This cohort study investigates the association of parental myopia with refractive error and ocular biometry in multiethnic children aged 6 to 72 months.
Key Points
Question
How is parental myopia associated with early-onset myopia in preschool children of different race/ethnicity?
Findings
In this cohort study of cycloplegic refraction data from 9793 children aged 6 to 72 months, maternal and paternal myopia were associated with a higher risk of myopia in children before school age. Even among children without myopia, parental myopia was associated with a greater ratio of axial length to corneal curvature radius and more myopic refractive error.
Meaning
Parental myopia, especially childhood-onset myopia, was associated with a greater risk of early-onset myopia in children regardless of race/ethnicity and as early as 1 year of age.
Abstract
Importance
Parental myopia is an important risk factor for preschool myopia in Asian children. Further investigation of the association between parental myopia and early-onset myopia risk in other racial/ethnic groups, such as African American and Hispanic white children, could improve understanding of the etiology and treatment of this condition.
Objective
To investigate the association of parental myopia with refractive error and ocular biometry in multiethnic children aged 6 to 72 months.
Design, Setting, and Participants
This cohort study pooled data from children in 3 population-based studies with comparable design from the US, Singapore, and Australia. Parental myopia was defined as the use of glasses or contact lenses for distance viewing by the child’s biological parent(s). Multivariable regressions were conducted to assess the association of parental myopia. Data were collected from 2003 to 2011 and analyzed from 2017 to 2019.
Main Outcomes and Measures
Cycloplegic refraction and prevalence of myopia (spherical equivalent refractive error of≤−0.5 diopters [D]) in the more myopic eye.
Results
The analysis cohort included 9793 children, including 4003 Asian, 2201 African American, 1998 Hispanic white, and 1591 non-Hispanic white participants (5106 boys [52.1%]; mean [SD] age, 40.0 [18.9] months). Compared with children without parental myopia, the odds ratios for early-onset myopia were 1.42 (95% CI, 1.20-1.68) for children with 1 parent with myopia, 2.70 (95% CI, 2.19-3.33) for children with 2 parents with myopia, and 3.39 (95% CI, 1.99-5.78) for children with 2 parents with childhood-onset myopia. Even among children without myopia, parental myopia was associated with a greater ratio of axial length to corneal curvature radius (regression coefficient for myopia in both parents, 0.023; P < .001) and more myopic refractive error (regression coefficient for myopia in both parents, −0.20 D; P < .001). Effects of parental myopia were observed in all 4 racial/ethnic groups and across age groups except those younger than 1 year. However, parental myopia was not associated with the age-related trends of refractive error (regression coefficient for children without parental myopeia, 0.08; for children with 2 parents with myopia, 0.04; P = .31 for interaction) and ratio of axial length to corneal curvature radius (regression coefficient for children without parental myopeia, 0.031; for children with 2 parents with myopia, 0.032; P = .89 for interaction) beyond infancy.
Conclusions and Relevance
Parental myopia, especially childhood-onset parental myopia, was associated with a greater risk of early-onset myopia in Asian, Hispanic, non-Hispanic white, and African American children. The observed associations of parental myopia in children as early as 1 year of age and in children without myopia suggests that genetic susceptibility may play a more important role in early-onset myopia and that parental myopia may contribute to myopia in children by setting up a more myopic baseline before school age.
Introduction
Early-onset myopia is an important factor associated with development of high myopia. A study in Singapore1 found that children with myopia at 3 to 6 years of age subsequently had a mean spherical equivalent (SE) refractive error of −5.5 diopters (D) at 11 years of age. Similar results were found in a Danish study.2 To reduce the future risk of high myopia,3 it is important to identify children at high risk for early-onset myopia so that they can receive close monitoring and early treatment to retard myopia progression.
Although major risk factors for school-age myopia have been identified, such as family history, race/ethnicity, and reduced time outdoors,4 fewer risk factors have been identified for myopia that develops before school age. Two studies of Singaporean children5,6 found that parental myopia was the strongest risk factor for myopia before school age, whereas levels of near work and outdoor activity were not associated with myopia in these children. The combined analysis of the Multi-Ethnic Pediatric Eye Disease Study (MEPEDS) and the Baltimore Pediatric Eye Disease Study (BPEDS)7 identified African American race or Hispanic ethnicity (compared with non-Hispanic white race), younger age (<36 months), and presence of astigmatism as the independent risk factors for myopia in preschool children. However, parental myopia was not evaluated in that analysis because these data were not available in the BPEDS. In school-age children, the association of parental myopia has been shown to vary by ethnicity.8 How parental myopia may influence myopia risk before school age in non-Asian children remains unclear. Furthermore, it is unknown whether the association between parental myopia and a child’s myopia is mediated through shared genetic susceptibility or environmental or behavioral factors (eg, time spent outdoors, time spent reading).8 Studies of preverbal children, before environmental or behavioral factors exert significant influences on the development of myopia, may provide better evidence for distinguishing the contribution of genetic susceptibility.9
In this report, we investigated the association of parental myopia and parental age of myopia onset with refractive error and related ocular biometry (ie, axial length, corneal curvature radius [CCR], and ratio of axial length to CCR) in multiethnic children aged 6 to 72 months using pooled individual participant data from 3 population-based studies.10,11,12 Studies have shown that the ratio of axial length to CCR may be a better marker of myopia progression compared with axial length.13 The large sample size (n = 9793) of this data set allowed us to address several important questions: Does the association of parental myopia depend on the age of onset of parental myopia? At what age may the association of parental myopia first present? Is parental myopia associated with age-related trends in refraction and ocular biometry within the preschool age range? Does parental myopia have similar associations among non-Asian children as among Asian children?
