This cross-sectional study investigates the association between myopia and primary open-angle glaucoma in the 2019 California Medicare population and whether there was evidence of effect measure modification of this association by race and ethnicity.
Key Points
Question
How is the association between myopia and primary open-angle glaucoma (POAG) modified by race and ethnicity?
Findings
In this cross-sectional study of 2 717 346 California Medicare beneficiaries, myopia was associated with higher adjusted odds of POAG. This association was stronger among Asian, Black, and Hispanic beneficiaries compared with non-Hispanic White beneficiaries.
Meaning
In the California Medicare population, racial and ethnic minority beneficiaries with myopia may have higher risk of POAG, suggesting possible need for earlier or more frequent screenings.
Abstract
Importance
Racial and ethnic differences in the association between myopia and primary open-angle glaucoma (POAG) are not well understood.
Objective
To investigate the association between myopia and POAG in the 2019 California Medicare population and to investigate whether there was evidence of effect measure modification of this association by race and ethnicity.
Design, Setting, and Participants
This cross-sectional study used administrative claims data from 2019 California Medicare beneficiaries 65 years or older with California residence and active coverage with Medicare parts A and B. Analysis took place between October 2021 and October 2023.
Exposures
The primary exposure was myopia, which was defined by International Statistical Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) diagnosis codes.
Main Outcomes and Measures
The outcome of interest was POAG, which was defined by ICD-10-CM code.
Results
Of 2 717 346 California Medicare beneficiaries in 2019, 1 440 769 (53.0%) were aged 65 to 74 years, 1 544 479 (56.8%) identified as female, 60 211 (2.2%) had myopia, and 171 988 (6.3%) had POAG. Overall, 346 723 individuals (12.8%) identified as Asian, 117 856 (4.3%) as Black, 430 597 (15.8%) as Hispanic, 1 705 807 (62.8%) as White, and 115 363 (4.2%) as other race and ethnicity. In adjusted logistic regression analyses, beneficiaries with myopia had higher odds of POAG compared with beneficiaries without myopia (odds ratio [OR], 2.41; 95% CI, 2.35-2.47). In multivariable models stratified by race and ethnicity, the association between myopia and POAG was stronger in Asian (OR, 2.74; 95% CI, 2.57-2.92), Black (OR, 2.60; 95% CI, 2.31-2.94), and Hispanic (OR, 3.28; 95% CI, 3.08-3.48) beneficiaries compared with non-Hispanic White beneficiaries (OR, 2.14; 95% CI, 2.08-2.21).
Conclusions and Relevance
In the 2019 California Medicare population, myopia was associated with greater adjusted odds of POAG. This association was stronger among Asian, Black, and Hispanic beneficiaries compared with non-Hispanic White beneficiaries. These findings suggest possible disparities in glaucoma risk by race and ethnicity in individuals with myopia and may indicate greater need for glaucoma screening in individuals with myopia from racial and ethnic minority backgrounds.
Introduction
Myopia, or nearsightedness, is one of the most common refractive disorders of the eye. The National Eye Institute estimates that almost 50% of the world’s population will have myopia by 2050.1 Myopia, defined as having a spherical equivalent of −0.5 diopters or less, is generally caused by excessive lengthening of the eye in the axial dimension or is less commonly due to progressive thinning of the cornea, leading to a distorted image on the retina.2 Myopia has been previously well studied as a possible risk factor for primary open-angle glaucoma (POAG).3,4,5 POAG is the most common type of glaucoma in the United States and the second leading cause of irreversible vision loss in the world.6,7 This disease process most commonly involves an increase in intraocular pressure, leading to the progressive loss of retinal ganglion cells and damage to the optic nerve.7 Myopic eyes seem to have greater deformity of the lamina cribrosa and weaker scleral support at the optic nerve, which may contribute to greater susceptibility of the optic nerve to glaucomatous damage.5,8,9
Although the association between myopia and increased risk of POAG has been well documented in previous literature, there is a gap in the current research on whether the magnitude of this association is different among various racial and ethnic groups. This is of particular concern given that prior studies have described that racial and ethnic disparities exist for both myopia and POAG, with higher prevalence of both conditions in individuals from Black and Hispanic backgrounds.10,11,12,13,14 Thus, the aims of this study were (1) to examine and quantify the association between myopia and POAG among 2019 Medicare beneficiaries in California and (2) to investigate whether race and ethnicity modified this association through an epidemiological phenomenon known as effect measure modification (EMM).15 In other words, we aimed to assess whether the association between myopia and POAG is different among various racial and ethnic groups, allowing for the exploration of the hypothesis that myopia may be a greater risk factor for POAG in certain racial and ethnic groups.
