Identifying, Understanding, and Addressing Disparities in Glaucoma Care in the United States

Abstract

Glaucoma is the leading cause of irreversible blindness worldwide, currently affecting around 80 million people. Glaucoma prevalence is rapidly rising in the United States due to an aging population. Despite recent advances in the diagnosis and treatment of glaucoma, significant disparities persist in disease detection, management, and outcomes among the diverse patient populations of the United States. Research on disparities is critical to identifying, understanding, and addressing societal and healthcare inequalities. Disparities research is especially important and impactful in the context of irreversible diseases such as glaucoma, where earlier detection and intervention are the primary approach to improving patient outcomes. In this article, we first review recent studies identifying disparities in glaucoma care that affect patient populations based on race, age, and gender. We then review studies elucidating and furthering our understanding of modifiable factors that contribute to these inequities, including socioeconomic status (particularly age and education), insurance product, and geographic region. Finally, we present work proposing potential strategies addressing disparities in glaucoma care, including teleophthalmology and artificial intelligence. We also discuss the presence of non-modifiable factors that contribute to differences in glaucoma burden and can confound the detection of glaucoma disparities.

Translational Relevance: By recognizing underlying causes and proposing potential solutions, healthcare providers, policymakers, and other stakeholders can work collaboratively to reduce the burden of glaucoma and improve visual health and clinical outcomes in vulnerable patient populations.

Introduction

Health disparities are defined as preventable differences in the burden of disease that are linked with social, economic, and/or environmental disadvantage. These disparities result from a wide range of underlying factors, including health insurance coverage and affordability, access to and utilization of care, and quality of care. Many health disparities are rooted in social and economic inequalities that lead to unequal distribution of resources and opportunities. Health disparities are prevalent in all medical disciplines but are commonly discussed in the context of chronic disease, such as diabetes, hypertension, and heart disease, as well as disability, cancer, substance use, infant and maternal mortality, and overall life expectancy.¹ Disparities in clinical care can magnify an underlying genetic predisposition for eye disease, such as the higher incidence of diabetic retinopathy and glaucoma among Black and Hispanic individuals compared to non-Hispanic White individuals.² 

Health disparities are closely interconnected to social determinants of health (SDOHs), defined by the US Department of Health and Social Services as “conditions in the environment where people are born, live, work, play, worship, age, and thrive that affect a wide range of health, functioning, and quality-of-life outcomes and risks.”³ Factors such as income, education, employment, housing, and access to healthcare services are all social determinants that can influence access to quality health care, susceptibility to illness, and ability to manage health conditions. In an investigation of differences in health outcomes on a county level, such as length and quality of life, clinical care accounted for 20% of the variation in outcomes, whereas SDOHs accounted for as much as 50%.¹ People's abilities to understand their health status and navigate health information are strongly influenced by socioeconomic status (SES), education, access to resources, and social support networks. Historically, lower health literacy has contributed to “suboptimal use of preventive services, delays in diagnosis, higher rates of hospitalization, and increased risk of mortality among adults.”^4,5 People's perceptions of their own health status also play a role in their health outcomes. Although self-reported health is generally correlated with actual health status, some research has demonstrated that a disconnect between the two can lead to a reduction in participation in follow-up care, adherence to medications, and lifestyle modifications.⁶ Furthermore, the patient–physician relationship is essential for optimal, equitable health care. Implicit or explicit provider biases associated with SDOHs, however, can compromise this relationship and deepen patient mistrust of medical care.⁷ 

Glaucoma is the leading cause of irreversible vision loss worldwide, currently affecting 80 million people total and 3 million people in the United States.⁸ The global prevalence of glaucoma among people 40 to 80 years old is predicted to rise rapidly from 76 million in 2020 to 111.8 million in 2040 due to aging of the population.⁹ Glaucoma costs the US healthcare system an estimated $2.5 billion annually, with $1.9 billion in direct costs and $0.6 billion in indirect costs.¹⁰ Glaucoma often serves as a representative disease for disparities research and healthcare models, given its broad impact and chronic nature. The disease course involves longitudinal care requiring regular clinical exams, diagnostic testing, and medical and surgical interventions, including intraocular pressure (IOP)-lowering eye drops, laser treatment, and glaucoma surgery. Studying glaucoma in this context may provide a global perspective on eyecare access and utilization, which are key factors that directly contribute to patient outcomes and healthcare disparities. 

