This population-based cross-sectional study reports the ocular determinants of refractive error in Chinese American adults in Monterey Park, California.
Key Points
Question
What are ocular determinants of refractive error in Chinese American adults?
Findings
In this population-based study in California that included 4582 participants, women had shorter axial length, shorter anterior chamber depth, and steeper corneal power compared with men, and older compared with younger individuals had shallower anterior chamber depth, greater lens thickness, and more nuclear opalescence. Overall, axial length was the strongest determinant of refractive error.
Meaning
Compared with US Latino or non-US Chinese populations, Chinese Americans appear to have longer axial length, with greater contribution to refractive error.
Abstract
Importance
Uncorrected refractive error (RE) is a leading cause of visual impairment, and variations in ocular anatomy determine RE. The unique ocular determinants of RE in Chinese American individuals have not been studied previously.
Objective
To report ocular determinants of RE in a Chinese American population 50 years and older in Monterey Park, California.
Design, Setting, and Participants
The Chinese American Eye Study, a population-based, cross-sectional study, was conducted from February 1, 2010, through October 31, 2013, in Monterey Park, with this particular data analysis performed from January 1 through December 31, 2016. This study included data from 4582 participants who underwent an eye examination to obtain axial length (AL), central corneal thickness, vitreous chamber depth (VCD), anterior chamber depth (ACD), lens thickness (LT), corneal power (CP), noncycloplegic subjective refraction, and lens nuclear opalescence (NOP) grading. Data from the right phakic eye of each participant were used. Multiple regression models (standardized regression coefficients [SRCs] and semipartial correlation coefficients squared [SPCCs2]) identified key determinants of RE.
Main Outcomes and Measures
Ocular determinants of RE.
Results
Among the 4071 participants eligible for analysis (1496 men [36.7%] and 2575 women [63.3%]; mean [SD] age, 60.5 [8.1] years), mean (SD) RE was −0.52 (2.95) diopters (D), with no sex-related difference. A hyperopic shift occurred in women from −0.62 (2.95) D at 50 to 59 years to 0.60 (1.62) D at 80 years or older and in men from −0.69 (3.00) D at 50 to 59 years to 0.40 (2.29) D at 80 years or older (P < .001 for both). Compared with men, women had shorter AL (mean [SD], 23.62 [1.34] vs 24.14 [1.27] mm; P = .006), shorter ACD (mean [SD], 3.33 [0.34] vs 3.44 [0.34] mm; P < .001), and steeper CP (mean [SD], 43.50 [1.52] vs 42.88 [1.45] D; P = .02), after adjusting for age and height. No sex differences were found in VCD, LT, and NOP after height adjustment. Compared with younger individuals, older individuals had shallower ACD, thicker LT, and more NOP compared with younger individuals (P < .001 for both), even after adjustment for height. Axial length was the strongest determinant of RE (SRC = −0.92; SPCC2 = 0.55), followed by CP (SRC = −0.43; SPCC2 = 0.15). When individual components of AL were evaluated, VCD had the greatest contributing effect (SRC = −0.99; SPCC2 = 0.52), followed by CP (SRC = −0.47; SPCC2 = 0.15) and LT (SRC = −0.29; SPCC2 = 0.06).
Conclusions and Relevance
These data suggest that Chinese American individuals have longer AL and greater contribution of AL to RE than do Latino and other Chinese populations. Future studies should explore risk factors for increased AL in Chinese Americans and potential interventions that may ultimately prevent myopia-related disease.
Introduction
Uncorrected refractive error (RE) remains the leading cause of visual impairment, accounting for 44% to 48% of cases worldwide. Refractive error results from variations in ocular anatomy, including corneal power (CP), lens nuclear opalescence (NOP) and shape, and axial length (AL), which is composed of central corneal thickness (CCT), anterior chamber depth (ACD), lens thickness (LT), and vitreous chamber depth (VCD). In addition, RE changes with age. Studying these ocular components in a population-based sample provides insight into the determinants and age- and sex-related variations of RE.
Several such studies exist for US white, black, and Latino but not Asian American populations. Studies of non-US Asian populations demonstrate unique frequencies of eye diseases. For example, persons of Chinese descent in China have a higher prevalence of myopia, myopic retinopathy, and angle-closure glaucoma than do many US populations. However, few data are available to reveal whether similar patterns affect Chinese Americans, the largest subgroup of the rapidly growing Asian American population.
The Chinese American Eye Study (CHES) was designed to determine cause-specific rates of blindness, visual impairment, and ocular disease in Chinese Americans 50 years and older. This investigation evaluates the ocular components and age- and sex-related variations of RE in CHES compared with other US ethnic groups and non-US Chinese populations. This investigation will inform future research aimed at preventing and treating conditions resulting from specific biometric characteristics in Chinese Americans.
Methods
Study Population
All self-identified Chinese Americans 50 years and older and living in 10 census tracts in Monterey Park, California, were invited to participate. By the study’s completion, 4582 of the 5782 eligible Chinese American adults (79.2%) had been interviewed and undergone a comprehensive eye examination. The institutional review boards at the University of Southern California, Los Angeles, and the University of Illinois at Chicago approved the study protocol. The study adhered to the tenets of the Declaration of Helsinki. All patients provided written informed consent.
Interview and Ocular Examination
The study design and sampling plan have been published separately. An in-home interview determined demographic characteristics, acculturation, ocular and medical histories, risk factors, and access to medical and eye care. Participants visited a local eye examination center for an assessment examination by 1 ophthalmologist (D.W.) and several technicians. The initial data collection occurred from February 1, 2010, through October 31, 2013.
Biometric and Lens Opalescence Measurements
Using an A-scan ultrasonographic device (4000B A-Scan/Pachymeter; DGH Technology, Inc), we obtained 3 measurements each for AL, ACD, CCT, and LT. Vitreous chamber depth was calculated by subtracting CCT, ACD, and LT from AL. Nuclear opalescence was graded at the slitlamp examination and categorized into grades N0, NI, NII, NIII, and NIV, with higher grades indicating increasing severity, according to the Lens Opacities Classification System II.
