Full-time administrative staff and three trained case managers were recruited. The principal investigators supervised the overall functioning of the study.
Recruitment and operational strategy All grade 2 schoolchildren (Fall 2013) in Taipei City were invited to participate in the MIT. We planned to follow-up with this school cohort for 3 years. The initial citywide survey was conducted between July 2013 and September 2013. Before the beginning of the survey, a coordination meeting was held between the MIT organizers and local education authorities to seek mutual understanding. Then, principals and the school staff, such as school nurses, section chiefs of hygiene and teachers in charge of grade 2 classes of all elementary schools, were contacted and informed of the MIT’s importance and significance. With their support, written informed consent forms were brought home by the schoolchildren to get their parents’ agreement to participate. The school then distributed a package including an eye care passport and parent-administered questionnaires to the family of each eligible schoolchild. Using MIT’s eye care passport, eligible schoolchildren can be taken to visit participating hospitals or eye clinics for free eye examinations once a semester for 3 consecutive years. Media reports and public events were used to publicize the campaign, arouse public awareness of childhood myopia, propagate the importance of myopic intervention, enhance the participation rate and encourage participants to stay in the 3-year follow-up.
Questionnaire survey The MIT parent questionnaire regarding potential myopia risk factors was composed of 38 questions in seven sections. Closed-ended questions with twooption (yes/no) responses or a list of ordered choices were adopted. Respondents were asked to check the choice they felt the most proper. Table 1 shows data items collected in questionnaire survey. The first section was concerned with participants’ demographic information and medical history, such as history of premature birth, asthma, atopic dermatitis or ocular diseases including myopia, strabismus and amblyopia, as well as age of near work initiation and past myopic interventions. In the second section, parents’ information, such as education, occupation and refractive status, was collected. The participants’ near work activities over the past year were assessed in the third section. Their parents were asked to
© 2015 Royal Australian and New Zealand College of Ophthalmologists
Method of myopia investigation in Taipei Table 1.
615
Data items collected in the questionnaire survey of the Myopic Investigation study in Taipei
Domains
Questions
Demographics and medical history of child Age, gender, school location History of premature birth, asthma, or atopic dermatitis History of myopia, strabismus, or amblyopia Treatment history of myopia (atropine, spectacles, contact lens) Treatment history of other ocular conditions Age of initiating near work activities Parents’ information Parental education and occupation Parental refractive status/ presence of high myopia Near work activities (over the past year) Eye-to-object distance when doing near work or watching television Use of mobile devices (smart phone, tablet computer, etc.) Average time spent on near work per day Use of table lamp when doing near work Poor reading posture (such as reading in bed) Adequate rest after doing near work more than 30 min After-school programme Attendance to after-school programme or cram school Average time spent in after-school programme per week Outdoor activities (over the past year) Average time spent on outdoor activities per day during the weekdays Average time spent on outdoor activities per day during the weekend Eye checks Eye examination by ophthalmologist over the past year Knowledge about high myopia Definition of high myopia Complications of high myopia
answer questions about eye-to-object distance, use of mobile devices, average daily quantity of near work activities, including reading, writing, painting, playing instruments, watching television and playing computer/video games, reading posture, lighting, as well as the presence of adequate rest. The fourth section focused on attendance at after-school programmes or cram schools and hours per week spent there. The next section enquired about average hours per day spent engaging in outdoor activities during both weekdays and weekends over the past year. Eye-care seeking behaviour and knowledge about high myopia and myopia-related complications were recorded in the final two sections, respectively. The survey questions and their responses are presented in Supporting Information Table S2. This questionnaire survey will be administered annually during the follow-up.
Clinical examination The parents/caregivers of eligible schoolchildren could choose any MIT-associated hospital/clinic at their convenience to receive the ocular examinations for their children. Figure 1 shows the flow chart of the examination procedures, which are described in the succeeding text.
Figure 1. Flowchart for the eye examination procedures in the Myopic Investigation study in Taipei.
© 2015 Royal Australian and New Zealand College of Ophthalmologists
616
Tsai et al.