Vision loss caused by glaucoma is irreversible but largely preventable, making early detection and management critical. This begs the question of why up to 8.9% individuals with primary open-angle glaucoma (POAG) and up to 27% with primary angle-closure glaucoma (PACG)¹¹ are affected by blindness despite screening efforts and effective medical and surgical interventions. Many studies have focused on identifying SDOHs that contribute to disparities in glaucoma care and disease outcomes. This article synthesizes recent research on disparities in glaucoma care and disease outcomes and highlights promising methods and approaches taken to address these disparities. 

Identifying Disparities in Glaucoma

Racial Disparities in Glaucoma Screening, Treatment, and Outcomes

Before reviewing glaucoma disparities among different racial groups, it is important to discuss the role of race as a social construct rather than biological or genetic attribute.¹² Most studies cited in this paper defined race based on participants’ self-reported identification, which is primarily based on the sociopolitical category with which they identify rather than their genetic background and heritage. These two definitions of race—as a sociopolitical category or marker of ancestry—are closely intertwined in the scientific literature, rendering the interpretation of racial associations with glaucoma outcomes inherently complex. It is possible that the use of more precise variables, such as genetic variants, will become more accessible in the future for studying intrinsic differences in glaucoma pathogenesis and risk. However, it makes sense to maintain race as a sociopolitical category when conducting disparities research, as this definition of race is likely associated with SDOHs that play key roles in glaucoma outcomes. 

Through this lens, it is widely recognized that Black, Hispanic, and Asian Americans experience a higher prevalence of glaucoma when compared to non-Hispanic White Americans.^13–15 Although a single, collective study assessing glaucoma prevalence across all races and ethnicities in the United States has yet to be conducted, multiple studies have compared glaucoma prevalence in minority races and ethnicities with those of non-Hispanic White individuals. The Baltimore Eye Study, for example, reported that Black participants have around a fourfold higher prevalence (4.97%) of POAG than age-matched non-Hispanic White participants (1.44%).¹³ The Los Angeles Latino Eye Study (LALES), a population-based epidemiology study of Hispanic individuals in Los Angeles, reported a similar prevalence of POAG in Latinos (4.74%).¹⁶ In addition, the prevalence of PACG and normotensive glaucoma was found to be higher in individuals of Asian descent.^9,14 Researchers have turned to genome-wide association studies to investigate the genetic underpinnings of these differences in glaucoma prevalence. Through these studies, a multitude of ancestry-specific genetic variants have been associated with glaucoma risk.¹⁷ Although these genetic associations are too numerous to include all of them in this review, these findings suggest that there are biological differences that contribute to race-associated glaucoma phenotypes and prevalence differences. For example, a specific variant at the APBB2 locus is associated with African ancestry and increased risk of POAG, correlating with the higher prevalence of POAG observed in this population.¹⁸ 

It is important to differentiate between intrinsic biological differences in glaucoma prevalence associated with race and ethnicity and extrinsic sociopolitical factors that contribute to disparities in disease burden and outcomes among racial minorities. Examples of early identified disparities include Black Americans having a higher risk of blindness at first glaucoma diagnosis compared to non-Hispanic White Americans.^19–21 In addition, Black and Hispanic patients are more likely to need glaucoma surgery or laser treatment at the time of diagnosis.^22,23 These disparities in visual morbidity and clinical outcomes may be related to disparities in rates of glaucoma detection. For example, LALES reported that 75% of participants with POAG were previously undiagnosed, a rate that is higher than rates reported in other US populations.¹⁶ This finding is consistent with results of the Proyecto VER project. In that study,²⁴ the rate of undiagnosed glaucoma was 62% among the Hispanic participants, which was higher than the 50% estimated occurrence among individuals of African and European ancestry.²⁵ Finally, Black Americans with anatomical narrow angles are more likely to go undetected until PACG develops, even after correcting for socioeconomic factors.²⁶ 

