Epidemiologic Reviews Copyright © 1996 by The Johns Hopkins University School of Hygiene and Public Health All rights reserved

Vol. 18, No. 2 Printed in U.S.A.

Epidemiology of Myopia

Seang-Mei Saw,1 Joanne Katz,2 Oliver D. Schein,3 Sek-Jin Chew,4 and Tat-Keong Chan 5

INTRODUCTION

holes and tears, as well as retinal detachment. Methods of correction of myopia are not without complications, including corneal infections due to contact lens wear and corneal scarring and persistent corneal haze from refractive surgery (5). The public health and economic impact of myopia, the most common eye condition in the world, is enormous. In the United States, the cost of correcting refractive errors with spectacles or contact lenses is estimated to be 2 billion dollars per year (6). The military spends large amounts of money on pilot training, but pilots may not be able to continue flying if they develop myopia. Thus, myopia is a condition with social, educational, and economic consequences. Over the past few decades, there has been an increase in the prevalence of myopia in some populations, leading to growing concern among the public and the scientific community. The Chinese and Japanese appear to have had escalations of myopia rates. There is no well established or universally accepted treatment for the prevention of myopia onset or progression. This review will summarize the descriptive epidemiology of myopia, possible risk factors for myopia, and the interventions available to prevent the onset and progression of myopia. The limitations of the existing research will be addressed, as well as suggested directions for further research.

Myopia is the state of refraction in which parallel rays of light are brought to focus in front of the retina of a resting eye (1). Myopia is measured by the spherical power in diopters of the diverging lens needed to focus light onto the retina, which can be expressed as the spherical equivalent or refraction in the least myopic meridian (2, 3). The clinical correlates of myopia include blurred distance vision, eye rubbing, and squinting. Myopia has been classified as either physiologic or pathologic. Physiologic myopia occurs due to an increase in the axial diameter of the eye over that which is attained by normal growth. Pathologic myopia is caused by an abnormal lengthening of the eyeball, and is often associated with thinning of the scleral wall (1). Another classification is based on age of onset. Congenital, or infantile, myopia occurs at birth, with a reported prevalence in the full-term newborn varying from 0.0 to 24.2 percent. This variability is due to technical difficulties in measuring refraction in newborns (4). School myopia occurs at approximately 7-17 years of age and stabilizes by the late teens or early twenties. Both school and adult-onset myopia are mainly the result of idiopathic causes, while congenital myopia is often associated with other abnormalities. Severe myopia may be associated with myopic macular degeneration, cataract, glaucoma, peripheral retinal changes (such as lattice degeneration), and retinal

DEFINITION OF MYOPIA IN EPIDEMIOLOGIC STUDIES

Different studies have adopted different definitions of myopia. The most common definitions are a refractive error greater than 0.25 diopter and a refractive error greater than 0.50 diopter. The lack of uniform criteria has led to difficulties in comparing prevalence rates in different studies. Cross-sectional and cohort studies use different criteria for defining persons as myopic. All studies should specify the definition of myopia used and the range of refractive error of the subjects in the study. The accuracy and reliability of ophthalmologic and refractive examinations is crucial in epidemiologic studies. The "gold standard" for measurement of re-

Received for publication January 4,1996, and accepted for publication July 16, 1996. Abbreviation: NHANES, National Health and Nutrition Examination Survey. 1 Department of Epidemiology, School of Hygiene and Public Health, The Johns Hopkins University, Baltimore, MD. 2 Department of International Health, School of Hygiene and Public Health, The Johns Hopkins University, Baltimore, MD. 3 Dana Center for Preventive Ophthalmology, Wilmer Eye Institute, The Johns Hopkins Hospital, Baltimore, MD. 4 Singapore Eye Research Institute, Singapore National Eye Center, Singapore. 5 Department of Ophthalmology, National University Hospital, Singapore. Reprint requests to Dr. Joanne Katz, Department of International Health, School of Hygiene and Public Health, The Johns Hopkins University, 615 North Wolfe Street, Baltimore, MD 21205.

175

176

Sawetal.

fractive error in children is cycloplegic refraction (1). Cycloplegia is the act of paralyzing the muscles of accommodation in the eye. Usually, cyclopentolate hydrochloride eye drops are instilled, which provides cycloplegia lasting for 1 hour. Cycloplegic refraction is especially important in children and infants, as they have strong accommodative responses which may lead to "pseudomyopia" (7). However, often cycloplegic refraction is not used for the diagnosis of myopia in children and young adults. Thus, myopia rates may be overestimated in the determination of refractive error in these studies. PREVALENCE AND DEMOGRAPHIC PATTERNS

There is considerable geographic variation in the reported prevalence of myopia (table 1). It is difficult to compare prevalence rates between countries based on previous studies; the definitions of myopia are not uniform, and refraction may have been performed without cycloplegia. Prevalence studies are not all population-based, with some studies being conducted on convenient select groups of individuals with limited generalizability. The prevalence of myopia varies with time and the age of the study population. From data gathered on 7,401 persons aged 12-54 years in the National Health and Nutrition Examination Survey (NHANES) in 1971 and 1972, the prevalence of myopia in the United States was estimated to be 25 percent (8). However, the exact criteria for myopia in this survey were not clearly defined. This population-based survey did not include cycloplegic refraction (thereby probably overestimating the rate and degree of myopia), and the nonparticipation rate was 30 percent. A more recent population-based survey in Beaver Dam, Wisconsin, of 4,926 adults between the ages of 43 and 84 years showed a decreasing prevalence of myopia with age, from 43.0 percent in the age group 43-54 years to 14.4 percent in subjects above the age of 75 years (9). Myopia was defined as more than 0.5 diopter; however, there was no mention of whether cycloplegia was used. TABLE 1.

In Scandinavia, most of the studies were not population-based (10). Myopia prevalence was reported to be 50.3 percent among 133 medical students in Norway (11). In Sweden, the prevalence of myopia among 2,616 Swedish conscripts aged 20 years was 8.9 percent. These studies defined myopia as more than 0.25 diopter, and no cycloplegia was used. Approximately 20.5 percent of 21,000 Icelanders refracted with cycloplegia in 1975 were myopic, defined as more than 0.5 diopter (10). In Asia, there is currently a high prevalence of myopia, especially among the Chinese and Japanese. As early as the 1930s, Rasmussen (12) estimated a prevalence of myopia of approximately 70 percent in China; however, the refraction procedures were not clearly described. A total of 4,000 schoolchildren aged 6-18 years were refracted with cycloplegia in an island-wide survey in Taiwan in 1983. There was an increasing prevalence of myopia with age, from 4 percent at age 6 years to 40 percent at age 12 years, more than 70 percent at age 15 years, and more than 75 percent at age 18 years (13). Three studies carried out in Singapore showed varying myopia rates: 24.9 percent in 10-year-old Chinese children, 63 percent in university freshmen aged 19 years, and 82 percent in medical students (14-16). However, the definitions of myopia differed. Various surveys in India have found myopia prevalences ranging from 6.9 percent to 19.7 percent (17, 18). The techniques used for refraction and the definition of myopia used were often not mentioned in the studies conducted in Asia. In agricultural countries, the prevalence of myopia has been low. There have been several populationbased studies. On the South Pacific island of Vanuatu, 788 Melanesian children aged 6-19 years were examined and refracted without cycloplegia. Only 2.9 percent were found to be myopic by 0.5 diopter or more (19). In the Solomon Islands, an ophthalmic survey conducted in 1966 found that only 0.8 percent of the study population from the islands of Bougainville and Malaita were myopic by 0.25 diopter or more. No

Summary of selected studies of myopia prevalence

oounuy Solomon Islands Vanuatu Sweden Iceland

Study (ref. no.) Verlee (20) Grosvenor(19) Str6mberg (cited by Fledelius (10)) Sveinsson (23)

Populationbased?

