Of 10 033 participants 40 to 80 years old, 362 were diagnosed as having glaucoma, of which 273 had primary glaucoma.
Glaucoma severity increases with age. A mean (SD) of 49.0% (14.0%) (n = 24) of adults 50 to 59 years old had moderate glaucomatous visual field loss or worse in at least 1 eye.
eTable. Characteristics of Participants With Undiagnosed Glaucoma by Ethnicity
Chua J, Baskaran M, Ong PG, Zheng Y, Wong TY, Aung T, Cheng C. Prevalence, Risk Factors, and Visual Features of Undiagnosed GlaucomaThe Singapore Epidemiology of Eye Diseases Study. JAMA Ophthalmol. 2015;133(8):938–946. doi:10.1001/jamaophthalmol.2015.1478
Copyright 2015 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
Glaucoma can remain asymptomatic until a severe stage, resulting in a high prevalence of undiagnosed glaucoma worldwide. Asia accounts for 60% of the world’s total glaucoma cases. To our knowledge, no epidemiological studies have assessed ethnic differences in undiagnosed glaucoma among various Asian subgroups.
To determine the prevalence of, risk factors for, and visual features of undiagnosed primary glaucoma in a multiethnic Asian population.
Design, Setting, and Participants
The Singapore Epidemiology of Eye Diseases Study is a population-based trial in which 3353 Chinese (2009-2011), 3280 Malays (2004-2006), and 3400 Indians (2007-2009) aged 40 to 80 years were invited for an eye examination, including visual field assessment, to establish glaucoma diagnosis. Participants with undiagnosed glaucoma (ie, answering no to whether they previously had been told by a physician that they had glaucoma, not using glaucoma medication, or not having undergone glaucoma surgery) were identified.
Main Outcomes and Measures
Prevalence of, risk factors for, and visual features of undiagnosed glaucoma.
Of 272 participants with primary glaucoma, 196 (72.1%) were previously undiagnosed. The overall prevalence of undiagnosed primary glaucoma was highest among Malays (2.65%; 95% CI, 2.10%-3.31%), followed by Chinese (1.51%; 95% CI, 1.13%-2.01%) and Indians (0.97%; 95% CI, 0.64%-1.43%). In multivariable analysis, variables associated with higher risk of undiagnosed glaucoma were younger age (odds ratio [OR], 1.04; 95% CI, 1.00-1.09; P = .04), Malay ethnicity (OR, 3.65; 95% CI, 1.31-10.13; P = .01), presence of primary open-angle glaucoma (OR, 3.82; 95% CI, 1.60-9.14; P = .003), absence of yearly eyeglass checks (OR, 9.29; 95% CI, 3.43-25.21; P < .001), and lack of cataract surgery (OR, 4.19; 95% CI, 1.68-10.48; P < .001). No patients were blind in both eyes. A mean (SD) of 4.1% (2.8%) (n = 8) of the newly diagnosed patients were blind in 1 eye, and a mean (SD) of 56.0% (7.2%) (n = 102) had noteworthy visual field damage (mean deviation worse than −6 dB) in at least 1 eye.
Conclusions and Relevance
The prevalence of undiagnosed primary glaucoma varied among ethnic populations in whom a mean (SD) of 49.0% (14.0%) (n = 24) of affected individuals 50 to 59 years old already had clinically significant visual field loss. Such data may assist policymakers in implementing cost-effective public health interventions to reduce the effect of blindness associated with undiagnosed glaucoma.