Refraction and CP Assessment
Visual acuity was measured for each eye at 4 m using a modified Early Treatment Diabetic Retinopathy Study distance chart and illuminator (Precision Vision). Participants with an unaided letter score of 85 or better (approximate Snellen equivalent, 20/20 or better) were considered to be emmetropic with an RE of zero. For participants with a score of less than 85 in either eye (the letter score was calculated as the number of letters correctly identified at 4 m plus 30; approximate Snellen equivalent, 20/25 or worse), an automated noncycloplegic refraction and CP measurement (to the closest 0.25 diopter [D], recorded in the negative) were performed (Humphrey Autorefractor Model 599; Carl Zeiss Meditec). Refraction was refined by subjective refraction. Spherical RE was measured to the closest 0.25 D and converted to the spherical equivalent (SE) (calculated as sphere plus half cylinder). Three CP measurements in 2 axes were recorded in terms of dioptric power, and the mean was used.
Statistical Analysis
Biometric data for each eye were analyzed, but the right eye was used to improve comparison with other Asian populations. Participants with a history of refractive or cataract surgery, phthisis, or no data on RE or biometric variables were excluded.
Mean, range, and SD were calculated for the ocular variables. Mean SE, biometric and other ocular variables, and Lens Opacities Classification System II grades were calculated for each 10-year age group. Age and sex differences were evaluated by multinomial logistic regression for NOP and by multivariate linear regression for SE and other measurements. To examine threshold and nonlinear associations of ocular and clinical variables with age for each sex, locally weighted scatter-plot smoothing used an iterative, least squares method to generate smooth fit lines to the data, with and without height adjustment.
Using multivariate linear regression models, we assessed the contribution of CP, NOP, and AL (model 1) or the components of AL (CCT, ACD, LT, and VCD) (model 2) to the dependent variable RE for all participants (age and sex adjusted) and for each decade from 50 to 80 years or older (sex adjusted). Each independent variable’s contribution was estimated by the magnitude of standardized regression coefficients (SRCs; influence of variable) and semipartial correlation coefficients squared (SPCCs; decrease in the R2 statistic without the variable). The R2 statistic indicates RE variation explained by the model’s variables, together with age and sex. Semipartial correlation coefficients were estimated from multivariate regressions using a stepwise approach, adding the variable with the largest SPCC first. Confidence intervals presented are 95%. P values are 2-sided, with P < .05 indicating statistical significance. Analyses used SAS software (version 9.4; SAS Institute, Inc).
Results
Study Cohort
Of 5782 eligible individuals, a total of 4582 Chinese American participants (79.2%) enrolled. Of these, 511 participants were excluded from analysis, including 379 for cataract surgery in the right eye, 88 for refractive surgery, 1 for phthisis, and 43 for lack of refraction, biometric, or NOP data owing to an incomplete examination. The 4071 remaining participants (88.8%) were younger (mean [SD] age, 60.5 [8.1] years; P < .001) and more likely to be women (2575 [63.3%] compared with 1496 men [36.7%]; P = .03) compared with the excluded group but similar in marital status, educational attainment, acculturation, and income level.
Refractive Error
Table 1 and Table 2 present mean (SD) and range of noncycloplegic RE, stratified by sex and age group. Overall, mean RE was myopic (−0.52 [2.95] D), ranging from −23.5 to 10.88 D in women (mean, −0.51 [2.97] D) and from −22.50 to 7.50 D in men (mean, −0.56 [2.90] D). We found no age-adjusted sex difference before (P = .21) or after (P = .12) height adjustment (Table 1). Mean RE increased from −0.62 (2.95) D at 50 to 59 years to 0.60 (1.62) D at 80 years or older in women (P < .001) (Table 2) and from −0.69 (3.00) D at 50 to 59 years to 0.40 (2.29) D at 80 years or older in men (P < .001) (Table 2).
Table 1. Overall and Sex-Stratified Ocular Biometry, Lens NOP, and Noncycloplegic Refraction in CHES.
| Variables | Overall (N = 4071) |
Women (n = 2575) |
Men (n = 1496) |
P Valuea | Adjusted P Valueb |
|---|---|---|---|---|---|
| Age, mean (SD), y | 60.5 (8.1) | 59.7 (7.7) | 61.9 (8.6) | NA | NA |
| SE, mean (SD), D | −0.52 (2.95) | −0.51 (2.97) | −0.56 (2.90) | .21 | .12 |
| CP, mean (SD), D | 43.27 (1.52) | 43.50 (1.52) | 42.88 (1.45) | <.001 | .02 |
| AL, mean (SD), mm | 23.81 (1.34) | 23.62 (1.34) | 24.14 (1.27) | <.001 | .006 |
| VCD, mean (SD), mm | 15.40 (1.27) | 15.27 (1.28) | 15.64 (1.20) | <.001 | .16 |
| ACD, mean (SD), mm | 3.37 (0.35) | 3.33 (0.34) | 3.44 (0.34) | <.001 | <.001 |
| CCT, µm | 559 (35) | 557 (34) | 564 (35) | <.001 | <.001 |
| LT, mm | 4.48 (0.38) | 4.47 (0.37) | 4.50 (0.39) | .94 | .68 |
| Lens NOP grade, No. (%)c | |||||
| N0 | 70 (1.7) | 47 (1.8) | 23 (1.5) | .09 | .18 |
| NI | 2670 (65.6) | 1707 (66.3) | 963 (64.4) | ||
| NII | 1162 (28.5) | 716 (27.8) | 446 (29.8) | ||
| NIII | 139 (3.4) | 87 (3.4) | 52 (3.5) | ||
| NIV | 27 (0.7) | 16 (0.6) | 11 (0.7) |
Abbreviations: ACD, anterior chamber depth; AL, axial length; CCT, central corneal thickness; CHES, Chinese American Eye Study; CP, corneal power; D, diopter; LT, lens thickness; NA, not applicable; NOP, nuclear opalescence; SE, spherical equivalent; VCD, vitreous chamber depth.