Methods
A cross-sectional study was conducted using the entire population of the 2019 CA Master Beneficiary Summary File and the Standard Analytic Files of part B Carrier Claim files from the Centers for Medicare & Medicaid Services.16,17 Inclusion criteria for this study were age 65 years or older, California residence, active coverage with both Medicare parts A and B, and at least 1 part B claim in 2019 as an indication of active Medicare usage. The study was approved by the institutional review board of the University of California, Los Angeles. Waiver of informed consent was approved by the University of California, Los Angeles institutional review board because the design of the study does not allow the possibility of obtaining consent. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
The outcome of interest was POAG and the primary exposure was myopia. POAG was examined as a binary variable of any POAG vs no POAG and also by severity levels of mild, moderate, and severe POAG. An additional other severity category included indeterminate stage (ie, when visual fields have not been performed or results are unreliable or uninterpretable) and unspecified (ie, stage not recorded). Myopia was assessed as the presence of any myopia and also by severity level of no myopia, myopia without degenerative changes, and degenerative myopia. Both POAG and myopia were defined by having International Statistical Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) diagnosis codes for their respective conditions in part B claims (eTables 1 and 2 in Supplement 1). Race and ethnicity were defined in the California Medicare population as Asian, Black, Hispanic, non-Hispanic White (hereafter White), and other (American Indian/Alaskan Native, unknown [did not report any race], and other race and ethnicity [not specified in more granular detail by Medicare]) from the enhanced race and ethnicity code within Medicare, which was based on taking the beneficiary race code that has historically been reported to the Social Security Administration by individuals applying for a Social Security number or benefits, and using first and last name algorithms developed by the Research Triangle Institute.18 Currently, data from Medicare do not separate multiracial beneficiaries.19 Sex is similarly self-identified by applicants to Social Security.20 Other covariates to further characterize beneficiaries included age (subdivided into the following age groups: 65-69, 70-74, 75-89, ≥90 years), sex (male vs female), and systemic disease burden defined by Charlson Comorbidity Index (CCI) score (categorized into scores of 0, 1-2, 3-4, ≥5). The CCI score, designed by Charlson and colleagues,21 is based on 19 medical conditions (eTable 3 in Supplement 1) and produces a morbidity score that reflects mortality risk, adjusting for variable morbidity rates within a patient population.
Beneficiary baseline characteristics were compared by myopia status (any myopia vs no myopia) using χ2 tests. Crude and adjusted logistic regression models were first created to evaluate the association between myopia and POAG for the entire analytic sample, the latter adjusted for age, sex, race and ethnicity, and CCI score. To test for EMM between myopia and race and ethnicity, a fully saturated multivariable logistic regression model was constructed with POAG as the outcome and myopia as the exposure and the covariates age, sex, race and ethnicity, CCI score and an interaction term between myopia and race and ethnicity. To compare the association between myopia and POAG modified by race and ethnicity, separate multivariable models were fitted for the association between myopia and POAG, adjusted for age, sex, and CCI score and stratified by race and ethnicity.
To further evaluate for EMM by race and ethnicity for associations between different levels of myopia and POAG severity, we performed additional analyses of (1) the association between levels of myopia severity and any POAG and (2) the association between any myopia and levels of POAG severity. To define myopia severity, the myopia exposure was stratified into 3 levels: no myopia, myopia without degenerative changes, and degenerative myopia. To assess the association between myopia severity and POAG, a multivariable logistic regression model was constructed with myopia severity as the exposure and any POAG as the outcome, where myopia severity was defined as a categorical exposure variable of no myopia, myopia without degenerative changes, and degenerative myopia; this model was adjusted for age, sex, race and ethnicity, and CCI score. Beneficiaries with both ICD-10-CM codes listed for myopia and degenerative myopia were categorized as having only degenerative myopia for this analysis. For the myopia and POAG severity analysis, POAG severity was defined as mild, moderate, severe, and other based on ICD-10-CM modifier codes (eTable 1 in Supplement 1). The other category included ICD-10-CM modifier codes for unspecified and indeterminate POAG. To evaluate the association between any myopia and POAG severity, multivariable multinomial logistic regressions were constructed to estimate the odds of (1) severe POAG vs no glaucoma, (2) moderate POAG vs no glaucoma, (3) mild POAG vs no glaucoma, and (4) other POAG vs no glaucoma, comparing beneficiaries with any myopia vs no myopia, adjusting for age, sex, race and ethnicity, and CCI score all in the same model. Finally, we evaluated for EMM by race and ethnicity in the above analyses using the same methods described in the primary analyses, including fitting a fully saturated regression model with an interaction term to assess for interaction and subsequently fitting separate regression models stratified by race and ethnicity. By default, all P values for χ2 tests were 2-sided. The significance threshold was .05, except for 99% CIs, for which it was .01. P values were not adjusted for multiple analyses. All statistical analyses were performed using SAS version 9.4 (SAS Institute). Analysis took place between October 2021 and October 2023.
Results
Table 1 describes baseline characteristics for the study population overall and by myopia status. The study population included 2 717 346 California Medicare beneficiaries. There were 60 211 beneficiaries (2.2%) with a diagnosis of myopia and 171 988 (6.3%) with a diagnosis of POAG. Overall, 346 723 individuals (12.8%) identified as Asian, 117 856 (4.3%) as Black, 430 597 (15.8%) as Hispanic, 1 705 807 (62.8%) as White, and 115 363 (4.2%) as other race and ethnicity (Table 1). The P values for χ2 tests comparing beneficiaries with vs without any myopia were P < .001 each for age (χ2 = 929.4), sex (χ2 = 473.4), CCI score (χ2 = 5167.5), race and ethnicity (χ2 = 1476.0), and POAG severity (χ2 = 5518.0). In beneficiaries who identified as White and other, there were a higher proportion who had vs did not have myopia, while the opposite was true for Asian, Black, and Hispanic beneficiaries (χ2 = 1476; P < .001). In beneficiaries with any myopia (60 711 [2.2%]), the highest proportions were aged 70 to 74 years (16 589 [27.6%]), female (36 840 [61.2%]), and had a CCI score of 1 to 2 (21 351 [35.5%]). In crude analyses, a higher proportion of individuals with myopia had all levels of glaucoma severity compared with those without myopia (χ2 = 5518.0; P < .001).