Registration, non-cycloplegic autorefraction and visual acuity After confirming the consented school child’s name and demographic details on his or her eye care passport, the refractive status was first measured with an autorefractor. At least three autorefraction readings were acquired and averaged. Unaided visual acuity was assessed monocularly using a Snellen visual acuity chart at a distance of 6 m. The left eye was occluded with an eye occluder when the right eye was tested and vice versa. If the schoolchildren could not read any letters at 6 m, they were moved to 1 m in front of the chart and tested again, allowing for measurement of visual acuity as low as 1/60. Subjective refraction was performed to measure the bestcorrected visual acuity for those whose unaided visual acuity was less than 6/6 in either eye.
Slit-lamp examination The conjunctiva, cornea, anterior chamber, pupil, iris, lens and anterior vitreous were examined using slit-lamp biomicroscopy.
Pupil dilation and cycloplegic autorefraction Three kinds of cycloplegic eye drops were approved for use in the MIT, including Cyclogyl (1% cyclopentolate), Mydriacyl (1% tropicamide) and Mydrin-P (0.5% phenylephrine hydrochloride/0.5% tropicamide). Among these three drugs, each hospital or clinic could use only one kind of cycloplegic eye drop to perform cycloplegia for the MIT participants. There were 20 hospitals/clinics using Cyclogyl, 27 using Mydriacyl and 27 using Mydrin-P. These drugs were chosen based on hospital/clinic preferences. Cycloplegia was achieved with two drops of the same cycloplegic drug (administered 10 min apart). Thirty minutes after the second drop of the cycloplegic agent, a penlight was used to check the pupil light reflex. If the pupil still responded to penlight, an additional 10-min wait was required before cycloplegic refraction. Then three consecutive autorefraction readings were taken using an autorefractor. The spherical equivalent (SE) of the refractive error was calculated as the spherical value plus one half of the cylindrical value.
Definition of refractive status Consistent with recent large epidemiological studies,20,25 we defined myopia as ≤−0.5D SE. Similarly, we used the equivalent definition for other refractive categories (e.g. moderate myopia [−6.0D < SE ≤−3.0D], high myopia [SE ≤−6.0D],
hyperopia [SE ≥+0.5D]). Those who had SE ≤−0.5D in either eye were classified into the myopic group. Anisometropia was defined as the SE difference ≥ 1.0D between eyes. Astigmatism was defined as cylinder power ≤−1.0D in either eye. Fast myopic progression was defined as changes in myopic refraction more than 1.0D per year and slow myopic progression as less than 0.5D per year.12 Among myopic schoolchildren at baseline, cross comparisons between their questionnaire data and clinical examinations were performed and those without history of reported myopia were identified as previously undiagnosed myopes. All the results of clinical examination were reported to the Department of Health, Taipei City Government for data entry and further analysis.
Telecoaching According to the cycloplegic refraction status measurements, the list of myopic schoolchildren was collected by thoroughly trained case managers. Case manager-led telecoaching for myopia health-care instructions and reminders were then attempted to be delivered to the parents/caregivers of the myopic schoolchildren. Up to three phone calls were made to reach each case’s parents/caregivers. On the phone, the case managers informed these parents/caregivers about the risk and potential complications of high myopia and the modifiable risk factors for myopia development, such as excessive near work activities and deficiency of outdoor activities. In addition, the case managers recorded myopic treatment history and treatment preferences, such as atropine, spectacles or contact lenses etc., and urged parents/ caregivers to adhere to pharmacological, optical and/or environmental interventions. The content of structured coaching session is summarized in Supporting Information Table S3.
Data entry and statistical analysis Databases of both questionnaire items and examination variables were constructed using Microsoft Access database software. Questionnaire items and examination variables were coded and two independent administrative staff performed double data entry. Participants who had missing data, misdata or logic errors in their questionnaires and examination records were excluded from further analysis. Database quality, such as consistency between paper documents and electronic records, was monitored and assessed by MIT committee epidemiologists. Statistical analysis was performed using Statistical Analysis System Software (Statistical Analysis System Software V9.3, SAS Institute, Cary, NC, USA). Only the more myopic eye in each child was
© 2015 Royal Australian and New Zealand College of Ophthalmologists
Method of myopia investigation in Taipei
617 complete and analysed (n = 11 590), those who responded to the questionnaire survey but did not undergo an eye examination (n = 4467) and those who did not complete the whole procedure or did not participate at all (n = 7784). More than half of the analysed participants were male (52.9%) and went to elementary schools in urban districts (51.9%). These demographic distributions were similar to those of all the grade 2 students in Taipei City (P = 0.690 and 0.170, respectively), although the analysed participants were less likely to attend schools in the urban districts when compared with the unanalysed and non-participants and the participants that did not undergo eye examination (51.9% vs. 53.9% and 55.6%). Results regarding myopia prevalence, incidence and risk factors will be analysed in subsequent reports.