Analysis of covariance tests were used to estimate sex differences in age-adjusted means; Cochran-Mantel-Haenszel test, for lens NOP adjusted for age groups.
Adjusted for age and height.
Graded using the Lens Opacities Classification System II, with higher grades indicating increasing severity. Percentages have been rounded and may not total 100.
Table 2. Age- and Sex-Stratified Ocular Biometry and Lens NOP in CHES.
| Variable | Age Group | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Women | Men | |||||||||
| 50-59 y (n = 1407) |
60-69 y (n = 884) |
70-79 y (n = 223) |
≥80 y (n = 61) |
P Valuea | 50-59 y (n = 676) |
60-69 y (n = 551) |
70-79 y (n = 194) |
≥80 y (n = 75) |
P Valuea | |
| Age, mean (SD), y | 54.4 (3.3) | 63.0 (2.6) | 73.7 (2.9) | 83.5 (3.5) | NA | 54.8 (2.9) | 63.4 (2.8) | 73.6 (3.0) | 83.7 (3.9) | NA |
| SE, mean (SD) [range], D | −0.62 (2.95) [−19.63 to 5.13] | −0.53 (3.18) [−23.50 to 10.88] | 0.02 (2.41) [−12.25 to 4.63] | 0.60 (1.62) [−4.38 to 4.13] | <.001 | −0.69 (3.00) [−22.50 to 7.50] | −0.77 (3.01) [−20.63 to 4.25] | 0.15 (2.18) [−11.75 to 5.38] | 0.40 (2.29) [−8.38 to 4.88] | <.001 |
| CP, mean (SD) [range], D | 43.45 (1.55) [33.49 to 48.43] | 43.52 (1.50) [36.27 to 48.83] | 43.64 (1.28) [40.51 to 47.16] | 43.76 (1.58) [40.26 to 47.50] | .03 | 42.88 (1.47) [36.82 to 47.34] | 42.83 (1.45) [35.82 to 47.85] | 43.03 (1.35) [39.59 to 46.72] | 42.89 (1.50) [39.28 to 46.44] | .56 |
| AL, mean (SD) [range], mm | 23.62 (1.34) [20.76 to 33.30] | 23.69 (1.40) [20.84 to 34.73] | 23.44 (1.13) [20.88 to 29.35] | 23.17 (0.9) [21.05 to 25.44] | .08 | 24.14 (1.27) [20.80 to 30.68] | 24.29 (1.34) [21.31 to 33.33] | 23.91 (1.08) [21.39 to 29.07] | 23.93 (1.12) [22.37 to 28.22] | .26 |
| VCD, mean (SD) [range], mm | 15.28 (1.27) [11.29 to 24.56] | 15.33 (1.35) [12.17 to 26.01] | 15.08 (1.11) [13.11 to 20.73] | 14.81 (0.95) [12.4 to 17.13] | .03 | 15.66 (1.21) [12.60 to 22.35] | 15.72 (1.25) [13.05 to 24.34] | 15.38 (1.05) [13.10 to 20.76] | 15.41 (1.01) [13.94 to 19.28] | .02 |
| ACD, mean (SD) [range], mm | 3.39 (0.32) [1.98 to 4.84] | 3.29 (0.34) [2.09 to 4.62] | 3.15 (0.33) [2.27 to 4.10] | 3.05 (0.44) [2.01 to 4.26] | <.001 | 3.50 (0.33) [2.26 to 4.54] | 3.43 (0.33) [2.32 to 4.60] | 3.29 (0.35) [2.38 to 4.49] | 3.25 (0.37) [2.41 to 4.08] | <.001 |
| CCT, mean (SD) [range], µm | 558 (35) [407 to 673] | 557 (33) [439 to 707] | 555 (33) [462 to 646] | 552 (28) [492 to 594] | .16 | 565 (36) [428 to 677] | 563 (33) [473 to 673] | 562 (35) [475 to 671] | 559 (38) [491 to 644] | .07 |
| LT, mean (SD) [range], mm | 4.39 (0.34) [3.16 to 7.85] | 4.53 (0.36) [2.94 to 6.85] | 4.66 (0.40) [2.84 to 6.21] | 4.76 (0.54) [3.55 to 6.40] | <.001 | 4.37 (0.34) [2.77 to 7.12] | 4.57 (0.36) [3.06 to 6.48] | 4.68 (0.41) [2.89 to 5.80] | 4.71 (0.56) [2.32 to 6.00] | <.001 |
| Lens NOP grade, No. (%)b | ||||||||||
| N0 | 44 (3.1) | 3 (0.3) | 0 | 0 | <.001 | 20 (3.0) | 3 (0.5) | 0 | 0 | <.001 |
| NI | 1119 (79.5) | 504 (57.1) | 71 (31.8) | 13 (21.7) | 544 (80.6) | 336 (61.0) | 69 (35.6) | 14 (18.7) | ||
| NII | 236 (16.8) | 343 (38.8) | 113 (50.7) | 24 (40.0) | 108 (16.0) | 192 (34.8) | 104 (53.6) | 42 (56.0) | ||
| NIII | 6 (0.4) | 27 (3.1) | 36 (16.1) | 18 (30.0) | 1 (0.1) | 15 (2.7) | 21 (10.8) | 15 (20.0) | ||
| NIV | 2 (0.1) | 6 (0.7) | 3 (1.3) | 5 (8.3) | 2 (0.3) | 5 (0.9) | 0 | 4 (5.3) | ||
Abbreviations: ACD, anterior chamber depth; AL, axial length; CCT, central corneal thickness; CHES, Chinese American Eye Study; CP, corneal power; D, diopter; LT, lens thickness; NA, not applicable; NOP, nuclear opalescence; SE, spherical equivalent; VCD, vitreous chamber depth.
Calculated using the test for trend by age groups.
