Retinal vein occlusion (RVO) is a fairly common cause of vision loss, with an estimated prevalence between 0.3%1 and 1.6%.2 More recently, a pooled analysis from 11 studies estimated that 16.4 million adults were affected by RVO worldwide.3 Macular edema resulting from RVO is often the primary cause of impaired vision, with vascular endothelial growth factor (VEGF) playing a critical role in the pathogenesis. Multiple studies have demonstrated that intravitreal anti-VEGF injections improve visual outcomes.4–6 In the CRYSTAL study for macular edema due to central retinal vein occlusion (CRVO), ranibizumab (Lucentis; Genentech, South San Francisco, CA) therapy resulted in significant visual acuity (VA) gains at 12 months with a mean of 8.1 injections.7 Similarly, in the BRAVO and CRUISE trials, patients with RVO with monthly ranibizumab injections achieved improved vision-related function compared with patients who received sham through 6 months.8 Regular follow-up is imperative for many of these cases in order to optimize outcomes. However, to date, few studies have investigated the frequency of loss to follow-up (LTFU) in patients with RVO receiving intravitreal injections, as well as the associated demographics and risk factors. In this study, we sought to determine the proportion of patients who were LTFU in a large retina practice and identify potential associated factors.
Patients and Methods
A retrospective cohort study based on billing codes was initiated among all patients with macular edema secondary to RVO who received at least one intravitreal anti-VEGF injection at Wills Eye Hospital and the offices of Mid-Atlantic Retina, a large retina practice with multiple offices in Pennsylvania, New Jersey, and Delaware, from January 1, 2012, to January 1, 2017. A flowchart with the description of included and excluded patients is shown in Figure 1. This study was approved by the institutional review board at Wills Eye Hospital.
Flowchart for inclusion of patients with retinal vein occlusion (RVO) in final analysis. A total of 3,400 patients with macular edema due to RVO receiving intravitreal anti-vascular endothelial growth factor injections were eligible for final analysis after the application of the inclusion and exclusion criteria.
Medical diagnoses for RVO were based on International Classification of Disease, 9th or 10th Revision Clinical Modification (ICD-9/ICD-10). Current Procedural Terminology 4 (CPT4) codes were used for identifying intravitreal injections. Bevacizumab (Avastin; Genentech, South San Francisco, CA), ranibizumab, and aflibercept (Eylea; Regeneron, Tarrytown, NY) were identified using the medication-specific Healthcare Common Procedure Coding System codes. Only those initiating intravitreal anti-VEGF injections between January 1, 2012, to January 1, 2016, were included. Patients who were found to be deceased during follow-up or received corticosteroid injections were excluded. Repeated intravitreal injections were performed on a treat-and-extend or pro re nata basis at the discretion of the injecting physician.
Information from patient records, including race / ethnicity, age, gender, address, VA, date of each office visit, and date of each injection, was entered into an electronic database. Best available Snellen VA measurements were collected using distance spectacle correction or best pinhole correction when available. Only a subgroup of our patient cohort had recorded VA that could be utilized in the final analysis due to the fact that, during the study period, patient charts were transitioned to an electronic health record system. As a result, some of the older clinical data were no longer accessible. The interval between each injection and the next subsequent visit was calculated. The longest interval was used for assessment of follow-up. Regional average adjusted gross income (AGI) was calculated using the Internal Revenue Service's AGI database, which provides the average AGI per zip code across the United States.9 Patients' addresses were translated into a coordinate format using the Bing maps application program interface (Microsoft, Redmond, WA). Spherical distance from patient residence to the retina clinic was calculated using a Haversine formula.10,11
Definition of LTFU
Follow-up intervals for each patient were measured from the date of each intravitreal anti-VEGF treatment to the next subsequent visit. LTFU was defined as a subsequent visit occurring more than 12 months after an intravitreal injection or no further visits after the last intravitreal injection with this last injection occurring at least 12 months prior and no records of death before the end of the study period. The analysis was closed 12 months before database closure, and the period from January 1, 2016, to January 1, 2017, was used as a window for observation.12,13 Any patients who initiated intravitreal injections during that period were excluded.