Yes Yes No, conscripts No, spectacle

Cycloplegic refraction?

Myopia definition

512 788 2,616

No No No

>0.25 diopter >0.5 diopter >0.25 diopter

1-69 6-19 20

21,000

Yes

>0.5 diopter

1-89

7,401 133

No No

>0.25 diopter

12-54 21-33

4,000

Yes

Sample size

Age (years)

Prevalence (%) 0.8 2.9 8.9 21

/MrtCI iftC

United States Norway

Sperduto et al. (8) Midelfart et al. (11)

Taiwan

Lin et al. (13)

Yes No, medical students Yes

6-18

25 50.3 75 at age 18 years

Epidemiol Rev Vol. 18, No. 2, 1996

Epidemiology of Myopia

cycloplegic refraction was done on the 512 subjects (20). Myopia not only shows regional variation in prevalence but also exhibits country-specific differences in secular trends as well. A possible reason for the increase in myopia rates in many countries is the increase in formal education, with more time being spent on closeup work, in the past few decades. The prevalence of myopia has increased over the past several decades in Singapore and Japan (21, 22). Similarly, the prevalence of myopia in Iceland increased from 3.6 percent in 1935 to 20.51 percent in 1975 (23). The Iceland study included the use of cycloplegic refraction and the same myopia definition of more than 0.5 diopter over the 50-year period. Sex and race also affect the distribution of myopia. The 1971 and 1972 NHANES data showed that prevalence rates were higher in females than in males and higher in whites than in blacks in the United States (8). Several other studies have found a slightly higher preponderance of myopia in females (9, 21). Certain ethnic groups, such as Asians and Jews, have a higher prevalence of myopia, whereas Africans and African Americans have a low myopia prevalence rate (8). In Hawaii, the prevalence of myopia varies among the different ethnic populations: 17 percent in Chinese, 13 percent in Koreans, 12 percent in Japanese, and 12 percent in Caucasians (24). In a Taiwanese survey, where the eyes of children in one school were refracted under cycloplegia, the prevalence of myopia among the purely aboriginal children was 13 percent, compared with 30 percent in the Chinese children (25). The prevalence of myopia changes considerably with age. Newborns are usually hyperopic. In subsequent years, the ocular axis elongates, with thinning of the lens and flattening of the cornea, which leads to emmetropia in children by age 8-10 years (22). When myopia occurs, it usually develops between the ages of 6 and 14 years. Thereafter, the prevalence of myopia remains relatively constant between the ages of 12 and 54, as reported in the US NHANES data (8). There is a decreasing prevalence of myopia with increasing age after age 40 years (9). To facilitate appropriate comparisons of the prevalence of myopia across different populations, studies should be population-based, have similar definitions of myopia, refract children under cycloplegia, and report findings by age. This will allow researchers to compare prevalence rates across geographic boundaries. Similarly, studies of secular trends in myopia rates and the sociodemographic characteristics of myopia should have the same definitions of myopia and should include refraction by cycloplegia. Epidemiol Rev Vol. 18, No. 2, 1996

177

Incidence and progression of myopia

There is a lack of adequate data on the incidence of myopia from population-based cohort studies. Over a 10-year period, the incidence of myopia among Israeli pilots was 7.4 percent in 991 pilots with 20/20 vision in each eye upon entry into the profession and 22.5 percent in 221 pilots with 20/25 vision in one eye upon entry into the profession (26). The results of this study are only generalizable to populations of pilots in Israel, who are varied ethnically (European, North African, Asian). This is also a very unusual definition of myopia; it is unclear how 20/25 vision relates to refractive error. Longitudinal studies have found that myopia stops increasing earlier in females than in males, and that mean cessation ages range from 14.44 to 15.28 years for females and 15.01 to 16.66 years for males (27). Lin et al. (28), however, showed that even after puberty, myopia continues to progress slowly, and the increase in axial length is the main component in myopia progression. Both Goss (29) and Chew et al. (30) have reported that a greater amount of myopia at the initial examination age is associated with a greater rate of progression. In a study of Finnish schoolchildren by Parssinen and Lyyra (31), myopia progressed faster in girls than in boys, in children with an earlier age of onset of myopia, and in children who had more severe myopia at initial examination. All of these studies have a potential bias in that they examined populations that self-referred for spectacle or contact lens correction of myopia (i.e., perhaps only certain types of people seek help when they first start to become myopic, while others wait longer before seeking correction of their myopia). RISK FACTORS FOR MYOPIA

Both environmental and genetic factors have been associated with the onset and progression of myopia. The use-abuse theory states that closeup work causes myopia, as seen in the higher prevalence of myopia among persons who are more highly educated and are in white collar occupations. The genetic theory, on the other hand, is based on the belief that natural individual variation in eye growth will produce myopia in certain individuals (3). The mechanisms underlying the environmental and genetic factors, and the nature of the interaction between the two factors, is not certain. Educational level, intelligence, certain personality traits, and socioeconomic status have all been associated with myopia. Premature and lowbirth-weight infants have a higher risk of developing myopia later in life (32-34). The effects of malnutrition and height on myopia are poorly substantiated

178

Saw et al.

(35-37). The strongest evidence for an environmental cause is the effect of closeup work on the onset and progression of myopia. Family history

There is a greater prevalence of myopia in children of myopic parents than in children of nonmyopic parents (38, 39). Genetic studies of myopia have mainly been twin studies, pedigree studies, and studies of familial correlation. Family studies by Sorsby et al. (40) and Keller (41) demonstrated significant parentchild correlations. However, it is difficult to separate hereditary factors from environmental factors such as similar work patterns in parents and their children (41). Initial cross-sectional results of the Orinda Longitudinal Study of volunteer schoolchildren showed that before the onset of myopia, the children of myopic parents had longer eyes, suggesting a possible hereditary predisposition to myopia. However, early environmental factors may also have led to longer eyes (42). The role of heredity is postulated to be more significant in persons with higher degrees of myopia. In a study of 258 myopic patients, the percentage of parents with myopia was 15 percent for those with myopia of less than 1.00 diopter versus 55 percent for patients with myopia of more than 7.00 diopters (43). Different modes of Mendelian inheritance, including autosomal dominant, autosomal recessive, and sex-linked, have been suggested by different authors (44, 45). In a study conducted in Hawaii of 185 families with both parents of Japanese ancestry and of 192 families with both parents of European ancestry, segregation analysis was performed (46). The results showed that there was little evidence for a Mendelian mode of genetic inheritance. Past twin studies have not defined the mode of inheritance but have provided evidence to support the heritability of myopia. Accurate classification of zygosity and the comparison of monozygotic and dizygotic twin populations of similar characteristics are important considerations in the design of twin studies (47). Similar results have been obtained from twin studies conducted in the United Kingdom, Finland, Taiwan, and Shanghai, where there were higher concordance rates of myopia in monozygotic twins than in dizygotic twins (48-51). In a study of Chinese twin pairs (52), there was a higher concordance rate of myopia (92.2 percent) for monozygotic twins with concordant close-work habits (differences of less than 1 hour per day spent studying and reading) than for monozygotic twins (79.3 percent) with discordant close-work habits. The authors concluded that there was significant additive interaction between zygosity and close-work habits.