Table 1. Baseline Characteristics of Study Population From the Centers for Medicare & Medicaid Services in 2019.
| Characteristic | Beneficiaries, No. (%) | P value | |
|---|---|---|---|
| With any myopia (n = 60 211)a | Without myopia (n = 2 657 135) | ||
| Age, y | |||
| 65-69 | 14 443 (23.99) | 744 619 (28.02) | <.001 |
| 70-74 | 16 589 (27.55) | 665 118 (25.03) | |
| 75-79 | 12 569 (20.87) | 486 863 (18.32) | |
| 80-84 | 8342 (13.85) | 345 920 (13.02) | |
| 85-89 | 4951 (8.22) | 228 777 (8.61) | |
| ≥90 | 3317 (5.51) | 185 838 (6.99) | |
| Sex | |||
| Male | 23 371 (38.82) | 1 149 496 (43.26) | <.001 |
| Female | 36 840 (61.18) | 1 507 639 (56.74) | |
| CCI score | |||
| 0 | 19 565 (32.49) | 9 145 944 (34.42) | <.001 |
| 1-2 | 21 351 (35.46) | 924 666 (34.80) | |
| 3-4 | 11 100 (18.44) | 456 270 (17.18) | |
| ≥5 | 8195 (13.61) | 361 605 (13.61) | |
| Race and ethnicity | |||
| Asian | 7176 (11.92) | 423 421 (15.94) | <.001 |
| Black | 1618 (2.69) | 116 238 (4.37) | |
| Hispanic | 7126 (11.84) | 339 697 (12.78) | |
| Non-Hispanic White | 41 111 (68.28) | 1665 696 (62.69) | |
| Otherb | 3180 (5.28) | 112 183 (4.22) | |
| POAG severity | |||
| No POAG | 50524 (83.91) | 2 446 160 (92.06) | <.001 |
| Mild | 2206 (3.66) | 47 748 (1.80) | |
| Moderate | 2612 (4.34) | 56 677 (2.13) | |
| Severe | 1886 (3.13) | 34 145 (1.29) | |
| Other | 2983 (4.95) | 72 405 (2.72) | |
| Myopia classification | |||
| Myopia | 51 565 (85.64) | 0 | NA |
| Degenerative myopia | 8646 (14.36) | 0 | |
Abbreviations: CCI, Charlson Comorbidity Index; NA, not applicable; POAG, primary open-angle glaucoma.
The classification any myopia includes both myopia and degenerative myopia, as indicated by International Statistical Classification of Diseases, Tenth Revision, Clinical Modification codes H52.1* and H44.2*, respectively.
Other included American Indian/Alaskan Native, unknown (did not report any race), and other race and ethnicity (not specified in more granular detail by Medicare).
Table 2 outlines results from logistic regression modeling of the association between myopia and POAG overall and by race and ethnicity. In fully adjusted analyses, beneficiaries with myopia had 2.41 times higher odds of POAG compared with beneficiaries without myopia (odds ratio [OR], 2.41; 95% CI, 2.35-2.47). The interaction term between myopia and race and ethnicity had a P value less than .001. In multivariable models stratified by race and ethnicity, myopia was associated with increased odds of POAG in all racial and ethnic groups, but these associations were stronger in beneficiaries who identified as Asian (OR, 2.74; 95% CI, 2.57-2.92), Black (OR, 2.60; 95% CI, 2.31-2.94), and Hispanic (OR, 3.28; 95% CI, 3.08-3.48) compared with the associations in those who identified as White (OR, 2.14; 95% CI, 2.08-2.21), which was indicated by the lack of overlap in confidence intervals. The Figure offers a forest plot for a graphical comparison of the ORs and 95% CIs for the race and ethnicity–stratified associations between myopia and POAG.
Table 2. The Associations Between Myopia and Primary Open-Angle Glaucoma, Stratified by Racial and Ethnic Group: Results From Logistic Regression Modelsa.
| Race and ethnicity | Modela | Odds ratio (95% CI) | 99% CI |
|---|---|---|---|
| All | Unadjusted | 2.35 (2.30-2.41) | 2.28-2.43 |
| Adjustedb | 2.41 (2.35-2.47) | 2.33-2.48 | |
| Asian | Unadjusted | 3.68 (3.47-3.91) | 2.44-2.87 |
| Adjustedb | 2.74 (2.57-2.92) | 2.52-2.97 | |
| Black | Unadjusted | 2.67 (2.37-3.01) | 2.28-3.12 |
| Adjustedb | 2.60 (2.31-2.94) | 2.22-3.06 | |
| Hispanic | Unadjusted | 2.64 (2.49-2.81) | 3.40-3.98 |
| Adjustedb | 3.28 (3.08-3.48) | 3.02-3.55 | |
| Non-Hispanic White | Unadjusted | 2.11 (2.05-2.18) | 2.03-2.20 |
| Adjustedb | 2.14 (2.08-2.21) | 2.06-2.24 | |
| Otherc | Unadjusted | 2.34 (2.10-2.61) | 2.03-2.70 |
| Adjustedb | 2.45 (2.20-2.73) | 2.13-2.83 |
Models examine odds of primary open-angle glaucoma comparing beneficiaries with myopia to those without myopia.
Adjusted for age, sex, and Charlson Comorbidity Index score.
