Optometry & Vision Science:
Myopia Prevalence in Chinese-Canadian Children in an Optometric Practice
CHENG, DESMOND DIPOPT, OD, MSc, FAAO; SCHMID, KATRINA L. BAppSc(Optom)(Hons), PhD; WOO, GEORGE C. OD, MSc, PhD, LOSc, FAAO
School of Optometry and Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Australia (DC, KLS) and School of Optometry, The Hong Kong Polytechnic University, Hong Kong SAR, China (GCW)
This research was completed as part of the PhD program at the School of Optometry and Institute of Health and Biomedical Innovation, Queensland University of Technology.
Received January 14, 2006; accepted October 12, 2006.
Purpose. The high prevalence of myopia in Chinese children living in urban East Asian countries such as Hong Kong, Taiwan, and China has been well documented. However, it is not clear whether the prevalence of myopia would be similarly high for this group of children if they were living in a Western country. This study aims to determine the prevalence and progression of myopia in ethnic Chinese children living in Canada.
Methods. Right eye refraction data of Chinese-Canadian children aged 6 to 12 years were collated from the 2003 clinical records of an optometric practice in Mississauga, Ontario, Canada. Myopia was defined as a spherical equivalent refraction (SER) equal or less than −0.50 D. The prevalence of myopia and refractive error distribution in children of different ages and the magnitude of refractive error shifts over the preceding 8 years were determined. Data were adjusted for potential biases in the clinic sample. A questionnaire was administered to 300 Chinese and 300 Caucasian children randomly selected from the clinic records to study lifestyle issues that may impact on myopia development.
Results. Optometric records of 1468 children were analyzed (729 boys and 739 girls). The clinic bias adjusted prevalence of myopia increased from 22.4% at age 6 to 64.1% at age 12 and concurrently the portion of the children that were emmetropic (refraction between −0.25 and +0.75 D) decreased (68.6% at 6 years to 27.2% at 12 years). The highest incidence of myopia for both girls (∼35%) and boys (∼25%) occurred at 9 and 10 years of age. The average annual refractive shift for all children was −0.52 ± 0.42 D and −0.90 ± 0.40 D for just myopic children. The questionnaire revealed that these Chinese-Canadian children spent a greater amount of time performing near work and less time outdoors than did Caucasian-Canadian children.
Conclusions. Ethnic Chinese children living in Canada develop myopia comparable in prevalence and magnitude to those living in urban East Asian countries. Recent migration of the children and their families to Canada does not appear to lower their myopia risk.
In urban East Asian countries, at least 50 to 60% of children by the age of 12 years are myopic,1–4 whereas the prevalence of myopia in Caucasian children of similar age is merely 10 to 20% (Europe,5 United States,6 and Australia7). This difference in prevalence appears a relatively recent occurrence, with more Chinese children today being myopic compared to in the past and also compared to the portion of older Chinese people that are myopic.8 For example, in Taiwan, using a myopia definition of ≤−0.25 D the prevalence of myopia in Chinese children has increased 6.7 times (from 3 to 20.2%) and doubled (from 29 to 60.7%) for children aged 7 and 12 years respectively (year 2000 compared with 1986).4,9,10 In Singapore, Chinese male conscripts in the late 1980s had a myopia prevalence of 40 to 50% with myopia defined as a visual acuity of <6/1811,12 and in a mid 1990s study13 with myopia defined as ≤−0.5 D, the prevalence had increased to 80%. Similarly, in family studies of myopia over three generations in Hong Kong, the prevalence of myopia increased significantly from one generation to the next.14,15 Such a difference is also found in Singapore, where the prevalence of myopia (<−1 D) in Chinese adults aged between 40 and 49 years is about 50%,16 which is lower than that of younger groups (e.g., 82.2% for young adults aged 17 to 19 years with a myopia cutoff ≤−0.5 D).13 In Europe and North America, a similar longitudinal increase in the prevalence of myopia has also been noticed, but the prevalence has not risen to the level seen in East Asian countries.8 Among the prevalence of myopia data for Europe and North America, the highest ever reported is 49.7% (cutoff ≤−0.5 D) in Sweden17 for children aged 12 to 13 years.
In addition to the high prevalence rate, the speed at which the refraction shifts in the myopic direction in Chinese children is particularly high. The average rate of refractive change of Hong Kong children aged 7 to 12 years was −0.32 D per year in the 1990s.3 This progression rate is more than three times the rate that is reported for Caucasian children of a similar age (6 to 11 years; −0.09 D per year).18 In a more recent Hong Kong study,19 the average rate of refractive change of children aged between 5 and 16 years was −0.40 D per year. In Singapore, the mean rate of refractive change in a group of primarily Chinese children aged 8 to 9 years was 0.57 to 0.80 D per year.20 Though direct comparison of the progression rate data from these studies is complicated by the differences in the age range of the children and other characteristics of the population, it appears in general that the rate of myopic shift in Chinese children has increased in recent times.