The following factors were evaluated for possible associations with LTFU: age (< 65 years, 65 years to 80 years, > 80 years), gender (male, female), race (white, black, Asian, Hispanic, other, unknown), AGI (< $50,000, $50,000 to $100,000, > $100,000), RVO types (CRVO, branch retinal vein occlusion [BRVO]), distance of residence from clinic (<10 miles, 10 miles to 20 miles, > 20 miles). The continuous variables of age, distance from clinic, and regional average AGI were converted to categorical variables based on the data distributions and biological factors. In the category of race, “Other” included American Indian, Pacific Islander, and multiple races. Patients who had no listed race in the practice database or who declined to be identified were classified as “Unknown.” Subgroup analysis was conducted based on a subset of patients with VA data at the baseline visit. It was converted to categorical variables based on the data distribution and classified as VA better than 20/50, 20/50 to 20/100, and worse than 20/100.
Data were analyzed using SPSS version 17.0 (SPSS, Chicago, IL). The chi-square test was used to compare the differences based on characteristics between patients with follow-up and LTFU. Variables with a P value of .2 or less in the univariate analysis were evaluated in the logistic regression model using a stepwise backward strategy to evaluate the independent association between variables and LTFU. Associations were estimated using odds ratio with 95% confidence intervals (CI). A P value of less than .05 was considered statistically significant.
A total of 4,462 patients with RVO undergoing intravitreal injections from January 1, 2012, to January 1, 2017, were identified (Figure 1). Since January 1, 2016, to January 1, 2017, was set as a window of observation for patients who had injections prior to that time period, 799 (17.9%) participants receiving injections for the first time during this period were excluded. An additional 134 (3.0%) patients receiving intravitreal corticosteroid injections were also excluded. 129 (2.9%) patients were excluded due to death. As a result, 3,400 patients who had 37,664 clinical visits were eligible. Mean (± standard deviation, [SD]) age of the study population was 75.3 years (± 13.2 years), with 72.8 years (± 13.6 years) for males and 77.5 years (± 12.5 years) for females. A total of 26,026 intravitreal injections were performed, with an average of 7.66 injections for each patient.
Following an injection, 863 (25.4%) patients were LTFU. Of those, the majority (789, 91.4%) had no further visits for more than 12 months after the last intravitreal injection and only 74 (8.6%) returned more than 1 year after an injection. Of the 789 patients, 271 (34.3%) only received one injection and never returned. Characteristics of the RVO subjects is shown in table 1. Among the three age groups, LTFU rate was lowest in patients between 65 and 80 years old (22.1% [298 of 1,347]), increasing to 25.5% (347 of 1,359) in those older than 80 years of age and 31.4% (218 of 694) in those younger than 65 years old (P < .001). Based on race / ethnicity, LTFU was observed in 22.3% (574 of 2,576) of white patients, 28.2% (97 of 344) of black patients, 16.7% (12 of 72) of Asian patients, 41.1% (51 of 124) of Hispanic patients, and 44.2% (119 of 269) in unknown races (P < .001). Four hundred seventy-four of 1,632 (29.0%) patients with CRVO and 389 of 1,768 (22.0%) patients BRVO were LTFU (P < .001). The highest proportion of LTFU in each group was found to be patients living more than 20 miles away from clinic (30.9%), those with AGI <$50,000 (29.2%), and those with no health insurance (50.0%) (P < .05 within each group).