Another factor hindering racial equity in glaucoma diagnosis and treatment is access to care. Discussion of this problem is nuanced, as socioeconomic issues are inherently associated with race in the United States. When compared to non-Hispanic White Medicare beneficiaries, Hispanic Medicare beneficiaries attended fewer outpatient visits and received fewer optical coherence tomography (OCT) retinal nerve fiber layer (RNFL) tests, but they had more inpatient/emergency department encounters and selective laser trabeculoplasty procedures.²³ Similarly, Black Medicare beneficiaries attended fewer outpatient visits and received fewer visual field (VF) tests, but had more inpatient/ED encounters and surgeries.^23,27 The differences between non-Hispanic Black and White Medicare beneficiaries persisted after correcting for SES, which was defined by the low-income indicators that determine different levels of Medicare eligibility.²³ Furthermore, non-Hispanic Black and Hispanic patients reported greater difficulty affording glaucoma medications than non-Hispanic White patients and were less likely to adhere to medication regimens.^28,29 It is also crucial to recognize that both overt and implicit biases of providers against racial minorities have impacted the patient–physician relationship and led to an entrenched mistrust in the healthcare system. As a result, patients of historically marginalized groups may have reduced rates of seeking and utilizing health care, influencing differences in visual outcomes among various racial and ethnic groups. 

The issue of racial disparities in glaucoma extends to the realm of scientific research and manifests as a lack of diversity in glaucoma studies. For example, in a meta-analysis that examined the demographics of 105 POAG studies conducted between 1994 and 2019, 103 of these studies consisted predominantly of non-Hispanic White participants.³⁰ Black and Hispanic populations, despite experiencing a greater disease burden, were significantly underrepresented in glaucoma clinical trials, comprising only 16.8% and 3.4% of participants, respectively, whereas non-Hispanic White individuals accounted for the majority at 70.7%.³⁰ 

Age

Age is an intrinsic risk factor for many eye diseases, including glaucoma. The Centers for Disease Control and Prevention (CDC) has stated that all people over the age of 60 and specifically Black Americans over age 40 are considered at high risk for developing glaucoma, thus highlighting differences in glaucoma risk at the intersection of race and age.³¹ Black Americans were also found to have the highest POAG prevalence at all ages, with 4% and 13% prevalence in Black Americans ages 50 to 59 years and 80 to 89 years, respectively.¹³ This pattern of increased prevalence with age is seen across all races, although to different degrees. Overall glaucoma prevalence, regardless of race or gender, has been found to increase with age, ranging from 0.6% in the age category of 40 to 49 years to 8.3% in the 80+ age category.³² This elevated prevalence associated with age can be attributed to various factors, such as accelerated RNFL thinning.³³ Consequently, glaucoma may develop at lower IOP levels in older patients.³³ The confounding effect of age-related disease risk poses a challenge when investigating age-related disparities in ocular health. The prevalence of PACG also increases with older age, likely related to cataract formation and subsequent angle narrowing and closure.³⁴ However, treatments for other age-related conditions, such as cataracts, can further confound detection of age-related disparities in glaucoma care. For example, cataract surgery is strongly protective against PACG; therefore, the higher risk of PACG with older age is counterbalanced by the protective effect of cataract surgery.³⁴ 

Beyond conferring a higher risk of glaucoma, older age has been identified as a risk factor for lower participation in glaucoma screening and adherence to treatment. Patients who are younger, male, and live in an urban area are more likely to receive VF tests and more likely to receive them frequently, defined as greater than one VF test per year, which is compliant with recommendations by the American Academy of Ophthalmology.³⁵ Similarly, 23% of patients 65 years or older have been found to be non-adherent (based on prescription filing and number of days without glaucoma therapy) to newly prescribed topical glaucoma agents.³⁶ When categorizing adherence patterns, those who were “never adherent” to their POAG medications had the highest average age at diagnosis at 64.9 years old.³⁷ The reasons for this drop in adherence with older age are multifactorial. Generally, medication adherence is inversely proportional to the number of medications.³⁸ Older patients with chronic medical conditions are likely to have other medications to manage and therefore may be less likely to adhere to a daily glaucoma treatment regimen in the setting of polypharmacy.^38,39 Cognitive impairment, associated with increasing age, affects memory and executive function and can contribute to misunderstanding of medication regimens and subsequent non-adherence.⁴⁰ 