The high prevalence of myopia in certain ethnic populations, such as Chinese and Japanese people, suggests that genetic input plays a role in myopia development, but the rapid increase in prevalence over the last few decades indicates that environmental factors are also important. It is generally believed that the genetic characteristics of East Asian children predispose them to myopia development and faster myopia progression,21 and that this genetic background coupled with an increasingly competitive educational environment has contributed to the upward trend in myopia prevalence. One piece of evidence for genetic and environment interaction is that male Chinese conscripts (82.2%) have a higher prevalence of myopia than that of male Indian conscripts (68.8%)13 in Singapore, but male Indian conscripts in Singapore have a higher prevalence of myopia than that of young adults in India (10.8%).22 The difference in prevalence between Chinese and Indian people living in a relatively common environment could be genetic or cultural (e.g., Chinese children perform more near work) in origin, but environmental influences appear to result in much higher prevalence levels for Indian people living in Singapore when compared with those living in India. Further support of the environmental influences on myopia, are reports that the prevalence of myopia in Chinese children living in urban Hong Kong1,3,19 is higher than that for children in rural China23 (by at least 20%) and this urban rural difference is also found in other countries e.g., in Malaysia,24 India,25,26 and Japan.27 These rural localities presumably have a less competitive educational environment than the large urban centers. The increased education demands especially in Asian urban cities have resulted in children spending longer hours reading and studying, and the prolonged periods of near work activity are believed to be related to the higher myopia prevalence.11,12,28,29 Although the process by which near work exacerbates myopia has yet to be fully determined, studies on different occupational groups, educational environments, and animal myopia models suggest that near work plays an important role in myopia development. A higher prevalence of myopia is observed in microscopists,30 in children who spent more time doing close work2,6,18 and in populations with higher levels of educational achievement.29,31
Two reports have suggested that the large number of Asian migrants in recent decades has contributed to an overall increase in the prevalence of myopia in children living in Australia32 and the United States.33 However, there have been limited studies on the effect of ethnic demographics on the prevalence of myopia in Western countries, and there have been no reports of prevalence data for ethnic Chinese children living in Western countries. A related study on university students in the United Kingdom reveals no significant difference in the prevalence of myopia (∼50%) between White and Asian (non-Chinese, South Asian)34 students educated exclusively under the UK educational system. These results suggest a predominant environmental influence in a sample of high educational achievement. A question that remains unanswered is whether the prevalence of myopia in Chinese children will be high regardless of where the urban environment is that they live in.
To determine meaningful comparison of myopia prevalence data, the refractive error information of a large group of ethnic Chinese children living in a Western country is required; finding such a population has its own difficulties. In the late 1980s and early 90s, a large number of Chinese people from Asian countries such as Hong Kong, China, and Taiwan migrated to Western countries for political and educational reasons. These 1st generation migrants provide an excellent opportunity to study myopia prevalence of ethnic Chinese people living in a Western influenced environment. This study aimed to determine whether Chinese children who have migrated to Western countries with less congested living space and presumably less intensive schoolwork systems will have lower myopia prevalence and slower progression. A questionnaire was also administered to determine if these Chinese children have adapted to a Western lifestyle (in terms of their visual activities).
Refraction data of Chinese children aged 6 to 12 years were collated from the clinic records of an optometric practice in Canada (Dr. Desmond Cheng & Associates, Mississauga, Ontario, Canada, 1997–2003). According to 2001 census data, the total population of Mississauga was 610,815; of this the ethnic Chinese population was 35,959 (5.9%), with 3700 Chinese children aged 6 to 12 years (10.3% of the total Chinese community). The majority (at least 80%) of the ethnic Chinese people were first generation migrants from Hong Kong with a first spoken language of Cantonese. There were 1489 optometric records of ethnic Chinese children identified, representing 40% of the Chinese children in this age group living in Mississauga. The 2211 Chinese-Canadian children not in the studied sample were examined by another 90 optometrists in the area. Of the 1489 records identified, 21 children had ocular pathology or other visual problems, and their data were excluded from the analysis. These 21 children included three children with esotropia, eight with exotropia, nine with anisometropia (>1.50 D) including eight with amblyopia, and one with corneal disease, leaving the records of 1468 children with normal vision or only refractive problems. The total number of children included 729 boys (49.66%) and 739 girls (50.34%).
In Ontario, children’s eye examinations are covered by the Ontario Health Insurance plan (OHIP), and the public is aware of the importance of an early eye examination for their children. Therefore, data collected from this optometric practice should avoid strong bias toward high income families or at risk populations. In predicting the potential bias of this clinic sample, refraction data of 6-year-old Canadian-Caucasian children were also collated from the practice to compare with that of a 1998 prevalence of myopia study on a province-wide sample of 6-year-old children in New Brunswick, Canada.35 Two hundred and fifty-two optometric records of 6-year-old Caucasian children with normal vision or only refractive problems were identified. The prevalence of myopia in Chinese children from the clinic sample was then adjusted by this bias factor to determine the prevalence of myopia of Chinese children living in Canada. In addition, information on the prevalence of parental myopia of the Chinese-Canadian children was also evaluated for comparison purposes. Refraction data of 665 male parents and 777 female parents of age between 40 and 49 years were collated from the clinical records to determine if the prevalence of myopia in Chinese-Canadian adults (i.e., the parents of the children) was also high.
In the annual comprehensive eye examination, the preliminary testing had included monocular and binocular unaided and/or corrected visual acuities, unilateral and alternate cover test, near-point of convergence, pupil reactions, and stereopsis (Stereofly test, Stereo Optical, Chicago). Refraction had been measured objectively using a non-cycloplegic auto-kerato-refractor (TOPCON KR7000, Topcon Corporation, Tokyo) followed by non-cycloplegic subjective refraction at 6 m. The subjective refraction had been carried out by the same practitioner (Dr. Cheng) throughout the study period. The end-point chosen for the subjective refraction was maximum plus power for best visual acuity. Intraocular pressure had been measured by a noncontact tonometer (Reichart AT550, Reichart Ophthalmic Instruments, Buffalo). Ocular health of the anterior segment had been assessed using slit lamp biomicroscopy and that of the posterior segment using fundus biomicroscopy (90 D lens with the slit lamp). Cycloplegic refraction and dilated pupil fundus examination had been performed on indication (e.g., hyperopia >1 D or a family history of retinal detachment). A total of 63 children had cycloplegic refraction.
To study issues surrounding the visual activities of the children, a questionnaire was administered by mail, with a follow-up phone call if required, to 300 Chinese children and 300 Caucasian children randomly selected from the clinic records,. The format of the questionnaire was adopted from the Orinda Longitudinal Study of Myopia6 except for additional questions on the time the child spent engaged in portable games and in computer work. The questionnaire has previously been reported to provide a valid assessment of near work.6 In addition to the 30 to 32 h per week Canadian children spent at school, the questionnaire asked parents on average how many hours per week their child spent performing each of the following activities outside of school: studies or does school assignments (computer work not included), engages in computer work (assignments, games, and internet), reads for pleasure, watches television or plays television based games (e.g., playstation), plays gameboy or other small hand held portable games, engages in outdoor/sport activities.