Other included American Indian/Alaskan Native, unknown (did not report any race), and other race and ethnicity (not specified in more granular detail by Medicare).
Figure. Association Between Myopia and Primary Open-Angle Glaucoma (POAG), Stratified by Racial and Ethnic Groups.
Adjusted odds ratio (OR) estimates for the association between myopia and POAG among all elderly California Medicare beneficiaries stratified by racial/ethnic groups. The OR estimates were obtained from multivariable logistic regression models adjusted for age, sex, race and ethnicity, and Charlson Comorbidity Index for all participants and adjusted for age, sex, and Charlson Comorbidity Index for each racial and ethnic group. The horizontal bars represent the 95% CIs of OR estimates.
aOther included American Indian/Alaskan Native, unknown (did not report any race), and other race and ethnicity (not specified in more granular detail by Medicare).
Table 3 describes results of analyses examining the association between myopia severity and POAG. Beneficiaries with myopia without degenerative changes had increased adjusted odds of POAG compared with beneficiaries without myopia (OR, 2.13; 95% CI, 2.07-2.19), as did beneficiaries with degenerative myopia compared with beneficiaries without myopia (OR, 4.19; 95% CI, 3.79-4.41). The P value for the statistical interaction term of myopia severity by race and ethnicity was P < .001. In multivariable models stratified by race and ethnicity, compared with beneficiaries without myopia, those with degenerative myopia consistently had greater odds of POAG although the magnitude of association did not significantly differ by race and ethnicity, as indicated by overlapping CIs. Conversely, myopia without degenerative changes vs no myopia was associated with greater odds of POAG among beneficiaries who identified as Black (OR, 2.49; 95% CI, 2.19-2.82) and Hispanic (OR, 3.15; 95% CI, 2.95-3.36) compared with beneficiaries who identified as White (OR, 1.84; 95% CI, 1.84-1.97).
Table 3. Dose-Response Association Between Myopia Severity and Primary Open-Angle Glaucoma, Stratified by Racial and Ethnic Group: Results From Logistic Regression Modelsa.
| Race and ethnicity | Myopia severity | Odds ratio (95% CI)a | 99% CI |
|---|---|---|---|
| All | Degenerative myopia | 4.19 (3.97-4.41) | 3.91-4.49 |
| Myopia | 2.13 (2.07-2.19) | 2.06-2.21 | |
| No myopia | 1 [Reference]b | [Reference]b | |
| Asian | Degenerative myopia | 3.92 (3.57-4.30) | 3.47-4.42 |
| Myopia | 2.14 (1.97-2.33) | 1.92-2.39 | |
| No myopia | 1 [Reference]b | [Reference]b | |
| Black | Degenerative myopia | 4.41 (2.95-6.95) | 2.60-7.47 |
| Myopia | 2.49 (2.19-2.82) | 2.10-2.94 | |
| No myopia | 1 [Reference]b | [Reference]b | |
| Hispanic | Degenerative myopia | 4.43 (3.74-5.24) | 3.55-5.52 |
| Myopia | 3.15 (2.95-3.36) | 2.89-3.43 | |
| No myopia | 1 [Reference]b | [Reference]b | |
| Non-Hispanic White | Degenerative myopia | 4.39 (4.08-4.73) | 3.98-4.84 |
| Myopia | 1.90 (1.84-1.97) | 1.82-1.99 | |
| No myopia | 1 [Reference]b | [Reference]b | |
| Otherc | Degenerative myopia | 3.73 (3.05-4.57) | 2.86-4.87 |
| Myopia | 2.14 (1.88-2.43) | 1.81-2.53 | |
| No myopia | 1 [Reference]b | [Reference]b |
Models examine odds of primary open-angle glaucoma comparing beneficiaries with either degenerative myopia or myopia to those without myopia, adjusted for age, sex, and Charlson Comorbidity Index score.
Reference for all odds ratios is no myopia.
Other included American Indian/Alaskan Native, unknown (did not report any race), and other race and ethnicity (not specified in more granular detail by Medicare).
Table 4 describes results from analyses of the association between myopia and POAG severity. Compared with beneficiaries without myopia, those with myopia had higher odds of all levels of POAG severity (other POAG: OR, 1.93; 95% CI, 1.86-2.01; mild POAG: OR, 2.22; 95% CI, 2.14-2.33; moderate POAG: OR, 2.32; 95% CI, 2.23-2.41; severe POAG: OR, 2.91; 95% CI, 2.78-3.05). The interaction term between myopia and race and ethnicity had a P value less than .001. In multivariable multinomial logistic regression models stratified by race and ethnicity, the magnitude of association was stronger for moderate and severe POAG for Asian (moderate POAG: OR, 2.64; 95% CI, 2.40-2.91; severe POAG: OR, 3.55; 95% CI, 3.18-3.96) and Hispanic (moderate POAG: OR, 2.84; 95% CI, 2.57-3.14; severe POAG: OR, 3.83; 95% CI, 3.45-4.26) beneficiaries compared with White beneficiaries (moderate POAG: OR, 2.12; 95% CI, 2.02-2.24; severe POAG: OR, 2.54; 95% CI, 2.39-2.71), and for severe POAG for Black beneficiaries (severe POAG: OR, 3.28; 95% CI, 2.70-4.00) compared with White beneficiaries.