The exact mode of inheritance and possible genetic markers for myopia have not been identified. Not all observations, such as the increase in myopia prevalence in Taiwan, Singapore, and Hong Kong, can be explained solely by genetic causes. There may be an interaction between genetic and environmental factors wherein some individuals have a genetic predisposition such that they are more susceptible to environmental influences causing myopia. More conclusive and well-designed studies of family pedigrees of individuals with high myopia that use genetic markers associated with myopia must be conducted. The markers for collagen metabolism, intelligence, and retinal neurotransmitters could provide clues to the location of possible myopia genes.

Education and intelligence

Several cross-sectional studies in Denmark, Israel, the United States, and Finland have shown a higher prevalence of myopia among individuals with higher educational levels (53-56). Other studies have shown an association between myopia and intelligence and socioeconomic status (57-60). Refractive error and intelligence have been compared in various studies, with inconsistent results. Positive associations were found in Ohio and in Auckland, New Zealand, when the California Test of Mental Maturity and the Otis Self-Administered Test, respectively, were used to evaluate intelligence (59, 60). However, no relation was found when the StanfordBinet Test was used in Ohio or when the Raven Matrix Test was used in Auckland (59, 60). Ashton (61), in Hawaii, measured the effects of both closeup work and intelligence on the onset of myopia. Although no association between myopia and closeup work was found, a relation between school achievement and myopia was noted. The results of this study may have been affected by the crude measure of closeup work (number of books and magazines read per month), refraction without the use of cycloplegia, and the cross-sectional nature of the study. Questions about the validity of intelligence testing and the omission of information on other confounding factors, such as closeup work, socioeconomic status, and educational level, limit conclusions from previous studies of intelligence and myopia (57-60). Hirsch (59) noted that intelligence test scores could be influenced by the amount of reading a child does or that a more intelligent child might read more and thus become more myopic. Educational level and intelligence are strongly related to amount of closeup work and are probably not independent risk factors but surrogates for closeup work. Epidemiol Rev Vol. 18, No. 2, 1996

Epidemiology of Myopia

Closeup work

Closeup work encompasses tasks of high accommodative demand, such as reading, writing, computer work, and close television viewing. It has been suggested that the side-to-side movement of the eyes during reading has a different effect on myopia than does close work without similar eye movement, such as sewing (31). The incidence of myopia increases at the time children start attending school, and this suggests that closeup work may be a cause of the development of myopia (62). The increase in myopia prevalence observed in Hong Kong, Taiwan, Japan, and Singapore over the past few decades suggests an environmental risk factor, since the gene pool has not changed. There has been an increase in educational attainment over the past several decades, with an accompanying increase in myopia incidence, in countries such as the United States (63). However, these observations have generally been ecologic rather than epidemiologic. An increased prevalence of myopia is observed in certain occupations, such as microscopy, sewing, and carpet weaving, that require a large amount of time spent in closeup work (64). However, it is difficult to separate cause from effect; the study of persons in select occupational groups who spend large amounts of time on close work may be part of a selection process whereby individuals with myopia may prefer these occupations. Further evidence for the close-work hypothesis is the higher prevalence of myopia among college graduates, with a higher number of new cases in the college years, compared with other adults in the same age group (65). In 1964, Sato (66) noted a higher incidence of myopia among US graduates after they studied Chinese in universities. In the native populations of the Arctic regions of Alaska and Canada there has been a notable increase in myopia in the younger generation. There is little parent-child correlation in refractive error, but a sibling-sibling correlation now exists. The prevalence of myopia was much higher in young persons compared with older individuals among Alaskan Eskimos, Canadian Inuit, members of a Labrador community, Yupik Eskimos, and American Indians in Ontario (6771). The increase in myopia incidence in Arctic regions has coincided with the establishment of compulsory schooling after World War II and with an increase in exposure to closeup work. Intermarriage with whites could also be contributing to a genetic change in the predisposition to myopia. However, a homogenous change in refractive error in different populations suggests that intermarriage is unlikely to be contributing substantially to the rising incidence of myopia and, more importantly, Caucasians do not have very high rates of myopia. Thus, closeup work has been Epidemiol Rev Vol. 18, No. 2, 1996

179

implicated as a risk factor for the onset of myopia (8). The mechanisms for myopia onset and progression may be similar, and the association between closeup work and myopia progression can provide evidence for the causation of myopia onset. Cross-sectional prevalence studies

Cross-sectional studies conducted in Newfoundland and Hong Kong have found positive associations between closeup work and the prevalence of myopia (72-74). The odds ratio for myopia in subjects who attended school in the Hong Kong study was 1.7 (95 percent confidence interval 1.0-3.0). However, refraction was measured without cycloplegia in these studies. The measures of closeup work were crude and were obtained from questions on the amount of reading and writing done. The effects of different types of closeup work, such as reading or watching television, and variations in levels over time were not assessed. Moreover, the studies did not account for variations in the amount of closeup work by age, or the distances used for various tasks. An interesting study was conducted in Israel in which orthodox schoolboys of identical ethnic background had a myopia prevalence of 81.3 percent, as compared with 27.4 percent among boys from general schools (myopia was defined as more than 0.50 diopter; cycloplegic refraction could not be performed on all subjects) (75). The authors of this study attributed this increased myopia in orthodox males to their unique study habits, and to the fact that the printed letters in the commentaries studied may be as small as 1 mm in height. In addition, there was a large difference in the amount of time spent reading and writing at school. The girls in the orthodox schools had rates of myopia comparable to those of girls in the nonorthodox schools. Again, the big difference was in the amount of closeup work, which was much less for girls than for boys in the orthodox schools. However, individual estimates of the amount of closeup work were not obtained. Cohort studies

Parssinen et al. (76) reported a faster rate of progression of myopia among children who spent a greater amount of time on closeup work. Refractive error was measured annually with cycloplegia. A questionnaire was designed to determine the amount of time spent on closeup work to the nearest half hour, with information obtained on closeup work done on both weekends and school days, as well as details on reading distance.

180

Sawetal.