Baseline Characteristics of Patients With RVO Follow-Up and Loss to Follow-Up After Intravitreal Anti-VEGF Injections
Table 2 shows the univariate and multivariate results for the predictive factors for LTFU. Based on the univariate analysis, age group, race / ethnicity, RVO type, distance of residence from clinic, insurance status, and regional average AGI were all associated with LTFU and were considered in the multivariate model. After adjusting for covariates, gender and AGI did not significantly predict LTFU or interact with the other predictors. Compared with white patients, patients who were black (odds ratio [OR]: 1.37; 95% confidence interval [CI], 1.06–1.78; P = .02), Hispanic (OR: 2.37; 95% CI, 1.62–3.47; P < .001), other (OR: 8.04; 95% CI, 2.71–23.67; P < .001), and unknown (OR: 2.81; 95% CI, 2.17–3.65; P < .001) were associated with a higher risk of LTFU. Those living more than 20 miles away from the clinic were also more likely to be LTFU (OR: 1.47; 95% CI, 1.17–1.86) when compared with patients living fewer than 10 miles away (P = .001). Patients with no insurance were at approximately two-fold increased odds of LTFU compared to those with government coverage for insurance (P = .05). In contrast, patients between the ages of 65 and 80 years had a 29% lower risk of LTFU compared to those younger than 65 years (P = .007). Compared to patients with CRVO, the BRVO group was less likely to be LTFU (OR: 0.70; 95% CI, 0.60–0.82; P < .001).
Potential Risk Factors for Loss to Follow-Up in Patients With Macular Edema Due to RVO Following Intravitreal Anti-VEGF Injections
Subgroup analysis was performed on 1,957 patients with RVO with available VA data. The lowest rate of LTFU (14.1%) was found in patients with a baseline VA better than 20/50, whereas the highest rate (20.1%) occurred in those with VA between 20/50 and 20/100. A statistically significant difference was detected between these three groups (P = .006). Multivariate analysis showed that patients with baseline VA 20/50 to 20/100 or worse than 20/100 were approximately 1.5-times more likely to be LTFU compared to those with VA better than 20/50.
Our study investigated LTFU immediately after an intravitreal anti-VEGF injection for macular edema due to RVO in a real-world setting. We found that approximately one in four patients did not return for a subsequent visit for at least 1 year, which is of potential concern given the importance of ongoing therapy to optimize visual outcomes. We also found that LTFU was significantly associated with race/ethnicity, age, RVO type, and distance from clinic.
The 25.4% rate of LTFU found in our study was higher than expected. Among those patients classified as LTFU, the vast majority (91.4%) never returned, whereas the remaining 8.6% returned for a visit sometime beyond 1 year. Moreover, approximately one in three patients who were LTFU only had one injection and never returned. Since these patients are likely at greatest risk for vision loss without appropriate therapy, it is concerning to see such a high rate. Although no prior published studies appear to have explored LTFU rates in RVO, one study has looked at this issue in patients with neovascular age-related macular degeneration (AMD) receiving intravitreal ranibizumab injections. In this study, 19.9% were noncompliant with follow-up. However, they only investigated 95 patients undergoing intravitreal ranibizumab on a pro re nata regimen with monthly follow-up and defined nonadherence as missing a scheduled visit. A questionnaire found that lack of motivation was the primary reason for the LTFU.14 Differences in the definition of LTFU, method of data analysis, and underlying diagnosis distinguishes their study from ours, limiting direct comparisons.
We also examined potential factors associated with LTFU. Patients with a diagnosis of CRVO were more likely to be LTFU than those with BRVO. One possible explanation is that patients with more severe ischemia and, therefore, poorer visual outcomes despite anti-VEGF injections may be less motivated to adhere to further therapy. Prior studies have demonstrated that patients with severe ocular disease and poor VA were less likely to follow-up.15,16 Since CRVO patients are more likely than BRVO to have severe macular ischemia and poorer vision,17,18 this may explain the greater rate of LTFU. Our subgroup analysis of patients with available baseline VA data appear to support this hypothesis, as those with VA of 20/50 or worse were more likely to be LTFU.
