Increasing Incidence and Prevalence of Common Retinal Diseases in Retina Practices Across the United States

Introduction

Understanding the retinal disease distribution in the United States can provide insight into the relative burden of each of the major retinal diseases and inform practices and policies regarding disease prevention, screening, evaluation, and treatment. Numerous prior studies have examined the prevalence and incidence of retinal diseases across the globe.^1–9 These historical studies, however, were based on disparate populations across non-overlapping time periods or were single-site studies. This study uses a broader cross-section of data from retina specialty clinics to provide an updated estimate of the current incidence and prevalence of the foremost retinal diseases in the United States: age-related macular degeneration (AMD), diabetic retinopathy (DR), diabetic macular edema (DME), branch retinal vein occlusion (BRVO), and central retinal vein occlusion (CRVO).

AMD is a leading cause of blindness in the Western world and the leading cause of blindness in people older than 65 years of age.^2,10 Estimates of the incidence and prevalence of wet and dry AMD vary; one study of beneficiaries in a large U.S. managed care network estimated the prevalence of dry AMD in the U.S. to be 5% and wet AMD 0.76%.¹ A 2005 study of a Wisconsin cohort found a 15-year cumulative incidence of early AMD to be 14.3% and 3.1% for late AMD, characterized by the presence of wet AMD or geographic atrophy (GA).³

Ocular complications of diabetes, namely DR and DME, are the leading cause of blindness in the U.S. in people ages 20 to 74 years old.^4,11–13 DR is thought to affect 30% to 50% of people with diabetes, translating to approximately 4.2 million people worldwide.^5,13,14 DME, a manifestation of DR, is estimated to affect 4% of all Americans with diabetes who are 40 years old or older, or approximately 746,000 people.¹¹ Another study approximates that 500,000 people in the U.S. have clinically significant DME with an annual incidence of 75,000.⁵

Retinal vein occlusions (RVOs) are the second most common retinal vascular disorder and often lead to severe vision loss.^6,7,15 An estimated 14 to 19 million adults are affected by RVOs worldwide, with approximately 13.9 million affected by branch retinal vein occlusion (BRVO) and 2.5 million affected by central retinal vein occlusion (CRVO); however, there is significant variation in the prevalence rates of RVOs reported in various major population-based studies, with estimates ranging from 0.3% to 1.6%.^7,8 Differences in reported prevalence rates are likely due to numerous factors including the relatively small number of RVO patients in any single study, different study methodologies, variations in population age, and potential variations in RVO prevalence across different ethnic groups.⁶ Furthermore, many of these epidemiologic studies lump CRVO and BRVO together despite the fact that these two diseases have significantly different prevalence rates.^15,16

The heterogeneous estimates of the prevalence of retinal diseases in the U.S. from previous studies may reflect true heterogeneity in the underlying incidence of disease across different patient populations, geographic regions, and time periods, but they may also reflect variation in study methodology, including variation due to results drawn from small samples. In this study we offer an updated estimate of the incidence and prevalence of key retinal diseases in retina specialty clinics from 2014 to 2019 using an extensive national database.

Patients and Methods

This was a retrospective study of electronic medical records from the Vestrum Health Retina Database, a geographically diverse sample of retina providers in private retina practices across the United States. Vestrum Health Retina Database receives deanonymized data at both the patient and visit level, which is tracked visit-by-visit for the entirety of each patient's history at a given retina practice. In total, 3,086,791 distinct eyes were evaluated from 58 retina practices across the U.S., including approximately 300 retina specialists, from January 2014 to December 2019. The same 58 retina practices were included in each year of the 6-year analysis, with practices requiring at least 6 years of data to be included in this study. We assessed all eyes that carried a diagnosis code (ICD-9: January 2014 – September 2015; ICD-10: October 15 – December 2019) for wet or dry AMD, DME, DR, BRVO, or CRVO. Patients with GA were included within the dry AMD group. Since DME is a manifestation of DR, patients were grouped based on those who had a diagnosis of DME (and therefore also have DR) and those who had an exclusive diagnosis of DR (DR without DME). Although the ICD-9 codes were not laterality-specific, physicians had to attach the code to one of the eyes in the database.

Patients with a diagnosis of dry AMD or GA who were later coded as wet AMD were counted as a “new diagnosis” under the wet AMD cohort, but patients initially coded as wet AMD who were later temporarily coded as dry AMD or GA remained categorized as part of the wet AMD cohort. Annual and total incidence were calculated as the distinct number of eyes first diagnosed with the associated condition, per year and during the 6-year period. Annual and total prevalence were calculated as the distinct number of eyes seen with an associated condition per year or during the 6-year period, as well as stratified by age and geographic region. Relative proportion of prevalence and incidence of these retinal diseases was also calculated as a proportion of all the eyes seen during the study period within these retina practices. Institutional Review Board (IRB) ruled that approval was not required for this study. The described research adhered to the tenets of the Declaration of Helsinki.