Table 4. Association Between Myopia and Primary Open-Angle Glaucoma (POAG) Severity, Stratified by Racial and Ethnic Group: Results From Multinomial Logistic Regression Modelsa.
| Race/ethnicity | POAG severity | Odds ratio (95% CI)a | 99% CI |
|---|---|---|---|
| All | Severe | 2.91 (2.78-3.05) | 2.74-3.09 |
| Moderate | 2.32 (2.23-2.41) | 2.20-2.44 | |
| Mild | 2.22 (2.14-2.33) | 2.11-2.36 | |
| Other | 1.93 (1.86-2.01) | 1.83-2.04 | |
| No POAG | 1 [Reference]b | [Reference]b | |
| Asian | Severe | 3.55 (3.18-3.96) | 3.07-4.10 |
| Moderate | 2.64 (2.40-2.91) | 2.33-3.00 | |
| Mild | 2.31 (2.06-2.59) | 1.99-2.69 | |
| Other | 1.77 (1.61-1.95) | 1.56-2.02 | |
| No POAG | 1 [Reference]b | [Reference]b | |
| Black | Severe | 3.32 (2.73-4.03) | 2.57-4.28 |
| Moderate | 2.29 (1.88-2.80) | 1.77-2.98 | |
| Mild | 2.36 (1.84-3.02) | 1.71-3.26 | |
| Other | 2.34 (1.92-2.85) | 1.81-3.03 | |
| No POAG | 1 [Reference]b | [Reference]b | |
| Hispanic | Severe | 3.83 (3.45-4.26) | 3.33-4.40 |
| Moderate | 2.84 (2.57-3.14) | 2.49-3.24 | |
| Mild | 4.09 (3.70-4.53) | 3.58-4.67 | |
| Other | 2.07 (1.85-2.32) | 1.78-2.41 | |
| No POAG | 1 [Reference]b | [Reference]b | |
| Non-Hispanic White | Severe | 2.54 (2.39-2.71) | 2.34-2.76 |
| Moderate | 2.12 (2.02-2.24) | 1.99-2.27 | |
| Mild | 1.93 (1.82-2.03) | 1.79-2.07 | |
| Other | 1.92 (1.82-2.03) | 1.79-2.06 | |
| No POAG | 1 [Reference]b | [Reference]b | |
| Otherc | Severe | 2.54 (2.04-3.18) | 1.90-3.41 |
| Moderate | 2.55 (2.15-3.04) | 2.03-3.21 | |
| Mild | 2.27 (1.89-2.74) | 1.78-2.90 | |
| Other | 2.06 (1.73-2.45) | 1.64-2.59 | |
| No POAG | 1 [Reference]b | [Reference]b |
Models examine odds of varying levels of POAG severity vs no glaucoma, comparing beneficiaries with any myopia to those without myopia, adjusted for age, sex, and Charlson Comorbidity Index score.
Reference for all odds ratios is no glaucoma.
Other included American Indian/Alaskan Native, unknown (did not report any race), and other race and ethnicity (not specified in more granular detail by Medicare).
Discussion
The findings of this study indicate that in the 2019 California Medicare population, beneficiaries with myopia had increased odds of receiving a diagnosis of POAG overall and also by racial and ethnic group, with greater strength of association in beneficiaries who identify as Asian, Black, and Hispanic compared with White beneficiaries. These results not only reinforce previous findings of increased risk of POAG in individuals with myopia, but also suggest that EMM exists by race and ethnicity in the association between myopia and POAG in the California Medicare population, where myopia may be a stronger risk factor for POAG in racial and ethnic minority groups.
Additional analyses showed associations between myopia and POAG at different levels of both myopia and POAG severity. The higher odds of POAG in those with degenerative myopia supports the mechanistic hypothesis described earlier involving structural changes in myopic eyes. Eyes with degenerative or pathologic myopia have greater axial lengths,22 thereby perhaps leading to greater deformities of the lamina cribrosa and weaker scleral support at the optic nerve, which may contribute to greater susceptibility of the optic nerve to glaucomatous damage.5,8,9 Furthermore, in the analysis of myopia and POAG severity, the association between myopia and POAG was strongest for severe POAG, which may suggest that myopia is a more potent risk factor for severe POAG than for mild/moderate POAG. While this association is true for all racial and ethnic groups, the strongest evidence for this finding is among Asian, Black, and Hispanic individuals. We show both an exposure-response association between increasing degrees of the myopia exposure (no myopia vs myopia vs degenerative myopia) and the binary POAG outcome as well as a potency association in a separate analysis of the binary myopia exposure and increasing severity of the POAG outcome (no POAG vs mild, moderate, severe, and other).
Most of the current literature on the association between myopia and POAG has shown that myopia is a risk factor for POAG across multiple racial and ethnic groups.10,11,12,13,14,23,24,25,26,27,28 Notably, a systematic review and meta-analysis on the association between myopia and POAG by Wu et al28 showed a positive association of risk estimates with an OR of 2.45 (95% CI, 1.80-3.33). Additionally, a dose-response meta-analysis by Ha et al23 showed that the association is even stronger between moderate to high myopia and POAG, with a pooled OR of 4.14 (95% CI, 2.57-6.68). However, there are a few prior studies that have reported a null or even negative association between myopia and POAG, particularly in select racial and ethnic groups of small sample sizes.10,27,28,29 Nonetheless, the results of our present study are consistent with the existing literature suggesting evidence for a positive association between myopia and POAG.