Another factor that appeared to impact LTFU was race / ethnicity. Compared with white patients, those who were black and Hispanic were at higher risk of LTFU. In previous diabetic retinopathy screening studies, African-Americans were found to be less likely to receive eye care.19,20 Another study also found that black patients with diabetes were only 70% as likely to have eye care visits compared with white patients.21 Similarly, the Los Angeles Latino Eye Study analyzed 821 adult Latinos with type 2 diabetes mellitus and discovered that only about 35% of Latinos followed the American Diabetes Association guidelines for eye care.22 Our study differed by looking at patients who had already started a treatment plan but failed to return in a timely fashion whereas the prior studies explored the likelihood of patients to initiate a care visit. Nevertheless, the fact that 28.2% of black patients and 41.1% of Hispanic patients compared to 22.3% of white patients were LTFU suggests that there may exist societal and / or cultural barriers to proper care.23 Of note, a high rate of LTFU (44.2%) was detected in patients who chose not to self-report their race in our study. The sociodemographic distribution in this population is not well understood due to the lack of supporting literature.
Age was another important factor associated with LTFU. Patients who were between 65 and 80 years old were less likely to be LTFU. The Diabetic Retinopathy Awareness Program found that younger age was one of the factors associated with diabetic vision care noncompliance.24 One explanation may be that most patients younger than 65 years are still working, making it more challenging to take time off for clinic visits. Another possible explanation may be related to medical insurance coverage. Since patients in the U.S. are covered under Medicare starting at age 65, these older patients would have at least 80% of medical expenses covered. Younger patients may be more likely to be uninsured or have high deductible plans, which may be another barrier to receiving appropriate care. Based on our findings, patients older than 80 years of age were also more likely to be LTFU. Although these older patients should have less problems with insurance coverage due to Medicare, they may need to rely more on others to transport them to office visits. In addition, with advancing age they likely have more comorbidities and therefore need to prioritize their care.
Distance also appeared to influence LTFU, with patients living more than 20 miles away being at highest risk. Some of these patients from more distant locations may have been coming for a one-time second opinion or were visiting the area when the RVO occurred. The longer driving time as well as higher expense for gas and possibly tolls may have also contributed. Moreover, approximately 40% of the patients in our study were older than 80 years of age. Many of these patients may not feel comfortable driving longer distances and may have to rely on others to bring them to the office, which is a greater burden to the caregiver if the office is farther away.
This study has several limitations. It was retrospective and relied on accurate demographic and billing data from a practice database. Our analysis did not include all of the potential factors that might explain the LTFU. Since we only had access to our practice database rather than all practice databases in our geographic area, it is possible that some patients may have followed up with another eye care specialist for ongoing care. Finally, the definition for LTFU focused on the time interval between an injection and next follow-up visit. We chose a period of 12 months for that interval, as it would be extremely unlikely that a patient would have been given an injection and then instructed to return a year or more later. Despite these limitations, this was a fairly large study that included 3,400 patients receiving 26,026 intravitreal injections over 37,664 visits during a 4-year period. In addition, we were able to perform a systematic analysis of these patients that highlights a real-world potentially alarming rate of LTFU in patients with RVO receiving intravitreal anti-VEGF injections.
In summary, our findings highlight the need for more effective interventions to encourage better follow-up in patients with RVO receiving intravitreal anti-VEGF injections. Patients who are younger than 65 years of age, black or Hispanic, diagnosed with CRVO, living in more distant locations, or have worse baseline VA may require special attention. Future studies looking at larger numbers of patients with varying demographic backgrounds and geographic locations will be helpful to confirm these results. Identifying additional risk factors will also be necessary in order to design risk prediction models that allow clinicians to focus interventions on those at highest risk of LTFU.