Occupational studies

Rapid industrialization and modernization has led to many workers' spending more time on closeup work with video display terminals and to children's using video display terminals for computer-aided instruction and video games, as well as increased television watching (77). Studies of persons in occupations involving long hours of closeup work, such as textile workers and visual display unit workers, show that the prevalence of myopia is higher in these occupations (78, 79). These studies often compare groups of people with different educational levels and socioeconomic status; such comparisons are difficult. There is a growing belief that both genetic and environmental factors, such as closeup work, play a part in the onset of myopia. Refraction is possibly a product of both genetic and environmental factors, with the environment modifying the genetically determined development of the eye. Biologic theories for closeup work

The growing eye of a child is sensitive to visual cues that could determine axial length and whether the eye grows in the direction of myopia or hyperopia (80). There are several theories which attribute closeup work to the increase in axial length that causes myopia. One of the most widely held theories is the accommodation theory, wherein there is an increase in pressure in the posterior part of the eye during accommodation which is poorly resisted by the sclera, resulting in increased ocular length (63). Although intraocular pressure plays a role in normal eye growth during development, there has been no documented increase in intraocular pressure in myopic eyes. Nonetheless, defective accommodation may cause retinal image defocus, which is increasingly regarded as a key factor in myopia development (81). Animal research showed that monkeys whose vision was restricted to a distance of 18 inches (46 cm) by drapes became myopic, and cage-reared animals had a higher prevalence of myopia than their wild counterparts (82). This supports the association of closeup work with increased accommodation and myopia. Experiments by Raviola and Wiesel (83) showed that monkeys with unilaterally surgically closed eyelids who were reared in lighted environments developed axial myopia in the closed eye and none in the other, open eye. This could be due to visual form deprivation, as animals with sutured lids who are reared in the dark do not become myopic. Another theory (84) postulates that the printed page provides an impoverished stimulus for nonfoveal retinal neurons, which have large receptive fields. Posterior poles of chicks

show a greater role in myopia development than equatorial areas. Recent studies have shown strong evidence that objects viewed nearby may cause the eye to elongate further than it does during normal growth to maximize the sharpness of images on the retina. This growing eye thus elongates and becomes myopic (85). In a study by Hung et al. (86) in Houston, Texas, refractive errors such as myopia were induced in monkeys by lenses. There was resultant compensating eye growth that reduced the effect of refractive errors produced by the spectacle lenses. This experiment supports the hypothesis that lens wearing affects the growth of the eye and that myopia progression may be hastened by focusing on close objects when wearing minus lenses, but this has not been demonstrated in humans. Further research is needed to bridge the gap between animal models and human eye physiology. Other risk factors

Other risk factors that have been explored as possibly contributing to myopia onset and progression include prematurity, low birth weight, height, personality, and malnutrition. There is strong evidence for a link between prematurity and low birth weight and myopia, but unconvincing evidence for any association between myopia and height, personality, or malnutrition. Past studies have reported a greater prevalence of myopia later in life in premature infants as compared with full-term infants (32-34). Myopia is especially common in premature infants with retinopathy of prematurity, which is caused by excessive exposure to oxygen during the first few weeks of life (1). Eye size may be linked to body stature, with taller individuals having longer axial lengths. Several studies have shown that myopic individuals are taller than nonmyopic individuals (35, 87). However, this difference is often explained by a difference in socioeconomic status. A Finnish case-control study by Teikari (35) showed that myopic persons were significantly taller than nonmyopic persons. However, refractive status was not directly examined, and height information was obtained indirectly from a questionnaire. There have been several studies which investigated the association between personality and myopia. Early studies showed that myopic individuals may be more introverted, reflective, self-confident, dominant, and sedentary than nonmyopic individuals (57, 58), while other studies, such as a cross-sectional study by Bullimore et al. (88), did not find any association between personality and myopia. These personality attributes of myopic individuals may be associated with other risk factors such as intelligence and large amounts of time spent on closeup work. Epidemiol Rev Vol. 18, No. 2, 1996

Epidemiology of Myopia

There is no evidence that specific vitamin deficiencies are associated with myopia (57). The evidence for nutritional causes for the onset of myopia has been unconvincing, as past studies showing an association have had methodological limitations. Studies in African tribal people and Lebanese Arab infants showed that malnourished individuals had higher myopia rates (36, 37). However, only limited conclusions may be made, as the cross-sectional studies do not allow direct analysis of the temporal nature of the relation and there may be more proximal causes of myopia that are associated with nutrition that have not been examined. In addition, there is a question as to why there would be an increase in myopia in Singapore, Taiwan, Japan, etc., at a time when people's diets were improving (in terms of calories and protein content). If there is any association, the attributable risk is probably very small. NEEDS FOR FURTHER EPIDEMIOLOGIC RESEARCH

Currently, there is no conclusive evidence for any of the myopia risk factors postulated above. Most of the observed associations have come from cross-sectional studies. There are very few cohort studies that have a sufficient sample size, accurate measurement of risk factors, adjustment for possible confounding factors, and measurement of the different refractive components in myopia development, which include refraction by cycloplegia, axial length, and corneal curvature. There is a need for further well-designed epidemiologic studies to provide us with information on risk factors for myopia onset and progression. From our assessment of the available literature, we must make inferences about the relative importance of the different risk factors in order to set directions for further epidemiologic research. It appears that there is an hereditary component of myopia, as seen in the many familial correlation, twin, and pedigree studies that have been conducted. However, the exact mode of interaction between genetic and environmental factors, the relative contribution of genetic factors as opposed to environmental factors, and the nature of the genetic markers is unknown. Time trends showing increased myopia rates in many countries point to an environmental cause for myopia. The most important environmental risk factor for myopia appears to be closeup work, for which several cross-sectional and cohort studies have shown an association. Other risk factors, such as intelligence, academic achievement, socioeconomic status, and educational level, are possible surrogates for closeup work. Myopia also varies with age, sex, race, and gestational age at birth. All of the above factors are potential confounders and should be meaEpidemiol Rev Vol. 18, No. 2, 1996

181

sured and appropriately adjusted for in studies examining the association between myopia and closeup work. There is no consistent evidence for height, personality, or malnutrition as risk factors for myopia. INTERVENTIONS

Visual corrective aids, such as spectacles and contact lenses, are established methods of correcting the defective distant vision arising from myopia. However, to date, there has not been any convincing or widely accepted method of preventing the onset of myopia or retarding the progression of myopia in humans. A variety of different methods to reduce the onset and progression of myopia have been described. These methods include visual training, biofeedback training, the use of bifocal spectacles, contact lenses, the instillation of atropine eyedrops, the instillation of betablocker eyedrops, lowering of the intraocular pressure, and surgery (89). Unfortunately, most of the results published have had limited validity. Some of the early intervention trials did not have a control group for comparison. Many clinical trials did not include randomization, thus allowing for selection bias by the investigators and participants. Furthermore, the treatment groups were not comparable with regard to measured confounding factors. The sample size and length of follow-up were often insufficient. In addition, large numbers of dropouts were common, and a difference in myopia progression among subjects lost to follow-up may have led to biased conclusions. Masking of subjects is almost impossible, and it is difficult to mask the technicians who refract the subjects with regard to intervention status. The trials discussed here are limited to those that utilized controls, as shown in table 2. Bifocal spectacles

Bifocal spectacles have been postulated to slow the progression of myopia by reducing accommodative demand. Clinical trials on the effects of bifocals are often not randomized, and there is no conclusive evidence for the effect of bifocals in the slowing of myopia progression (90). In 1975, Oakley and Young (91) conducted a clinical trial which assigned bifocals to volunteers and spectacles to subjects who refused to wear bifocals. The study population of 156 Native Americans and 441 Caucasians aged 6-21 years was followed for 2-4 years, and an average of three cycloplegic refractions were performed. The results showed a significant difference in the rate of progression of myopia of -0.04 diopter in the bifocal group compared with —0.51 diopter in the control group. No

182

Saw et al.