Glaucoma is the leading cause of blindness, affecting 64 million people worldwide.1 Because glaucoma is asymptomatic in the early stages, it is often undetected until extensive visual field (VF) loss. Glaucoma blindness is largely preventable if detected and treated earlier. Significant proportions of undiagnosed glaucoma have been reported.2- 5 Undiagnosed glaucoma is associated with risk factors such as younger age,2 less frequent eye care visits,2 eye care provided by optometrists instead of ophthalmologists,3,4 male sex,5 and ethnicity.6 Variation in undiagnosed glaucoma also has been reported by ethnicities and by countries.2,7- 17 Asia accounts for 60% of the world’s total glaucoma cases.1 From a public health perspective, it is important to know the prevalence, features, and predictors of undiagnosed glaucoma in Asians. Studies indicate that Asians in developing countries, such as China7- 12 and India,9,11,12,18 have higher rates of undiagnosed glaucoma than individuals of white race/ethnicity in developed countries.2,13- 17
Singapore is an urbanized state consisting of immigrants of Chinese, Malaysian, and Indian ancestries, where patients with suspected eye conditions can be referred from community clinics for ophthalmological services that are part of the government health subsidy plan, Medisave.19 The proportions of undiagnosed glaucoma among Malays20 and Indians21 living in Singapore have been reported to be as high as 90% and 70%, respectively. To our knowledge, the age-specific prevalence and severity of undiagnosed glaucoma and its associated risk factors among Chinese, Malays, and Indians have not been reported. It is unclear if ethnic variations reflect only the access to eye care in different countries or other factors related to awareness of disease, such as cultural characteristics. The objective of this study was to determine the prevalence of undiagnosed primary glaucoma and its risk factors and visual features among a multiethnic population in Singapore.
This cross-sectional study examines ethnic differences in undiagnosed glaucoma among Chinese, Malay, and Indian participants living in Singapore.
The risk factors for undiagnosed glaucoma are younger age, presence of primary open-angle glaucoma, absence of yearly glasses checks, lack of cataract surgery, and Malay ethnicity.
Half of the participants 40 to 80 years old with undiagnosed glaucoma already had visual field loss (mean deviation worse than −6 dB).
The Singapore Epidemiology of Eye Diseases Study is a population-based trial that included the 3 major ethnic groups in Singapore aged 40 to 80 years: Chinese (2009-2011), Malays (2004-2006), and Indians (2007-2009).22,23 An age-stratified (by 10-year age groups) random sampling in each ethnic group was used to select 4606 Chinese, 4168 Malays, and 4497 Indians who were 40 to 80 years old. Of these, 3353 Chinese, 3280 Malays, and 3400 Indians participated. The overall response rate was 75.6%. Participants were older than nonparticipants (P < .001), but there was no sex difference (P = .68). This study adhered to the Declaration of Helsinki, ethics approval was obtained from the Singapore Eye Research Institute Institutional Review Board, and written informed consent was obtained from all participants.
Each participant underwent a standard ophthalmological examination, including visual acuity (VA) and subjective refraction, slitlamp biomicroscopy, gonioscopy, intraocular pressure measurement using Goldmann applanation tonometry, and fundoscopy.20,21,24 After pupil dilation, the optic disc was evaluated using a +78-diopter (D) lens at ×16 magnification, with a measuring graticule. Suspected glaucoma was defined as the presence of any of the following: (1) intraocular pressure exceeding 21 mm Hg; (2) vertical cup-disc ratio exceeding 0.6 or vertical cup-disc ratio asymmetry exceeding 0.2; (3) anterior segment deposit consistent with pseudoexfoliation or pigment dispersion syndrome; (4) morphological features, such as disc hemorrhage, notching of the neuroretinal rim, and defects of the retinal nerve fiber layer; (5) narrow anterior chamber angle during gonioscopy; or (6) known history of glaucoma.
Automated perimetry (SITA 24-2, Humphrey VF Analyzer II; Carl Zeiss Meditec, Inc) was performed in all individuals with suspected glaucoma. Visual field testing was repeated if the test reliability was not satisfactory (fixation loss >20%, false-positive rate >33%, or false-negative rate >33%). Visual field testing was also repeated if a glaucomatous VF defect was confirmed on 2 consecutive VF examinations (defined as the presence of ≥3 [P < .05] nonedge contiguous points with at least 1 at the P < .01 level on the same side of the horizontal meridian in the pattern deviation plot and classified as outside normal limits on the Glaucoma Hemifield Test).