- Wong TY, Larsen EK, Klein R, et al. Cardiovascular risk factors for retinal vein occlusion and arteriolar emboli: The Atherosclerosis Risk in Communities & Cardiovascular Health studies. Ophthalmology. 2005;112(4):540–547. doi:10.1016/j.ophtha.2004.10.039 [CrossRef]
- Mitchell P, Smith W, Chang A. Prevalence and associations of retinal vein occlusion in Australia. The Blue Mountains Eye Study. Arch Ophthalmol. 1996;114(10):1243–1247. doi:10.1001/archopht.1996.01100140443012 [CrossRef]
- Rogers S, McIntosh RL, Cheung N, et al. The prevalence of retinal vein occlusion: Pooled data from population studies from the United States, Europe, Asia, and Australia. Ophthalmology. 2010;117(2):313–319.e1. doi:10.1016/j.ophtha.2009.07.017 [CrossRef]
- Campochiaro PA, Wykoff CC, Singer M, et al. Monthly versus as-needed ranibizumab injections in patients with retinal vein occlusion: The SHORE study. Ophthalmology. 2014;121(12):2432–2442. doi:10.1016/j.ophtha.2014.06.011 [CrossRef]
- Korobelnik JF, Holz FG, Roider J, et al. Intravitreal aflibercept injection for macular edema resulting from central retinal vein occlusion: One-year results of the phase 3 GALILEO Study. Ophthalmology. 2014;121(1):202–208. doi:10.1016/j.ophtha.2013.08.012 [CrossRef]
- Heier JS, Clark WL, Boyer DS, et al. Intravitreal aflibercept injection for macular edema due to central retinal vein occlusion: Two-year results from the COPERNICUS study. Ophthalmology. 2014;121(7):1414–1420.e1. doi:10.1016/j.ophtha.2014.01.027 [CrossRef]
- Larsen M, Waldstein SM, Boscia F, et al. Individualized ranibizumab regimen driven by stabilization criteria for central retinal vein occlusion: Twelve-month results of the CRYSTAL Study. Ophthalmology. 2016;123(5):1101–1111. doi:10.1016/j.ophtha.2016.01.011 [CrossRef]
- Varma R1, Bressler NM, Suñer I, et al. . Improved vision-related function after ranibizumab for macular edema after retinal vein occlusion: Results from the BRAVO and CRUISE trials. Ophthalmology. 2012;119(10):2108–2118. doi:10.1016/j.ophtha.2012.05.017 [CrossRef]
- SOI Tax Stats - Individual Income Tax Statistics - ZIP Code Data (SOI). IRS website. https://www.irs.gov/uac/soi-tax-stats-individual-income-tax-statistics-zip-code-data-soi. Updated November 5, 2018.
- ARblast: All Resources on Blasting. Office of Surface Mining Reclamation and Enforcement website. https://www.osmre.gov/resources/blasting/arblast.shtm. Updated October 4, 2018.
- Robusto CC. The Cosine-Haversine Formula. Am Math Mon. 1957;64(1):38–40. doi:10.2307/2309088 [CrossRef]
- Obeid A, Gao X, Ali FS, et al. Loss to follow-up in patients with proliferative diabetic retinopathy after panretinal photocoagulation or intravitreal anti-VEGF injections. Ophthalmology. 2018;125(9):1386–1392. doi:10.1016/j.ophtha.2018.02.034 [CrossRef]