TABLE 2.

Clinical trials of interventions to decrease the rate of progression of myopia Study (ref. no.)

Oakley and Young (91)

Bifocal lenses

Goss and Grosvenor (92)

Bifocal lenses

Grosvenor et al. (93)

Bifocal lenses

Parssinen et al. (76)

Bifocal lenses

Stone (94)

Contact lenses

Andreo (95)

Hydrophilic contact lenses Gas-permeable contact lenses

Grosvenor et al. (97)

Result

Intervention

Perrigin et al. (98)

Silicone-acrylate contact lenses

Bedrossian (99)

1 % atropine eyedrops

Kaoetal. (102)

1 % atropine ointment

Hosaka (104)

Labetalol and timolol eyedrops

Jensen (105)

0.25% timolol maleate

Limitations

Significant difference in annual rate of myopia progression of -0.12 diopter in the bifocal group compared with -0.38 diopter in the control group No significant difference in myopia progression between different groups No significant difference in myopia progression between different groups No significant difference in myopia progression between different groups Significant difference of annual myopia progression of 0.1 diopter in contact lens wearers compared with 0.36 diopter in spectacle wearers No significant difference in myopia progression in different groups Significant difference in annual myopia progression of 0.14 diopter in the contact lens group versus 0.40 diopter in the spectacle group Significant difference in annual myopia progression of 0.16 diopter in the contact lens group compared with 0.51 diopter in the spectacles group No myopia progression in 74% of atropine treated eyes versus 4% of untreated fellow eyes Significant difference in myopia progression of 0.17 diopter in the atropine group compared with 0.75 diopter in the control group Significant difference in myopia progression between eyes treated with labetalol and placebo but no difference for eyes treated with timolol and placebo No statistically significant difference in myopia progression

masking was done, and this could have led to investigator bias wherein favorable refractive measurements were made in the bifocal group. An analysis of three studies by Grosvenor et al., Roberts and Banford, and Goss showed decreased rates of progression of myopia in patients with convergent strabismus who wore bifocals, but no difference in rates in patients with no strabismus or divergent strabismus who wore bifocals (92). The Grosvenor and Goss (90) bifocal study of 112 myopic patients from three optometry practices in the central United States showed no statistically significant difference in the rate of progression of myopia of —0.44 diopter per year for wearers of single-vision spectacles and —0.37 diopter per year for wearers of bifocals. The treatment assignment was not randomized, and refractive measurements were

No randomization; investigators measuring outcome not masked No randomization; refractive outcomes from medical records Large number of dropouts

No randomization; refraction measured without cycloplegia

No randomization No randomization

No randomization; large number of dropouts

Fellow eye used as control

No randomization

No randomization; small sample size

Small sample size

elicited from past medical records. A randomized clinical trial (93) in Houston, Texas, placed subjects into three groups consisting of children wearing singlevision lenses, +1.00 diopter added bifocals, or +2.00 diopters added bifocals based on a table of random numbers. The mean increase of myopia in the 124 participants was —0.34 diopter per year for the singlevision subjects, —0.36 diopter per year for the +1.00 diopter added bifocal subjects, and —0.34 diopter per year for the +2.00 diopters added bifocal subjects. The differences in the rates were not statistically significant. There was a large number of dropouts, with only 124 of the 207 subjects remaining in the study after 3 years. In Finland, a randomized clinical trial in which children aged 8-13 years were assigned to the use of bifocal lenses, continuous use of single-vision Epidemiol Rev Vol. 18, No. 2, 1996

Epidemiology of Myopia

spectacles, or use of single-vision spectacles only for distant vision showed no significant difference between rates of progression in the three groups (76). Contact lenses

Rigid contact lenses have been used in several clinical trials, as it is postulated that these lenses retard myopia progression by causing corneal flattening. One of the first studies to assess' the possible effects of contact lenses on the rate of progression of myopia was conducted by Stone in the London Refraction Hospital, where 120 children were followed for 5 years (94). However, the subjects were not randomized into contact lens and spectacle groups, and myopia was measured with noncycloplegic refraction. The increase in myopia among the contact lens wearers was 0.10 diopter per year as compared with 0.36 diopter per year for the spectacle wearers. Andreo (95) studied a small sample of 56 patients who were wearing spectacles or hydrophilic contact lenses over a period of approximately 12 months, and the results showed no statistically significant difference in the rates of progression between the two groups. As with Stone's study, the subjects were not randomized to the two different groups. A study by Grosvenor et al. (96) used gas-permeable contact lenses in 100 myopic children and compared them with another nonrandomized age-matched group of spectacle-wearers. They found an increase in myopia of 0.14 diopter in the contact lens group compared with 0.40 diopter in the spectacle group in this nonrandomized study. Grosvenor et al. noted that upon discontinuation of contact lens wear, myopia progression increased. However, the reduction in myopia progression was not accounted for entirely by corneal flattening as measured by the keratometer. The researchers concluded that the keratometer did not provide an accurate assessment of corneal flattening from contact lens wear (97). However, another Houston study (98) fitted 100 children with silicone-acrylate contact lenses and made comparisons with 20 spectacle-wearing children matched by age and initial amount of myopia over a 3-year period. The myopia of the contact lenses wearers progressed at a statistically significantly slower rate of 0.16 diopter per year, compared with 0.51 diopter per year in the spectacle wearers. However, there was a large dropout rate, with only 56 of the original 100 children fitted with contact lenses remaining in the study at the end of 3 years, and there was no randomization of treatment assignments. Atropine eyedrops

Another putative method of myopia control is the daily instillation of a long-acting cycloplegic agent Epidemiol Rev Vol. 18, No. 2, 1996

183

such as atropine to decrease ocular accommodation. Several past clinical trials did not randomize subjects, and dropout rates were high. The findings were often equivocal and inconclusive (99-101). Bedrossian's study (99) involving 75 subjects aged 7-13 years used the other eye as a control. Bedrossian found that 112 of the 150 atropine-treated eyes had no change or a decrease in myopia, whereas in the control eyes, only four had no change or a decrease in myopia. Kao et al. (102) studied the effect of 1 percent atropine ointment on the progression of myopia in Taiwanese schoolchildren with myopia of more than —0.5 diopter. A total of 40 schoolchildren received 1 percent atropine ophthalmic ointment in both eyes every night for the duration of 1 year; 40 similarly myopic schoolchildren wearing spectacles but not receiving atropine treatment served as controls. The authors found that 51.3 percent of the treated group showed no progression of myopia, and only 10 percent showed progression of greater than 0.5 diopter. By contrast, in the control group, 12.5 percent showed no myopia progression and 62.5 percent showed progression of greater than 0.5 diopter. Intraocular pressure reduction using betablocking agents