Age-related differences in health have also been analyzed in the context of frailty. Frailty, defined as a “state of increased vulnerability,” is a concept that helps quantify the overall impact of disease in older individuals.²³ Studies have found that higher frailty levels are associated with older age, increased chronic conditions, and lower SES.⁴¹ Frailty is also heavily intertwined with ocular health; visual impairment can influence fall risk, inpatient hospitalization, and the need for additional support with activities of daily living.²³ Studies found that Medicare recipients with higher levels of frailty had fewer outpatient visits, less glaucoma testing, and lower rates of surgical glaucoma treatment than non-frail or pre-frail recipients.²³ Thus, moderately to severely frail patients are less likely to be seen regularly in person for glaucoma and may present for acute inpatient rather than routine outpatient care. 

Gender

Similar to age, gender differences in glaucoma can be attributed to a combination of intrinsic, non-preventable risk factors and extrinsic, preventable social or environmental factors associated with disease. Before exploring these disparities, it is important to distinguish between biological sex and gender. Given that our aim is to evaluate modifiable factors that may influence glaucoma burden within a social context, we are considering gender in terms of the sociopolitical category with which one identifies. Studies have attempted to explore differences in the pathophysiology underlying glaucoma between males and females. For example, female sex has been associated with faster rates of ganglion cell complex thinning over time.⁴² Other research in mice models has found that estrogen could have a protective effect on retinal ganglion cell death in vivo.⁴³ However, basic science research has yet to produce definitive results that are clearly translatable to glaucoma risk or progression in humans. 

It is well established that women are at higher risk of PACG than men due to anatomic differences.⁴⁴ However, evidence supporting the role of gender as a risk factor for POAG is mixed, even within specific racial and ethnic groups.⁴⁵ For example, LALES reported a 1.73 times higher risk for POAG in Hispanic males,¹⁶ whereas Proyecto VER found no gender difference among Hispanic patients with POAG in Tucson and Nogales, Arizona.²⁴ In contrast, in another study, the prevalence of diagnosed POAG in women was higher than in men even after adjusting for factors such as age, race, and ethnicity.⁴⁶ It is important to note that, given the role of older age as a glaucoma risk factor and given that women tend to live longer than men, the unadjusted prevalence of glaucoma tends to be higher in women. 

An important contributor to gender-related disparities in glaucoma care and outcomes is that women are more likely to seek health care than men.⁴⁷ As a result, male patients may present with disease at a more severe stage due to lower utilization of screening and preventive care earlier on in the disease course; however, young men living in urban areas are more likely to receive VF testing and at a higher frequency than young women.³⁴ 

The gender differences extend beyond discrepancies in testing and eyecare visits. After adjusting for the number of eyecare visits in a year and the type of provider seen, men were more likely than women to be diagnosed with POAG.⁴⁶ Given variations in gender differences across studies, it may be useful to focus on understanding the underlying effects of gender on healthcare-seeking behaviors. 

Understanding Disparities in Glaucoma Care

Socioeconomic Status and Education

Socioeconomic status has been widely accepted as a determinant of health outcomes, with lower economic opportunities being linked to poorer health.⁴⁸ This unfortunate reality is complicated and multifactorial in origin and is propagated by economic, physical, and sociopolitical forces such as profit-driven healthcare systems, public health disparities, neighborhood segregation, and food insecurity. Although it falls outside the scope of this review to delve deeply into the underlying causes of health inequality in the United States it is important to provide a brief discussion on key socioeconomic variables, such as income and education, that significantly impact individual health outcomes. 