Methods
Study Cohort
Details of the Population-Based Pediatric Eye Disease Study (POPEYE) Consortium have been published previously.14 Briefly, the consortium is composed of 4 population-based studies of preschool-aged children: MEPEDS in Los Angeles, California; BPEDS in Baltimore, Maryland15; the Strabismus, Amblyopia, and Refractive Error in Singaporean Children (STARS) study in Singapore11; and the Sydney Pediatric Eye Disease Study (SPEDS) in Sydney, Australia.12 The BPEDS did not collect data on parental myopia and therefore was not included in the present analysis. Data were collected for this cohort study from 2003 to 2011. Written informed consent was obtained from a parent or guardian of participating children in the original studies. Only deidentified individual participant data were pooled. The protocols of this investigation adhered to the tenets of the Declaration of Helsinki16 and were reviewed and approved by the institutional review board at the University of Southern California, Los Angeles. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Ocular Examination and Interview
Details of the original studies have been reported previously.7,10,11,12,15 Briefly, comprehensive eye examinations using standardized protocols were performed by optometrists or ophthalmologists who were trained and certified. Refractive error was measured by cycloplegic autorefraction or streak retinoscopy 30 to 60 minutes after the administration of 2 or 3 drops of cyclopentolate hydrochloride, 1% (0.5% for children aged ≤12 months). No clinically significant differences in measurements were found between different cycloplegic autorefraction methods and cycloplegic retinoscopy.17 For the current analysis, only participants who received at least 2 drops of cyclopentolate and completed a cycloplegic refraction were included. Axial length was measured in children 2 years and older using a noncontact partial coherence interferometer (IOLMaster; Carl Zeiss Meditec), and keratometry was conducted using a handheld refractometer (Retinomax; Righton Manufacturing).
Parent(s) of the participants (90.4% mothers, 6.3% fathers, and 3.3% guardians) were interviewed to collect data on demographic characteristics, family history of eye diseases, and participant’s ocular and medical history. In particular, parents were asked whether the child’s birth parents wore glasses or contact lenses and, if so, to report the age at first use and whether the refractive correction was worn for distance viewing, close work, or both. Data on the severity of parental myopia were not collected. Parents who had undergone laser, photorefractive keratectomy, or laser in situ keratomileusis surgical procedures were asked to answer for the time before the surgery in MEPEDS. Similar instruction was not specified in the other 2 studies. In MEPEDS, data on parental history of use of glasses or contact lenses were collected formally after March 1, 2005. Compared with the children without parental myopia data, the children with parental myopia data were similar in sex, refractive error, and the educational attainment of the primary caregiver, but were younger, more likely to be participants of STARS and MEPEDS than SPEDS, and less likely to be non-Hispanic white than other racial/ethnic groups.
Statistical Analysis
Data were analyzed from 2017 to 2019. The analysis cohort consisted of children in whom a reliable cycloplegic refraction was obtained and information on parental myopia was available. Children with a reported diagnosis of Down syndrome or cerebral palsy were excluded. Nine children with retinopathy of prematurity, cataract, or glaucoma were included and did not change our results materially. As previously reported,14 race/ethnicity was identified by parental report as Hispanic white, non-Hispanic white, African American, and Asian. Children from other or mixed racial/ethnic backgrounds were not included because of small sample size.
We calculated SE refractive error as the sphere power plus one-half of the cylinder power. Myopia was defined as SE refractive error of −0.5 D or greater in the eye with higher myopic SE refractive error (the worse eye). If both eyes had the same SE refractive error, data from the right eye were chosen. If only 1 eye had refractive error data, that eye was denoted as the worse eye. Parental myopia was defined as use of glasses or contact lenses for distant viewing by the child’s biological parent(s).
Multivariable logistic regression and linear regression were performed to assess whether parental myopia was associated with the risk of myopia and differences in SE refractive error and ocular biometry in their child. Sociodemographic, behavioral, and clinical risk factors that have been previously suspected or associated with refractive error in young children7,18,19,20,21 and collected in the participating studies were included as covariates; these included study (MEPEDS, STARS, and SPEDS), age group (6-11, 12-23, 24-35, 36-47, 48-59, and 60-72 months), sex, racial/ethnic group, low income (yes or no), educational attainment (less than high school graduate, high school graduate, college/university graduate, or more) of the primary caregiver, maternal smoking during pregnancy (yes or no), maternal age at childbirth (<30, 30-34, and ≥35 years), gestational age (≤36, 37-41, and ≥42 weeks), birth weight (<2.5, 2.5-4.2, and >4.2 kg), and breastfeeding (yes or no). Data on near work and outdoor activity were not available for approximately 50% of the cohort and therefore were not included. For analyses of age at onset of parental myopia, we presented results from the full data set with missing indicator methods, which were similar to those from complete case analyses and analyses with imputed data. Interactions of parental myopia with other factors were tested by including proper cross-product terms in the regression models. To evaluate possible clustering effects from including siblings from the same household, mixed-effect models with household as the random factor were also performed. Quantile regression was used to explore the association of parental myopia with different levels of refractive error using cutoff points determined by percentile ranks. All statistical analyses were conducted using SAS, version 9.4 (SAS Institute, Inc). All reported P values are 2 sided, and P < .05 indicated significance.
Results
A total of 9793 children with cycloplegic refraction and parental myopia data were pooled (5106 boys [52.1%] and 4687 girls [47.9%]; mean [SD] age, 40.0 [18.9] months) (eFigure 1 in the Supplement). Of these participants, 1591 were non-Hispanic white (16.2%), 1998 were Hispanic white (20.4%), 2201 were African American (22.5%), and 4003 were Asian (40.9%) (2500 Chinese Singaporean, 1108 Asian American, and 395 Asian Australian). Table 1 presents the demographic characteristics of the analysis cohort by parental myopia. Paternal myopia was associated with maternal myopia, even after adjustment for participant’s race/ethnicity (odds ratio, 2.40; 95% CI, 2.18-2.65; P < .001).