Intraocular pressure could be an important mediator of scleral stress, causing axial elongation of the eyeball and resultant myopia (103). On the basis of this hypothesis, pharmacologic agents which lower the intraocular pressure may have an effect in retarding the progression of myopia. Hosaka (104) conducted a small study in which 20 Japanese children aged 6-14 years were treated with 0.25 percent timolol maleate (a beta-blocker) twice daily, another 50 subjects were treated with 0.5 percent or 0.25 percent labetalol eyedrops (another beta-blocker) twice daily, and other subjects were treated with placebo. With a short follow-up period of only 2-5 months, Hosaka found a statistically significant difference in the progression of myopia between the labetalol-treated eyes and the eyes treated with placebo, whereas no such difference was found in timolol maleate-treated eyes and placebotreated eyes. Jensen (105), in a preliminary report published in 1988, studied the effect of timolol maleate in the control of myopia in 9- to 12-year-old schoolchildren in Denmark. A total of 159 schoolchildren were randomly allocated to one of three groups: a control group, a group with bifocal spectacles, and a group with 0.25 percent timolol eyedrops instilled twice daily. Timolol maleate was found not to have any statistically significant effect in slowing the progression of myopia in these schoolchildren (106). Thus, it can be inferred that there has been no conclu-

184

Saw et al.

sive evidence that beta-blocking agents help to retard myopia progression. The available interventions are limited by their side effects, and there has been inconclusive evidence from present intervention studies. Atropine instillation may occasionally result in side effects such as atropine dermatitis, allergic reactions to atropine, and chronic pupillary dilation leading to cataract, and it has been reported that the myopia tends to resume at a faster rate once the eyedrops are withdrawn (107). Furthermore, the compliance rate is low, as the individual has to instill eyedrops daily over long periods of time and is unable to read without bifocals if the drops are instilled in both eyes. Beta-blocking agents need to be instilled in the eye daily, with possible side effects and a low compliance rate. The results of clinical trials using beta-blocking agents have not been conclusive. Bifocals do not cause much discomfort for wearers. However, the randomized trials of bifocals have not showed any slowing of myopia progression. There was some slowing of myopia progression with the use of contact lenses, but the trials were not randomized. Future research should be directed at interventions such as the use of rigid gas-permeable contact lenses, with the emphasis on well-designed randomized clinical trials with adequate sample sizes and accurate refractive measurements.

CONCLUSIONS

Myopia is an ocular condition with a high prevalence in many parts of the world. The relative contribution of genetic and environmental factors to the development and progression of myopia is not fully understood (108). There are several questions that remain unanswered. To what extent does closeup work contribute to the increased prevalence of myopia in Japan, Taiwan, Hong Kong, Singapore, and the United States? Is the difference in myopia prevalence in different ethnic groups due primarily to genetic factors or to environmental influences? How much of myopia is genetically determined, and how do environmental factors alter the onset and progression of myopia? Is closeup work an equal risk factor for both the onset and the progression of myopia? Is the age of onset of myopia important? Are there different risk factors for high and low myopia? Recent studies have shown that a family history of myopia and closeup work are the two strongest risk factors. The relation between closeup work and genetic factors, as well as the interaction between these two variables, should be further studied. It has been suggested that in populations with little exposure to closeup work, genetic factors play an important part in

the development of myopia, while in populations where closeup work is common, there is a high prevalence of myopia and genetic factors do not have a large influence (52). Over the past few decades, epidemiologic studies have been mainly cross-sectional in nature, with poor documentation of the temporal relation between risk factors and myopia. Confounding variables were not examined, refraction was measured without cycloplegia, and the different components of refraction, such as axial length and corneal curvature, were not measured directly. The definition of myopia has varied widely, sample sizes have been insufficient, and longitudinal follow-up has been poor. Well-designed concurrent cohort studies with accurate instruments for measuring closeup work, other risk factors, and refractive outcomes will provide us with further insights into the environmental causes of myopia. Closeup work is difficult to quantify, and much more study is needed to obtain precise estimates of amounts and types of closeup work and the environmental conditions under which closeup work is done. Future studies may examine the effects of reading English and Chinese characters, as well as the direction of eye movements, whether vertical or right-to-left. Tools for closeup work assessment, mainly questionnaires and diaries, may be administered repeatedly over different time periods in order to document seasonal variations in closeup activities that result from factors such as school examinations or vacations. In children with active accommodation reflexes, refraction with cycloplegia is essential. The availability of instruments for biometric measurements of the eye will enable us to better understand mechanisms of myopia onset and progression. Twin and familial correlation studies have supported the hypothesis of a genetic component of myopia causation. However, the exact mode of inheritance is uncertain, and marker studies have been few. Further research should be directed at linkage-analysis studies and the identification of myopia gene markers. A better understanding of the risk factors for myopia would enable better public health interventions, such as health education efforts, to advise the public about the types and circumstances under which closeup work could accelerate myopia onset and progression. Cohort studies examining the effects of changes in lighting, types of closeup work, distance from reading material, or type sizes could provide a basis for specific closeup work interventions in the future. Potential interventions for the prevention of the onset and progression of myopia should be subjected to rigorously performed randomized clinical trials. Epidemiol Rev Vol. 18, No. 2, 1996

Epidemiology of Myopia REFERENCES 1. Curtin BJ. The myopias: basic science and clinical management. Philadelphia, PA: Harper and Row, 1985. 2. Curtin B. Topics to be considered when establishing standards for clinical myopia studies. Acta Ophthalmol Suppl 1988;185:61-2. 3. Angle J, Wissmann DA. The epidemiology of myopia. Am J Epidemiol 1980;ll 1:220-8. 4. Goldschmidt E. Refraction in the newborn. Acta Ophthalmol (Copenh) 1969;47:570-8. 5. Ruben M, Khoo CY. Contact lenses: medical aspects. Singapore: PG Publishers, 1989. 6. National Advisory Eye Council (US). Vision research: a national plan, 1983-1987. US Department of Health and Human Services, Public Health Service, National Institutes of Health, 1983. (NIH publication no. 83-2469). 7. Mutti DO, Zadnik K, Egashira S, et al. The effect of cycloplegia on measurement of the ocular components. Invest Ophthalmol Vis Sci 1994;35:515-27. 8. Sperduto RD, Seigel D, Roberts J, et al. Prevalence of myopia in the United States. Arch Ophthalmol 1983; 101: 405-7. 9. Wang Q, Klein BE, Klein R, et al. Refractive status in the Beaver Dam Eye Study. Invest Ophthalmol Vis Sci 1994;35: 4344-7. 10. Fledelius H. Myopia prevalence in Scandinavia: a survey, with emphasis on factors of relevance for epidemiological studies in general. Acta Ophthalmol Suppl 1988; 185:44-50. 11. Midelfart A, Aamo B, Sjohaug KA, et al. Myopia among medical students in Norway. Acta Ophthalmol (Copenh) 1992;70:317-22. 12. Rasmussen OD. Incidence of myopia in China. Br J Ophthalmol 1936,20:350-60. 13. Lin LL, Chen CJ, Hung PT, et al. Nation-wide survey of myopia among schoolchildren in Taiwan, 1986. Acta Ophthalmol Suppl 1988;185:29-33. 14. Ling SL, Chen AJ, Rajan U, et al. Myopia in ten year old children—a case-control study. Singapore Med J 1987;28: 288-92. 15. Chow YC, Dhillon B, Chew PT, et al. Refractive errors in Singapore medical students. Singapore Med J 1990;31: 472-3. 16. Law NM, Chew SJ, Ritch R, et al. Survey of refraction in a Chinese population shows that myopia severity can be predicted from its age of onset. (Abstract). Invest Ophthalmol Vis Sci 1992;33:709. 17. Jain IS, Jain S, Mohan K. The epidemiology of high myopia—changing trends. Indian J Ophthalmol 1983;31: 723-8. 18. Mohan M, Pakrasi S, Zutshi R. Myopia in India. Acta Ophthalmol Suppl 1988;185:19-23. 19. Grosvenor T. Myopia in Melanesian school children in Vanuatu. Acta Ophthalmol Suppl 1988;185:24-8. 20. Verlee DL. Ophthalmic survey in the Solomon Islands. Am J Ophthalmol 1968;66:304-19. 21. Rajan U, Tan FT, Chan TK, et al. Increasing prevalence of myopia in Singapore school children. In: Chew SJ, Weintraub J, eds. Proceedings of the Fifth International Conference on Myopia, Toronto, Ontario, Canada, June 22-24, 1994. New York, NY: Myopia International Research Foundation, 1995:41-6. 22. Hosaka A. The growth of the eye and its components: Japanese studies. Acta Ophthalmol Suppl 1988; 185:65-8. 23. Sveinsson K. The refraction of Icelanders. Acta Ophthalmol (Copenh) 1982;60:779-87. 24. Crawford HE, Hamman GC. Racial analysis of ocular deformities in schools of Hawaii. Hawaii Med J 1949,9:90-3. 25. Lin LL, Hung PT, Ko LS, et al. Study of myopia among aboriginal school children in Taiwan. Acta Ophthalmol Suppl 1988; 185:34-6. 26. Froom P, Biger Y, Erel J, et al. The incidence of myopia in Epidemiol Rev Vol. 18, No. 2, 1996