Detailed interviewer-administered questionnaires were used to collect demographic (age, sex, and ethnicity) and socioeconomic (educational level, individual monthly income, and housing type) characteristics and medical and ocular history.22,23 Ethnicities were defined by the Singapore census and as indicated on the National Registration Identity Card.25 All interviewers were bilingual, and participants were given a choice to be interviewed in English, Chinese, Malay, or Tamil. Educational level was categorized as primary or lower (≤6 years), secondary (7-10 years), or postsecondary (≥11 years, including university education). Individual monthly income was based on Singapore dollars (SGD) and was categorized in US dollars as low (SGD <$2000 [US <$1492]), middle (SGD $2000 to <$3000 [US $1492 to <$2237]), or (3) high (SGD ≥$3000 [US ≥$2237]). Housing type was categorized as small (public housing of 1-2 rooms), middle (public housing of 3-4 rooms), or (3) big (public or private housing of ≥5 rooms). Myopia was defined as a spherical equivalent (sphere plus half cylinder) exceeding −0.50 D. With regard to eye care use patterns, participants were asked if they used any form of visual aids, and those who answered yes were then asked how often they visited any optometrists, opticians, or ophthalmologists to check their eyeglasses or contact lenses. Clinical assessment of lens status was determined by slitlamp examination. Patients with any cataract surgery were defined as having lens extraction in either eye.
Systolic and diastolic blood pressure were measured twice using a digital automatic blood pressure monitor (Dinamap Pro Series DP110X-RW; GE Medical Systems Information Technologies, Inc). A third measurement was obtained if 2 previous systolic blood pressure readings differed by more than 10 mm Hg or if previous diastolic blood pressure readings differed by more than 5 mm Hg. Blood samples were collected for biochemistry analysis and DNA extraction. Diabetes mellitus was defined as a random glucose level of at least 200 mg/dL (to convert glucose level to millimoles per liter, multiply by 0.0555), diabetic medication use, or a physician diagnosis of diabetes mellitus. Hypertension was defined as a systolic blood pressure of at least 140 mm Hg or a diastolic blood pressure of at least 90 mm Hg, physician-diagnosed hypertension, or self-reported history of hypertension. Hyperlipidemia was defined as a total cholesterol level of at least 239 mg/dL (to convert cholesterol level to millimoles per liter, multiply by 0.0259) or as self-reported use of lipid-lowering drugs.
Glaucoma was diagnosed according to the International Society for Geographical and Epidemiological Ophthalmology scheme.26 Primary angle-closure glaucoma was defined as the presence of narrow angles and features of trabecular obstruction by peripheral anterior synechiae, elevated intraocular pressure, iris whorling, glaukomflecken lens opacities, or excessive pigment deposition on the trabecular surface. Participants with glaucoma and an open, normal drainage angle with no secondary pathologic conditions were defined as having primary open-angle glaucoma (POAG). Final identification, adjudication, and classification of glaucoma cases were reviewed by the senior researcher (T.A.) and by a glaucoma fellowship–trained ophthalmologist (M.B.).
The magnitude of VF damage was defined by the mean deviation value from the VF test. The stage of glaucoma was obtained from the worse eye and was categorized as early (at least −6.00 dB), moderate (−6.01 to −12.00 dB), advanced (−12.01 to −20.00 dB), severe (≤−20.01 dB), or end stage or blind based on a simplified version of the Bascom Palmer glaucoma staging system.27,28 Blindness was defined according to World Health Organization criteria29 as greater than 1.30 logMAR (worse than 20/400 Snellen) using best possible correction obtained with subjective refraction.
In the interview, participants were asked, “Have you ever been told by a doctor that you have glaucoma?” Patients were coded based on their response of yes, no, or do not know or if a response was unobtainable. History of glaucoma medication use was ascertained during the personal interview. Those who answered yes to the question were asked to show the interviewer the medication containers of all the products used. The interviewer entered the product names into a computer. Participants with glaucoma were identified as having undiagnosed glaucoma if they (1) responded no to the question and (2) were not using glaucoma medication and had not undergone any glaucoma-related laser treatment or surgery.