Table 1. Demographic Characteristics of Participants With Cycloplegic Refraction by Parental Myopiaa.
| Study characteristic | No. with parental myopia data | Maternal myopia | Paternal myopia | All parents with myopia, No. (%) | ||||
|---|---|---|---|---|---|---|---|---|
| No. (%) | P valueb | No. (%) | P valueb | 1 | 2 | P valueb | ||
| Study (location) | ||||||||
| STARS (Singapore) | 2500 | 1455 (59.0) | <.001 | 1256 (51.3) | <.001 | 989 (39.6) | 861 (34.4) | <.001 |
| SPEDS (Sydney, Australia) | 1069 | 371 (34.9) | 293 (27.9) | 416 (38.9) | 124 (11.6) | |||
| MEPEDS (Los Angeles, California) | 6224 | 1901 (30.6) | 1227 (19.9) | 1934 (31.1) | 597 (9.6) | |||
| Child’s race/ethnicity | ||||||||
| Asian | 4003 | 2156 (54.4) | <.001 | 1856 (47.3) | <.001 | 1608 (40.2) | 1202 (30.0) | <.001 |
| African American | 2201 | 631 (28.7) | 262 (12.1) | 705 (32.0) | 94 (4.3) | |||
| Hispanic white | 1998 | 368 (18.4) | 225 (11.3) | 451 (22.6) | 71 (3.6) | |||
| Non-Hispanic white | 1591 | 572 (36.1) | 433 (27.5) | 575 (36.1) | 215 (13.5) | |||
| Child’s sex | ||||||||
| Male | 5106 | 1971 (38.8) | .35 | 1420 (28.2) | .16 | 1753 (34.3) | 819 (16.0) | .94 |
| Female | 4687 | 1756 (37.7) | 1356 (29.3) | 1586 (33.8) | 763 (16.3) | |||
| Child’s age, mo | ||||||||
| 6-11 | 782 | 289 (37.1) | .99 | 206 (26.6) | .81 | 259 (33.1) | 118 (15.1) | .97 |
| 12-23 | 1693 | 670 (39.8) | 459 (27.4) | 563 (33.3) | 283 (16.7) | |||
| 24-35 | 1749 | 630 (36.3) | 514 (29.7) | 590 (33.7) | 277 (15.8) | |||
| 36-47 | 1828 | 692 (38.0) | 518 (28.9) | 644 (35.2) | 283 (15.5) | |||
| 48-59 | 1841 | 724 (39.5) | 559 (30.8) | 637 (34.6) | 323 (17.5) | |||
| 60-72 | 1900 | 722 (38.2) | 520 (27.8) | 646 (34.0) | 298 (15.7) | |||
| Child’s birth weight, kg | ||||||||
| <2.5 | 810 | 326 (40.4) | .04 | 216 (27.2) | .49 | 298 (36.8) | 122 (15.1) | .04 |
| 2.5-4.2 | 8088 | 3169 (39.4) | 2384 (29.9) | 2773 (34.3) | 1390 (17.2) | |||
| >4.2 | 330 | 86 (26.1) | 58 (17.7) | 106 (32.1) | 19 (5.8) | |||
| Unknown | 565 | 146 (26.0) | 118 (21.2) | 162 (28.7) | 51 (9.0) | |||
| Child’s gestational age, wk | ||||||||
| <37 | 770 | 315 (41.0) | .20 | 218 (28.8) | .26 | 283 (36.8) | 125 (16.2) | .02 |
| 37-41 | 8283 | 3168 (38.5) | 2380 (29.1) | 2794 (33.7) | 1377 (16.6) | |||
| ≥42 | 223 | 76 (34.1) | 47 (21.6) | 89 (39.9) | 17 (7.6) | |||
| Unknown | 517 | 168 (32.8) | 131 (25.9) | 173 (33.5) | 63 (12.2) | |||
| Maternal age at pregnancy, y | ||||||||
| <30 | 4994 | 1631 (32.8) | .002 | 1102 (22.4) | <.001 | 1589 (31.8) | 572 (11.5) | <.001 |
| 30-34 | 2591 | 1205 (46.7) | 956 (37.4) | 977 (37.7) | 592 (22.9) | |||
| ≥35 | 1633 | 746 (45.9) | 600 (37.4) | 592 (36.3) | 377 (23.1) | |||
| Unknown | 575 | 145 (25.4) | 118 (20.8) | 181 (31.5) | 41 (7.1) | |||
| Educational attainment of primary caregiver | ||||||||
| Less than high school | 2297 | 449 (19.6) | <.001 | 314 (13.8) | <.001 | 551 (24.0) | 106 (4.6) | <.001 |
| High school graduate | 5062 | 1909 (37.9) | 1356 (27.1) | 1851 (36.6) | 707 (14.0) | |||
| College graduate or more | 2295 | 1320 (58.0) | 1070 (47.4) | 884 (38.5) | 753 (32.8) | |||
| Unknown | 139 | 49 (35.8) | 36 (26.9) | 53 (38.1) | 16 (11.5) | |||
Abbreviations: BPEDS, the Baltimore Pediatric Eye Disease Study; MEPEDS, the Multiethnic Pediatric Eye Disease Study; SPEDS, Sydney Pediatric Eye Disease Study; STARS, the Strabismus, Amblyopia and Refractive Error in Singaporean Children Study.
Data were missing for 49 children for maternal myopia and 133 children for paternal myopia.
Estimated from multivariable logistic regression with all listed factors included as covariates. Age group was analyzed as an ordinal variable and all other factors were analyzed as categorical variables.