27. 28.

29. 30.

31. 32. 33.

34. 35. 36. 37. 38.

39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52.

185

the Israel Air Force rated population: a 10-year prospective study. Aviat Space Environ Med !992;63:299-301. Goss DA, Winkler RL. Progression of myopia in youth: age of cessation. Am J Optom Physiol Opt 1983;60:651-8. Lin LK, Shih YF, Lee YC, et al. Changes of the ocular refraction and its components among the medical students—5 years' longitudinal study. (Abstract). Invest Ophthalmol Vis Sci 1995;36:S947. Goss DA. Variables related to the rate of childhood myopia progression. Optom Vis Sci 1990;67:631-6. Chew SJ, Ritch R, Leong YK, et al. The age of onset of myopia is a predictor of adult myopia severity. In: Shimizu K, ed. Current aspects of ophthalmology: proceedings of the XIII Congress of the Asia-Pacific Academy of Ophthalmology, Kyoto, Japan, May 12-17, 1991. Amsterdam, The Netherlands: Elsevier Science Publishers BV, 1992:680-5. Parssinen O, Lyyra AL. Myopia and myopic progression among schoolchildren: a three-year follow-up study. Invest Ophthalmol Vis Sci 1993;34:2794-802. Graham MV, Gray OP. Refraction of premature babies' eyes. BrMed J 1963; 1:1452-4. Quinn GE, Dobson V, Repka MX, et al. Development of myopia in infants with birth weights less than 1251 grams. The Cryotherapy for Retinopathy of Prematurity Cooperative Group. Ophthalmology 1992;99:329-40. Lue CL, Hansen RM, Reisner DS, et al. The course of myopia in children with mild retinopathy of prematurity. Vision Res 1995;35:1329-35. Teikari JM. Myopia and stature. Acta Ophthalmol (Copenh) 1987;65:673-6. McLaren DS. Nutrition and eye disease in East Africa: experience in Lake and Central Provinces, Tanganyika. J Trop Med Hyg 1960;63:101-22. Halasa AH, McLaren DS. The refractive state of malnourished children. Arch Ophthalmol 1964;71:827-31. Krause UH, Rantakallio PT, Koiranen MJ, et al. The development of myopia up to the age of twenty and a comparison of refraction in parents and children. Arctic Med Res 1993; 52:161-5. Hui J, Peck L, Howland HC. Correlations between familial refractive error and children's non-cycloplegic refractions. Vision Res 1995,35:1353-8. Sorsby A, Leary GA, Fraser GR. Family studies on ocular refraction and its components. J Med Genet 1966;3:269-73. Keller JT. A comparison of the refractive status of myopic children and their parents. Am J Optom Arch Am Acad Optom 1973;50:206-ll. Zadnik K, Satariano WA, Mutt DO, et al. The effect of parental history of myopia on children's eye size. JAMA 1994;271:1323-7. Hirsch MJ, Ditmars DL. Refraction of young myopes and their parents—a reanalysis. Am J Optom Arch Am Acad Optom 1969;46:30-2. Bartsocas CS, Kastrantas AD. X-linked form of myopia. Hum Hered 1981 ;31:199-200. Karlsson JL. Evidence of recessive inheritance of myopia. Clin Genet 1975;7:197-202. Ashton GC. Segregation analysis of ocular refraction and myopia. Hum Hered 1985,35:232-9. Schwartz JT. Twin studies in ophthalmology: hereditary and environmental determinants of eye disease. Am J Ophthalmol 1968;66:323-7. Sorsby A, Fraser GR. Statistical note on the components of ocular refraction in twins. J Med Genet 1964; 1:47-9. Teikari JM, O'Donnell JJ, Kaprio J, et al. Impact of heredity in myopia. Hum Hered 1991 ;41:151—6. Lin LL, Chen CJ. Twin study on myopia. Acta Genet Med Gemellol (Roma) 1987;36:535-4O. Hu DN. Twin study on myopia. Chin Med J (Engl) 1981;94: 51-5. Chen CJ, Cohen BH, Diamond EL. Genetic and environmental effects on the development of myopia in Chinese twin

186

Saw et al.