In the analysis of refraction data, only refractive error data for right eyes were used. For the purpose of determining myopia prevalence, myopia was defined as a spherical equivalent refraction (SER) ≤−0.50 D. Since retrospective cross-sectional refraction data from clinical records included both dependent (data from the same child appeared in more than one age group) and independent (data from some children only appeared in one age group) data, the results pertaining to the prevalence of myopia and refractive error distributions have been restricted to graphical analysis and descriptive statistics. The refractive error distribution is described by skewness and kurtosis. The skewness for a normal distribution is zero, and any symmetric data should have a skewness near zero. Kurtosis is a measure of whether the data are peaked or flat relative to a normal distribution. That is, data sets with high kurtosis tend to have a distinct peak near the mean, decline rather rapidly, and have heavy tails. A standard normal distribution has a kurtosis of zero. Data were grouped into six age bands [6 to 7 (n = 196), 7 to 8 (n = 244), 8 to 9 (n = 304), 9 to 10 (n = 306), 10 to 11 (n = 313), and 11 to 12 years (n = 289)] to determine yearly rate of refractive change and the yearly cumulative incidence of myopia. Refraction data of children who had two years consecutive measurements were used. The yearly rate of refractive change in a given age band (e.g., 6 to 7) was the difference between the refraction measured in the “second” year (e.g., 7 years) to that measured in the “first” year (e.g., 6 years). The 1-year cumulative incidence of myopia for a given age band (e.g., 6 to 7) was the percentage of previously non-myopic children (e.g., 6 years) who became myopic (≤−0.50 D) in the following year (e.g., 7 years). With this method of analysis, some children did not have their refraction data included in any of the age bands (e.g., if their refractive error was not measured in consecutive years), some children would have their refraction data only in one age band, whereas others would have refraction data in more than one age band (e.g., if their refractive error was measured for more than 2 years consecutively.) There were also some children with refraction data in two or more discrete age bands (e.g., age band 6 to 7 and age band 10 to 11). As a result, the refraction data across age bands includes both dependent and independent data; this is unavoidable in a retrospective analysis. To satisfy statistical laws of comparison, only independent data were used for statistical analysis; those dependent data were randomly removed when the significance of gender and age on refractive error was determined by the analysis of variance (ANOVA); p < 0.05 indicated that the effect was significant. The normality of refractive error distributions was determined using the Shapiro–Wilk test.
Prevalence of Myopia
The prevalence of myopia (≤−0.50 D) showed a positive correlation with age for both girls and boys (Fig. 1) (r2 = 0.97 and p < 0.05 for girls and r2 = 0.99 and p < 0.05 for boys). The myopia prevalence increased from 26.6% for girls and 23.3% for boys at age 6 to 73.2% for girls and 69.2% for boys at age 12. Girls had a slightly higher prevalence of myopia than boys except for a similar prevalence value of 41.5% at age 8. When data for girls and boys were grouped together, the prevalence of myopia increased from 24.9% at age 6 to 71.2% at age 12. The prevalence of hyperopia (>+0.75 D) was initially low and also decreased; from 3.5% for girls and 7.4% for boys at age 6 years to 1.4% for girls and 1.8% for boys at age 12. Girls had a slightly lower prevalence of hyperopia.
Refractive Error Distribution
At each age the refractive error distribution was significantly different (p < 0.001) from a normal distribution (Figs. 2A and B, 3A and B). The percentage of children with myopia has been shown rather than the actual numbers to better reflect the pattern of change. At age 6, the curve was leptokurtic with more data concentrated about the mean (value for kurtosis was 13.0 for girls and 20.4 for boys) and the most common error was between +0.12 and +0.87 D for both girls and boys. The value representing skewness of the refractive distribution at age 6 was more negative (−1.3) for girls due to a high myopic data point at −6 to −6.87 D and more positive (2.8) for boys due to a high hyperopic data point at +7 to +7.87 D. If these two data points were removed, the value became 0.1 for girls and 0.4 for boys indicating only a slight skew toward hyperopia. At age 12, the leptokurtosis of the distribution decreased (value for kurtosis was 0.81 for girls and 0.84 for boys) as the refractive data became more spread. At this later age (12 years), the distribution showed a greater skewness toward myopic refractive errors, giving a negative tail with a skew value of −0.87 for girls and −0.93 for boys.
The leptokurtosis and skewness in refractive error distribution were similar for girls and boys, except that starting from age 9 slightly more girls than boys become myopic and by 12 years 19.7% of girls but only 13.4% of boys were highly myopic (≥4 D of myopia). When the girls and boys data were grouped together, the proportion of children with emmetropia (refractions between −0.25 and +0.75 D) was greatest at age 6, with 68.6%, and decreased at older ages, with 27.2% at age 12. At 6 years of age, 0.3% of children had ≥4 D of myopia and no children had myopia ≥8 D, by 9 years of age 5.2% had ≥4 D of myopia and 0.2% had myopia ≥8 D. By 12 years, these numbers had increased such that 16.7% of children had ≥4 D of myopia and 0.9% more than ≥8 D of myopia.
The average refractive error of all children at age 6 years was −0.02 ± 1.02 D and for just the myopic children was −1.12 ± 0.86 D, whereas at 12 years the average refractive error of all children was −1.97 ± 2.04 D and it was −2.97 ± 1.82 D for just the myopic children. There were very few hyperopes (>+0.75 D) (∼1.5% for age 12 years) but for those that were hyperopic, the average amount of hyperopia was +1.73 ± 1.57 D at 6 years and +1.95 ± 0.87 D at 12 years.
Incidence of Myopia
The 1-year cumulative incidence of myopia for a given age band (e.g., 6 to 7) was the percentage of non-myopic children in the “first” year (e.g., 6 years) who became myopic (≤−0.50 D) in the “second” year (e.g., 7 years) (Fig. 4). Girls and boys showed a slightly different age-related change in the cumulative incidence. For girls, the incidence increased with age from age 6 at 19.4% to age 10 at 35.1% and then decreased to 19.0% at age 11. Boys, however, showed a less consistent change in incidence with age. The incidence increased from 18.4% at 6 years to 22.5% at 7 years, followed by a slight drop in incidence at 8 years (17.2%), and then an increase again to 10 years (25.8%) before the incidence declined again (to 17.6% at 11 years). For all ages, girls had a higher incidence of myopia than boys. When the data of boys and girls were pooled, the incidence of myopia peaked at 9 and 10 years of age at around 30% and then declined to 18.3% at 11 years of age. This decline was presumably due to the fact that so many children were by then already myopic.