Results

A total of 3,086,791 eyes of 1,547,048 patients were examined. Figures 1 to 3 illustrate the annual incidence, annual prevalence, and total prevalence stratified by age for each of the three major disease categories based on the number of eyes: AMD (Figure 1), diabetic eye disease (Figure 2), and retinal vein occlusion (RVO) (Figure 3). Dry AMD had the largest incidence and prevalence (Figure 1B), followed by diabetic eye disease (Figures 2A and 2B). RVO had the lowest incidence (Figure 3A) and prevalence (Figure 3B). All diseases showed a steadily increasing annual incidence and prevalence, with the exception of dry AMD and wet AMD, which both showed a stabilization in prevalence from 2018 to 2019 (Figure 1B), and CRVO, which showed a stabilization in incidence from 2018 to 2019 (Figure 3A).

Figure 1. Incidence and prevalence of dry and wet age-related macular degeneration (AMD). (A) Annual incidence as distinct number of eyes seen per year in which the patient was new to the practice and first diagnosed with the associated condition. (B) Annual prevalence as distinct number of eyes with AMD seen in the given year. (C) Total prevalence as distinct number of eyes with AMD seen during the entire 6-year period, stratified by age group.

Figure 2. Incidence and prevalence of diabetic macular edema (DME) and diabetic retinopathy (DR) without DME. (A) Annual incidence as distinct number of eyes seen per year in which the patient was new to the practice and first diagnosed with the associated condition. (B) Annual prevalence as distinct number of eyes with diabetic eye disease seen in the given year. (C) Total prevalence as distinct number of eyes with diabetic eye disease seen during the entire 6-year period, stratified by age group.

Figure 3. Incidence and prevalence of branch retinal vein occlusion (BRVO) and central retinal vein occlusion (CRVO). (A) Annual incidence as distinct number of eyes seen per year in which the patient was new to the practice and first diagnosed with the associated condition. (B) Annual prevalence as distinct number of eyes with BRVO or CRVO seen in the given year. (C) Total prevalence as distinct number of eyes with RVO seen during the entire 6-year period, stratified by age group.

The prevalence of both wet and dry AMD generally increased with age, with a peak prevalence in patients 80 to 84 years old (dry AMD) and 85 to 89 years old (wet AMD) followed by a decline in the older age groups (Figure 1C). For diabetic eye disease, the 60-to-69-year-old age group had the largest number of cases, with only a small number of patients older than 89 years of age with DME or DR (Figure 2C). BRVO prevalence peaked in patients 70 to 74 years old and CRVO prevalence peaked in patients 75 to 79 years old (Figure 3C). Similar trends in prevalence by age group were seen when analyzed as the number of eyes (Figures 1C, 2C, and 3C) and when analyzed as the number of patients (Figure A, images A–C, available in the online version of this article).

Figure A. (A) Total prevalence as distinct number of patients with age-related macular degeneration (AMD) seen during the entire 6-year period. (B) Total prevalence as distinct number of patients with diabetic macular edema (DME) and diabetic retinopathy (DR) without DME seen during the entire 6-year period. (C) Total prevalence as distinct number of patients with branch retinal vein occlusion (BRVO) and central retinal vein occlusion (CRVO) seen during the entire 6-year period.

Figure B. (A) Average annual incidence as number of patients for the retinal diseases included in this study during the 6-year period. (B) Average annual prevalence as number of patients for the retinal diseases included in this study during the 6-year period.

Figure C. (A) Total incidence for each of the retinal diseases for the entire 6-year study period as a relative proportion of the total number of new patients examined by a retina specialist during the study period. (B) Total prevalence for each of the retinal diseases for the entire 6-year study period as a relative proportion of the total number of patients examined by a retina specialist during the study period.

Dry AMD had the largest average annual incidence in terms of number of eyes (57,698), followed by DME (31,511), DR without DME (29,895), wet AMD (27,175), BRVO (8,731), and lastly CRVO (5,960) (Figure 4A). Dry AMD also had the largest average annual prevalence (180,390), followed by wet AMD (145,838), DME (111,050), DR without DME (86,944), BRVO (28,880), and lastly CRVO (19,838) (Figure 4B). Similar trends were observed when analyzed by number of patients rather than number of eyes (Figure B, available in the online version of this article).

Figure 4. (A) Average annual incidence, in terms of number of eyes, for the retinal diseases included in this study during the 6-year period. (B) Average annual prevalence, in terms of number of eyes, for the retinal diseases included in this study during the 6-year period.

