The economic costs of age-related ocular diseases and vision loss are increasing rapidly as our society ages. In addition to the direct costs of treating age-related eye diseases, elderly persons with vision loss are at significantly increased risk for falls and fractures, experiencing social isolation, and suffering from an array of comorbid medical conditions compared with individuals with normal vision. Recent studies estimate the total economic burden (direct and indirect costs) of adult vision impairment in the United States at $51.4 billion. This figure is expected to increase as the baby boomer generation continues to age. While a number of highly effective new therapies have caused a paradigm shift in the management of several major age-related ocular diseases in recent years, these treatments come at a substantial cost. This article reviews the economic burdens and treatment-related costs of 4 major ocular diseases of aging—glaucoma, age-related macular degeneration, diabetic retinopathy, and dry eye disease—and the implications for managed care.
(Am J Manag Care. 2013;18:S85-S91)
EDITOR'S NOTE: A clarification has been made within the supplement entitled: Addressing the Clinical and Managed Care Challenges in Treating Diseases of the Aging Eye. Within the section entitled “Managed Care Implications of Age-Related Ocular Conditions,” a statement on page s89 has been corrected to state the following: “Topical cyclosporine comprised 68% of dry eye medication prescriptions in 2003 to 2004, increasing to 84% of prescriptions in 2005 to 2006.”The original statement referred to topical cyclosporine as comprising 68% of glaucoma medication prescriptions.
Economic Burden of Vision Loss in the United States
Visual impairment and blindness has a profound impact on individuals, caregivers, and society as a whole. Older adults with impaired vision are at least twice as likely as those with normal vision to report difficulty walking or getting into or out of bed, and at least 3 times as likely to have difficulty going outside, preparing meals, grocery shopping, and managing medications.1 Visually impaired older adults are also significantly more likely to have 1 or more major comorbidities.2 Of the estimated 5.7 million people with vision loss in the United States, 3.3 million are at risk for mild to moderate depression, 350,000 are at risk for severe depression, 1.2 million have diabetes, and 3 million have impaired hearing.2 Older adults who have vision loss and a major age-related comorbidity (eg, breathing problems, heart disease, diabetes, joint/back pain) report substantially greater degrees of physical and social impairment, difficulty with shopping, and an increased likelihood of poor overall health.2
The economic costs of age-related ocular diseases and vision loss are increasing rapidly as our society ages. A 1981 economic analysis of vision loss conducted by researchers at Penn State University reported a total cost from vision problems of $14.1 billion; updates to this analysis conducted in 1991 and 2003 by researchers at the National Eye Institute put the inflation-adjusted costs at $38.4 billion and $67.6 billion for those 2 years, respectively.3 In a widely cited paper, Rein et al estimated the total economic burden of major visual disorders in 2004 dollars at $35.4 billion—$16.2 billion in direct medical costs, $11.1 billion in indirect costs, and $8 billion in lost productivity.4 Outpatient health services and pharmaceutical costs comprised the majority of direct medical costs.
A second study by Frick et al used data from the Medical Expenditure Panel Survey (1996-2002) to examine the financial costs of vision impairment for individuals and caregivers.5 They reported a total cost of $5.5 billion in direct medical expenditures, informal care, and health utility, which amounted to almost $1400 per year for each of the 3.7 million visually impaired or blind adults in the United States. Utilization of home healthcare services, the single greatest medical expenditure, was significantly increased in visually impaired individuals. Compared with persons without vision impairment who used home health agencies, visually impaired individuals spent an extra $4900 in home health services per year. The researchers also projected that visually impaired individuals lost a total of more than 209,000 quality-adjusted life-years (QALYs) each year, adding another $10 billion or more to the annual economic impact, and bringing the total cost to nearly $16 billion.