Graded using the Lens Opacities Classification System II, with higher grades indicating increasing severity. Percentages have been rounded and may not total 100.
eFigure 1 in the Supplement shows the association among age, sex, and mean RE using locally weighted scatterplot smoothing regression lines and associated 95% CIs. The SE was higher with each year beyond 60 to 65 years of age. Table 3 gives the association of biometric variables with age as the independent variable. Each year, SE increased by 0.033 D (P < .001).
Table 3. Multivariate Linear Regression Models Demonstrating Association of Ocular Variables With Age for All Participantsa.
| Biometric or Clinical Variable | Regression Coefficient | P Value |
|---|---|---|
| SE, D | 0.033 | <.001 |
| AL, mm | −0.006 | .04 |
| VCD, mm | −0.009 | .001 |
| ACD, mm | −0.011 | <.001 |
| CCT, µm | −0.159 | .02 |
| LT, mm | 0.015 | <.001 |
| CP, D | 0.008 | .01 |
| Lens NOP gradeb | 1.127c | <.001d |
Abbreviations: ACD, anterior chamber depth; AL, axial length; CCT, central corneal thickness; CP, corneal power; D, diopter; LT, lens thickness; NOP, nuclear opalescence; SE, spherical equivalent; VCD, vitreous chamber depth.
All data were adjusted for sex. eTable 1 in the Supplement provides age-stratified results.
Graded using the Lens Opacities Classification System II (LOCSII), with higher grades indicating increasing severity.
Indicates odds ratio for LOCSII grade of NII or greater.
Associated with each year of older age.
Ocular Components
Data on the mean (SD) and range of the ocular components described below, stratified by sex and age group, are presented in Tables 1 and 2 and in eFigures 2 and 3 in the Supplement. Additional data are presented in Table 3 and eTable 1 and eFigure 4 in the Supplement.
Corneal Power
Overall mean CP was 43.27 (1.52) D, with a range of 33.49 to 48.83 D (mean, 43.50 [1.52] D) for women and 35.82 to 47.85 D (mean, 4.28 [1.45] D) for men. We found an age-adjusted sex difference before (P < .001) and after (P = .02) height adjustment (Table 1). Mean CP increased slightly from 43.45 (1.55) D at 50 to 59 years to 43.76 (1.58) D at 80 years or older among women (P = .03) but not among men (42.88 [1.47] at 50-59 years to 42.89 [1.50] D at 80 years or older; P = .56) (Table 2). Corneal power was 0.008 D higher for each year older (P = .01) (Table 3).
Axial Length
Overall mean AL was 23.81 (1.34) mm, with a range of 20.76 to 34.73 mm (mean, 23.62 [1.34] mm) in women and 20.80 to 33.33 mm (mean, 24.14 [1.27] mm) in men. Mean AL decreased from 23.62 (1.34) mm at 50 to 59 years to 23.17 (0.90) mm at 80 years or older in women and from 24.14 (1.27) mm at 50 to 59 years to 23.93 (1.12) mm at 80 years or older in men. Differences across age groups were not significant for women (P = .08) or men (P = .26) (Table 2). However, men’s age-adjusted AL was longer before (P < .001) and after (P = .006) height adjustment (Table 1). In eFigure 2 in the Supplement, this sex difference is demonstrated by the relatively parallel and separate locally weighted scatterplot smoothing regression lines across age distributions. Axial length was 0.006 mm decreased for each year older (P = .04) (Table 3).
Vitreous Chamber Depth
Overall mean VCD was 15.40 (1.27) mm, with a range of 11.29 to 26.01 mm (mean, 15.27 [1.28] mm) in women and 12.60 to 24.34 mm (mean, 15.64 [1.20] mm) in men. Mean VCD was 15.28 (1.27) mm at 50 to 59 years and 14.81 (0.95) mm at 80 years or older in women (P = .03) but 15.66 (1.21) mm at 50 to 59 years and 15.41 (1.01) mm at 80 years or older in men (P = .02). Mean VCD among men was 0.3 to 0.6 mm longer than that among women for each age group (Table 2), which was significant before (P < .001) but not after (P = .16) height adjustment (Table 1). Mean VCD was lower by 0.009 mm for each year older (P = .001) (Table 3 and eFigures 2 and 3 in the Supplement).
Anterior Chamber Depth
Mean ACD was 3.37 (0.35) mm (range, 1.98-4.84 mm in women and 2.26-4.60 mm in men). Mean ACD decreased from 3.39 (0.32) mm at 50 to 59 years to 3.05 (0.44) mm at 80 years or older in women (P < .001) and from 3.50 (0.33) mm at 50 to 59 years to 3.25 (0.37) mm at 80 years or older in men (P < .001). We found an age-adjusted sex difference before (P < .001) and after (P < .001) height adjustment (Table 1 and eFigure 4 in the Supplement). Mean ACD was 0.011 mm shallower for each year older (Table 3) (P < .001).
Central Corneal Thickness
Central corneal thickness ranged from 407 to 707 µm in women and 428 to 677 µm in men. We found a small age-adjusted sex difference in mean CCT before and after height adjustment of 557 µm in women and 564 µm in men (P < .001). No age-related difference in CCT was noted for women (mean, 558 [35] mm at 50-59 years to 552 [28] mm at 80 years or older; P = .16) or men (mean, 565 [36] mm at 50-59 years to 559 [38] mm at 80 years or older; P = .07).
Lens Thickness
Mean LT for women was 4.47 (0.37) mm; for men, 4.50 (0.39) mm. We found no age-adjusted sex difference in LT before (P = .94) or after (P = .68) height adjustment (Table 1). Mean LT increased from 4.39 (0.34) mm at 50 to 59 years to 4.76 (0.54) mm at 80 years or older in women (P < .001) and from 4.37 (0.34) mm at 50 to 59 years to 4.71 (0.56) mm at 80 years or older in men (P < .001) (Table 2 and eFigure 5 in the Supplement). The mean yearly increase was 0.015 mm (P < .001) (Table 3).