The plot for the mean spherical equivalent refractive error (SER) in a given age band (e.g., 6 to 7) has two bars for each gender representing the “first” (e.g., 6 years) and the “second” year (e.g., 7 years) refraction data (Fig. 5A). The pattern of the age-related change in mean (SER) was similar for girls and boys. The mean SER at 6 years for girls was −0.21 ± 0.83 D and for boys was +0.19 ± 1.00 D and at 12 years was −2.21 ± 2.08 D and −1.98 ± 2.06 D for girls and boys, respectively. Girls had a higher degree of myopia than boys and this gender difference was statistically significant (p < 0.01).
The 1-year rate of refractive change in a given age band (e.g., 6 to 7) was the difference between the refraction measured in the “second” year (e.g., 7 years) to that measured in the “first” year (e.g., 6 years) (Fig. 5B). Girls showed a higher annual myopic shift than boys (p < 0.05), but the difference was not significant at ages 6 and 8 (Bonferroni t-test). Compared with age 6, the annual refractive shift for all other ages was significantly higher (p < 0.01). The highest annual change for girls occurred at 9 years (−0.71 ± 0.42 D per year) and for boys occurred 1 year later (10 years, −0.51 ± 0.32 D per year). The average annual rate of refractive change for all ages was −0.52 ± 0.42 D per year. For just the myopic children, their average annual myopia progression was −0.90 ± 0.40 D per year and the highest annual progression of −1.15 ± 0.51 D per year occurred at age seven. There were 14 children with refraction measured every single year from 6 to 12 years and their mean annual refractive change was −0.48 ± 0.45 D per year.
Bias of the Clinic Sample
Of the refraction data of 6-year-old Caucasian children collated from the 252 clinical records in the practice, 18 had myopia less than −0.25 D, giving a prevalence of myopia of 7.1%. The mean spherical equivalent was +0.55 ± 0.95 D. The definition of myopia of <−0.25 D was chosen to match that of a Canadian screening study.35 In a province-wide vision screening program using non-cycloplegic retinoscopy in New Brunswick, Canada, the prevalence of myopia (<−0.25 D) of 10,616 children aged 6 years was 6.4% and the mean refractive error was +0.62 ± 0.91 D.35 Using the prevalence of 6.4% as the population reference, the bias factor for this clinic sample was 6.4/7.1 giving a value of 0.90. The adjusted prevalence of myopia for the general population of Chinese-Canadian children was then 22.4% (6 years), 28.5% (7 years), 37.5% (8 years), 46.4% (9 years), 53.3% (10 years), 60.22% (11 years), and 64.1% (12 years).
Based on the refraction data collated from the clinical records of the parents (aged 40 to 49 years) of the Chinese-Canadian children, 60.3% of fathers (n = 665) and 58.2% of mothers (n = 777) were myopic (≤−0.50). If the bias factor of 0.90 were to apply for this clinic sample, the adjusted prevalence of myopia was 54.3% for male and 52.4% for female parents. In Japan,36 the prevalence of myopia was 70% for male and 60% for female adults in the same age range and using the same myopia criteria. Reported data from Singapore,16 gives the prevalence of myopia of 45.2% for male and 51.7% for female adults in the age range 40 to 49 years, but using a myopia cutoff of <−0.50 D. Thus the prevalence of myopia of the parents of the Chinese-Canadian children was similar to that of similarly aged adults in East Asian countries.
The mean age of the randomly selected children that completed the questionnaire was 9.8 ± 2.1 years for Chinese-Canadian children and 9.4 ± 2.4 years for Caucasian-Canadian children and their mean refractive error was −1.06 ± 1.70 D and +0.39 ± 0.79 D, respectively. The questionnaire on weekly visual activities showed that Chinese-Canadian children spent significantly more time on homework, computer, leisure reading, and portable games, but significantly less time on outdoor activities than did Caucasian-Canadian children. Chinese-Canadian children spent an average of 23.9 h per week performing near work, whereas Caucasian-Chinese children spent 17.8 h per week giving a difference of 6.1 h per week (Table 1). The data from a study in an urban city Tianjin, China2 have been included for comparison purposes. In that year of the study (1994), computer and portable games were not common and therefore data were not available. Overall, Chinese-Canadian children spent an additional 2.8 h per week on near work when compared with the children in Tianjin, China who spent only 21.1 h per week on near work. Although statistical comparison between the two groups was not possible, as only the average values were reported in the Tianjin study, the high amount of near work and lack of outdoor activities were similar findings for the two groups of children.
The main findings of this research are that many ethnic Chinese children living in Canada are myopic and that high levels of myopia are common. Chinese-Canadian children develop myopia comparable to their East Asian counterparts and their myopia progresses rapidly. The implications of these data in terms of causes of myopia are discussed.
The Chinese-Canadian sample drawn from the clinical records in the optometric practice introduces a potential bias of the data toward more children with refractive problems. To measure the bias effect, the refraction data of a control group of Caucasian children in the clinic was also analyzed and compared with that of the children in a province-wide sample in New Brunswick, Canada.35 This adjustment for clinical bias in the Chinese-Canadian sample is based on the assumption that the prevalence data in this neighborhood province is similar to that in Ontario, Canada, and that the same factors influence the Caucasian- and Chinese-origin population. The latter assumption is highly unlikely given the markedly different prevalence of myopia in the two groups. This comparison yielded a bias factor of 0.90 indicating that the bias of this clinic sample was relatively low. The estimated bias factor was used for adjusting prevalence values of all age groups as only data of 6-year-old children was reported in the province-wide sample. This may underestimate the bias effect of older age groups as it is possible that emmetropic children are less likely to return for regular eye examinations. Although there are weaknesses in the ability of current method to determine bias, the approach represents a way to deal with the clinical bias to some degree using the only population data available in Canada.