Maternal and paternal myopia were associated with a greater prevalence of myopia (Table 2). Early-onset (before 12 years of age) parental myopia was associated more strongly with myopic refractive error in children than adult-onset parental myopia (P < .001 for trend). When maternal and paternal myopia were analyzed jointly (eTable 1 in the Supplement), they were associated similarly with participant myopia, and having 2 parents with myopia was consistently associated with a higher risk of participant myopia, regardless of the age at onset of parental myopia. Compared with children without parental myopia, the odds ratios for early-onset myopia were 1.42 (95% CI, 1.20-1.68) for children with 1 parent with myopia, 2.70 (95% CI, 2.19-3.33) for children with 2 parents with myopia, and 3.39 (95% CI, 1.99-5.78) for children with 2 parents with childhood-onset myopia. Children with 2 parents with myopia in general had more myopic refractive error (regression coefficient, −0.39 [95% CI, −0.47 to −0.31]) than children of parents without myopia. This effect of parental myopia was seen across the whole spectrum of refractive error (eFigure 2 in the Supplement). Specifically, parental myopia remained associated with more myopic refractive error (regression coefficient, −0.20 D [95% CI, −0.27 to −0.13 D]) among children without myopia. Parental myopia, especially myopia in both parents, was also associated with longer axial length (regression coefficient, 0.102 [95% CI, 0.044-0.159]), smaller CCR (regression coefficient, −0.041 [95% CI, −0.060 to −0.021]), and greater ratio of axial length to CCR (regression coefficient, 0.029 [95% CI, 0.022-0.036]; P < .001 for all) (eTable 2 in the Supplement). Results were similar when children with myopia were excluded.
Table 2. Association of Parental Myopia With Cycloplegic SE Refractive Error and Prevalence of Myopia in Children Aged 6 to 72 Months.
| Characteristic | No. | SE refractive error | SE myopia (≤−0.5 D) | ||||
|---|---|---|---|---|---|---|---|
| Mean (SD), D | Regression coefficient (95% CI)a | P valuea | Prevalence, % | OR (95% CI)a | P valuea | ||
| Maternal myopiab | |||||||
| No | 6017 | 1.01 (1.28) | 1 [Reference] | NA | 8.0 | 1 [Reference] | NA |
| Yes | 3727 | 0.69 (1.29) | −0.21 (−0.27 to −0.15) | <.001 | 12.5 | 1.50 (1.29 to 1.76) | <.001 |
| Age at onset of maternal myopia, y | |||||||
| ≥21 | 498 | 0.95 (1.20) | −0.11 (−0.23 to 0.00) | .06 | 8.0 | 1.26 (0.88 to 1.79) | .20 |
| 12-20 | 1034 | 0.78 (1.21) | −0.24 (−0.33 to −0.16) | <.001 | 10.9 | 1.53 (1.21 to 1.94) | <.001 |
| ≤11 | 641 | 0.76 (1.62) | −0.26 (−0.37 to −0.15) | <.001 | 12.8 | 1.79 (1.36 to 2.34) | <.001 |
| Unknown age | 1554 | 0.53 (1.18) | −0.19 (−0.28 to −0.09) | <.001 | 14.7 | 1.43 (1.11 to 1.84) | .006 |
| P value for trendc | NA | NA | NA | <.001 | NA | NA | <.001 |
| Paternal myopiab | |||||||
| No | 6884 | 0.98 (1.26) | 1 [Reference] | NA | 8.2 | 1 [Reference] | NA |
| Yes | 2776 | 0.69 (1.33) | −0.17 (−0.23 to −0.11) | <.001 | 13.3 | 1.78 (1.51 to 2.10) | <.001 |
| Age at onset of paternal myopia, y | |||||||
| ≥21 | 353 | 1.03 (1.36) | −0.06 (−0.20 to 0.07) | .36 | 8.2 | 1.45 (0.97 to 2.18) | .07 |
| 12-20 | 431 | 0.78 (1.54) | −0.23 (−0.35 to −0.10) | <.001 | 12.1 | 2.07 (1.49 to 2.89) | <.001 |
| ≤11 | 701 | 0.75 (1.30) | −0.20 (−0.30 to −0.10) | <.001 | 13.4 | 1.95 (1.51 to 2.53) | <.001 |
| Unknown age | 1291 | 0.53 (1.24) | −0.17 (−0.27 to −0.07) | <.001 | 15.1 | 1.69 (1.32 to 2.15) | <.001 |
| P value for trendc | NA | NA | NA | <.001 | NA | NA | <.001 |
| No. of parents with myopia | |||||||
| 0 | 4872 | 1.04 (1.26) | 1 [Reference] | NA | 7.6 | 1 [Reference] | NA |
| 1 | 3339 | 0.82 (1.28) | −0.18 (−0.24 to −0.12) | <.001 | 9.7 | 1.42 (1.20 to 1.68) | <.001 |
| 2 | 1582 | 0.55 (1.32) | −0.39 (−0.47 to −0.31) | <.001 | 16.1 | 2.70 (2.19 to 3.33) | <.001 |
| P value for trend | NA | NA | NA | <.001 | NA | NA | <.001 |
Abbreviations: D, diopter; NA, not applicable; OR, odds ratio; SE, spherical equivalent.
Estimated from linear and logistic regressions adjusted for study, race/ethnicity, child’s age, sex, educational attainment of primary caregiver, maternal smoking during pregnancy, gestational age, birth weight, maternal age at pregnancy, breastfeeding, and poverty. Estimates for maternal myopia and paternal myopia were generated from the same model with both variables included as covariates.
Data were missing for 49 children for maternal myopia and 133 children for paternal myopia.