Lens NOP
We found no sex difference in the frequency distribution for Lens Opacities Classification System II grades (eg, 716 women [27.8%] vs 446 men [29.8%] had grade NII; P = .09) (Table 1). Greater frequency of higher grades was noted with older age. A grade of NII or higher was present in 244 women (17.3%) and 111 men (16.5%) aged 50 to 59 years and in 47 women (78.3%) and 61 men (81.3%) 80 years or older (P < .001) (Table 2).
RE Determinants
We constructed 2 multivariate linear regression models to evaluate the independent association of ocular anatomical variables with noncycloplegic RE (age and sex adjusted) overall and for specific age groups (sex adjusted), as described in the Methods (Table 4 and eTable 2 in the Supplement). In model 1, CP, NOP, and AL explained 71% of RE variation (R2 = 0.71). The most important contributor was AL, followed by CP (SRC = −0.43; SPCC = 0.15). Nuclear opalescence only contributed significantly in the groups aged 60 to 69 (SRC = −0.046; SPCC = 0.002) and 80 years or older (SRC = −0.17; SPCC = 0.03).
Table 4. Overall and Age-Stratified Multiple Linear Regression Models Demonstrating Ocular Determinants of RE in CHESa.
| Variable | All Participants | P Value | ||
|---|---|---|---|---|
| Regression (95% CI) | SRC | SPCC | ||
| Model 1a | ||||
| AL | −2.01 (−2.05 to −1.97) | −0.92 | 0.55 | <.001 |
| CP | −0.82 (−0.86 to −0.78) | −0.43 | 0.15 | <.001 |
| NOP | −0.15 (−0.24 to −0.06) | −0.03 | 0.0008 | .002 |
| Model R2 value | 0.71 | NA | NA | NA |
| Model 2a | ||||
| VCD | −2.28 (−2.32 to −2.23) | −0.99 | 0.52 | <.001 |
| CP | −0.90 (−0.93 to −0.86) | −0.47 | 0.15 | <.001 |
| LT | −2.24 (−2.38 to −2.09) | −0.29 | 0.06 | <.001 |
| ACD | −0.50 (−0.66 to −0.34) | −0.059 | 0.002 | <.001 |
| CCT | −0.003 (−0.004 to −0.002) | −0.035 | 0.001 | <.001 |
| NOP | −0.07 (−0.15 to 0.02) | −0.01 | 0.0002 | .13 |
| Model R2 value | 0.74 | NA | NA | NA |
Abbreviations: ACD, anterior chamber depth; AL, axial length; CCT, central corneal thickness; CHES, Chinese American Eye Study; CP, corneal power; LT, lens thickness; NA, not applicable; NOP, nuclear opalescence; RE, refractive error; SRC, standardized regression coefficient; SPCC, semipartial correlation coefficient squared; VCD, vitreous chamber depth.
Model 1 uses AL and model 2 uses component parts of AL (VCD, LT, ACD, and CCT). Both were adjusted for age and sex. eTable 2 in the Supplement provides age-stratified results.
In model 2, VCD became the most important determinant (SRC = −0.99; SPCC = 0.52), followed by CP (SRC = −0.47; SPCC = 0.15) and LT (SRC = −0.29; SPCC = 0.06). Anterior chamber depth (SRC = −0.059; SPCC = 0.002) and CCT (SRC = −0.035; SPCC = 0.001) had smaller effects. When the models were adjusted for sex, VCD, LT, CP, and ACD were associated with RE across all ages (Table 4 and eTable 2 in the Supplement). Central corneal thickness was a smaller contributor in the groups aged 50 to 59 and 60 to 69 years. Only in the group 80 years or older did NOP contribute (SRC = −0.17; SPCC = 0.007; P = .007). In CHES, AL was longer compared with data from US Latino, Singaporean and mainland Chinese, and Mongolian populations (Table 5).
Table 5. Comparison of Mean AL Among Various Studies.
| Studya | Ethnicity/Race | Age Group, y | ||||
|---|---|---|---|---|---|---|
| 40-49 | 50-59 | 60-69 | 70-79 | ≥80 | ||
| Women | ||||||
| LALES, 2003 | Latino | 23.2 | 23.2 | 23.1 | 23.1 | 23.0 |
| CHES | Chinese | NA | 23.6 | 23.7 | 23.4 | 23.2 |
| Tanjong Pagar Survey,2001 | Singaporean Chinese | 23.4 | 23.0 | 22.7 | 22.7b | NA |
| Liwan Eye Study, 2009 | Mainland Chinese | 22.8 | 22.9 | 22.8 | 23.0 | |
| Men | ||||||
| LALES, 2003 | Latino | 23.7 | 23.6 | 23.6 | 23.5 | 23.7 |
| CHES | Chinese | NA | 24.1 | 24.3 | 23.9 | 23.9 |
| Tanjong Pagar Survey,2001 | Singaporean Chinese | 23.8 | 23.5 | 23.4 | 23.4b | NA |
| Liwan Eye Study, 2009 | Mainland Chinese | NA | 23.4 | 23.4 | 23.3 | 23.7 |
Abbreviations: AL, axial length; CHES, Chinese American Eye Study; LALES, Los Angeles Latino Eye Study; NA, not applicable.
All listed studies measured AL with A-scan ultrasonography.
Age range is 70 to 81 years.
Discussion
CHES is a population-based study of ocular conditions in aging Chinese Americans. A prior CHES report demonstrated a higher burden of myopia and high myopia compared with other US populations, and the present CHES study reports ocular determinants of RE.
Although a sex difference in RE was reported in some previous studies, this was not seen in other studies or in CHES. In CHES, we found a significant sex difference in CP, AL, ACD, and CCT after age adjustment and before and after height adjustment, but not in LT, NOP, or VCD. Women had steeper corneas. Women’s shorter AL and ACD (also observed in other studies) likely represent inherent anatomical differences. Women’s slightly thinner CCT may not have clinical significance. Of interest, women’s shorter age-adjusted VCD was not significant after height adjustment. Thus, women’s shorter ACD may explain their shorter AL.