The normality of the data was assessed with the Shapiro-Wilk test. To compare characteristics of participants among groups, the Kruskal-Wallis test was used for nonnormally continuous variables, and the χ2 test or Fisher exact test was used for categorical variables. A prevalence estimate of undiagnosed glaucoma and its 95% CI were calculated and standardized to the 2010 Singapore census.30 Associations of demographics and medical and ocular factors with undiagnosed glaucoma were assessed using univariate and multivariable logistic regression models. In addition to age, sex, educational level, and individual monthly income, factors with P < .10 in the univariate model were included in the multivariable model. Data were analyzed with statistical software (STATA, version 12.1; StataCorp LP).
Of 10 033 participants 40 to 80 years old, 362 were diagnosed as having glaucoma, of which 273 had primary glaucoma (Figure 1). It was previously reported by our group that 112 Malays had primary glaucoma.20 However, 2 of us (M.B. and T.A.) reclassified 3 of 8 previously unclassifiable glaucoma cases as POAG, changing the total number of Malays with primary glaucoma to 115. We excluded 89 secondary glaucoma cases because these patients would be aware of their ocular condition.
Of 273 participants with primary glaucoma, we excluded 1 individual because he did not respond to the questionnaire. Among 272 who responded, 222 (81.6%) were unaware that they had glaucoma, representing 71.4% (75 of 105) of Chinese, 94.8% (109 of 115) of Malays, and 73.1% (38 of 52) of Indians.20,21 However, among 222 individuals who were unaware of having glaucoma, 26 were receiving topical glaucoma therapy. Therefore, we recategorized them as having diagnosed glaucoma because they were being treated for the disease. Previously, higher rates of undiagnosed glaucoma had been reported for Malays20 and Indians21 because the rates were based solely on self-reported glaucoma awareness. By considering both awareness (self-reported glaucoma) and treatment (presence or absence of glaucoma medication or any glaucoma-related laser treatment or surgery), undiagnosed primary glaucoma was present in 55.2% (58 of 105) of Chinese, 90.4% (104 of 115) of Malays, and 65.4% (34 of 52) of Indians.
The overall age and sex–standardized prevalence of undiagnosed glaucoma was 1.76% (95% CI, 1.51%-2.05%). These prevalences were 1.51% (95% CI, 1.13%-2.01%) among Chinese, 2.65% (95% CI, 2.10%-3.31%) among Malays, and 0.97% (95% CI, 0.64%-1.43%) among Indians (Table 1). The prevalence of undiagnosed glaucoma increased with age (P < .001 for trend). We also observed higher standardized prevalences for men than women among Chinese (2.0% vs 1.1%, P < .001) and among Indians (1.2% vs 0.7%, P < .001). However, there was no significant sex difference among Malays (0.1% difference in the age-standardized prevalence rate [P = .30]).
Characteristics of participants with diagnosed vs undiagnosed glaucoma are summarized in Table 2. The mean (SD) age of all the participants with glaucoma was 64.8 (10.6) years, where 39.0% were female and 67.8% had an educational level of primary or lower. Persons with undiagnosed glaucoma were younger, were primarily of Malay ethnicity, tended to have POAG or unilateral glaucoma, and had poorer eye care use patterns. In addition, the undiagnosed group had less VF damage (ie, better mean deviation values) in the worse eye than the diagnosed group (P = .01), while VA (as expected) showed no intergroup difference.
Characteristics of 196 participants with undiagnosed glaucoma among the 3 ethnic groups are summarized in the eTable in the Supplement. There were differences in educational level, housing type, and glaucoma subtype across ethnicities. No difference in the mean deviation (P = .96) or the VA (P = .22) was observed in the better eye among the ethnic groups.
In the regression analysis, adjusted for age and sex (model 1 in Table 3), younger age, Malay ethnicity, small housing type, presence of POAG or unilateral glaucoma, and poorer eye care use patterns were associated with undiagnosed glaucoma. In the multivariable model, the risk of undiagnosed glaucoma increased with younger age (odds ratio [OR], 1.04; 95% CI, 1.00-1.09; P = .04) and was 3-fold higher in participants 50 to 59 years old (OR, 2.97; 95% CI, 1.01-8.77; P = .049) compared with those 70 to 80 years old (P = .03 for trend). Participants with undiagnosed glaucoma tended to be of Malay ethnicity (OR, 3.65; 95% CI, 1.31-10.13; P = .01), have POAG (OR, 3.82; 95% CI, 1.60-9.14; P = .003), and receive no yearly eyeglass checks (OR, 9.29; 95% CI, 3.43-25.21; P < .001) and were less likely to have undergone cataract surgery (OR, 4.19; 95% CI, 1.68-10.48; P < .001) (model 2 in Table 3).