Estimated from regression of the continuous variable for age at onset of parental myopia, coded as 0 for no parental myopia, 1 for parental myopia with an age at onset of 21 years or older, 2 for parental myopia with age at onset of 12 to 20 years, and 3 for parental myopia with age at onset of 11 years or younger.
There was no between-study difference in the association of parental myopia with participant myopia (odds ratios associated with having 2 parents with myopia, 2.96 [95% CI, 2.20-4.00] in MEPEDS; 6.48 [95% CI, 2.89-14.54] in SPEDS; and 2.11 [95% CI, 1.53-2.92] in STARS; P = .10). Estimates of the effect of parental myopia were similar with and without controlling for familial clustering or parental history of strabismus and amblyopia. Results were also similar when the presence of myopia in children was defined based on the child’s better eye or the right eye (eTable 3 in the Supplement).
When stratified by race/ethnicity (Table 3), parental myopia was consistently associated with a higher risk of preschool myopia in all 4 racial/ethnic groups (P = .35 for interaction with race/ethnicity). Results were similar when limiting to children 36 to 72 months of age, in whom early emmetropization has already occurred (Figure 1). Differences by respondent type are presented in eTable 4 in the Supplement. Residual racial/ethnic differences in myopia risk remained after adjusting for parental myopia (eTable 5 in the Supplement).
Table 3. Association of Parental Myopia With Prevalence of Myopia in Children Aged 6 to 72 Months by Race/Ethnicity and Child’s Age.
| Stratification | 0 Parents with myopia | 1 Parent with myopia | 2 Parents with myopia | P value for interaction | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Total No. of children | Prevalence of myopia in children, % | OR (95% CI)a | Total No. of children | Prevalence of myopia in children, % | OR (95% CI)a | Total No. of children | Prevalence of myopia in children, % | OR (95% CI)a | ||
| Race/ethnicity | ||||||||||
| Asian | 1193 | 8.4 | 1 [Reference] | 1608 | 10.3 | 1.30 (1.00 to 1.70) | 1202 | 17.5 | 2.43 (1.86 to 3.18) | .35 |
| African American | 1402 | 11.4 | 1 [Reference] | 705 | 14.2 | 1.36 (1.04 to 1.79) | 94 | 20.2 | 2.23 (1.29 to 3.84) | |
| Hispanic white | 1476 | 6.8 | 1 [Reference] | 451 | 9.5 | 1.52 (1.04 to 2.22) | 71 | 21.1 | 4.60 (2.48 to 8.53) | |
| Non-Hispanic white | 801 | 1.0 | 1 [Reference] | 575 | 2.6 | 2.61 (1.10 to 6.21) | 215 | 5.1 | 5.55 (2.19 to 14.06) | |
| Child's age, mo | ||||||||||
| 6-11 | 405 | 13.3 | 1 [Reference] | 259 | 15.4 | 1.37 (0.87 to 2.15) | 118 | 14.4 | 1.31 (0.71 to 2.42) | .04 |
| 12-23 | 847 | 11.8 | 1 [Reference] | 563 | 12.3 | 1.08 (0.77 to 1.51) | 283 | 20.5 | 2.15 (1.46 to 3.15) | |
| 24-35 | 882 | 9.2 | 1 [Reference] | 590 | 13.7 | 1.73 (1.23 to 2.42) | 277 | 23.8 | 3.46 (2.36 to 5.08) | |
| 36-47 | 901 | 5.6 | 1 [Reference] | 644 | 8.7 | 1.75 (1.17 to 2.62) | 283 | 12.4 | 2.64 (1.65 to 4.24) | |
| 48-59 | 881 | 5.0 | 1 [Reference] | 637 | 6.9 | 1.44 (0.93 to 2.24) | 323 | 13.0 | 3.09 (1.94 to 4.90) | |
| 60-72 | 956 | 4.2 | 1 [Reference] | 646 | 5.3 | 1.29 (0.80 to 2.07) | 298 | 12.4 | 3.70 (2.28 to 6.00) | |
Abbreviation: OR, odds ratio.
Estimated from logistic regressions adjusted for study, race/ethnicity, child’s age group, sex, educational attainment of primary caregiver, maternal smoking during pregnancy, gestational age, birth weight, maternal age at pregnancy, breastfeeding, and poverty, with a product term between the number of parents with myopia and race/ethnicity or child’s age group (as a continuous variable).
Figure 1. Prevalence of Different Levels of Refractive Error by Parental Myopia and Race/Ethnicity.
Participants include children aged 36 to 72 months. Error bars indicate 95% CIs. D indicates diopter.
aP < .05 for difference in the prevalence of myopia (≤−0.5 D) and premyopia (+1.0 to −0.49 D) when comparing children with 1 or 2 parents with myopia with children without a parent with myopia. In multinomial logistic regression adjusted for study, age of the child, sex, educational attainment of primary caregiver, maternal smoking during pregnancy, gestational age, birth weight, maternal age at pregnancy, breastfeeding, and poverty.
We also assessed the age at which the effect of parental myopia may first present (Table 3). For children with 2 parents with myopia, prevalence of myopia was higher than for children without parents with myopia in all groups older than 12 months, and the associated odds ratios tended to be greater in the older age groups. Mean SE refractive error was also consistently more myopic in children with 2 parents with myopia in all groups older than 12 months (Figure 2). For age-specific biometry measurements, the association of parental myopia was more pronounced for the ratio of axial length to CCR than axial length or CCR alone. Compared with children whose parents had no myopia, children with 2 parents with myopia had a greater ratio of axial length to CCR in all age groups. There was no difference in the age-related trends of SE refractive error (regression coefficient, 0.08 vs 0.04; P = .31 for interaction), axial length (regression coefficient, 0.25 vs 0.28; P = .44 for interaction), CCR (regression coefficient, 0.008 vs 0.009; P = .89 for interaction), and the ratio of axial length to CCR beyond infancy (from 12 to 72 months of age) (regression coefficient, 0.031 vs 0.032; P = .89 for interaction) for children without parental myopia compared with children with 2 parents with myopia.