In the Los Angeles Latino Eye Study (LALES), with protocols identical to those of CHES, women also had steeper CP and shorter ACD, VCD, and AL but thinner LT and more hyperopia compared with men. Singaporean Chinese women in the Tanjong Pagar Survey had a shorter ACD as in CHES but also had shorter VCD, flatter CP, and thicker LT. In the Liwan Eye Study in mainland China, women had steeper CP and shorter AL and ACD, similar to CHES, but had thicker LT, similar to Singaporean Chinese. The Tanjong Pagar Survey and Liwan Eye Study did not adjust for height; thus, we cannot comment on the difference that height adjustment would have made to their findings. Also, the older Tanjong Pagar Survey and Liwan Eye Study may be skewed by having a different proportion of excluded patients with pseudophakia. When comparing AL of different age groups between each of these studies with CHES results (without height adjustment), CHES participants have the longest AL in each age group and a longer AL among male compared with female participants in each group. In CHES, women’s shorter mean AL (predisposing to hyperopia) may be cancelled out by the steeper CP and shorter ACD (supporting myopia), resulting in RE similar to that of men.
In CHES, being older was correlated with more hyperopic refraction, shallower ACD, and increased LT and NOP, stratified by age and sex (Table 2). Vitreous chamber depth was shorter with increased age. The age-related lens changes of smaller ACD, thicker LT, and higher NOP were also seen among Singaporean and mainland Chinese and in US Latinos. The shorter VCD and AL (Tables 2 and 3) with being older was also observed in Singaporean Chinese (aged 40-59 years). This change may result because younger persons now have different environmental factors (eg, near-work activities during ocular growth) predisposing to myopia. Age-related AL variation was not present in Mongolian, US Latino, or mainland Chinese populations.
Many population-based studies report a hyperopic shift in noncycloplegic RE with being older followed by a myopic shift among the oldest participants. This shift has been noted in US and non-US white, Afro-Caribbean, white and African American, Mongolian, US Latino, Burmese, Singaporean Chinese and Malay, and mainland Chinese populations. The hyperopic shift is attributed to loss of accommodation and unmasking of hyperopia and the myopic shift to NOP. In CHES, we observed hyperopic shift without significant subsequent myopic shift.
Stepwise regression models revealed that AL and CP were the biggest determinants of RE in this population, as in US Latino, Singaporean Chinese, mainland Chinese, and Mongolian populations. To observe a myopic shift in the oldest groups, we would expect NOP to be an important determinant of RE. However, in CHES, NOP was a significant independent contributor only in the groups 60 to 69 years and 80 years or older but not in the group aged 70 to 79 years, in whom NOP played a much smaller role (SRC = −0.15; SPCC = 0.0008) than in Singaporean Chinese and Latino populations. With use of a Lens Opacities Classification System II grade of NII or higher as a criterion for clinically significant cataract, the age-adjusted prevalence of nuclear opacities was 38.1% in CHES vs 13.1% in LALES. In every age group, cataract prevalence in CHES was similar to or higher than that in LALES; however, we found a lack of cataract-induced myopic shift in CHES in the oldest groups. This finding may be explained by AL’s stronger contribution or by cataract prevalence increasing at the same time as the loss-of-accommodation–driven hyperopic shift, which could mask the cataract-induced myopic shift seen in other studies. Finally, the high prevalence of excluded participants with pseudophakia (454 [9.9%] in CHES vs 242 [3.9%] in LALES) may contribute to not seeing the late myopic shift.
Additional analysis revealed the relative importance of AL in determining RE in CHES compared with the similarly designed LALES and Tanjong Pagar Survey. The overall R2 value was 0.71 for model 1 in CHES compared with 0.52 in LALES, 0.61 in the Liwan Eye Study, and 0.75 for the group aged 40 to 59 years and 0.48 for the group aged 60 to 81 years in the Tanjong Pagar Survey. Of interest, the relative contribution of AL was higher in CHES compared with Los Angeles Latinos, urban mainland Chinese, or Singaporean Chinese (assessed by SRC values). This finding may be partly explained by the greater mean and variation of AL in CHES (23.81 [1.32] mm) compared with LALES (23.38 [1.01] mm), the Tanjong Pagar Survey (23.27 [1.17] mm), the Liwan Eye Study (23.11 [0.65] mm), and studies of other populations.
In model 2, VCD was more important and NOP and ACD less so, in determining RE in CHES than in LALES or the Tanjong Pagar Survey. The greater NOP contribution in the Tanjong Pagar Survey may result from its greater burden of NOP compared with CHES. Of interest, LALES had more NOP, ACD, and LT contribution to RE than did CHES despite CHES having a similar to greater prevalence of NOP. The lens shape in LALES may be more convex, representing a larger proportion of power for overall refraction. However, in CHES, the greater importance of AL and VCD was likely attributable to longer overall AL, VCD, and their variations. The longer mean AL in CHES suggests more axial elongation in childhood and beyond compared with non-US counterparts. In the future, we plan to investigate genetic and environmental risk factors affecting myopia in CHES.
Greater AL and axial myopia among Chinese Americans are important public health concerns contributing to greater myopic RE and related diseases, such as primary open-angle glaucoma, retinal detachment, and myopic retinopathy. A disproportionate burden of myopic retinopathy and associated visual impairment has been demonstrated. Clinicians may need to use population-specific screening strategies, such as more frequent eye examinations, and researchers should investigate interventions targeting environmental risk factors for axial elongation (ie, near correction to reduce accommodation or outdoor activities).
Limitations
The main limitation of this study is its cross-sectional nature. Because of this, we inferred the effects that being older may have on biometry, but this inference should be interpreted with caution, because cohort effects may result from things such as environmental differences between the younger and older Chinese American participants rather than a true age-related change.