Figure 2. Flowchart for ascertainment of the Myopic Investigation study in Taipei.
included for myopia analysis in the present study. For myopia prevalence and incidence estimations to be completed in the future, overall and genderspecific estimates will be calculated and the difference between genders will be analysed using chi-square test. For risk factor analysis, frequency of categorical variables in schoolchildren with and without the outcome of interest (e.g. prevalent myopia, incident myopia, myopic progression, previously undiagnosed myopia) will be compared using chi-square tests. For continuous variables, t-tests and analysis of variance will be used. We will perform multiple logistic regressions to control for potential confounders and determine the adjusted odds ratios. Risk factors on time to event (incident myopia) will be analysed using the Cox regression model. Furthermore, eye-specific analyses will be conducted using generalized estimating equation models.
RESULTS The recruitment process is shown in Figure 2. Among all the 19 374 grade 2 schoolchildren in Taipei City, a total of 16 486 (85.1%) students’ parental consent forms and parent-administered questionnaires were collected, 12 019 (62.0%) actually underwent eye examinations during the period from July 2013 to September 2013 and 11 590 (59.8%) had completed cycloplegic autorefraction measurements on both eyes. The data of these 11 590 participants were further analysed. Table 2 displays the demographic characteristics of all grade 2 schoolchildren (n = 19 374), the participants whose data were
DISCUSSION The MIT is intended to provide large-scale population-based information concerning the frequency and associated factors of childhood myopia in a metropolitan area of northern Taiwan. In addition, the MIT will report the 3-year incidence data of myopia based on semiannual follow-up examinations of participants. In order to facilitate appropriate comparisons of the myopia prevalence and incidence rate across different studies, we adopted the population-based approach, utilized a narrow agerange cohort design, had an equivalent definition for myopia and refracted schoolchildren under cycloplegia, which are steps similar to other major epidemiological studies.20,25 Data from the MIT will give us insight into myopia epidemiology among young, ethnic Chinese schoolchildren in the modern era of pervasive mobile device usage. Unlike many other large-scale studies of myopia epidemiology,1–6,9,20,21,25 the MIT did not use a sampling frame design to estimate myopia prevalence. Aiming to identify every undiagnosed myope in early childhood, the MIT invited all grade 2 schoolchildren in Taipei City to participate with the help of school authorities. Thanks to the compulsory education system and high attendance rate, the MIT was able to analyse the data of 11 590 students, making it one of the largest epidemiological studies of childhood myopia. With the MIT study design, potential selection bias occurring in the sampling procedure could be minimized. Table 2 shows no significant difference in the baseline demographic characteristics between the 11 590 analysed participants and all 19 374 eligible students. The high degree of representativeness strengthens our confidence that the evaluations of associations, including those of confounders to the exposures and outcomes and among variables of interest, are more likely to be unbiased.
© 2015 Royal Australian and New Zealand College of Ophthalmologists
618
Tsai et al.