Figure 2. Age-Specific Spherical Equivalent Refractive Error and Ocular Biometry by Parental Myopia Among All Children.
Dots represents least squares means and error bars, 95% CI, estimated from multivariable linear regression adjusted for study, sex, educational attainment of the primary caregiver, maternal smoking during pregnancy, gestational age, birth weight, maternal age at pregnancy, breastfeeding, and poverty.
aP < .05 for children with 1 vs no parent with myopia.
bP < .05 for children with 2 parents vs no parent with myopia.
Discussion
Using population-based data from 9793 children, we found that parental myopia, especially in both parents, was associated with a greater risk of myopia, less hyperopic refraction, and a greater ratio of axial length to CCR before school age. Younger age at onset in parents was associated with an even greater risk. This association of parental myopia was found not only in Asian children but also in non-Hispanic white, Hispanic white, and African American children, and in children as young as 12 months. These results suggest that children who have parents with myopia are at a higher risk of having myopia much earlier than school age. Furthermore, the biometric differences seen even among children without myopia, some of whom likely will develop myopia, suggest that parental myopia is associated with ocular biometry at very early ages and before the actual onset of myopia.
Our observation of a strong association between parental myopia and myopia in children before school age is consistent with the findings of a high heritability of myopia in other studies.22,23 Genetic susceptibility probably plays a more important role for this association than shared behavioral or environmental factors, given the following findings from our study. First, parental myopia was associated with a child’s myopia risk in a dose-response manner, with stronger associations observed for early-onset than late-onset parental myopia and for having 2 parents than 1 parent with myopia. Second, the effect of parental myopia was observed in children as young as 12 months, for whom factors such as outdoor activity or reading habits most likely have not yet had a big influence. Although the present analysis did not control for factors such as levels of outdoor activities and near work, some studies5,6 have reported that near work and outdoor activities were not associated with myopia before school age.
The association of parental myopia with a child’s risk for myopia before school age did not differ substantially across racial/ethnic groups. Some evidence suggests that potential racial/ethnic differences exist in the relative contribution of genetic and environmental causes to myopia. For example, the rapid historical increase in myopia prevalence and greater urban-rural differences in myopia prevalence observed in Asian individuals24 suggest that myopia in Asian people is unlikely dominated by genetic causes. If parental myopia in Asian individuals resulted more from environmental causes (eg, reduced outdoor play), it might be less likely to be associated with early-onset myopia in their children. However, no significant racial/ethnic differences in the association with parental myopia was observed in this study. Nonetheless, early screening and intervention may be particularly important for African American and Hispanic white children, who are not traditionally thought to have a high risk for myopia.24 Our data show that 16% of Hispanic white children and 24% of African American children with 2 parents with myopia already had myopia at 36 to 72 months of age.
Our data suggest that parental myopia may contribute to a child’s myopia by setting up a more myopic baseline at preschool age. The influence of parental myopia was seen across the whole spectrum of refractive error and from very young ages (as early as 12 months). We also considered whether children of parents with myopia may undergo a greater refractive or biometric shift toward myopia during the preschool years than children of parents without myopia. Although research of myopia progression in school-age children1 suggests that age of myopia onset can be a proxy for duration of myopic progression, whether this assumption can be extrapolated to preschool ages is unknown. Our analysis of trends in SE refraction and ratio of axial length to CCR across age groups suggest that parental myopia is not associated with a more myopic refractive shift or a steeper slope of axial elongation as a function of preschool age. Therefore, we believe that parental myopia may contribute to a child’s myopia in at least 2 different ways: first by setting up a more myopic baseline from very young ages, as suggested by the present analysis, and later by increasing susceptibility to myopic shift during school age, as seen in studies of school-age myopia progression.25,26,27,28,29,30
Strengths and Limitations
This study has unique strengths, including a large sample size from a diverse group of children, allowing for comparisons by race/ethnicity and age of the child. However, this study has some limitations as well. First, our definitions of myopia as an SE refractive error of −0.5 D or worse may have led to the misclassification of some children, especially those younger than 36 months whose myopia might resolve with emmetropization. However, the influence of parental myopia was also found in older children. Also, measurement errors from autorefraction and retinoscopy may be present, and such errors are likely to be nondifferential by parental myopia. Consequently, our estimated effects of parental myopia may have been biased toward null. Second, parental myopia was ascertained based on reports by 1 parent, usually the mother, or sometimes a guardian; therefore, there might be misclassifications of parental myopia, leading to an underestimation of the influence of parental myopia. Indeed, we found stronger associations between parental myopia and participant myopia when the information regarding parental myopia was provided by the relevant parent. Misclassification of parental myopia may be reduced if parents who reported wearing glasses for distant and near vision equally are required to have started before 17 years of age to be classified as myopic.31 Our estimates of the effect of parental myopia were not materially changed using this definition. Third, the severity of parental myopia was not available for this pooled sample, and the age at onset of parental myopia was unknown for a large proportion of the participants. Fourth, because the children in our study were recruited from selected urban sites, our results may be not generalizable to children from other geographic areas, such as rural areas. Further studies, especially longitudinal studies with details on the severity of parental myopia, are needed.