Notably, there is currently minimal literature about the association between myopia and POAG among specific racial and ethnic groups. The present study fills that gap by showing that patients with myopia from racial and ethnic minority groups in a large and diverse population have increased odds of POAG and particularly, severe POAG. There are several factors that may explain these findings. Studies have demonstrated that Black and Hispanic individuals may have thinner corneas, which is a known risk factor for glaucoma, and that Hispanic patients may be more susceptible to vascular problems, such as severely high blood pressures and poor optic nerve flow, which can both contribute to glaucomatous optic nerve damage.30,31 While some of these factors may reflect anatomic differences between racial and ethnic groups, it is important to note that race is a sociopolitical construct and not a genetic/biologic determinant of disease.32 Additionally, in racial and ethnic minority groups, studies have demonstrated lower rates of glaucoma testing relative to disease burden, inconsistent follow-up for glaucoma, underuse of low-vision devices, and surgical undertreatment of individuals with glaucoma.33,34,35,36 While these results do not imply a causative mechanism, they may serve as a foundation to consider more frequent or earlier glaucoma screening for racial and ethnic minority individuals with myopia and also merit further investigation of potential racial and ethnic disparities in the diagnosis of glaucoma among patients with myopia.
Limitations
There are several limitations of this study. Given the nature of the Medicare claims data, we defined myopia by its respective ICD-10-CM diagnosis codes and were thereby restricted to 2 levels of myopia: (1) myopia and (2) degenerative myopia, rather than looking at the specific diopters of refractive error. Furthermore, there is a possibility of misclassification with the use of administrative claims data. Beneficiaries with refractive errors of any type are included in Medicare, but whether these diagnoses are captured depends on clinican coding. Our observed myopia prevalence of 2.2% is substantially less than the 14.8% found in those aged 65 to 74 years in the Beaver Dam Eye Study,37 probably due to undercoding during encounters. Myopia is likely further underreported in patients with other primary diagnoses, such as POAG, or in older adults who may have pseudophakia or long-time spectacle users. The diagnosis of POAG can be more difficult in patients with high myopia who have anomalous optic discs and visual field loss not associated with glaucoma.38 Nevertheless, our observed POAG prevalence of 6.3% in the entire California Medicare population 65 years and older is similar to the pooled estimate of 6.0% (95% CI, 4.7%-7.6%) from a subgroup meta-analysis of 19 estimates for adults aged 70 to 79 years.39 There exists the possibility of diagnostic suspicion or detection bias, which results from conditioning on factors caused by disease and that affect diagnosis of the disease.40 Eye care utilization may represent such a factor, with individuals with myopia having greater care utilization, leading to more cases of diagnosed POAG. However, several population-based studies24,25,27,41,42 that are not subject to detection bias, given that testing is performed uniformly for the entire population, have reported similar estimates for the association between myopia and POAG, tempering concerns about the magnitude of possible detection bias in our study. Additionally, this study was cross-sectional, rather than longitudinal in design; thus, temporality between myopia and POAG cannot be established. In most cases, however, myopia has a childhood onset while POAG develops later in life; therefore, one can plausibly assume that myopia occurs before POAG.14,43,44,45 Finally, uncontrolled confounding may persist, particularly from socioeconomic status variables that are unavailable in Medicare data.
Conclusions
In summary, this study found an association between myopia and increased odds of POAG in the 2019 California Medicare population, which is consistent with findings from previous population-based studies. Our findings suggest EMM by race and ethnicity on the association between myopia and POAG, with higher risk of POAG and possibly severe POAG in beneficiaries with myopia from Asian, Black, and Hispanic individuals. Further research is needed to examine these associations and address any potential health or health care disparities that may contribute to these observations and to consider earlier or more frequent screening for individuals from racial and ethnic minority backgrounds.
eTable 1. ICD-10-CM Codes for POAG Severity
eTable 2. ICD-10-CM Codes for Myopia Severity
eTable 3. ICD-10-CM Codes for CCI Comorbidities
Data sharing statement
References
- 1.Myopia: a close look at efforts to turn back a growing problem. National Eye Institute . Published October 3, 2017. Accessed December 6, 2022. https://www.nei.nih.gov/about/news-and-events/news/myopia-close-look-efforts-turn-back-growing-problem