Table 2. Demographic characteristics of study population Analysed participants (n = 11 590)
Urbanization Urban Suburban Gender Female Male
Unanalysed and non-participants (n = 7784)
n
(%)
n
(%)
6011 5579
51.9 48.1
4193 3591
53.9 46.1
5455 6135
47.1 52.9
3709 4075
47.6 52.4
Participants without eye examinations (n = 4467) P value†
n
(%)
2485 1982
55.6 44.4
2085 2382
46.7 53.3
All Grade 2 schoolchildren (n = 19 374) P value‡
n
(%)
10 204 9170
52.7 47.3
9164 10 210
47.3 52.7
Clinical and Experimental Ophthalmology 2015; 43: 612–620 doi: 10.1111/ceo.12532
Original Article Study design, rationale and methods for a population-based study of myopia in schoolchildren: the Myopia Investigation study in Taipei Der-Chong Tsai MD PhD,1,2* Li-Ju Lin PhD,3* Nicole Huang PhD,4 Chih-Chien Hsu MD,5,6 Shing-Yi Chen MS,3 Allen Wen-Hsiang Chiu MD PhD1 and Catherine Jui-Ling Liu MD1,6 1
School of Medicine, 4Institute of Hospital and Health Care Administration, 5Institute of Clinical Medicine, National Yang-Ming University, 3Department of Health, Taipei City Government, 6Department of Ophthalmology, Taipei Veterans General Hospital, Taipei, 2 Department of Ophthalmology, National Yang-Ming University Hospital, Yilan, Taiwan
ABSTRACT Background: To describe the study design, rationale and methodology of the Myopia Investigation Study in Taipei (MIT). Design: The MIT was a citywide, population-based cohort study. Participants: Participants were grade 2 students (Fall 2013) of all 153 elementary schools in Taipei City. Methods: The baseline data on the risk factors for myopia development was collected by parentadministered questionnaire surveys covering demographics, medical history, parental myopia, time spent on near work and outdoor activities, reading habits and eye care-seeking behaviour. Ocular examinations focused on the measurement of visual acuity (unaided and best-corrected) and refractive status (before and after cycloplegia), which will be carried out for the eligible schoolchildren biannually for 3 years consecutively. Once myopic children are identified, case manager-led telecoaching for healthcare instructions and reminders will be delivered to parents or caregivers.
Main Outcome Measures: To build a comprehensive database for prevalence, incidence and risk factors of early childhood myopia over a 3-year follow-up period. Results: Of all 19 374 eight-year-old schoolchildren (10 210 [52.7%] boys) eligible for the MIT, 16 486 (85.1%) responded to the questionnaire, 12 019 (62.0%) were examined during the third quarter of 2013 and 11 590 (59.8%) (6267 [52.9%] boys) completed cycloplegic autorefraction on both eyes and were enrolled for further data analysis. There was no significant difference in terms of demographics between the analysed participants and all grade 2 students in Taipei City. Conclusions: Data from the MIT will provide population-based information concerning the prevalence, incidence and risk factors for myopia development among young schoolchildren in a metropolitan area of Taiwan. Key words: cohort study, methodology, myopia, population-based study, schoolchildren.
■ Correspondence: Prof Catherine Jui-Ling Liu, Department of Ophthalmology, Taipei Veterans General Hospital, No. 201, Sec. 2, Shih-Pai Road, Taipei 112, Taiwan. Email: [email protected] Received 1 January 2015; accepted 1 April 2015. *These authors contributed equally to this work. Conflicts of interest: None. Funding sources: The present study was supported by grants from H10237 and P10303 from the Taipei City Government. © 2015 Royal Australian and New Zealand College of Ophthalmologists
Method of myopia investigation in Taipei
INTRODUCTION Myopia is one of the most common eye disorders worldwide, especially in the industrialized regions of East Asia where more than 80% are affected in some populations.1–7 Myopes with an earlier age of onset may suffer from faster myopic progression and eventually be more likely to have high myopia (≤−6.0 diopter [D]).2,8 Although the refractive vision problem is usually correctable with the aid of spectacles, contact lenses or refractive surgeries, the ocular comorbidities associated with high myopia, such as rhegmatogenous retinal detachment, degenerative maculopathy and glaucoma may contribute to an increased risk of irreversible loss of vision and impose heavy socioeconomic burdens to both family and community.2 Several population-based prevalence surveys revealed that the prevalence of myopia has escalated considerably over the past few decades.1,2,4,9 This secular trend is evident in Taiwan, where there was a 3.6-fold increase in myopia prevalence from 1983 (5.8%) to 2000 (21%) among the 7-year-old schoolchild population.4 An epidemiological study recently reported a prevalence of myopia as high as 86.1% among young Taiwanese men aged 18 to 24 years.7 Similarly, the prevalence of myopia in the United States appears to be substantially higher from 1999– 2004 than 30 years earlier.9 Intensive environmental pressures, such as rapid urbanization and a rigorous education system, are considered the driving forces behind the changes in prevalence of myopia over time.10 Given the swift rise in popularity and deep penetration of mobile devices, such as smartphones and tablet computers, over the last few years, the environmental pressures would have become more intensive than ever and this public health issue clearly warrants further epidemiological investigations. Although exact mechanism of myopic onset and progression is still not well understood, pharmacological, environmental and optical interventions have been used for myopia control among schoolchildren.11 Numerous studies have disclosed that atropine is effective in slowing the progression of childhood myopia.12–16 Despite potentially annoying side effects, including photophobia and blurred near vision, there is an increasing trend of atropine eye drop prescription for myopic children in Taiwan.17 Increasing time spent outdoors and reducing peripheral hyperopia with optical devices also have been proven to protect against the development of myopia.18,19 Although the optimal mode of myopia prevention has not yet been made clear, it is desirable to have myopic children and their parents adhere to these interventions until stabilization of refractive status by the late teen years.