Of 196 participants with undiagnosed glaucoma, 130 (66.3%) had unilateral glaucoma. The proportions of participants with unilateral glaucoma were similar across ethnicities (62.1% of Chinese, 69.2% of Malays, and 64.7% of Indians [P = .64]) (eTable in the Supplement) and across age groups, ranging from 64.3% in the youngest group (40-49 years) to 67.7% in the oldest group (70-80 years).
Figure 2 shows the severity of undiagnosed glaucoma based on the worse eye stratified by age. In 92.9% (n = 182) of participants, the diagnosis was based on the presence of repeatable VF loss compatible with glaucoma. A mean (SD) of 4.1% (2.8%) (n = 8) of the newly diagnosed patients were blind in 1 eye as a result of end-stage glaucoma. None of them were blind in both eyes. Glaucoma severity in individuals with undiagnosed glaucoma increased with age (R = 3.40, P < .001). Visual field loss (mean deviation worse than −6 dB) was found in a mean (SD) of 33.4% (17.8%) (n = 9) of those 40 to 49 years old and in a mean (SD) of 76.4% (11.2%) (n = 42) of those 70 to 80 years old.
In this multiethnic Asian population of adults 40 to 80 years old, more than two-thirds of glaucoma cases were undiagnosed. Malays had the highest prevalence of undiagnosed glaucoma, whereas Indians had the lowest prevalence. The risk factors for undiagnosed glaucoma were younger age, presence of POAG, limited eye care use patterns (absence of yearly eyeglass checks and cataract surgery), and Malay ethnicity. A mean (SD) of 56.0% (7.2%) (n = 102) of participants with undiagnosed glaucoma had moderate or worse VF loss (mean deviation worse than −6 dB) in at least 1 eye.
Contrary to data indicating that Asians have higher rates of undiagnosed glaucoma than individuals of white race/ethnicity,31 we found that the prevalence varied among Asians. Chinese (55%) and Indians (65%) exhibit rates of undiagnosed glaucoma similar to those of white residents in industrialized countries.31 Population-based data on undiagnosed glaucoma indicate the rate to be as high as 90% for Indians living in India9,11,12,18 and for Chinese living in China,7- 12 whereas it is approximately 50% for white residents living in developed countries.2,13- 17 One might propose that improvements in health care services (ie, subsidized medical insurance or better accessibility) may decrease the prevalence of undiagnosed glaucoma among developing nations. Nevertheless, similar high proportions of undiagnosed glaucoma are reported among rural and urban Indians, even though more ophthalmologists are practicing in urban regions.12
The prevalence of undiagnosed glaucoma increases with age.6,17 Although the prevalence is highest among older participants, they are in fact at lower risk of being undiagnosed.2,6,32 This finding may be due to the fact that considerable visual damage may have occurred in older individuals, and hence they are more symptomatic.33 Furthermore, ophthalmologists may suspect glaucoma more frequently in older cohorts and be more proactive in conducting glaucoma checks. However, these findings are in contrast to a study by Topouzis et al,5 who found no age difference between undiagnosed and diagnosed glaucoma groups. The discrepancy may be attributed to their primary selection of patients aged 60 to 74 years old, omitting younger adults.
The higher proportion of men among the undiagnosed category herein is unexpected. Although such a sex difference was not detected in individuals of white race/ethnicity,6,17 it may originate from the known tendency of women to be more proactive in visiting their physicians for annual check-ups than men.34,35 Moreover, women were more likely to obtain an eye examination following announcements by the Glaucoma EyeCare Program.36
Primary angle-closure glaucoma is the major form of glaucoma in Asia.37 Although this form of glaucoma is often related to an acute, painful episode, most patients do not experience such an attack.38 Instead, they have an asymptomatic course with progressive loss of the VF, similar to that in patients with POAG.39 Our study shows that Asians with POAG have a significantly higher risk of being undiagnosed. Although the prevalence of primary angle-closure glaucoma is high in Asia, POAG continues to be a more significant health problem.