- Grimsrud AT, Cornell M, Egger M, et al. Impact of definitions of loss to follow-up (LTFU) in antiretroviral therapy program evaluation: Variation in the definition can have an appreciable impact on estimated proportions of LTFU. J Clin Epidemiol. 2013;66(9):1006–1013. doi:10.1016/j.jclinepi.2013.03.013 [CrossRef]
- Droege KM, Muether PS, Hermann MM, et al. Adherence to ranibizumab treatment for neovascular age-related macular degeneration in real life. Graefes Arch Clin Exp Ophthalmol. 2013;251(5):1281–1284. doi:10.1007/s00417-012-2177-3 [CrossRef]
- Thompson AC, Thompson MO, Young DL, et al. Barriers to follow-up and strategies to improve adherence to appointments for care of chronic eye diseases. Invest Ophthalmol Vis Sci. 2015;56(8):4324–4331. doi:10.1167/iovs.15-16444 [CrossRef]
- Ung C, Murakami Y, Zhang E, et al. The association between compliance with recommended follow-up and glaucomatous disease severity in a county hospital population. Am J Ophthalmol. 2013;156(2):362–369. doi:10.1016/j.ajo.2013.03.005 [CrossRef]
- Ehlers JP, Fekrat S. Retinal vein occlusion: Beyond the acute event. Surv Ophthalmol. 2011;56(4):281–299. doi:10.1016/j.survophthal.2010.11.006 [CrossRef]
- Mir TA, Kherani S, Hafiz G, et al. Changes in retinal nonperfusion associated with suppression of vascular endothelial growth factor in retinal vein occlusion. Ophthalmology. 2016;123(3):625–634.e1. doi:10.1016/j.ophtha.2015.10.030 [CrossRef]
- Owsley C, McGwin G, Scilley K, et al. Perceived barriers to care and attitudes about vision and eye care: Focus groups with older African Americans and eye care providers. Invest Ophthalmol Vis Sci. 2006;47(7):2797–2802. doi:10.1167/iovs.06-0107 [CrossRef]
- Walker EA, Basch CE, Howard CJ, et al. Incentives and barriers to retinopathy screening among African-Americans with diabetes. J Diabetes Complications. 1997;11(5):298–306. doi:10.1016/S1056-8727(96)00121-3 [CrossRef]
- Wang F, Javitt JC. Eye care for elderly Americans with diabetes mellitus. Failure to meet current guidelines. Ophthalmology. 1996;103(11):1744–1750. doi:10.1016/S0161-6420(96)30432-6 [CrossRef]
- Paz SH, Varma R, Klein R, et al. Noncompliance with vision care guidelines in Latinos with type 2 diabetes mellitus: The Los Angeles Latino Eye Study. Ophthalmology. 2006;113(8):1372–1377. doi:10.1016/j.ophtha.2006.04.018 [CrossRef]
- Nsiah-Kumi P, Ortmeier SR, Brown AE. Disparities in diabetic retinopathy screening and disease for racial and ethnic minority populations — a literature review. J Natl Med Assoc. 2009;101(5):430–437. doi:10.1016/S0027-9684(15)30929-9 [CrossRef]
- Schoenfeld ER, Greene JM, Wu SY, Leske MC. Patterns of adherence to diabetes vision care guidelines: Baseline findings from the Diabetic Retinopathy Awareness Program. Ophthalmology. 2001;108(3):563–571. doi:10.1016/S0161-6420(00)00600-X [CrossRef]
Baseline Characteristics of Patients With RVO Follow-Up and Loss to Follow-Up After Intravitreal Anti-VEGF Injections
|Variable||Follow-Up (N = 2,537)||Loss to Follow-Up (N = 863)||P Value|
|Gender, n (%)||.06|
| Male||1,167 (46.0%)||429 (49.7%)|