613 For early identification of myopes and retardation of myopic progression among the youth population, the Myopia Investigation study in Taipei (MIT) was designed to provide a citywide eye examination programme for grade 2 students of all 153 elementary schools in Taipei. Once myopic schoolchildren are identified, telecoaching service will be provided to promote their adherence to myopic intervention. Furthermore, the MIT will follow-up the refractive status of this school cohort biannually for 3 years to bridge the knowledge gap concerning the myopia incidence and risk factors of progression in Taiwan. Compared with numerous cross-sectional studies of school myopia prevalence, there is relatively fewer data on myopia incidence from longitudinal cohort studies.1,2,20,21 The purpose of this paper is to summarize the study design and rationale of the MIT and to describe the procedures used to carry out this citywide programme.
METHODS Study design The MIT is a population-based cohort study initiated and funded through a programme grant from the Taipei City Government. It was set up in 2013 and designed to investigate prospectively the development of myopia among the 8-year-old schoolchildren (born mainly between September 2005 and August 2006) in Taipei City, which is one of the areas with the highest prevalence of myopia in the world.2–4 A previous survey in Taiwan revealed that the mean refractive status becomes myopic at the age of 9 years.3 For early identification and intervention of myopia in young children, we chose grade 2 students and will continue to follow their refractive changes for 3 years. Written informed consent was obtained from either parent of each child and the Declaration of Helsinki was adhered to throughout. The protocol and consent procedure were approved by the Institutional Review Board of the Taipei City Hospital (TCHIRB-1020501).
Specific aims In addition to the baseline examination in 2013, there will be up to five episodes of citywide refractive status surveys performed biannually (i.e. once per semester) in the following 3 years. The specific aims of the MIT include the following: 1
2
© 2015 Royal Australian and New Zealand College of Ophthalmologists
Determining the baseline prevalence of myopia in the population of aforementioned young schoolchildren (primary outcome); Determining the 1-year to 3-year incidence of myopia among those non-myopic participants at baseline (primary outcome);
614 3
4
5
6
Tsai et al. Determining the degree of myopia progression during the up to 3-year follow-ups among those myopic participants at baseline (secondary outcome); Examining the associations between both modifiable (environmental) and non-modifiable (hereditary) risk factors and myopia development (secondary outcome); Determining the prevalence and associated factors of previously undiagnosed myopia (secondary outcome). Determining the prevalence and associated factors of anisometropia and astigmatism (secondary outcome).
Study area Taipei City, the capital of Taiwan situated in the northern part of the island, has a population of approximately 2.6 million and an area of 271.7 km2.22 It is the political, economic and cultural center of Taiwan. There are 12 administrative districts in Taipei City. Of those, seven districts are defined as urban and five districts as suburban (Supporting Information Table S1), according to the criteria of urbanization classification based mainly on population density, medical resources, age and residents’ education levels.23 There are 153 elementary schools across the city.24 Due to the well-executed compulsory education system, the student attendance rate at elementary schools is greater than 99% in Taipei City and there were 19 374 schoolchildren in grade 2 during the 2013 Fall semester.24 The distribution of these 19 374 students by administrative districts is provided in Supporting Information Table S1. In addition, the nationalized health-care system is also well established in Taiwan. A total of 24 hospitals and 123 private clinics provide eye health-care services in Taipei City.