The healthcare-related cost of visually impaired individuals is projected to increase as the society ages and newer, more costly treatments for ocular diseases are developed.4 In their 2006 paper, Rein et al projected that the cost of treatment for age-related macular degeneration (AMD) would increase from $575 million to $845 million over the following 15 years due to the aging of the population alone. Similar trends are expected for glaucoma, diabetic retinopathy (DR), and other age-related ocular diseases. The introduction of new medications, such as intravitreal anti-vascular endothelial growth factor (VEGF) agents for wet AMD and diabetic macular edema (DME), and prostaglandins and topical cyclosporine for glaucoma and dry eye disease, respectively, has accelerated this trend. Moreover, enactment of the Medicare Part D prescription drug program in 2006 was expected to result in greater use of prescription medications in patients at least 65 years of age with glaucoma, and this may apply to other eye diseases.
The following sections review the economic burden of the 4 major age-related eye diseases discussed in this supplement—glaucoma, wet AMD, DR/DME, and dry eye disease—and implications for managed care, including reimbursement issues and comparative efficacy.
Open-angle glaucoma (OAG) affected an estimated 2.2 million US residents 40 years and older according to 2004 data, and that figure is projected to increase by 50% to 3.36 million Americans in 2020.6,7 As glaucoma is often asymptomatic in its early stages, much of the medical resource use for this disease has historically occurred during the later stages, at which point aggressive treatment is indicated to preserve vision.8 It should be noted that, with greater awareness, improved diagnostic instruments, and greater access to care, OAG is now diagnosed at reasonably early stages and pharmaceutical treatment is far more common than surgical intervention. A retrospective study of medical records of 151 individuals found that the average direct cost of glaucoma treatment increased from $623 per patient per year for those with suspected or early-stage glaucoma (stage 0-1) to $2511 per patient per year for patients with endstage disease (stage 5; blindness).8 Medications comprised the largest proportion of total direct costs for all glaucoma stages; using real-world compliance estimates, medication use accounted for 38% of total costs for patients with stage 5 glaucoma, and at least 44% for those with stage 1 to 3 glaucoma. One potential mitigating factor in these costs is the introduction of generic latanoprost in 2011.7 In another retrospective analysis of Medicare beneficiaries with glaucoma, patients with vision loss were significantly more likely to be placed in a nursing home (odds ratio [OR] = 2.18; 95% confidence interval [CI], 2.06-2.31) and develop depression (OR = 1.63; 95% CI, 1.54-1.73) than those with no vision loss.9 Additionally, patients with vision loss were 67.4% more likely to fracture a femur (OR = 1.67; 95% CI, 1.53-1.83) and 58.6% more likely to experience a fall or accident (OR = 1.59; 95% CI, 1.50-1.68) versus the reference group (Figure).9 The mean total and component annual healthcare costs increased incrementally with increasing severity of vision loss, ranging from $8157 for no vision loss to $18,670 for blindness.
The most comprehensive analysis of glaucoma medication expenditures comes from the 2001 to 2006 Medical Expenditure Panel Survey, which compiled data on glaucoma medication from interviews and pharmacy prescribing records for 1404 participants at least 18 years of age.7 Mean glaucoma medication expenditure increased from $445 in 2001 to $557 in 2006 (slope = 20.8; P <.001). Significant increases in expenditures were found for women (P = .02), Caucasians (P =.02), middle-aged participants (45-64 years; P = .001), older age participants (≥65 years; P = 0.02), and those with less than a high school education (slope = 55.1; P = .008).
In terms of specific medications, expenditures increased significantly for prostaglandin analogues (from $168 in 2001 to $271 in 2006) but decreased for b-blockers (from $167 in 2001 to $69 in 2006).7 The majority of patients took only 1 medication class per year (range, 61%-68% for 2001- 2006). While only 130 of the study participants enrolled in the Medicare Part D prescription drug plan in 2006, these patients demonstrated a significant increase in glaucoma medication expenditures relative to participants with private insurance ($772 vs $440; P <.001), but not versus participants with public insurance without Part D coverage ($613; P = .12).