children. Ophthalmic Paediatr Genet 1985;6:353-9. 53. Teasdale TW, Goldschmidt E. Myopia and its relationship to education, intelligence and height: preliminary results from an on-going study of Danish draftees. Acta Ophthalmol Suppl 1988; 185:41-3. 54. Rosner M, Belkin M. Intelligence, education, and myopia in males. Arch Ophthalmol 1987,105:1508-11. 55. Angle J, Wissmann DA. Age, reading, and myopia. Am J Optom Physiol Opt 1978;55:302-8. 56. Parssinen TO. Relation between refraction, education, occupation, and age among 26- and 46-year-old Finns. Am J Optom Physiol Opt 1987;64:136-143. 57. Baldwin WR. A review of statistical studies of relations between myopia and ethnic, behavioral, and physiological characteristics. Am J Optom Physiol Opt 1981 ;58:516—27. 58. Young FA, Singer RM, Foster D. The psychological differentiation of male myopes and nonmyopes. Am J Optom Physiol Opt 1975;52:679-86. 59. Hirsch MJ. The relationship between refractive state of the eye and intelligence test scores. Am J Optom Arch Am Acad Optom 1959;36:12-21. 60. Grosvenor T. Refractive state, intelligence and refractive errors. Am J Optom Arch Am Acad Optom 1963;40:257-64. 61. Ashton GC. Nearwork, school achievement and myopia. J Biosoc Sci 1985; 17:223-33. 62. Whitmore WG. Congenital and developmental myopia. Eye 1992;6:361-5. 63. Goldschmidt E. The importance of heredity and environment in the etiology of low myopia. Acta Ophthalmol (Copenh) 1981;59:759-62. 64. Adams DW, McBrien NA. Prevalence of myopia and myopic progression in a population of clinical microscopists. Optom Vis Sci 1992;69:467-73. 65. Dunphy EB, Stoll MR, King SH. Myopia among American male graduate students. Am J Ophthalmol 1968;65:518-21. 66. Sato T. The cause and prevention of school myopia. Tokyo, Japan: Excerpta Medica, 1993:106-7. 67. Young FA, Leary GA, Baldwin WR, et al. The transmission of refractive errors within Eskimo families. Am J Optom Arch Am Acad Optom 1969:46:676-85. 68. Johnson GJ. Myopia in arctic regions: a survey. Acta Ophthalmol Suppl 1988; 185:13-18. 69. Johnson GJ, Matthews A, Perkins ES. Survey of ophthalmic conditions in a Labrador community. I. Refractive errors. Br J Ophthalmol 1979;63:440-8. 70. Alward WL, Bender TR, Demske JA, et al. High prevalence of myopia among young adult Yupik Eskimos. Can J Ophthalmol 1985;20:241-5. 71. Boniuk V. Refractive problems in native peoples (the Sioux Lookout Project). Can J Ophthalmol 1973,8:229-33. 72. Richler A, Bear JC. Refraction, nearwork and education: a population study in Newfoundland. Acta Ophthalmol (Copenh) 1980;58:468-78. 73. Bear JC, Richler A, Burke G. Nearwork and familial resemblances in ocular refraction: a population study in Newfoundland. Clin Genet 1981; 19:462-72. 74. Wong L, Coggon D, Cruddas M, et al. Education, reading, and familial tendency as risk factors for myopia in Hong Kong fishermen. J Epidemiol Community Health 1993;47: 50-3. 75. Zylbermann R, Landau D, Berson D. The influence of study habits on myopia in Jewish teenagers. J Pediatr Ophthalmol Stabismus 1993;30:319-22. 76. Parssinen O, Hemminke E, Klemetti A. Effect of spectacle use and accommodation on myopic progression: final results of a three-year randomised clinical trial among schoolchildren. Br J Ophthalmol 1989,73:547-51. 77. Tokoro T. Effect of visual display terminal (VDT) work on myopia progression. Acta Ophthalmol Suppl 1988;185: 172-4. 78. Nyman KG. Occupational near-work myopia. Acta Ophthalmol Suppl 1988; 185:167-71.

79. Simensen B, Thorud LO. Adult-onset myopia and occupation. Acta Ophthalmol (Copenh) 1994;72:469-71. 80. Wallman J. Nature and nurture of myopia. (News). Nature 1994,371:201-2. 81. Gwiazda J, Thorn F, Bauer J, et al. Myopic children show insufficient accommodative response to blur. Invest Ophthalmol Vis Sci 1993;34:690-4. 82. Young FA. The effect of restricted visual space on the primate eye. Am J Ophthalmol 1961 ;52:799-806. 83. Raviola E, Wiesel TN. Effect of dark-rearing on experimental myopia in monkeys. Invest Ophthalmol Vis Sci 1978; 17: 485-8. 84. Wallman J, Gottlieb MD, Rajaram V, et al. Local retinal regions control local eye growth and myopia. Science 1987; 237:73-7. 85. Wallman J, McFadden S. Monkey eyes grow into focus. Nature Med 1995; 1:737-9. 86. Hung LF, Crawford MLJ, Smith EL. Spectacle lenses alter eye growth and the refractive status of young monkeys. Nature Med 1995;f:761-5. 87. Johansen EV. Simple myopia in schoolboys in relation to body height and weight. Acta Ophthalmol 1949;28:355-61. 88. Bullimore M, Conway R, Nakash A. Myopia in optometry students: family history, age of onset and personality. Ophthalmic Physiol Opt 1989;9:284-8. 89. Grosvenor T. Myopia: what can we do about it clinically? Optom Vis Sci 1989;66:415-19. 90. Grosvenor T, Goss DA. The role of bifocal and contact lenses in myopia control. Acta Ophthalmol Suppl 1988; 185: 162-6. 91. Oakley KH, Young FA. Bifocal control of myopia. Am J Optom Physiol Optics 1975,52:758-64. 92. Goss DA, Grosvenor T. Rates of childhood myopia progression with bifocals as a function of near point phoria: consistency of three studies. Optom Vis Sci 1990;67:637-40. 93. Grosvenor T, Perrigin DM, Perrigin J, et al. Houston Myopia Control Study: a randomized clinical trial. Part II. Final report by the patient care team. Am J Optom Physiol Opt l987;64:482-98. 94. Stone J. The possible influence of contact lenses on myopia. Br J Physiol Opt 1976;31:89-114. 95. Andreo LK. Long-term effects of hydrophilic contact lenses on myopia. Ann Ophthalmol 1990;22:224-7. 96. Grosvenor T, Perrigin J, Perrigin DM, et al. The use of silicone-acrylate contact lenses for the control of myopia: results after two years of lens wear. Optom Vis Sci 1989;66: 41-7. 97. Grosvenor T, Perrigin DM, Perrigin J, et al. Rigid gaspermeable contact lenses for myopia control: effects of discontinuation of lens wear. Optom Vis Sci 1991 ;68: 385-9. 98. Perrigin J, Perrigin DM, Quintero S, et al. Silicone-acylate contact lenses for myopia control: 3-year results. Optom Vis Sci 1990;67:764-9. 99. Bedrossian RH. The effect of atropine on myopia. Ann Ophthalmol 1971 ;3:891-7. 100. Gimbel HV. The control of myopia with atropine. Can J Ophthalmol 1973;8:527-32. 101. Dyer JA. Role of cycloplegics in progressive myopia. Ophthalmology 1979,86:692-4. 102. Kao SC, Lu HY. Liu JH. Atropine effect on school myopia: a preliminary report. Acta Ophthalmol (Suppl) 1988;185: 132-3. 103. Pruett RC. Progressive myopia and intraocular pressure: what is the linkage? A literature review. Acta Ophthalmol (Suppl) 1988; 185:117-27. 104. Hosaka A. Myopia prevention and therapy: the role of pharmaceutical agents. Japanese studies. Acta Ophthalmol (Suppl) 1988;185:130-l. 105. Jensen H. Timolol maleate in the control of myopia: a preliminary report. Acta Ophthalmol (Suppl) 1988; 185: 128-9. Epidemiol Rev Vol. 18, No. 2, 1996

Epidemiology of Myopia 106. Jensen H. Myopia progression in young school children: a prospective study of myopia progression and the effect of a trial with bifocal lenses and beta blocker eye drops. Acta Ophthalmol (Suppl) 1991 ;200:1-79. 107. Brodstein RS, Brodstein DE, Olson RJ, et al. The treatment

Epidemiol Rev Vol. 18, No. 2, 1996

187

of myopia with atropine and bifocals: a long-term prospective study. Ophthalmology 1984;91:1373-9. 108. Mutti DO, Zadnik K, Adams AJ. Myopia: the nature versus nurture debate goes on. Invest Ophthalmol Vis Sci 1996;37: 952-7.