Organizational structure National Yang-Ming University has a mandate from the Taipei City Government to help organize and monitor the MIT. A total of 20 hospitals and 54 private clinics across the 12 districts have joined the MIT, providing hospital-based or clinic-based eye examinations for eligible schoolchildren. To meet a consistent standard operating procedure, the MIT organizing committee formulated a standardized protocol for staff training and held a series of coordination courses for participating ophthalmologists and other registered medical practitioners. The principal investigators of the MIT also invited 11 senior ophthalmologists and four epidemiologists to build a committee to monitor the quality of the study execution. They regularly visited the MIT-associated
hospitals/clinics to evaluate eye examination procedures during the survey campaign. It was requested that all examinations be carried out according to the standardized protocol. Full-time administrative staff and three trained case managers were recruited. The principal investigators supervised the overall functioning of the study.
Recruitment and operational strategy All grade 2 schoolchildren (Fall 2013) in Taipei City were invited to participate in the MIT. We planned to follow-up with this school cohort for 3 years. The initial citywide survey was conducted between July 2013 and September 2013. Before the beginning of the survey, a coordination meeting was held between the MIT organizers and local education authorities to seek mutual understanding. Then, principals and the school staff, such as school nurses, section chiefs of hygiene and teachers in charge of grade 2 classes of all elementary schools, were contacted and informed of the MIT’s importance and significance. With their support, written informed consent forms were brought home by the schoolchildren to get their parents’ agreement to participate. The school then distributed a package including an eye care passport and parent-administered questionnaires to the family of each eligible schoolchild. Using MIT’s eye care passport, eligible schoolchildren can be taken to visit participating hospitals or eye clinics for free eye examinations once a semester for 3 consecutive years. Media reports and public events were used to publicize the campaign, arouse public awareness of childhood myopia, propagate the importance of myopic intervention, enhance the participation rate and encourage participants to stay in the 3-year follow-up.
Questionnaire survey The MIT parent questionnaire regarding potential myopia risk factors was composed of 38 questions in seven sections. Closed-ended questions with twooption (yes/no) responses or a list of ordered choices were adopted. Respondents were asked to check the choice they felt the most proper. Table 1 shows data items collected in questionnaire survey. The first section was concerned with participants’ demographic information and medical history, such as history of premature birth, asthma, atopic dermatitis or ocular diseases including myopia, strabismus and amblyopia, as well as age of near work initiation and past myopic interventions. In the second section, parents’ information, such as education, occupation and refractive status, was collected. The participants’ near work activities over the past year were assessed in the third section. Their parents were asked to
© 2015 Royal Australian and New Zealand College of Ophthalmologists
Method of myopia investigation in Taipei Table 1.
615
Data items collected in the questionnaire survey of the Myopic Investigation study in Taipei
Domains
Questions
Demographics and medical history of child Age, gender, school location History of premature birth, asthma, or atopic dermatitis History of myopia, strabismus, or amblyopia Treatment history of myopia (atropine, spectacles, contact lens) Treatment history of other ocular conditions Age of initiating near work activities Parents’ information Parental education and occupation Parental refractive status/ presence of high myopia Near work activities (over the past year) Eye-to-object distance when doing near work or watching television Use of mobile devices (smart phone, tablet computer, etc.) Average time spent on near work per day Use of table lamp when doing near work Poor reading posture (such as reading in bed) Adequate rest after doing near work more than 30 min After-school programme Attendance to after-school programme or cram school Average time spent in after-school programme per week Outdoor activities (over the past year) Average time spent on outdoor activities per day during the weekdays Average time spent on outdoor activities per day during the weekend Eye checks Eye examination by ophthalmologist over the past year Knowledge about high myopia Definition of high myopia Complications of high myopia
answer questions about eye-to-object distance, use of mobile devices, average daily quantity of near work activities, including reading, writing, painting, playing instruments, watching television and playing computer/video games, reading posture, lighting, as well as the presence of adequate rest. The fourth section focused on attendance at after-school programmes or cram schools and hours per week spent there. The next section enquired about average hours per day spent engaging in outdoor activities during both weekdays and weekends over the past year. Eye-care seeking behaviour and knowledge about high myopia and myopia-related complications were recorded in the final two sections, respectively. The survey questions and their responses are presented in Supporting Information Table S2. This questionnaire survey will be administered annually during the follow-up.
Clinical examination The parents/caregivers of eligible schoolchildren could choose any MIT-associated hospital/clinic at their convenience to receive the ocular examinations for their children. Figure 1 shows the flow chart of the examination procedures, which are described in the succeeding text.
Figure 1. Flowchart for the eye examination procedures in the Myopic Investigation study in Taipei.