A limitation of the study is that it did not account for medication compliance, which is a major problem in patients with glaucoma. It is possible that poor compliance with medications could keep costs down initially, but increase costs later on as patients undergo more expensive procedures for late-stage disease.7 On the other hand, newer treatments that improve compliance (ie, require less frequent dosing) or those that more effectively lower intraocular pressure may reduce the number of patients who progress to more severe stages of vision impairment.
The direct cost of treatment of wet AMD in 2004 dollars was estimated at $575 million, which is expected to increase as the society ages and as more costly treatments are introduced.4,10 While there are a substantial number of published cost-effectiveness studies of wet AMD therapies, many of these looked at older treatments, such as laser photocoagulation, photodynamic therapy (PDT), and pegaptanib sodium, a first-generation intravitreal anti-VEGF agent.11 A 2009 review of 44 cost-effectiveness studies in AMD found that, overall, older treatments (eg, PDT) were more cost-effective compared with best supportive care or no treatment, while newer therapies (eg, ranibizumab) were more cost-effective than older treatments.11 Across the studies analyzed, a treatment time horizon of at least 5 years was generally needed to demonstrate cost-effectiveness for the newer, more clinically efficacious AMD therapies at standard willingness-topay (WTP) thresholds compared with older, less-effective therapies.
Following publication of the CATT (Comparison of Agerelated macular degeneration Treatment Trials)12 and IVAN (Inhibit VEGF in Age-related choroidal Neovascularisation)13 studies demonstrating clinical equivalence of ranibizumab and bevacizumab for wet AMD, much of the cost discussions have focused on these 2 agents. A 2011 survey of retina specialists (Patterns and Trends Survey) found that bevacizumab was chosen over ranibizumab 70% of the time to treat wet AMD based on its lower cost.14 The per unit dose cost of ranibizumab (approximately $2000) is as much as 50 times that of bevacizumab (around $50), the latter of which is typically split up into 1.25-mg single injections from the original 100-mg vial licensed for cancer treatment.15,16
As bevacizumab is not licensed to treat wet AMD, there are few published studies on the cost-effectiveness of this agent for this condition. However, a British study published in 2007 found that bevacizumab would need to be approximately 40% as effective as ranibizumab for the treatment of predominately classic AMD for the latter to achieve £30,000 per QALY, a standard cost-effectiveness threshold.17 (It should be noted that the 2010 Patient Protection and Affordable Care Act prohibits the use of QALYs for making reimbursement decisions in the United States, but QALYs are widely used outside the United States to measure costeffectiveness.)18
More recently, a group of researchers with the Veterans Affairs San Diego Healthcare System and University of California San Diego published a cost-effectiveness analysis of ranibizumab and bevacizumab from the US payer per spective.16 The researchers employed a simplified Markov model using visual acuity data from published randomized clinical trials of the 2 agents to simulate treatment of 1000 patients over a 20-year time span. The total direct cost for bevacizumab was $30,349 per patient with a mean average of 21.6 QALYs over 20 years compared with $220,649 for ranibizumab with a mean average of 18.1 QALYs. Compared with ranibizumab, treatment with bevacizumab resulted in an incremental cost-effectiveness ratio (ICER) of —$54,649 to gain 1 additional QALY. Based on a WTP of $50,000, bevacizumab was predicted to be more cost-effective versus ranibizumab 95% of the time.16
A limitation of the study was that it assumed continuous treatment with both agents throughout the 20-year time horizon; published data from clinical trials of ranibizumab and bevacizumab in AMD extend to only about 2 years.16 In addition, the model did not account for treat-and-extend or treat-and-observe strategies, which are used by more than 90% of clinicians who use anti-VEGF agents to treat wet AMD, according to the 2011 Patterns and Trends Survey.14 However, the regimens used in the analysis were similar, so the difference in costs should be minimal.