Conclusions
Our data provide RE distribution and its ocular determinants in this Chinese American population. In CHES, we identified greater mean AL compared with other Chinese and American populations and a stronger AL contribution to RE compared with US Latino and Singaporean Chinese populations, suggesting population-specific environmental influences. In addition to serving as normative values, these detailed biometric data highlight the importance of better understanding the role of AL and AL elongation in myopia-related disease in this at-risk population.
eFigure 1. Association Among Age, Sex, and Spherical Equivalent (SE) Plotted Using LOWESS
eFigure 2. Association Among Age, Sex, and Mean Axial Length (AL) Plotted Using LOWESS
eFigure 3. Association Among Age, Sex, and Mean Vitreous Chamber Depth Plotted Using LOWESS
eFigure 4. Association Among Age, Sex, and Mean Anterior Chamber Depth (ACD) Plotted Using LOWESS
eFigure 5. Association Among Age, Sex, and Mean Lens Thickness Plotted Using LOWESS
eTable 1. Age-Stratified Multivariate Linear Regression Models Demonstrating Association of Ocular Variables with Age
eTable 2. Age-Stratified Multiple Linear Regression Models Demonstrating Ocular Determinants of Refractive Error in CHES
References
- 1.Naidoo KS, Leasher J, Bourne RR, et al. ; Vision Loss Expert Group of the Global Burden of Disease Study . Global vision impairment and blindness due to uncorrected refractive error, 1990-2010. Optom Vis Sci. 2016;93(3):227-234. [DOI] [PubMed] [Google Scholar]
- 2.Vitale S, Cotch MF, Sperduto RD. Prevalence of visual impairment in the United States. JAMA. 2006;295(18):2158-2163. [DOI] [PubMed] [Google Scholar]
- 3.Hashemi H, Khabazkhooh M, Iribarren R, Emamian MH, Fotouhi A. Five year change in refraction and its ocular components in the 40-64 year old population of the Shahroud Eye Cohort Study. Clin Experiment Ophthalmol. 2016;44(8):669-677. [DOI] [PubMed] [Google Scholar]
- 4.Shufelt C, Fraser-Bell S, Ying-Lai M, Torres M, Varma R; Los Angeles Latino Eye Study Group . Refractive error, ocular biometry, and lens opalescence in an adult population: the Los Angeles Latino Eye Study. Invest Ophthalmol Vis Sci. 2005;46(12):4450-4460. [DOI] [PubMed] [Google Scholar]
- 5.Broman AT, Munoz B, Rodriguez J, et al. . The impact of visual impairment and eye disease on vision-related quality of life in a Mexican-American population: proyecto VER. Invest Ophthalmol Vis Sci. 2002;43(11):3393-3398. [PubMed] [Google Scholar]
- 6.Wang Q, Klein BE, Klein R, Moss SE. Refractive status in the Beaver Dam Eye Study. Invest Ophthalmol Vis Sci. 1994;35(13):4344-4347. [PubMed] [Google Scholar]
- 7.Katz J, Tielsch JM, Sommer A. Prevalence and risk factors for refractive errors in an adult inner city population. Invest Ophthalmol Vis Sci. 1997;38(2):334-340. [PubMed] [Google Scholar]
- 8.Wickremasinghe S, Foster PJ, Uranchimeg D, et al. . Ocular biometry and refraction in Mongolian adults. Invest Ophthalmol Vis Sci. 2004;45(3):776-783. [DOI] [PubMed] [Google Scholar]
- 9.Pan CW, Wong TY, Chang L, et al. . Ocular biometry in an urban Indian population: the Singapore Indian Eye Study (SINDI). Invest Ophthalmol Vis Sci. 2011;52(9):6636-6642. [DOI] [PubMed] [Google Scholar]
- 10.Wong TY, Foster PJ, Hee J, et al. . Prevalence and risk factors for refractive errors in adult Chinese in Singapore. Invest Ophthalmol Vis Sci. 2000;41(9):2486-2494. [PubMed] [Google Scholar]
- 11.Casson RJ, Newland HS, Muecke J, et al. . Prevalence and causes of visual impairment in rural Myanmar: the Meiktila Eye Study. Ophthalmology. 2007;114(12):2302-2308. [DOI] [PubMed] [Google Scholar]
- 12.Tan CS, Chan YH, Wong TY, et al. . Prevalence and risk factors for refractive errors and ocular biometry parameters in an elderly Asian population: the Singapore Longitudinal Aging Study (SLAS). Eye (Lond). 2011;25(10):1294-1301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.He M, Foster PJ, Ge J, et al. . Prevalence and clinical characteristics of glaucoma in adult Chinese: a population-based study in Liwan District, Guangzhou. Invest Ophthalmol Vis Sci. 2006;47(7):2782-2788. [DOI] [PubMed] [Google Scholar]
- 14.Lavanya R, Jeganathan VS, Zheng Y, et al. . Methodology of the Singapore Indian Chinese Cohort (SICC) Eye Study: quantifying ethnic variations in the epidemiology of eye diseases in Asians. Ophthalmic Epidemiol. 2009;16(6):325-336. [DOI] [PubMed] [Google Scholar]
- 15.Cheung CM, Tai ES, Kawasaki R, et al. . Prevalence of and risk factors for age-related macular degeneration in a multiethnic Asian cohort. Arch Ophthalmol. 2012;130(4):480-486. [DOI] [PubMed] [Google Scholar]
- 16.Yang K, Liang YB, Gao LQ, et al. . Prevalence of age-related macular degeneration in a rural Chinese population: the Handan Eye Study. Ophthalmology. 2011;118(7):1395-1401. [DOI] [PubMed] [Google Scholar]
- 17.Liang YB, Friedman DS, Wong TY, et al. ; Handan Eye Study Group . Prevalence and causes of low vision and blindness in a rural chinese adult population: the Handan Eye Study. Ophthalmology. 2008;115(11):1965-1972. [DOI] [PubMed] [Google Scholar]
- 18.Xu L, Wang Y, Li Y, et al. . Causes of blindness and visual impairment in urban and rural areas in Beijing: the Beijing Eye Study. Ophthalmology. 2006;113(7):1134.e1-1134.e11. [DOI] [PubMed] [Google Scholar]
- 19.US Census Bureau. 2010 Census shows Asians are fastest-growing race group. 2012. https://www.census.gov/newsroom/releases/archives/2010_census/cb12-cn22.html. Published March 12, 2012. Accessed October 1, 2015.