As the refractive error data in this retrospective study were derived from non-cycloplegic subjective refraction, the prevalence of myopia might have been overestimated. In previous studies comparing non-cycloplegic and cycloplegic refraction using retinoscopy or subjective refraction,37–39 the difference in the two refractive measurements was within ±0.50 D for myopic patients. For hyperopes, cycloplegic refraction revealed a significant increase in measured hyperopia, but only in younger patients (aged 6 to 10 years)39 and children with high levels of hyperopia (+4.00 to +8.00 D).37 For myopes and low-level ametropes of all types, cycloplegia failed to reveal an increase in measured hyperopia or a decrease in myopia.38,39 Therefore, the risk for inaccuracies in the myopia prevalence estimates resulting from a lack of routine cycloplegia and thus low-grade hyperopes being classified as low myopes should be very low, particularly given the ≤−0.50 D requirement for myopia categorization. Since the influence of cycloplegia was of least concern for myopes and low-grade hyperopes, and Chinese-Canadian children suspected of being hyperopic (>+1 D) were cyclopleged, the prevalence of myopia would not have been significantly different if cycloplegic refractions had been used for all children. In support of this are two recent studies in Sydney,7,40 one using non-cycloplegic retinoscopy7 with careful fogging techniques and the other using cycloplegic autorefraction40 both yielded similar myopia prevalence results for children aged 6 to 7 years. In contrast, if the method of refraction is autorefraction, the prevalence of myopia is found to be overestimated when cycloplegics are not used.41 The inaccuracies with non-cycloplegic autorefraction are due to inadequate fogging methods and therefore accommodation of the subjects. Nonetheless, the accuracy of subjective refraction is highly dependent on the technique, skill, and experience of the examiner, which are likely to vary from site to site; cycloplegic autorefraction is still the recommended refraction method for future prevalence studies.
Twenty-two percent (22.4%) of the Chinese-Canadian children were already myopic by 6 years; this was only slightly less than the 28% reported for children aged 6 to 7 years in a Hong Kong based study conducted in 19911 that used the same definition of myopia. In a province wide Canadian study with the definition of myopia <−0.25 D,35 the prevalence of myopia of 6-year-old Canadian children was merely 6.4%, although the data might also include some Chinese children. In other countries such as Australia,32 America,33 and South Africa,42 the prevalence of myopia (cutoff ≤−0.5 D) in 6-year-old children has usually been reported to be <6%. The high prevalence of myopia in Chinese-Canadian children at age six means that the age of onset of myopia for many Chinese-Canadian children is earlier than 6 years of age. Chinese-Canadian children also have a higher risk of developing high myopia presumably because of the earlier onset. From age six onwards, more children become myopic and this change coincides with the age of primary school entry in Canada at which schooling with intensive near work begins. The adjusted prevalence of myopia in the Chinese-Canadian children increased to 64.1% at age 12. This trend of increasing myopia prevalence with age is similar to that reported for Chinese children living in Hong Kong and Taiwan, but more of the Chinese-Canadian children appear to be myopic by 12 years. The reported prevalence of myopia (cutoff ≤−0.5 D) at the age of 12 years varies from 40 to 50% for the Chinese children living in Tianjin China (study conducted in 1993)2 to 55 to 58% for Chinese children living in Hong Kong (study conducted from 1991–1996).3 More recently (1998–2000), the prevalence of myopia in Hong Kong schoolchildren has been quoted to be 48.2% at the age of 10 years.19 In another study comparing prevalence of myopia in local and international schools in Hong Kong, 88.2% (2001) of Hong Kong children aged 13 years43 were myopic. These data may thus indicate a real prevalence difference between these locations or possible variations in the study design.
Another potential reason for differences in myopia prevalence between the published myopia prevalence studies may be when the studies were conducted. A question related to this, is whether the greater prevalence of myopia found here indicates that myopia is becoming even more common? To further investigate this, the adjusted prevalence data of the present study (Chinese-Canadian children, the majority are Hong Kong migrants) were plotted along with data of previous Hong Kong studies3,19,43 for ease of comparison (Fig. 6). These prevalence plots have a similar trend of greater myopia prevalence with increasing age, but results from latter studies consistently show higher prevalence values. For example, the prevalence of myopia of children at age seven through to age 12 is least for the study that was conducted the longest time ago (Hong Kong 1996, 11 to 57%).3 This increase in myopia prevalence over time is also observed in Taiwan where the myopia prevalence (cutoff ≤−0.25 D) at 12 years has increased from 29.0% in 1986 to 55.4% in 1995 to 60.7% in 2000, and such increases are also seen in children aged 6 to 15 years.4,9,10
From 6 to 12 years of age, the average refractive error of the Chinese-Canadian children becomes more myopic; this trend has been shown in many previous studies of Chinese children in East Asia.1,3,4,9,10 The refractive error distribution was significantly different from a normal distribution for all ages. The distribution was highly leptokurtic and slightly skewed toward hyperopia at age 6. The leptokurtosis of the distribution decreased at age 12 as more children became myopic. These values of kurtosis and skewness are comparable to that of a Hong Kong study.3 At age 12, the average amount of myopia was −2.97 ± 1.82 D with 16.7% have ≥4 D myopia. Since myopia tends to continue to progress into adolescence and sometimes beyond, this implies that a large percentage of Chinese-Canadian children will develop high myopia (≥6 D) by the time they reach adulthood. These high myopes are more susceptible to macular degeneration, glaucoma, cataract, and retinal detachment.44 Such a concern is also found in Singapore20; where 16.8% of children aged 9 to 11 years have myopia ≥6 D myopia. In contrast, only 0.4% children at age 12 have myopia ≥4 D in Sydney, Australia.32
The 1-year cumulative myopia incidence for the children aged 6 to 12 years in this study should not be viewed as being the same as a cumulative incidence for children who are studied longitudinally from age 6 to 12 years. The latter type of study will have the total cumulative incidence add to <100% and is referred to as 6-year cumulative incidence of myopia, an example of such a longitudinal study is Edwards (1999).3 In contrast, the total incidence for studies conducted using an initial cross-sectional survey (e.g., children aged 6 to 12 years) and then longitudinal follow-up (e.g., 1-year) could be added up to more than 100%, examples of such studies are Fan et al. (2004)19 and Zhao et al. (2002).45 The yearly cumulative incidence of myopia for the Chinese-Canadian children reached its highest level at 9 and 10 years of age for both girls (∼35%) and boys (∼25%). Girls had a slightly higher incidence of myopia than boys for all ages. Since many children have already become myopic before the age of 11 years, the incidence then decreases to 19.0% for girls and 17.6% for boys at 11 years; high incidence values cannot be sustained once the majority of children are already myopic. An earlier 5-year longitudinal study based in Hong Kong showed that the myopia incidence increased with increasing age, from 9% at 7 years to 18 to 20% at 12 years.3 A similarly high incidence of myopia to that found in this study has been reported in 10-year-old Chinese boys and 11-year-old Chinese girls living in Hong Kong, with a reported annual incidence of 20 and 27.6%, respectively.19 Collectively this shows that the incidence of myopia in these children is greatest between the ages of 9 and 10 years, suggesting that any treatment designed to prevent myopia in these groups should commence well before this.