- 2.Young TL, Metlapally R, Shay AE. Complex trait genetics of refractive error. Arch Ophthalmol. 2007;125(1):38-48. doi: 10.1001/archopht.125.1.38 [DOI] [PubMed] [Google Scholar]
- 3.Bullimore MA, Ritchey ER, Shah S, Leveziel N, Bourne RRA, Flitcroft DI. The risks and benefits of myopia control. Ophthalmology. 2021;128(11):1561-1579. doi: 10.1016/j.ophtha.2021.04.032 [DOI] [PubMed] [Google Scholar]
- 4.Haarman AEG, Enthoven CA, Tideman JWL, Tedja MS, Verhoeven VJM, Klaver CCW. The complications of myopia: a review and meta-analysis. Invest Ophthalmol Vis Sci. 2020;61(4):49. doi: 10.1167/iovs.61.4.49 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Marcus MW, de Vries MM, Junoy Montolio FG, Jansonius NM. Myopia as a risk factor for open-angle glaucoma: a systematic review and meta-analysis. Ophthalmology. 2011;118(10):1989-1994.e2. doi: 10.1016/j.ophtha.2011.03.012 [DOI] [PubMed] [Google Scholar]
- 6.Bertaud S, Aragno V, Baudouin C, Labbé A. [Primary open-angle glaucoma]. Rev Med Interne. 2019;40(7):445-452. doi: 10.1016/j.revmed.2018.12.001 [DOI] [PubMed] [Google Scholar]
- 7.Types of glaucoma. National Eye Institute . Accessed December 6, 2022. https://www.nei.nih.gov/learn-about-eye-health/eye-conditions-and-diseases/glaucoma/types-glaucoma
- 8.Fong DS, Epstein DL, Allingham RR. Glaucoma and myopia: are they related? Int Ophthalmol Clin. 1990;30(3):215-218. doi: 10.1097/00004397-199030030-00009 [DOI] [PubMed] [Google Scholar]
- 9.Scott R, Grosvenor T. Structural model for emmetropic and myopic eyes. Ophthalmic Physiol Opt. 1993;13(1):41-47. doi: 10.1111/j.1475-1313.1993.tb00424.x [DOI] [PubMed] [Google Scholar]
- 10.Chen SJ, Lu P, Zhang WF, Lu JH. High myopia as a risk factor in primary open angle glaucoma. Int J Ophthalmol. 2012;5(6):750-753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Chiang SY, Weng TH, Lin CM, Lin SM. Ethnic disparity in prevalence and associated risk factors of myopia in adolescents. J Formos Med Assoc. 2020;119(1 Pt 1):134-143. doi: 10.1016/j.jfma.2019.03.004 [DOI] [PubMed] [Google Scholar]
- 12.The Collaborative Longitudinal Evaluation of Ethnicity and Refractive Error (CLEERE) Study. ClinicalTrials.gov . Accessed December 6, 2022. https://clinicaltrials.gov/ct2/show/NCT00000169
- 13.Luong TQ, Shu YH, Modjtahedi BS, et al. Racial and ethnic differences in myopia progression in a large, diverse cohort of pediatric patients. Invest Ophthalmol Vis Sci. 2020;61(13):20. doi: 10.1167/iovs.61.13.20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Vajaranant TS, Wu S, Torres M, Varma R. The changing face of primary open-angle glaucoma in the United States: demographic and geographic changes from 2011 to 2050. Am J Ophthalmol. 2012;154(2):303-314.e3. doi: 10.1016/j.ajo.2012.02.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Corraini P, Olsen M, Pedersen L, Dekkers OM, Vandenbroucke JP. Effect modification, interaction and mediation: an overview of theoretical insights for clinical investigators. Clin Epidemiol. 2017;9:331-338. doi: 10.2147/CLEP.S129728 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Master Beneficiary Summary File (MBSF) LDS. Centers for Medicare & Medicaid Services . Accessed December 6, 2022. https://www.cms.gov/Research-Statistics-Data-and-Systems/Files-for-Order/LimitedDataSets/MBSF-LDS
- 17.Standard Analytical Files (Medicare Claims): LDS. Centers for Medicare & Medicaid Services . Accessed December 6, 2022. https://www.cms.gov/Research-Statistics-Data-and-Systems/Files-for-Order/LimitedDataSets/StandardAnalyticalFiles
- 18.Waldo DR. Accuracy and bias of race/ethnicity codes in the Medicare enrollment database. Health Care Financ Rev. 2004;26(2):61-72. [PMC free article] [PubMed] [Google Scholar]
- 19.Inaccuracies in Medicare's race and ethnicity data hinder the ability to assess health disparities. US Department of Health and Human Services Office of Inspector General . Published June 15, 2022. Accessed March 20, 2023. https://oig.hhs.gov/oei/reports/OEI-02-21-00100.asp
- 20.How do I change the sex identification on my Social Security record? Social Security Administration . Accessed February 14, 2023. https://www.ssa.gov/people/lgbtq/gender-identity.html
- 21.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373-383. doi: 10.1016/0021-9681(87)90171-8 [DOI] [PubMed] [Google Scholar]
- 22.Ohno-Matsui K. What is the fundamental nature of pathologic myopia? Retina. 2017;37(6):1043-1048. doi: 10.1097/IAE.0000000000001348 [DOI] [PubMed] [Google Scholar]
- 23.Ha A, Kim CY, Shim SR, Chang IB, Kim YK. Degree of myopia and glaucoma risk: a dose-response meta-analysis. Am J Ophthalmol. 2022;236:107-119. doi: 10.1016/j.ajo.2021.10.007 [DOI] [PubMed] [Google Scholar]
- 24.Mitchell P, Hourihan F, Sandbach J, Wang JJ. The relationship between glaucoma and myopia: the Blue Mountains Eye Study. Ophthalmology. 1999;106(10):2010-2015. doi: 10.1016/S0161-6420(99)90416-5 [DOI] [PubMed] [Google Scholar]
- 25.Sommer A, Tielsch JM. Risk factors for open-angle glaucoma: the Barbados Eye Study. Arch Ophthalmol. 1996;114(2):235. doi: 10.1001/archopht.1996.01100130229029 [DOI] [PubMed] [Google Scholar]