© 2015 Royal Australian and New Zealand College of Ophthalmologists
616
Tsai et al.
Registration, non-cycloplegic autorefraction and visual acuity After confirming the consented school child’s name and demographic details on his or her eye care passport, the refractive status was first measured with an autorefractor. At least three autorefraction readings were acquired and averaged. Unaided visual acuity was assessed monocularly using a Snellen visual acuity chart at a distance of 6 m. The left eye was occluded with an eye occluder when the right eye was tested and vice versa. If the schoolchildren could not read any letters at 6 m, they were moved to 1 m in front of the chart and tested again, allowing for measurement of visual acuity as low as 1/60. Subjective refraction was performed to measure the bestcorrected visual acuity for those whose unaided visual acuity was less than 6/6 in either eye.
Slit-lamp examination The conjunctiva, cornea, anterior chamber, pupil, iris, lens and anterior vitreous were examined using slit-lamp biomicroscopy.
Pupil dilation and cycloplegic autorefraction Three kinds of cycloplegic eye drops were approved for use in the MIT, including Cyclogyl (1% cyclopentolate), Mydriacyl (1% tropicamide) and Mydrin-P (0.5% phenylephrine hydrochloride/0.5% tropicamide). Among these three drugs, each hospital or clinic could use only one kind of cycloplegic eye drop to perform cycloplegia for the MIT participants. There were 20 hospitals/clinics using Cyclogyl, 27 using Mydriacyl and 27 using Mydrin-P. These drugs were chosen based on hospital/clinic preferences. Cycloplegia was achieved with two drops of the same cycloplegic drug (administered 10 min apart). Thirty minutes after the second drop of the cycloplegic agent, a penlight was used to check the pupil light reflex. If the pupil still responded to penlight, an additional 10-min wait was required before cycloplegic refraction. Then three consecutive autorefraction readings were taken using an autorefractor. The spherical equivalent (SE) of the refractive error was calculated as the spherical value plus one half of the cylindrical value.
Definition of refractive status Consistent with recent large epidemiological studies,20,25 we defined myopia as ≤−0.5D SE. Similarly, we used the equivalent definition for other refractive categories (e.g. moderate myopia [−6.0D < SE ≤−3.0D], high myopia [SE ≤−6.0D],
hyperopia [SE ≥+0.5D]). Those who had SE ≤−0.5D in either eye were classified into the myopic group. Anisometropia was defined as the SE difference ≥ 1.0D between eyes. Astigmatism was defined as cylinder power ≤−1.0D in either eye. Fast myopic progression was defined as changes in myopic refraction more than 1.0D per year and slow myopic progression as less than 0.5D per year.12 Among myopic schoolchildren at baseline, cross comparisons between their questionnaire data and clinical examinations were performed and those without history of reported myopia were identified as previously undiagnosed myopes. All the results of clinical examination were reported to the Department of Health, Taipei City Government for data entry and further analysis.
Telecoaching According to the cycloplegic refraction status measurements, the list of myopic schoolchildren was collected by thoroughly trained case managers. Case manager-led telecoaching for myopia health-care instructions and reminders were then attempted to be delivered to the parents/caregivers of the myopic schoolchildren. Up to three phone calls were made to reach each case’s parents/caregivers. On the phone, the case managers informed these parents/caregivers about the risk and potential complications of high myopia and the modifiable risk factors for myopia development, such as excessive near work activities and deficiency of outdoor activities. In addition, the case managers recorded myopic treatment history and treatment preferences, such as atropine, spectacles or contact lenses etc., and urged parents/ caregivers to adhere to pharmacological, optical and/or environmental interventions. The content of structured coaching session is summarized in Supporting Information Table S3.