The yearly rate of refractive change was the highest at 9 years of age for girls (−0.71 ± 0.42 D), 10 years for boys (−0.51 ± 0.42 D), and 7 years for just the myopic children (−1.15 ± 0.51 D). Girls had a significantly higher refractive shift than boys and a similar gender difference was also found in 7- to 9-years-old Singaporean children (majority Chinese),20 but there was no gender difference reported in a Hong Kong study of children aged 6 to 17 years.46 The fact that the peak progression rate for myopic children occurs at 7 years implies that children with early onset myopia should begin prevention treatment well before 7 years. The average refractive shift of all the Chinese-Canadian children aged 6 to 12 years was −0.52 ± 0.42 D per year and myopia progressed by −0.90 ± 0.40 D per year for children who were already myopic (≤−0.50 D) at the beginning of the study. Refractive shifts of −0.32 D per year3,46 to −0.63 D19 per year for children who were already myopic (≤−0.5 D) have been reported in Hong Kong. In a myopia control clinical trial in Singapore,47 a myopia progression rate of −0.56 D per year was found in children aged 6 to 12 years with <2.00 D myopia and −0.65 D per year for those with myopia ≥2.00 D. In another school based study in Singapore, the reported rate of annual refractive change was −1.03 D per year for the 7-year-olds when compared with −0.49 D per year in 12-year-olds.48 The authors attribute the high myopic shift to the predominance of ethnic Chinese children in their subject population. The data of these studies suggest that the rate of refractive change in Chinese-Canadian children is comparable to that found in Chinese children in East Asian countries.
From the questionnaire data, these ethnic Chinese children who are mostly 1st generation migrants still spend an average of 23.9 h per week on near work, which is significantly greater than that of Caucasian-Canadian children and slightly more than that of Chinese children living in an urban city, Tianjin, in China2 (Table 1). The idea that the Canadian based Chinese children may perform less near work than those children living in Asia was found not to be true. Chinese-Canadian children may even need to work harder to compensate for their language barrier and ensure they meet their parent’s scholastic expectations. Some of these parents even teach their children how to read numbers and alphabets and/or send them to private school as early as 2 years of age, although formal schooling for Canadian children starts at age four. These children also spent a large amount of time using computers and portable games, and these types of near activities could also impact on myopia progression. Outdoor activities that might be considered antimyopiagenic,6,18,28 were performed infrequently by these children; 6.1 ± 4.5 h/week compared to 10.5 ± 6.1 h/week for similar age Caucasian-Canadian children. We surmise that the longer and colder winter in Canada in some way may preclude these Chinese children whose families may be used to living in areas with milder climates, from participating in outdoor winter sports like skiing, skating, and snowboarding. Consequently, these children spend more time on indoor activities such as reading and using computers during winter months (4 to 6 months of the year). In support of this idea, a follow-up phone survey of the children who had not developed myopia by age 12 revealed that 16 out of 20 of these non-myopic children performed much less computer or reading tasks (∼15 h/week), but participated more in outdoor activities (∼11 h/week) than their myopic counterparts.
The findings of this study are different to those of the Sydney Myopia Study, where the prevalence of myopia in children of East Asian origin at both 6 years (3.6%)40 and 12 years (39.8%)49 has been reported to be higher than that of the Caucasian group, but very much lower than that reported for similar children living in urban East Asia, suggesting a predominant influence of environmental exposures rather than genetic input. This is related to the interpretation of the data from Singapore on differences between ethnic groups in the prevalence of myopia. Environmental influences appear to result in much higher prevalence of myopia in male Indian conscripts (68.8%)13 in Singapore than those young adults in India (10.8%),22 but their prevalence is not quite as high as that of the male Chinese conscripts (82.2%) in Singapore. The difference in prevalence between Chinese and Indian people living in a relatively common environment could be genetic, but given the available evidence that Chinese are more successful than Indians in education in Singapore,13 there is a perfectly plausible “environmental” explanation as well. A related study on university students in the United Kingdom reveals no significant difference in the prevalence of myopia (∼50%) between White and Asian (non-Chinese, South Asian) students educated exclusively under the United Kingdom education system from the start of their schooling.34 The similar prevalence values suggest susceptibility of South Asian students to environmental influence in the United Kingdom, although this study concerns a sample selected for educational success. In contrast, the results of this study suggest little change in both myopia and risk factors, and thus are compatible with both genetic and environmental aetiologies. Chinese-Canadian children may have brought both their genes and their families’ cultural attitudes with them. Despite the fact that Chinese-Canadian children commence schooling at an early age in Canada, their even earlier age of onset of myopia (22.4% at the age of 6 years are already myopic) also provides some evidence that Chinese children have a stronger genetic predisposition to myopia. Since the ethnic Chinese children living in Canada have experienced the same or more academic pressures compared with those living in Asian countries, the idea that a “healthier” environment in Canada can reduce myopia progression does not really hold. Perhaps, only if the Western lifestyle becomes more influential in the 3rd or 4th generation Chinese-Canadian population, will these children become less myopic than those living in East Asian countries.