| Female||1,370 (54.0%)||434 (50.3%)|
|Age (Years), n (%)||< .001|
| < 65||476 (18.8%)||218 (25.3%)|
| 65–80||1,049 (41.3%)||298 (34.5%)|
| > 80||1,012 (39.9%)||347 (40.2%)|
|Race, n (%)||< .001|
| White||2,002 (78.9%)||574 (66.5%)|
| Black||247 (9.7%)||97 (11.2%)|
| Asian||60 (2.4%)||12 (1.4%)|
| Hispanic||73 (2.9%)||51 (5.9%)|
| Othera||5 (0.2%)||10 (1.2%)|
| Unknown||150 (5.9%)||119 (13.8%)|
|RVO Type, n (%)||< .001|
| CRVO||1,158 (45.6%)||474 (54.9%)|
| BRVO||1,379 (54.4%)||389 (45.1%)|
|Distance From Clinics (Miles), n (%)||.02|
| < 10||1,751 (69.0%)||570 (66.1%)|
| 10–20||488 (19.2%)||160 (18.5%)|
| > 20||298 (11.8%)||133 (15.4%)|
|Regional Average AGI, n (%)||.006|
| < $50,000||621 (24.5%)||256 (29.7%)|
| $50,000–100,000||1,415 (55.8%)||462 (53.5%)|
| > $100,000||501 (19.7%)||145 (16.8%)|
|Health Insurance Status, n (%)||< .001|
| Government coverageb||1,776 (70.0%)||557 (64.5%)|
| Private||744 (29.3%)||289 (33.5%)|
| No insurance||17 (0.7%)||17 (2.0%)|
Potential Risk Factors for Loss to Follow-Up in Patients With Macular Edema Due to RVO Following Intravitreal Anti-VEGF Injections
|Univariate Analysis||Multivariate Analysis|
|Variable||Loss to Follow-Up, n (%)||OR (95% CI)||P Value||OR (95% CI)||P Value|
| Male||429 (26.9%)||1.00 [Ref]||1.00 [Ref]|
| Female||434 (24.1%)||0.86 (0.74–1.01)||.06||0.89 (0.76–1.05)||.17|
| < 65||218 (31.4%)||1.00 [Ref]||1.00 [Ref]|
| 65–80||298 (22.1%)||0.62 (0.51–0.76)||< .001||0.71 (0.55–0.91)||.007|
| > 80||347 (25.5%)||0.75 (0.61–0.92)||.005||0.91 (0.71–1.18)||.49|
| White||574 (22.3%)||1.00 [Ref]||1.00 [Ref]|
| Black||97 (28.2%)||1.37 (1.06–1.76)||.02||1.37 (1.06–1.78)||.02|
| Asian||12 (16.7%)||0.70 (0.37–1.31)||.26||0.65 (0.35–1.23)||.19|
| Hispanic||51 (41.1%)||2.44 (1.68–3.53)||< .001||2.37 (1.62–3.47)||< .001|
| Othera||10 (66.7%)||6.98 (2.38–20.49)||< .001||8.04 (2.71–23.67)||< .001|
| Unknown||119 (44.2%)||2.77 (2.14–3.58)||< .001||2.81 (2.17–3.65)||< .001|
| CRVO||474 (29.0%)||1.00 [Ref]||1.00 [Ref]|
| BRVO||389 (22.0%)||0.69 (0.59–0.81)||< .001||0.70 (0.60–0.82)||< .001|
|Distance From Clinics (Miles)|
| < 10||570 (24.6%)||1.00 [Ref]||1.00 [Ref]|
| 10–20||160 (24.7%)||1.01 (0.82–1.23)||.95||1.03 (0.84–1.27)||.80|
| > 20||133 (30.9%)||1.37 (1.10–1.72)||.006||1.47 (1.17–1.86)||.001|
|Regional Average AGI|
| < $50,000||256 (29.2%)||1.00 [Ref]||1.00 [Ref]|
| $50,000–100,000||462 (24.6%)||0.79 (0.66–0.95)||.01||0.90 (0.74–1.10)||.31|
| > $100,000||145 (22.4%)||0.70 (0.56–0.89)||.003||0.86 (0.67–1.12)||.25|
|Health Insurance Status|
| Government coverageb||557 (23.9%)||1.00 [Ref]||1.00 [Ref]|
| Private||289 (28.0%)||1.24 (1.05–1.46)||.01||1.16 (0.94–1.43)||.17|
| No insurance||17 (50.0%)||3.19 (1.62–6.29)||.001||2.06 (1.00–4.26)||.05|
|Baseline Snellen VAc|
| > 20/50||91 (14.1%)||1.00 [Ref]||1.00 [Ref]|
| 20/50–20/100||129 (20.1%)||1.525 (1.137–2.046)||.005||1.526 (1.129–2.063)||.006|
| < 20/100||134 (20.0%)||1.519 (1.135–2.033)||.005||1.401 (1.027–1.911)||.033|