When total incidence and prevalence of eyes across the 6-year period were analyzed as relative proportions of all eyes seen in the retina practices, the diseases analyzed made up 41.5% of total incidence (Figure 5A) and 46.5% of total prevalent cases seen across the retina practices surveyed, with approximately half of patients being managed for other retina conditions (Figure 5B). However, when the total incidence and prevalence of patients across the 6-year period were analyzed as relative proportions of all patients seen in the retina practices, the diseases analyzed made up 54.3% of total incident cases and 61.0% of total prevalent cases seen in the retina practices surveyed (Figure C, available in the online version of this article).

Figure 5. (A) Total incidence for each of the retinal diseases for the entire 6-year study period as a relative proportion of the total number of new eyes examined by a retina specialist during the study period. (B) Total prevalence for each of the retinal diseases for the entire 6-year study period as a relative proportion of the total number of eyes examined by a retina specialist during the study period.

Most of the patients in our sample were concentrated in the Northeast (n = 441,186; 28.5%) and Southeast (n = 423,541; 27.4%) regions during the 6-year period. Each region showed the same general disease prevalence trends in terms of number of eyes (Figure 6) and number of patients (Figure D) and mirrored the prevalence patterns seen when the country was examined as a whole. Similar trends were seen when the geographic regions were examined on an individual year-by-year basis.

Figure 6. Total prevalence as number of eyes for each of the retinal diseases included in this study for the entire 6-year study period, stratified by geographic region. Number of retina specialists included in the study from each region is displayed below the x-axis.

Figure D. Total prevalence as number of patients for each of the retinal diseases included in this study for the entire 6-year study period, stratified by geographic region. Number of retina specialists included in the study from each region is displayed below the x-axis.

Discussion

Overall, AMD had the highest prevalence among the U.S. retinal practices analyzed, with a total prevalence of 784,922 eyes (25.4%) across the 6-year period (490,881 eyes with dry AMD; 294,041 eyes with wet AMD), followed by diabetic eye disease with a total prevalence of 525,393 eyes (17.0%) (270,703 eyes with DME; 254,690 eyes with DR only). AMD also had the highest incidence among the retinal practices analyzed, with a total incidence of 509,236 eyes (21.9%) across the 6-year period (346,186 eyes with dry AMD; 163,050 eyes with wet AMD), followed by diabetic eye disease with a total incidence of 368,438 eyes (15.8%) (189,068 eyes with DME; 179,370 eyes with DR only). The retinal diseases studied collectively made up a significant portion of total prevalent (41.6%) and incident (46.5%) cases by eye and the majority of total prevalent (61.0%) and incident (54.3%) cases by patient at the retina practices analyzed.

Each of the retinal diseases sampled in our study showed a steadily increasing annual incidence across the 6-year period. Since incidence and prevalence was calculated based off of the total eyes and visits seen in the given year, our analysis accounted for general growth of physicians and practices in the Vestrum Health Retina Database from 2014 to 2019. It is possible that the increasing annual incidences for the diseases surveyed are due to a growing population as a whole, increasing prevalence of risk factors and co-morbid conditions that can predispose patients to these retinal diseases, or increased access to care and higher rates of patient referral for specialized care at these retina practices. It was notable that the fastest rising incidence among the major retinal diagnoses was diabetic eye disease, including both DME and DR without DME, which could be due to the growing rates of diabetes in the U.S. population, resulting in a greater proportion of the population affected by diabetic eye disease.

This study uses current data and has the strength of having analyzed a large sample from diverse retina practices across the U.S. to provide a broad, updated epidemiologic analysis of retinal diseases in the U.S. at the patient and eye level. The external validity of this study is limited, however, by the fact that we do not know exactly how similar our sample is to the overall makeup of all retina practices in the U.S. or patients seen in academic centers, and may further be influenced by insurance patterns that may impact the information available in large databases. Furthermore, this study does not benefit from a lookback period to confirm that cases recorded as new diagnoses are truly incident cases; this type of review is difficult to do in a subspecialty database in which patients are, by definition, only enrolled after they carry a diagnosis code. Nonetheless, the breadth of our database across U.S. retina clinics nonetheless provides a valuable assessment of incidence and prevalence despite these limitations.

In conclusion, based on analysis of a geographically diverse database of retina clinics across the U.S., AMD is the most common diagnosis followed by diabetic eye disease. The retinal conditions studied made up the majority of the incident and prevalent cases for retinal diseases in the past six years, with less than half of patients being managed for other retina conditions. Over a 6-year period, the annual incidence and prevalence of all of the retinal diseases evaluated steadily increased. This analysis provides an updated evaluation of the current landscape of retinal diseases across the U.S. and contributes towards a better understanding of the potential trajectory of these diseases in the future, which will inform efforts towards retinal disease prevention and treatment.