Household income alone has been correlated with health outcomes. In general, it is accepted that people's class is a mediator of potential longevity, with income levels having a positive correlation with life expectancy.⁴⁹ This is because one's income is a primary determinant of one's ability to access care and the quality of care one receives.³⁷ Therefore, it is not surprising that an association between health expectancy and income exists even when correcting for unhealthy behavior, such as smoking,^50,51 which suggests that income alone acts as a modifier of health outcomes. With regard to glaucoma treatment and income specifically, lower annual income levels (defined as less than $60,000/year) are associated with poor adherence to treatment regimens.³⁷ A study in India found that low-income individuals spend nearly 26% of their income on glaucoma treatment.⁵² The ability to afford transportation is also inherently tied to income. The most common reason for eligible individuals not showing up for a free eye clinic appointment was a lack of transportation.⁵³ A similar study, which assessed factors preventing follow-up for patients enrolled in the Philadelphia Telemedicine and Glaucoma Detection and Follow-up Study, found that the main reasons for missed appointments were feeling ill (38.1%), forgetting the appointment (34.2%), lack of transportation (13.5%), and inability to miss work (7.1%).⁵⁴ 

Education plays a similarly important role in shaping health outcomes. People who have completed at least a high school education exhibit higher utilization of eyecare services compared to those without this academic background.⁵⁵ In contrast, lower education rates are commonly associated with lower SES, less robust health infrastructure, and lower levels of health literacy.^56,57 Beyond education level, physicians are significantly less likely to educate Black patients about their glaucoma than non-Black patients.⁵⁸ This is particularly concerning given that Black patients are more likely to have severe glaucoma and VF defects compared to other racial populations.^59,60 Thus, progressing toward equal educational effort and opportunities for all patients regardless of age, gender, and race will be essential for addressing health disparities at the individual level rather than at the community or institutional level. In the Medication Adherence in Glaucoma to Improve Care (MAGIC) trial, individuals with the highest adherence to topical glaucoma medications endorsed greater knowledge about the disease as being a facilitating factor for their adherence, demonstrating how vital this physician–patient interaction is in promoting better outcomes.⁶¹ Other factors such as lower income and older age have also been implicated in contributing to poor health literacy rates.^62–64 As a result, health literacy is recognized as a critical determinant of health and is a target of interventions designed to improve the health education, adherence, and outcomes of patients. 

These same mediators (namely, access to care and education/health literacy) are also responsible for propagating glaucoma disparities. Prior research has linked lower education with a higher risk of visual impairment from cataracts, macular degeneration, and glaucoma.⁶⁵ Poor health literacy has also been associated with poor medication adherence among glaucoma patients.^21,66 Similarly, poor comprehension of the extent or severity of one's glaucomatous disease has been associated with poor follow-up.^53,54 In the Philadelphia Telemedicine and Glaucoma Detection and Follow-up Study, 29.2% of the participants who missed their follow-up appointments reported being unaware of either their diagnosis of glaucoma or its severity.⁵⁴ 

These factors help explain why individuals of lower SES have higher rates of glaucoma. Demographic data from the Michigan Screening and Intervention for Glaucoma and Eye Health Through Telemedicine (MI-SIGHT) program found that the area deprivation index—or the composite measure of neighborhood deprivation based on an individual's address—corresponded with higher levels of positive glaucoma screening tests.⁶⁷ Similarly, individuals with glaucoma and lower SES were less likely to have seen a physician within 12 months.⁶⁸ This pattern is not limited to the United States. In the United Kingdom, rates of glaucoma were highest (2.4%) among those with the lowest annual income and decreased as income increased, with the lowest rate of glaucoma (0.9%) being reported in the highest income category.⁶⁹ This discrepancy in income implies that underlying, non-modifiable factors are at play, such as race, age, or gender-related biases. Lower socioeconomic scores were also associated with later detection of glaucoma.⁷⁰ Ultimately, affordability, continuity, and regular sources of follow-up are essential in ensuring access to eyecare services.⁷⁰ 