- 26.Wang YX, Yang H, Wei CC, Xu L, Wei WB, Jonas JB. High myopia as risk factor for the 10-year incidence of open-angle glaucoma in the Beijing Eye Study. Br J Ophthalmol. Published online February 22, 2022. doi: 10.1136/bjophthalmol-2021-320644 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wong TY, Klein BE, Klein R, Knudtson M, Lee KE. Refractive errors, intraocular pressure, and glaucoma in a white population. Ophthalmology. 2003;110(1):211-217. doi: 10.1016/S0161-6420(02)01260-5 [DOI] [PubMed] [Google Scholar]
- 28.Wu J, Hao J, Du Y, et al. The association between myopia and primary open-angle glaucoma: a systematic review and meta-analysis. Ophthalmic Res. 2022;65(4):387-397. doi: 10.1159/000520468 [DOI] [PubMed] [Google Scholar]
- 29.Daniel E, Addis V, Maguire MG, et al. Prevalence and factors associated with optic disc tilt in the Primary Open-Angle African American Glaucoma Genetics Study. Ophthalmol Glaucoma. 2022;5(5):544-553. doi: 10.1016/j.ogla.2022.02.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Francis BA, Varma R, Chopra V, Lai MY, Shtir C, Azen SP; Los Angeles Latino Eye Study Group . Intraocular pressure, central corneal thickness, and prevalence of open-angle glaucoma: the Los Angeles Latino Eye Study. Am J Ophthalmol. 2008;146(5):741-746. doi: 10.1016/j.ajo.2008.05.048 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Zangwill LM, Ayyagari R, Liebmann JM, et al. ; African Descent and Glaucoma Evaluation Study III Genomics Study Group . The African Descent and Glaucoma Evaluation Study (ADAGES) III: Contribution of Genotype to Glaucoma Phenotype in African Americans: study design and baseline data. Ophthalmology. 2019;126(1):156-170. doi: 10.1016/j.ophtha.2017.11.031 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Witzig R. The medicalization of race: scientific legitimization of a flawed social construct. Ann Intern Med. 1996;125(8):675-679. doi: 10.7326/0003-4819-125-8-199610150-00008 [DOI] [PubMed] [Google Scholar]
- 33.Coleman AL, Yu F, Rowe S. Visual field testing in glaucoma Medicare beneficiaries before surgery. Ophthalmology. 2005;112(3):401-406. doi: 10.1016/j.ophtha.2004.09.034 [DOI] [PubMed] [Google Scholar]
- 34.Elam AR, Tseng VL, Rodriguez TM, Mike EV, Warren AK, Coleman AL; American Academy of Ophthalmology Taskforce on Disparities in Eye Care . Disparities in vision health and eye care. Ophthalmology. 2022;129(10):e89-e113. doi: 10.1016/j.ophtha.2022.07.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Siegfried CJ, Shui YB. Racial disparities in glaucoma: from epidemiology to pathophysiology. Mo Med. 2022;119(1):49-54. [PMC free article] [PubMed] [Google Scholar]
- 36.Wang F, Javitt JC, Tielsch JM. Racial variations in treatment for glaucoma and cataract among Medicare recipients. Ophthalmic Epidemiol. 1997;4(2):89-100. doi: 10.3109/09286589709057101 [DOI] [PubMed] [Google Scholar]
- 37.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]
- 38.Han JC, Lee EJ, Kim SH, Kee C. Visual field progression pattern associated with optic disc tilt morphology in myopic open-angle glaucoma. Am J Ophthalmol. 2016;169:33-45. doi: 10.1016/j.ajo.2016.06.005 [DOI] [PubMed] [Google Scholar]
- 39.Zhang N, Wang J, Li Y, Jiang B. Prevalence of primary open angle glaucoma in the last 20 years: a meta-analysis and systematic review. Sci Rep. 2021;11(1):13762. doi: 10.1038/s41598-021-92971-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Snoep JD, Morabia A, Hernández-Díaz S, Hernán MA, Vandenbroucke JP. Commentary: a structural approach to Berkson’s fallacy and a guide to a history of opinions about it. Int J Epidemiol. 2014;43(2):515-521. doi: 10.1093/ije/dyu026 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Perera SA, Wong TY, Tay WT, Foster PJ, Saw SM, Aung T. Refractive error, axial dimensions, and primary open-angle glaucoma: the Singapore Malay Eye Study. Arch Ophthalmol. 2010;128(7):900-905. doi: 10.1001/archophthalmol.2010.125 [DOI] [PubMed] [Google Scholar]
- 42.Xu L, Wang Y, Wang S, Wang Y, Jonas JB. High myopia and glaucoma susceptibility the Beijing Eye Study. Ophthalmology. 2007;114(2):216-220. doi: 10.1016/j.ophtha.2006.06.050 [DOI] [PubMed] [Google Scholar]
- 43.Cumberland PM, Peckham CS, Rahi JS. Inferring myopia over the lifecourse from uncorrected distance visual acuity in childhood. Br J Ophthalmol. 2007;91(2):151-153. doi: 10.1136/bjo.2006.102277 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Jones LA, Mitchell GL, Mutti DO, Hayes JR, Moeschberger ML, Zadnik K. Comparison of ocular component growth curves among refractive error groups in children. Invest Ophthalmol Vis Sci. 2005;46(7):2317-2327. doi: 10.1167/iovs.04-0945 [DOI] [PubMed] [Google Scholar]
- 45.Morgan IG, Ohno-Matsui K, Saw SM. Myopia. Lancet. 2012;379(9827):1739-1748. doi: 10.1016/S0140-6736(12)60272-4 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. ICD-10-CM Codes for POAG Severity
eTable 2. ICD-10-CM Codes for Myopia Severity
eTable 3. ICD-10-CM Codes for CCI Comorbidities
Data sharing statement