Conclusions
Based on the pooled data that included Asian, Hispanic white, non-Hispanic white, and African American children, we found that parental myopia was strongly associated with a higher risk of early-onset myopia before school age, particularly when the parental myopia was itself of earlier onset or when both parents had myopia. The association with parental myopia may present at a very early age, manifesting as more prevalent myopia and as less hyperopic or /more myopic refractive error and ocular biometry, regardless of race/ethnicity.
eFigure 1. Flowchart of Participants Eligible for the Current Analysis
eFigure 2. Quantile Regression Plots Demonstrating the Estimated Effect of Parental Myopia Across the Distribution of Refractive Errors (in diopters) in Children 6 to 72 Months of Age
eTable 1. Joint Effect of Paternal Myopia and Maternal Myopia on the Prevalence of Myopia in 6- to 72-Month-Old Children
eTable 2. Ocular Biometry by Parental Myopia
eTable 3. Association of Parental Myopia With Prevalence of Myopia in the Better Eye and the Right Eye in Children 6 to 72 Months of Age
eTable 4. Association of Parental Myopia With Prevalence of Myopia in Children 6 to 72 Months of Age Stratified by Respondent Type
eTable 5. Association of Race/Ethnicity With Prevalence of Myopia in Children 6 to 72 Months of Age With and Without Adjustment for Parental Myopia
References
- 1.Chua SYL, Sabanayagam C, Cheung Y-B, et al. . Age of onset of myopia predicts risk of high myopia in later childhood in myopic Singapore children. Ophthalmic Physiol Opt. 2016;36(4):388-394. doi: 10.1111/opo.12305 [DOI] [PubMed] [Google Scholar]
- 2.Jensen H. Myopia in teenagers: an eight-year follow-up study on myopia progression and risk factors. Acta Ophthalmol Scand. 1995;73(5):389-393. doi: 10.1111/j.1600-0420.1995.tb00294.x [DOI] [PubMed] [Google Scholar]
- 3.Wu PC, Chuang MN, Choi J, et al. . Update in myopia and treatment strategy of atropine use in myopia control. Eye (Lond). 2019;33(1):3-13. doi: 10.1038/s41433-018-0139-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ramamurthy D, Lin Chua SY, Saw S-M. A review of environmental risk factors for myopia during early life, childhood and adolescence. Clin Exp Optom. 2015;98(6):497-506. doi: 10.1111/cxo.12346 [DOI] [PubMed] [Google Scholar]
- 5.Low W, Dirani M, Gazzard G, et al. . Family history, near work, outdoor activity, and myopia in Singapore Chinese preschool children. Br J Ophthalmol. 2010;94(8):1012-1016. doi: 10.1136/bjo.2009.173187 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Chua SY, Ikram MK, Tan CS, et al. ; Growing Up in Singapore Towards Healthy Outcomes Study Group . Relative contribution of risk factors for early-onset myopia in young Asian children. Invest Ophthalmol Vis Sci. 2015;56(13):8101-8107. doi: 10.1167/iovs.15-16577 [DOI] [PubMed] [Google Scholar]
- 7.Borchert MS, Varma R, Cotter SA, et al. ; Multi-Ethnic Pediatric Eye Disease Study and the Baltimore Pediatric Eye Disease Study Groups . Risk factors for hyperopia and myopia in preschool children: the Multi-Ethnic Pediatric Eye Disease and Baltimore Pediatric Eye Disease studies. Ophthalmology. 2011;118(10):1966-1973. doi: 10.1016/j.ophtha.2011.06.030 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ip JM, Huynh SC, Robaei D, et al. . Ethnic differences in the impact of parental myopia: findings from a population-based study of 12-year-old Australian children. Invest Ophthalmol Vis Sci. 2007;48(6):2520-2528. doi: 10.1167/iovs.06-0716 [DOI] [PubMed] [Google Scholar]
- 9.Chew SJ, Ritch R. Parental history and myopia: taking the long view. JAMA. 1994;272(16):1255. doi: 10.1001/jama.272.16.1255b [DOI] [PubMed] [Google Scholar]
- 10.Varma R, Deneen J, Cotter S, et al. ; Multi-Ethnic Pediatric Eye Disease Study Group . The Multi-Ethnic Pediatric Eye Disease Study: design and methods. Ophthalmic Epidemiol. 2006;13(4):253-262. doi: 10.1080/09286580600719055 [DOI] [PubMed] [Google Scholar]
- 11.Dirani M, Chan YH, Gazzard G, et al. . Prevalence of refractive error in Singaporean Chinese children: the Strabismus, Amblyopia, and Refractive Error in Young Singaporean Children (STARS) study. Invest Ophthalmol Vis Sci. 2010;51(3):1348-1355. doi: 10.1167/iovs.09-3587 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Pai AS-I, Wang JJ, Samarawickrama C, et al. . Prevalence and risk factors for visual impairment in preschool children: the Sydney Paediatric Eye Disease Study. Ophthalmology. 2011;118(8):1495-1500. doi: 10.1016/j.ophtha.2011.01.027 [DOI] [PubMed] [Google Scholar]
- 13.Scheiman M, Gwiazda J, Zhang Q, et al. ; COMET Group . Longitudinal changes in corneal curvature and its relationship to axial length in the Correction of Myopia Evaluation Trial (COMET) cohort. J Optom. 2016;9(1):13-21. doi: 10.1016/j.optom.2015.10.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Jiang X, Tarczy-Hornoch K, Stram D, et al. ; Population-Based Pediatric Eye Disease Study Consortium . Prevalence, characteristics, and risk factors of moderate or high hyperopia among multiethnic children 6 to 72 months of age: a pooled analysis of individual participant data. Ophthalmology. 2019;126(7):989-999. doi: 10.1016/j.ophtha.2019.02.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Giordano L, Friedman DS, Repka MX, et al. . Prevalence of refractive error among preschool children in an urban population: the Baltimore Pediatric Eye Disease Study. Ophthalmology. 2009;116(4):739-746.e4. doi: 10.1016/j.ophtha.2008.12.030 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.World Medical Association. Declaration of Helsinki: ethical principles for medical research involving human subjects JAMA. 2013;310(20):2191-2194. doi: 10.1001/jama.2013.281053 [DOI] [PubMed] [Google Scholar]