Insurance Product Disparities

Insurance plays a critical role in a patient's quality of and access to eyecare. Medicaid status correlates with other socioeconomic factors, such as financial status, housing, education, assets, and occupation. As a result, Medicaid can serve as a surrogate for assessing an individual's SES when economic status, such as income or net worth, is unavailable.⁷¹ Patients with Medicaid are 2.34 times more likely to not receive any glaucoma testing in the 15 months following their first glaucoma diagnosis compared to patients with commercial health insurance.⁷¹ In addition, almost half (48.6%) of Medicaid recipients with POAG did not receive VF and/or OCT testing, compared to only 21.5% of commercial health insurance recipients.⁷¹ Among glaucoma suspects identified at free health fairs for underserved communities in South Florida, those with health insurance were 1.74 times more likely seek follow-up care compared to those without insurance, adjusted for age, gender, race/ethnicity, and education level.⁷² Out of those lost to follow-up, 57% of participants cited a lack of insurance as the primary reason,⁷² highlighting the vital role insurance plays in ensuring access to health care. Additionally, Medicaid or self-paying patients incur significantly higher total and glaucoma-related costs in the first year of diagnosis compared to patients with commercial insurance.⁷³ This financial burden is a major issue, particularly for Medicaid recipients who are typically representative of lower SES. 

Regional Disparities

Density and type of providers can contribute to regional variations in the quality of eye care and glaucoma outcomes. A nationwide analysis found that counties in the South have the lowest eyecare provider availability even when adjusting for population density.⁷⁴ This may explain why patients in Southern and Pacific regions are less likely to be detected with anatomical narrow angles prior to developing PACG.²⁶ Meanwhile, the incidence and prevalence of glaucoma based on healthcare claims data are higher in New England and Mid-Atlantic regions and lower in the East South Central and Mountain Regions, even after controlling for factors such as race/ethnicity, age, access to care, number and type of providers, and number of eyecare visits.⁴⁶ Such findings suggest that provider density does not entirely explain observed regional differences in glaucoma; overdiagnosis or underdiagnosis by providers and their respective “gold standards” of diagnosis in specific regions may also play a role.⁴⁶ 

The nature of the community that a patient lives in, such as urban, suburban, or rural, also affects that patient's ability to access glaucoma care. Living in an urban area is associated with an increased likelihood and increased frequency of receiving VF tests.³⁵ Transportation presents a greater barrier in non-urban communities with lower access to affordable, convenient public transportation. A study in Florida found that only 30.5% of inhabitants live within 15 minutes of ophthalmologists who are members of the American Glaucoma Society, contributing to a significant travel burden for people over the age of 65 to reach one.⁷⁵ 

Addressing Disparities in Glaucoma Care

Telemedicine and Online Health Education

Disparities exist across all levels of glaucoma care in the United States; therefore, promoting equity and pioneering solutions to overcome healthcare barriers are of the utmost importance. The most recent stance in 2022 by the US Preventive Services Task Force (USPSTF) maintains that there is insufficient evidence to evaluate the benefits versus harms of population-based glaucoma screenings in asymptomatic adult patients.⁷⁶ However, earlier detection of glaucoma facilitates earlier intervention, prevention of disease progression, and better quality of life and life expectancy.^77–81 Although the lack of ample research such as randomized controlled trials comparing screened and unscreened populations contributes to the current uncertainty about the benefit of glaucoma screening for the general public, early screening for at-risk populations may provide a possible solution to mitigate disparities. 

Telemedicine, a healthcare modality that has been gaining traction over the past decade, especially following the onset of COVID-19,^82–87 provides an alternative approach to glaucoma care delivery and screening at-risk populations. The American Telemedicine Association defines telemedicine as the use of communication technologies to improve patient health outcomes, increase access to health information, and obtain care.⁸⁸ One example is the MI-SIGHT program, which aims to provide telemedicine-based glaucoma screening by partnering primary care–based community clinics (Hope Clinic and Hamilton Community Health Network in Michigan) with ophthalmologists from the University of Michigan.⁸⁹ Ophthalmic technicians at community-based clinics perform glaucoma diagnostic testing on patients and send the results via electronic health records to ophthalmologists, who then provide recommendations for appropriate eye care. Patients diagnosed with glaucoma or identified as glaucoma suspects are also enrolled in a randomized controlled trial of personalized glaucoma coaching in which they receive motivational interviewing (a conversational style or technique that encourages patients to change their behaviors to benefit their health), education on glaucoma, and guidance on creating a question list for future ophthalmology appointments. 