Chinese children living in Canada develop myopia comparable to those living in Asian countries; migration to Canada does not lower their myopia risk.
We thank Elaine Chan, Annie Wong, Alice Leung, and Natalie Leung for their assistance in the collection of the questionnaire data. We also thank the topical editor and reviewers for their helpful comments.
204-719 Central Parkway West
Mississuaga, ON L5B 4L1, Canada
1. Lam CSY, Goh WSH. The incidence of refractive errors among school children in Hong Kong and its relationship with the optical components. Clin Exp Optom 1991;74:97–103.
2. Yap M, Wu M, Wang SH, Lee FL, Liu ZM. Environmental factors and refractive error in Chinese children. Clin Exp Optom 1994;77:8–14.
3. Edwards MH. The development of myopia in Hong Kong children between the ages of 7 and 12 years: a five-year longitudinal study. Ophthalmic Physiol Opt 1999;19:286–94.
4. Lin LL, Shih YF, Hsiao CK, Chen CJ, Lee LA, Hung PT. Epidemiologic study of the prevalence and severity of myopia among school children in Taiwan in 2000. J Formos Med Assoc 2001;100:684–91.
5. Goldschmidt E. [On the etiology of myopia. An epidemiological study]. Acta Ophthalmol (Copenh) 1968;1 (Suppl):11–172.
6. Zadnik K, Satariano WA, Mutti DO, Sholtz RI, Adams AJ. The effect of parental history of myopia on children’s eye size. JAMA 1994;271:1323–7.
7. Junghans BM, Crewther SG. Little evidence for an epidemic of myopia in Australian primary school children over the last 30 years. BMC Ophthalmol 2005;5:1.
8. Morgan I, Rose K. How genetic is school myopia? Prog Retin Eye Res 2005;24:1–38.
9. Lin LL, Chen CJ, Hung PT, Ko LS. Nation-wide survey of myopia among schoolchildren in Taiwan, 1986. Acta Ophthalmol Suppl 1988;185:29–33.
10. Lin LL, Shih YF, Tsai CB, Chen CJ, Lee LA, Hung PT, Hou PK. Epidemiologic study of ocular refraction among schoolchildren in Taiwan in 1995. Optom Vis Sci 1999;76:275–81.
11. Au Eong KG, Tay TH, Lim MK. Race, culture and Myopia in 110,236 young Singaporean males. Singapore Med J 1993;34:29–32.
12. Au Eong KG, Tay TH, Lim MK. Education and myopia in 110,236 young Singaporean males. Singapore Med J 1993;34:489–92.
13. Wu HM, Seet B, Yap EP, Saw SM, Lim TH, Chia KS. Does education explain ethnic differences in myopia prevalence? A population-based study of young adult males in Singapore. Optom Vis Sci 2001;78:234–9.
14. Goh WS, Lam CS. Changes in refractive trends and optical components of Hong Kong Chinese aged 19–39 years. Ophthalmic Physiol Opt 1994;14:378–82.
15. Lam CS, Goh WS, Tang YK, Tsui KK, Wong WC, Man TC. Changes in refractive trends and optical components of Hong Kong Chinese aged over 40 years. Ophthalmic Physiol Opt 1994;14:383–8.
16. Wong TY, Foster PJ, Hee J, Ng TP, Tielsch JM, Chew SJ, Johnson GJ, Seah SK. Prevalence and risk factors for refractive errors in adult Chinese in Singapore. Invest Ophthalmol Vis Sci 2000;41:2486–94.
17. Villarreal MG, Ohlsson J, Abrahamsson M, Sjostrom A, Sjostrand J. Myopisation: the refractive tendency in teenagers. Prevalence of myopia among young teenagers in Sweden. Acta Ophthalmol Scand 2000;78:177–81.
18. Zadnik K, Mutti DO, Friedman NE, Adams AJ. Initial cross-sectional results from the Orinda Longitudinal Study of Myopia. Optom Vis Sci 1993;70:750–8.
19. Fan DS, Lam DS, Lam RF, Lau JT, Chong KS, Cheung EY, Lai RY, Chew SJ. Prevalence, incidence, and progression of myopia of school children in Hong Kong. Invest Ophthalmol Vis Sci 2004;45:1071–5.
20. Saw SM, Tong L, Chua WH, Chia KS, Koh D, Tan DT, Katz J. Incidence and progression of myopia in Singaporean school children. Invest Ophthalmol Vis Sci 2005;46:51–7.
21. Saw SM. A synopsis of the prevalence rates and environmental risk factors for myopia. Clin Exp Optom 2003;86:289–94.
22. Murthy GV, Gupta SK, Ellwein LB, Munoz SR, Pokharel GP, Sanga L, Bachani D. Refractive error in children in an urban population in New Delhi. Invest Ophthalmol Vis Sci 2002;43:623–31.
23. Zhan MZ, Saw SM, Hong RZ, Fu ZF, Yang H, Shui YB, Yap MK, Chew SJ. Refractive errors in Singapore and Xiamen, China–a comparative study in school children aged 6 to 7 years. Optom Vis Sci 2000;77:302–8.
24. Chung KM, Mohidin N, Yeow PT, Tan LL, O’Leary D. Prevalence of visual disorders in Chinese schoolchildren. Optom Vis Sci 1996;73:695–700.
25. Dandona R, Dandona L, Naduvilath TJ, Srinivas M, McCarty CA, Rao GN. Refractive errors in an urban population in Southern India: the Andhra Pradesh Eye Disease Study. Invest Ophthalmol Vis Sci 1999;40:2810–18.
26. Dandona R, Dandona L, Srinivas M, Sahare P, Narsaiah S, Munoz SR, Pokharel GP, Ellwein LB. Refractive error in children in a rural population in India. Invest Ophthalmol Vis Sci 2002;43:615–22.