Further complicating the picture is the recent approval of aflibercept for wet AMD by the US Food and Drug Administration. At $1850 per dose, aflibercept costs about $100 less than ranibizumab and roughly $1800 more than bevacizumab.14,19 Although results from the phase 3 clinical trials of aflibercept suggest that this agent might have a longer duration of action relative to ranibizumab or bevacizumab, these data will need to be confirmed in post-marketing studies before clinicians are likely to appreciably change their treatment patterns for wet AMD.14 Clinicians using bevacizumab for cost reasons are likely to continue using this agent, while those using ranibizumab will probably exercise caution in adopting aflibercept until they become more comfortable with the use of this agent for wet AMD and are assured of timely reimbursement from Medicare.14
As with wet AMD, treatment and testing modalities for proliferative DR and DME have shifted over time—use of laser photocoagulation is decreasing, while more patients are receiving intravitreal injections (eg, intravitreal triamcinolone [IVTA], anti-VEGF therapy). The use of optical coherence tomography has also increased dramatically—from 2.5% in 2000 to more than 40% in 2004.20 Not surprisingly, a recent cost-effectiveness study that used data from published randomized clinical studies of the major therapies for DME showed a pattern similar to wet AMD, with the newer anti-VEGF agents producing greater clinical benefit but at a significantly higher cost than older therapies.21 As shown in the Table, lines of vision saved ranged from 0.26 (grid laser) on the low end to 2.02 (bevacizumab) on the high end.21 The cost per line of vision saved ranged from $1329 (bevacizumab) to $11,609 (ranibizumab).
A follow-up to this study that used a similar methodology found that in many instances, less costly therapies or dosing strategies could be used selectively in certain DME patient subgroups or clinical situations without significantly compromising efficacy.22 For example, patients with DME and good visual acuity (>20/32), who have typically been excluded from clinical studies of IVTA and anti-VEGF agents, could potentially benefit from early intervention with focal laser therapy and avoid later intravitreal anti-VEGF injections, which the authors calculated would save approximately 90% in treatment costs. On the opposite end of the spectrum, patients with DME and particularly poor vision (20/200—20/320) had a slightly more favorable response to IVTA versus focal laser therapy in a Diabetic Retinopathy Clinical Research Network (DRCRN) study. While the costs of 1 year of laser treatment versus IVTA were similar ($2330 vs $1907), the greater benefit of IVTA (particularly at 4 mg) in the poor vision subgroup yielded a 62% lower cost per line of vision saved.
Another DME subgroup of interest is patients with pseudophakic eyes, as this group is not at risk for cataract development from exposure to ocular steroids.23 Cost modeling of the DRCRN randomized trial of ranibizumab plus prompt or deferred laser versus IVTA plus prompt laser for DME showed that for that subgroup of patients with pseudophakic eyes, the ICER for ranibizumab plus deferred laser was more than $14,690 per letter of vision gained compared with IVTA plus laser.23 For the overall study population, the authors calculated an incremental cost utility ratio (ICUR) of $171,285 per QALY for ranibizumab relative to IVTA at 2 years; at 10 years the ICUR for ranibizumab dropped to $87,584/QALY, which suggests that ranibizumab becomes cost-effective for DME versus IVTA only after prolonged use (≥10 years). While formal comparative studies between ranibizumab and either bevacizumab or aflibercept have not yet been conducted, randomized clinical studies of these agents in DME have yielded similar efficacy results, with mean gains in best corrected visual acuity of approximately 9 letters.24-26 Therefore, as in wet AMD, cost will likely be a major determining factor for clinicians selecting treatment regimens for their patients with DME.