Implementation of patient coaching and empowerment is imperative to improve screening outcomes, given that follow-up appointments after initial detection are crucial for continual health management. For example, Black patients tend to ask fewer questions during appointments than non-Hispanic White patients,⁹⁰ which can be rectified by creating question prompt lists to promote active participation during appointments.⁹¹ Therefore, addressing disparities requires not only increased access to glaucoma screening but also educating patients on how to take full advantage of this benefit. After 1 year of operation, the MI-SIGHT program screened over 1000 patients for glaucoma, visual impairment, and diabetic retinopathy, with an overall patient satisfaction rating of 99%.⁹² It remains to be seen whether this model of free, individualized eye care and education can improve glaucoma disparities in underserved populations relative to standard methods of care delivery.⁹³ Another telemedicine-based program, Alabama Screening and Intervention for Glaucoma and Eye Health Through Telemedicine (AL-SIGHT), has adopted a similar approach by providing glaucoma screening through federally qualified health centers in rural counties of Alabama for patients regardless of insurance status.⁹⁴ The program provides patients with IOP values > 30 mmHg with a referral to an eyecare provider within 2 weeks. After a 6-week trial, there was a 56% improvement in glaucoma knowledge and 9% improvement in attitudes toward frequent follow-up with ophthalmologists.⁹⁵ Similarly, the New York Screening and Intervention for Glaucoma and Eye Health Through Telemedicine (NYC-SIGHT) program directly brings eye health screening into the underserved regions of Washington Heights and Harlem.⁹⁶ This program addresses barriers to eye care by providing free screenings and remote image reading by ophthalmologists, in addition to setting up appropriate follow-up appointments based on those reads.⁹⁶ After 15 months, 66% of participants whose screening demonstrated a visual acuity of 20/40 or worse, IOP of 23 to 29 mmHg, or an unreadable fundus image were referred to ophthalmology for follow-up appointments, and 20% were diagnosed as glaucoma suspects or with manifest glaucoma.⁹⁷ Moving forward, utilizing telemedicine and free community-based health screenings^98–101 to bridge disparities in underserved or rural regions of the United States may significantly improve glaucoma outcomes and overall public understanding of this disease. 

Health education delivered through the Internet can also enhance understanding about glaucoma and adherence to treatments. For example, one study reported that Black patients under the age of 70 would prefer to have online glaucoma educational programs (with topics such as explaining glaucoma, its implications, and the role of medications) led by ophthalmologists rather than ophthalmologic technicians or pharmacists as an option to learn about their glaucoma.¹⁰¹ However, an important consideration when providing Internet-based educational materials is that they need to be interpretable by a wide demographic. Online glaucoma materials are commonly written at the 10th- to 11th-grade reading level, which is markedly higher than the American Medical Association's recommendation to compose patient education materials at or below the seventh-grade reading level.¹⁰² This becomes more relevant given that patients who use outside educational sources or feel empowered to play an active role in their own care have better adherence to medications.¹⁰³ As a result, making glaucoma education not only accessible online but also easier to understand can encourage better understanding and compliance with glaucoma treatments. For individuals who do not have readily available access to Internet-based educational modalities, free health education in the local community is also a viable option. When asked about where glaucoma educational programs should be offered, nearly 40% of respondents voted for programs at public community centers, senior citizen centers, or on television, and nearly 30% requested to have programs at local churches.¹⁰¹ Collectively, implementing both online and offline educational programs either virtually or in person within accessible local communities may reach a much broader audience. 

When discussing these solutions to alleviate disparities, it is important to acknowledge the financial costs and logistical aspects that are involved in implementing these programs. The cost of implementation is not a negligible amount; for example, the MI-SIGHT program costs over $100,000 per clinic for the facility, provider and technicians, and diagnostic equipment. Although these costs may seem high in isolation, these programs are actually cost effective when considered in the context of cost per case and patient volume.⁸⁹ By strategically targeting higher volume, under-resourced communities, these programs can deliver care to those who face the highest levels of poverty and greatest need for care.⁸⁹ At the same time, glaucoma screening must be conducted in a deliberate and appropriate manner. The USPSTF states that the benefit of glaucoma screening is not clear⁷⁶; therefore, screening should remain mindful of costs and the financial burden associated with false-positive tests, especially in resource-limited safety-net healthcare settings that are unable to sustain non-productive financial burdens. 