Data entry and statistical analysis Databases of both questionnaire items and examination variables were constructed using Microsoft Access database software. Questionnaire items and examination variables were coded and two independent administrative staff performed double data entry. Participants who had missing data, misdata or logic errors in their questionnaires and examination records were excluded from further analysis. Database quality, such as consistency between paper documents and electronic records, was monitored and assessed by MIT committee epidemiologists. Statistical analysis was performed using Statistical Analysis System Software (Statistical Analysis System Software V9.3, SAS Institute, Cary, NC, USA). Only the more myopic eye in each child was
© 2015 Royal Australian and New Zealand College of Ophthalmologists
Method of myopia investigation in Taipei
617 complete and analysed (n = 11 590), those who responded to the questionnaire survey but did not undergo an eye examination (n = 4467) and those who did not complete the whole procedure or did not participate at all (n = 7784). More than half of the analysed participants were male (52.9%) and went to elementary schools in urban districts (51.9%). These demographic distributions were similar to those of all the grade 2 students in Taipei City (P = 0.690 and 0.170, respectively), although the analysed participants were less likely to attend schools in the urban districts when compared with the unanalysed and non-participants and the participants that did not undergo eye examination (51.9% vs. 53.9% and 55.6%). Results regarding myopia prevalence, incidence and risk factors will be analysed in subsequent reports.
Figure 2. Flowchart for ascertainment of the Myopic Investigation study in Taipei.
included for myopia analysis in the present study. For myopia prevalence and incidence estimations to be completed in the future, overall and genderspecific estimates will be calculated and the difference between genders will be analysed using chi-square test. For risk factor analysis, frequency of categorical variables in schoolchildren with and without the outcome of interest (e.g. prevalent myopia, incident myopia, myopic progression, previously undiagnosed myopia) will be compared using chi-square tests. For continuous variables, t-tests and analysis of variance will be used. We will perform multiple logistic regressions to control for potential confounders and determine the adjusted odds ratios. Risk factors on time to event (incident myopia) will be analysed using the Cox regression model. Furthermore, eye-specific analyses will be conducted using generalized estimating equation models.
RESULTS The recruitment process is shown in Figure 2. Among all the 19 374 grade 2 schoolchildren in Taipei City, a total of 16 486 (85.1%) students’ parental consent forms and parent-administered questionnaires were collected, 12 019 (62.0%) actually underwent eye examinations during the period from July 2013 to September 2013 and 11 590 (59.8%) had completed cycloplegic autorefraction measurements on both eyes. The data of these 11 590 participants were further analysed. Table 2 displays the demographic characteristics of all grade 2 schoolchildren (n = 19 374), the participants whose data were
DISCUSSION The MIT is intended to provide large-scale population-based information concerning the frequency and associated factors of childhood myopia in a metropolitan area of northern Taiwan. In addition, the MIT will report the 3-year incidence data of myopia based on semiannual follow-up examinations of participants. In order to facilitate appropriate comparisons of the myopia prevalence and incidence rate across different studies, we adopted the population-based approach, utilized a narrow agerange cohort design, had an equivalent definition for myopia and refracted schoolchildren under cycloplegia, which are steps similar to other major epidemiological studies.20,25 Data from the MIT will give us insight into myopia epidemiology among young, ethnic Chinese schoolchildren in the modern era of pervasive mobile device usage. Unlike many other large-scale studies of myopia epidemiology,1–6,9,20,21,25 the MIT did not use a sampling frame design to estimate myopia prevalence. Aiming to identify every undiagnosed myope in early childhood, the MIT invited all grade 2 schoolchildren in Taipei City to participate with the help of school authorities. Thanks to the compulsory education system and high attendance rate, the MIT was able to analyse the data of 11 590 students, making it one of the largest epidemiological studies of childhood myopia. With the MIT study design, potential selection bias occurring in the sampling procedure could be minimized. Table 2 shows no significant difference in the baseline demographic characteristics between the 11 590 analysed participants and all 19 374 eligible students. The high degree of representativeness strengthens our confidence that the evaluations of associations, including those of confounders to the exposures and outcomes and among variables of interest, are more likely to be unbiased.
© 2015 Royal Australian and New Zealand College of Ophthalmologists
618
Tsai et al.
Table 2. Demographic characteristics of study population Analysed participants (n = 11 590)
Urbanization Urban Suburban Gender Female Male
Unanalysed and non-participants (n = 7784)
n
(%)
n
(%)
6011 5579
51.9 48.1
4193 3591
53.9 46.1
5455 6135
47.1 52.9
3709 4075
47.6 52.4
Participants without eye examinations (n = 4467) P value†
n
(%)
2485 1982
55.6 44.4
2085 2382
46.7 53.3
All Grade 2 schoolchildren (n = 19 374) P value‡
n
(%)
10 204 9170
52.7 47.3
9164 10 210
47.3 52.7