The association between better eye care use patterns (yearly eyeglass checks and cataract surgery) and glaucoma detection may be attributed to increased contact with ophthalmologists.2,5 However, participants herein who wore eyeglasses or who had myopia showed no increase in glaucoma detection, which could be explained by their wearing over-the-counter eyeglasses or alternatively having been examined for eyeglasses by opticians who are not licensed to perform comprehensive ocular assessment in Singapore.
Despite residence within the same health care system, ethnic disparities in undiagnosed glaucoma still exist. Ninety percent of Malays live with undiagnosed glaucoma, whereas the proportion is halved for Chinese and Indians. After adjusting for individual monthly income and educational level, Malays were 3.65 times more likely to have undiagnosed glaucoma than Chinese despite similar glaucoma prevalences among both groups.20,40 This finding may be related to their distinct cultural health beliefs, whereby most Malays would seek traditional healers for their illnesses before using Western medicine, thereby delaying clinical treatment.41 Unfortunately, eye care access data were unavailable, and this hypothesis must be left to future research.
Our study demonstrated that 66.3% of participants with undiagnosed glaucoma had unilateral glaucoma across ethnicities and age groups. It is expected that most undiagnosed glaucoma cases would be unilateral because a damaged VF in 1 eye will not necessarily affect the overall visual function of a patient.42 Therefore, disease in patients may remain undetected until bilateral visual disability occurs.33 It is unclear why a Swedish study17 showed that the proportion of patients with unilateral glaucoma decreased with age, ranging from 72% in those 60 to 64 years old to 58% in those 75 to 79 years old.
Clinic-based studies43,44 have indicated that patients with glaucoma who are at greatest risk of blindness had moderate VF loss at presentation. Our data show that a mean (SD) of 49.0% (14.0%) (n = 24) of Asians with undiagnosed glaucoma had at least moderate VF loss from 50 years onward and are at greatest risk of blindness.
Strengths of our study include a population-based sample of 3 ethnic groups and VF assessments. Patient-reported undiagnosed glaucoma is commonly used to assess an individual’s disease status.2,13- 17 However, these reports may be subject to bias (Figure 1). Our study provided a robust definition of undiagnosed glaucoma by supplementing patient report with the presence or absence of glaucoma treatment. The study also has limitations. While the ORs herein were statistically significant, the wide 95% CIs suggest that the associations observed should be interpreted with caution because there were few outcome events, whereas a larger study could generate a more precise estimate of the effect. We did not assess types of eye care providers (ie, ophthalmologists or optometrists) or explore whether yearly eyeglass checks entailed eye health examination. Furthermore, we could not examine family history of glaucoma because this information was not obtained for Malays and Indians.
To our knowledge, ours is the first epidemiological study to assess ethnic differences in undiagnosed glaucoma cases among Asian subgroups using the same method to collect data. In Asian countries without universal health care, the burden of glaucoma blindness may be even greater.
Submitted for Publication: January 22, 2015; final revision received March 25, 2015; accepted April 5, 2015.
Corresponding Author: Ching-Yu Cheng, MD, PhD, Singapore Eye Research Institute, Singapore National Eye Centre, 20 College Rd, The Academia, Level 6, Discovery Tower, Singapore 169856 (firstname.lastname@example.org).
Published Online: June 4, 2015. doi:10.1001/jamaophthalmol.2015.1478.
Author Contributions: Drs Chua and Cheng had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Zheng, Wong, Cheng.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Chua, Zheng, Cheng.
Critical revision of the manuscript for important intellectual content: All authors.
Obtained funding: Wong, Cheng.
Administrative, technical, or material support: All authors.
Study supervision: Cheng.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest, and none were reported.
Funding/Support: This study was funded by grants 0796/2003, IRG07nov013, IRG09nov014, STaR/0003/2008, and CG/SERI/2010 from the Singapore National Medical Research Council and by grants 08/1/35/19/550 and 09/1/35/19/616 from the Singapore Biomedical Research Council.
Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.