- 17.Prabakaran S, Dirani M, Chia A, et al. . Cycloplegic refraction in preschool children: comparisons between the hand-held autorefractor, table-mounted autorefractor and retinoscopy. Ophthalmic Physiol Opt. 2009;29(4):422-426. doi: 10.1111/j.1475-1313.2008.00616.x [DOI] [PubMed] [Google Scholar]
- 18.Sham WK, Dirani M, Chong YS, et al. . Breastfeeding and association with refractive error in young Singapore Chinese children. Eye (Lond). 2010;24(5):875-880. doi: 10.1038/eye.2009.198 [DOI] [PubMed] [Google Scholar]
- 19.Castagno VD, Fassa AG, Carret ML, Vilela MA, Meucci RD. Hyperopia: a meta-analysis of prevalence and a review of associated factors among school-aged children. BMC Ophthalmol. 2014;14:163. doi: 10.1186/1471-2415-14-163 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Shulman JP, Weng C, Wilkes J, Greene T, Hartnett ME. Association of maternal preeclampsia with infant risk of premature birth and retinopathy of prematurity. JAMA Ophthalmol. 2017;135(9):947-953. doi: 10.1001/jamaophthalmol.2017.2697 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Dai AI, Demiryürek S, Aksoy SN, Perk P, Saygili O, Güngör K. Maternal iron deficiency anemia as a risk factor for the development of retinopathy of prematurity. Pediatr Neurol. 2015;53(2):146-150. doi: 10.1016/j.pediatrneurol.2015.04.002 [DOI] [PubMed] [Google Scholar]
- 22.Dirani M, Chamberlain M, Garoufalis P, Chen C, Guymer RH, Baird PN. Refractive errors in twin studies. Twin Res Hum Genet. 2006;9(4):566-572. doi: 10.1375/twin.9.4.566 [DOI] [PubMed] [Google Scholar]
- 23.Lopes MC, Andrew T, Carbonaro F, Spector TD, Hammond CJ. Estimating heritability and shared environmental effects for refractive error in twin and family studies. Invest Ophthalmol Vis Sci. 2009;50(1):126-131. doi: 10.1167/iovs.08-2385 [DOI] [PubMed] [Google Scholar]
- 24.Rudnicka AR, Kapetanakis VV, Wathern AK, et al. . Global variations and time trends in the prevalence of childhood myopia, a systematic review and quantitative meta-analysis: implications for aetiology and early prevention. Br J Ophthalmol. 2016;100(7):882-890. doi: 10.1136/bjophthalmol-2015-307724 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Lam DSC, Fan DSP, Lam RF, et al. . The effect of parental history of myopia on children’s eye size and growth: results of a longitudinal study. Invest Ophthalmol Vis Sci. 2008;49(3):873-876. doi: 10.1167/iovs.06-1097 [DOI] [PubMed] [Google Scholar]
- 26.Loh KL, Lu Q, Tan D, Chia A. Risk factors for progressive myopia in the atropine therapy for myopia study. Am J Ophthalmol. 2015;159(5):945-949. doi: 10.1016/j.ajo.2015.01.029 [DOI] [PubMed] [Google Scholar]
- 27.Guo Y, Liu LJ, Tang P, et al. . Outdoor activity and myopia progression in 4-year follow-up of Chinese primary school children: the Beijing Children Eye Study. PLoS One. 2017;12(4):e0175921. doi: 10.1371/journal.pone.0175921 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Pärssinen O, Kauppinen M, Viljanen A. The progression of myopia from its onset at age 8-12 to adulthood and the influence of heredity and external factors on myopic progression: a 23-year follow-up study. Acta Ophthalmol. 2014;92(8):730-739. doi: 10.1111/aos.12387 [DOI] [PubMed] [Google Scholar]
- 29.Saw SM, Nieto FJ, Katz J, Schein OD, Levy B, Chew SJ. Familial clustering and myopia progression in Singapore school children. Ophthalmic Epidemiol. 2001;8(4):227-236. doi: 10.1076/opep.8.4.227.1609 [DOI] [PubMed] [Google Scholar]
- 30.Wu LJ, Wang YX, You QS, et al. . Risk factors of myopic shift among primary school children in Beijing, China: a prospective study. Int J Med Sci. 2015;12(8):633-638. doi: 10.7150/ijms.12133 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Walline JJ, Zadnik K, Mutti DO. Validity of surveys reporting myopia, astigmatism, and presbyopia. Optom Vis Sci. 1996;73(6):376-381. doi: 10.1097/00006324-199606000-00004 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eFigure 1. Flowchart of Participants Eligible for the Current Analysis
eFigure 2. Quantile Regression Plots Demonstrating the Estimated Effect of Parental Myopia Across the Distribution of Refractive Errors (in diopters) in Children 6 to 72 Months of Age
eTable 1. Joint Effect of Paternal Myopia and Maternal Myopia on the Prevalence of Myopia in 6- to 72-Month-Old Children
eTable 2. Ocular Biometry by Parental Myopia
eTable 3. Association of Parental Myopia With Prevalence of Myopia in the Better Eye and the Right Eye in Children 6 to 72 Months of Age
eTable 4. Association of Parental Myopia With Prevalence of Myopia in Children 6 to 72 Months of Age Stratified by Respondent Type
eTable 5. Association of Race/Ethnicity With Prevalence of Myopia in Children 6 to 72 Months of Age With and Without Adjustment for Parental Myopia