27. Watanabe S, Yamashita T, Ohba N. A longitudinal study of cycloplegic refraction in a cohort of 350 Japanese schoolchildren. Cycloplegic refraction. Ophthalmic Physiol Opt 1999;19:22–9.
28. Parssinen O, Lyyra AL. Myopia and myopic progression among schoolchildren: a three-year follow-up study. Invest Ophthalmol Vis Sci 1993;34:2794–802.
29. Saw SM, Hong CY, Chia KS, Stone RA, Tan D. Nearwork and myopia in young children. Lancet 2001;357:390.
30. Adams DW, McBrien NA. Prevalence of myopia and myopic progression in a population of clinical microscopists. Optom Vis Sci 1992;69:467–73.
31. Zylbermann R, Landau D, Berson D. The influence of study habits on myopia in Jewish teenagers. J Pediatr Ophthalmol Strabismus 1993;30:319–22.
32. Junghans BM, Crewther SG. Prevalence of myopia among primary school children in eastern Sydney. Clin Exp Optom 2003;86:339–45.
33. Zadnik K. The Glenn A. Fry Award Lecture (1995). Myopia development in childhood. Optom Vis Sci 1997;74:603–8.
34. Logan NS, Davies LN, Mallen EA, Gilmartin B. Ametropia and ocular biometry in a U.K. university student population. Optom Vis Sci 2005;82:261–6.
35. Robinson BE. Factors associated with the prevalence of myopia in 6-year-olds. Optom Vis Sci 1999;76:266–71.
36. Shimizu N, Nomura H, Ando F, Niino N, Miyake Y, Shimokata H. Refractive errors and factors associated with myopia in an adult Japanese population. Jpn J Ophthalmol 2003;47:6–12.
37. Bannon RE. The use of cycloplegics in refraction. Am J Optom Arch Am Acad Optom 1947;24:513–68.
38. Rengstorff RH. Observed effects of cycloplegics in refractive findings. J Am Optom Assoc 1966;37:360.
39. Hiatt RL, Braswell R, Smith L, Patty JW. Refraction using mydriatic, cycloplegic, and manifest techniques. Am J Ophthalmol 1973;76:739–44.
40. Rose KA, Mai TQ, Ojiami E, Huynh S, Robaei D, Rochtchina E, Smith W, Morgan I, Mitchell P; for Sydney Myopia Study. Distribution of myopia in Australian school children of Caucasian and East Asian Origin: the Sydney Myopia Study (abstract). Invest Ophthalmol Vis Sci 2005;46:E-abstract 4621.
41. Zhao J, Mao J, Luo R, Li F, Pokharel GP, Ellwein LB. Accuracy of noncycloplegic autorefraction in school-age children in China. Optom Vis Sci 2004;81:49–55.
42. Naidoo KS, Raghunandan A, Mashige KP, Govender P, Holden BA, Pokharel GP, Ellwein LB. Refractive error and visual impairment in African children in South Africa. Invest Ophthalmol Vis Sci 2003;44:3764–70.
43. Lam CS, Goldschmidt E, Edwards MH. Prevalence of myopia in local and international schools in Hong Kong. Optom Vis Sci 2004;81:317–22.
44. Saw SM, Katz J, Schein OD, Chew SJ, Chan TK. Epidemiology of myopia. Epidemiol Rev 1996;18:175–87.
45. Zhao J, Mao J, Luo R, Li F, Munoz SR, Ellwein LB. The progression of refractive error in school-age children: Shunyi district, China. Am J Ophthalmol 2002;134:735–43.
46. Lam CS, Edwards M, Millodot M, Goh WS. A 2-year longitudinal study of myopia progression and optical component changes among Hong Kong school children. Optom Vis Sci 1999;76:370–80.
47. Saw SM, Nieto FJ, Katz J, Schein OD, Levy B, Chew SJ. Factors related to the progression of myopia in Singaporean children. Optom Vis Sci 2000;77:549–54.
48. Tan NW, Saw SM, Lam DS, Cheng HM, Rajan U, Chew SJ. Temporal variations in myopia progression in Singaporean children within an academic year. Optom Vis Sci 2000;77:465–72.
49. Rose KA, Ip J, Robaei D, Huynh SC, Kifley A, Smith W, Morgan IG, Mitchell P; for The Sydney Myopia Study and Sidney Childhood Eye Study. Near-work and outdoor activities and the prevalence of myopia in Australiian school students aged 12–13 years: the Sydney Myopia Study (abstract). Invest Ophthalmol Vis Sci 2006;47:E-abstract 5453.
This article has been cited 6 time(s).
Clinical and Experimental OphthalmologyPrevalence of refractive errors in school-age children in MoroccoClinical and Experimental Ophthalmology
Investigative Ophthalmology & Visual ScienceIncidence of Myopia in High School Students with and without Red-Green Color Vision DeficiencyInvestigative Ophthalmology & Visual Science
Ophthalmic and Physiological OpticsThe effect of positive-lens addition and base-in prism on accommodation accuracy and near horizontal phoria in Chinese myopic childrenOphthalmic and Physiological Optics
Non-Obvious Relationships between Anthropometry and Cognitive Ability
Mankind Quarterly, 49(2):
Journal of Pediatric Ophthalmology & Strabismus
Evaluation of Refraction in a Statistically Significant Sample: Changes According to Age and Strabismus
Journal of Pediatric Ophthalmology & Strabismus, 46(5):
Archives of Ophthalmology
Randomized Trial of Effect of Bifocal and Prismatic Bifocal Spectacles on Myopic Progression Two-Year Results
Archives of Ophthalmology, 128(1):
children; Chinese; myopia; prevalence
© 2007 American Academy of Optometry
Highlight selected keywords in the article text.
- Articles in PubMed by DESMOND CHENG, DIPOPT, OD, MSc, FAAO
- This article in PubMed
- Articles in Google Scholar by DESMOND CHENG, DIPOPT, OD, MSc, FAAO
- Other articles in this journal by DESMOND CHENG, DIPOPT, OD, MSc, FAAO