Artificial Intelligence

Artificial intelligence (AI), a technology that simulates intelligent behavior and analyzes data, can be utilized within the medical field to diagnose or manage diseases efficiently and can potentially address disparities in access to care and health outcomes.^104,105 Ophthalmology is a medical specialty that relies on manual evaluation of image-based data to guide patient care; therefore, the potential for AI implementation within ophthalmology is high.¹⁰⁶ In fact, numerous studies have demonstrated the utility of AI in diagnosing, detecting, and monitoring the progression of ocular disease, including glaucoma and diabetic retinopathy, based on fundus and optic disc images.^107–112 With this context, it is evident that AI offers a tremendous opportunity to improve glaucoma care. When this technology is eventually integrated into standard care, such as diagnostic devices and electronic medical record systems, it will be possible to more efficiently replicate time- and cost-intensive tasks for which we currently rely on human providers. Overall, there is a large potential for this technology to bridge economic gaps and reach a wider audience.^113,114 The virtual nature of telemedicine also opens additional avenues to address eyecare access disparities using AI.¹¹⁵ For example, predictive AI models can evaluate VF tests to diagnose glaucoma with increased sensitivity and equivalent specificity compared with physicians and could ultimately be used to support or even automate the glaucoma decision-making process.¹¹⁶ 

Despite its immense promise, implementation of AI in clinical practice faces several challenges. First, AI models are trained on datasets derived from specific patient populations that may differ from those of a provider’s end-users; overgeneralizing AI models can produce biased recommendations and conclusions that perpetuate or even exacerbate healthcare disparities.¹¹⁷ Second, AI models are also not immune to errors in interpreting data and providing accurate recommendations¹¹⁸; therefore, providers should be mindful and continue to exercise sound clinical judgment when adopting novel AI tools. Third, each healthcare system differs in its capacity and resources to accommodate newly detected glaucoma patients; therefore, model sensitivity should be balanced by the need to ensure timely care for more urgent patients. Finally, issues beyond the AI program itself, such as physician and patient technology literacy, legal liability, and overall public confidence in AI, will all have to be thoroughly addressed.^113,117,119 Although AI has great potential to revolutionize glaucoma care, there are many technical and ethical questions that must first be addressed. 

Conclusions

The current literature has identified consistent disparities in glaucoma care that are evident across race, age, and gender. Although underlying genetic factors may determine intrinsic disease risk, the social context of these factors strongly influences glaucoma detection and management, thereby impacting patient outcomes. These factors are also closely interwoven with SES, insurance product, and geographic region, ultimately shaping the disease course and trajectory. It is necessary to gain a holistic understanding of existing disparities in order to delineate solutions to bridge existing gaps in glaucoma care. Proposed interventions, including teleglaucoma and community-based screening initiatives, are focused on increasing access to eye care to facilitate glaucoma detection and improve health education. In the future, AI may provide an affordable and accessible solution to glaucoma detection and monitoring, but the technology also carries an inherent risk of bias. Ultimately, all proposed solutions to address glaucoma disparities will benefit from promoting patient-centered care and fostering the patient–physician relationship. We hope improved awareness about glaucoma disparities will motivate healthcare providers, policymakers, and other stakeholders to work collaboratively on reducing disease burden and improving clinical outcomes in vulnerable patient populations. 

Acknowledgments

Supported by a grant from the National Eye Institute, National Institutes of Health (K23 EY029763) and by an unrestricted grant to the Department of Ophthalmology from Research to Prevent Blindness. 

Disclosure: S.S. Davuluru, None; A.T. Jess, None; J.S.B. Kim, None; K. Yoo, None; V. Nguyen, None; B.Y. Xu, None