- 20.Asian & Pacific Islander American Health Forum. Demographic and socioeconomic profiles of Asian Americans, Native Hawaiians, and Pacific Islanders in the United States. 1986–2011. http://www.apiahf.org/resources/resources-database/profilesofaanhpi. Published July 2011. Accessed October 1, 2015.
- 21.World Medical Association 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]
- 22.Varma R, Hsu C, Wang D, Torres M, Azen SP; Chinese American Eye Study Group . The Chinese American Eye Study: design and methods. Ophthalmic Epidemiol. 2013;20(6):335-347. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Chylack LT Jr, Leske MC, McCarthy D, Khu P, Kashiwagi T, Sperduto R. Lens Opacities Classification System II (LOCS II). Arch Ophthalmol. 1989;107(7):991-997. [DOI] [PubMed] [Google Scholar]
- 24.Ferris FL III, Kassoff A, Bresnick GH, Bailey I. New visual acuity charts for clinical research. Am J Ophthalmol. 1982;94(1):91-96. [PubMed] [Google Scholar]
- 25.Cleveland WS. Robust locally weighted regression and smoothing scatterplots. J Am Stat Assoc. 1979;74:829-836. [Google Scholar]
- 26.Wong TY, Foster PJ, Ng TP, Tielsch JM, Johnson GJ, Seah SK. Variations in ocular biometry in an adult Chinese population in Singapore: the Tanjong Pagar Survey. Invest Ophthalmol Vis Sci. 2001;42(1):73-80. [PubMed] [Google Scholar]
- 27.He M, Huang W, Li Y, Zheng Y, Yin Q, Foster PJ. Refractive error and biometry in older Chinese adults: the Liwan Eye Study. Invest Ophthalmol Vis Sci. 2009;50(11):5130-5136. [DOI] [PubMed] [Google Scholar]
- 28.Varma R, Torres M, McKean-Cowdin R, et al. ; Chinese-American Eye Study Group . Prevalence of refractive error in adult Chinese Americans: the Chinese American Eye Study. Am J Ophthalmol. 2016;175:201-212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Attebo K, Ivers RQ, Mitchell P. Refractive errors in an older population: the Blue Mountains Eye Study. Ophthalmology. 1999;106(6):1066-1072. [DOI] [PubMed] [Google Scholar]
- 30.Saw SM, Zhang MZ, Hong RZ, Fu ZF, Pang MH, Tan DT. Near-work activity, night-lights, and myopia in the Singapore-China Study. Arch Ophthalmol. 2002;120(5):620-627. [DOI] [PubMed] [Google Scholar]
- 31.You X, Wang L, Tan H, et al. . Near work related behaviors associated with myopic shifts among primary school students in the Jiading District of Shanghai: a school-based one-year cohort study. PLoS One. 2016;11(5):e0154671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Wensor M, McCarty CA, Taylor HR. Prevalence and risk factors of myopia in Victoria, Australia. Arch Ophthalmol. 1999;117(5):658-663. [DOI] [PubMed] [Google Scholar]
- 33.Wu SY, Nemesure B, Leske MC. Refractive errors in a black adult population: the Barbados Eye Study. Invest Ophthalmol Vis Sci. 1999;40(10):2179-2184. [PubMed] [Google Scholar]
- 34.Warrier S, Wu HM, Newland HS, et al. . Ocular biometry and determinants of refractive error in rural Myanmar: the Meiktila Eye Study. Br J Ophthalmol. 2008;92(12):1591-1594. [DOI] [PubMed] [Google Scholar]
- 35.Saw SM, Chan YH, Wong WL, et al. . Prevalence and risk factors for refractive errors in the Singapore Malay Eye Survey. Ophthalmology. 2008;115(10):1713-1719. [DOI] [PubMed] [Google Scholar]
- 36.Varma R, Torres M; Los Angeles Latino Eye Study Group . Prevalence of lens opacities in Latinos: the Los Angeles Latino Eye Study. Ophthalmology. 2004;111(8):1449-1456. [DOI] [PubMed] [Google Scholar]
- 37.Varma R, Sun J, Torres M, et al. ; Chinese American Eye Study Group . Prevalence of lens opacities in adult Chinese Americans: the Chinese American Eye Study (CHES). Invest Ophthalmol Vis Sci. 2016;57(15):6692-6699. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Seah SK, Wong TY, Foster PJ, Ng TP, Johnson GJ. Prevalence of lens opacity in Chinese residents of Singapore: the Tanjong Pagar Survey. Ophthalmology. 2002;109(11):2058-2064. [DOI] [PubMed] [Google Scholar]
- 39.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] [PubMed] [Google Scholar]
- 40.Verhoeven VJ, Buitendijk GH, Rivadeneira F, et al. ; Consortium for Refractive Error and Myopia (CREAM) . Education influences the role of genetics in myopia. Eur J Epidemiol. 2013;28(12):973-980. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Varma R, Kim JS, Burkemper BS, et al. ; Chinese American Eye Study Group . Prevalence and causes of visual impairment and blindness in Chinese American adults: the Chinese American Eye Study. JAMA Ophthalmol. 2016;134(7):785-793. [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. Association Among Age, Sex, and Spherical Equivalent (SE) Plotted Using LOWESS
eFigure 2. Association Among Age, Sex, and Mean Axial Length (AL) Plotted Using LOWESS
eFigure 3. Association Among Age, Sex, and Mean Vitreous Chamber Depth Plotted Using LOWESS
eFigure 4. Association Among Age, Sex, and Mean Anterior Chamber Depth (ACD) Plotted Using LOWESS
eFigure 5. Association Among Age, Sex, and Mean Lens Thickness Plotted Using LOWESS
eTable 1. Age-Stratified Multivariate Linear Regression Models Demonstrating Association of Ocular Variables with Age
eTable 2. Age-Stratified Multiple Linear Regression Models Demonstrating Ocular Determinants of Refractive Error in CHES