Dry Eye Disease
Prevalence estimates for dry eye disease vary widely, ranging from 3.5% to more than 30% of adults over age 50 years.27,28 The use of dry eye medication and related costs have been steadily increasing for at least the past decade. A retrospective study found that between 2001 and 2006, medication expenditures for dry eye increased from $55 per patient year to $299, with much of the increase being driven by the introduction of topical cyclosporine 0.05% in 2003.28 The manufacturer of topical cyclosporine initiated a major direct-to-consumer marketing campaign about dry eye, and also heavily promoted the drug to eye care professionals; these activities increased physician and patient awareness of dry eye disease.28 Topical cyclosporine comprised 68% of dry eye medication prescriptions in 2003 to 2004, increasing to 84% of prescriptions in 2005 to 2006.28
A retrospective analysis of pharmacy claims data that evaluated costs in 9065 patients who used topical cyclosporine for dry eye disease from January 1, 2004, to December 31, 2005, found a mean healthcare cost of $336 per patient, with a total cost of $3.05 million.29 Several demographic features had a significant effect on dry eye medication expenditures. Mean spending for women was double that of men ($244 vs $122; P <.0001); similarly, spending for non-Hispanics was double that for the Hispanic population ($228 vs $106; P <.0001).28 Education level was also important: spending was more than double in patients with more than a high school education versus those with less education ($250 vs $100; P <.0001).28 Rising medication costs also contributed to overall cost increases, according to the analysis. The mean cost of sulfacetamide sodium—prednisolone acetate ophthalmic suspension increased from $36.27 in 2001 to 2002 to $54.56 in 2003 to 2004, and to $64.43 in 2005 to 2006, while the average cost of topical cyclosporine increased from $98.98 in 2003 to 2004 to $113.06 in 2005 to 2006.28
Yu et al published a more comprehensive study of the total economic burden of dry eye disease using 2008 economic and survey data.27 Their analysis encompassed 5 categories of dry eye healthcare resource use: ocular lubricant treatment, cyclosporine, punctal plugs, physician visits, and nutritional supplements. The average annual direct cost for patients seeking medical care was calculated to be $783 per patient—$678 for patients with mild dry eye, $771 for moderate dry eye, and $1267 for severe dry eye. Accounting for the prevalence of dry eye disease among adults at least 50 years of age (1.68 million men, 3.23 million women), the overall burden of dry eye disease for the US healthcare system was estimated at $3.84 billion. For each symptom severity group, patients with dry eye who were treated with punctal plugs plus topical cyclosporine had the highest annual direct expenditures: $2964 for mild disease, $2959 for moderate disease, and $2698 for severe disease. The total annual direct costs to the payer for a cohort of 1000 patients with varying severity of symptoms would be $782,673.
The indirect costs of dry eye disease considerably outweighed direct costs, ranging from $12,569 to $18,168 per year.27 Most of these costs were attributable to reduced work performance and diminished productivity rather than increased absenteeism. The mean number of lost work days per year was 8.4 for mild, 3.7 for moderate, and 14.2 for severe dry eye. By contrast, the equivalent of lost work days due to affected performance was calculated at 91 days for patients with mild dry eye, and 94.9 and 128.2 for patients with moderate and severe dry eye, respectively. The total cost of dry eye disease from a societal perspective was $11,302 per patient, with a total societal burden of $55.4 billion.
Of interest, approximately 40% of patients with dry eye symptoms in the analysis did not seek medical care.27 Although the direct costs for these patients were quite low ($126), indirect costs for this group were substantial due to diminished productivity and absenteeism.
The trend of rising costs associated with visual impairment and the treatment of age-related ocular diseases in the United States is having a tremendous impact on patients’ caregivers, third-party payers, and society. These costs will almost certainly continue to increase in the coming decades as the baby boomer generation reaches old age. While new treatments for eye diseases are benefiting thousands of patients, the costs of these treatments are a major cause of concern. Managed care organizations have not aggressively reviewed this category of drugs for many years, partially due to a lack of published payer-oriented studies on ophthalmologic agents. Moving forward, comparative effectiveness and cost-effectiveness studies, such as those discussed in this supplement, will be crucial in determining which treatments (and in what doses and combinations) are likely to provide the best outcomes for patients at the least cost. Managed care pharmacists can also contribute to improving outcomes by training patients on the proper installation of eye drops and encouraging patients to adhere to their treatment regimens. The development of generic equivalents of some ocular medications represents another potential avenue for cost savings. However, ophthalmic drugs are more difficult to manufacture than solid forms, which may hinder the introduction of generics to the marketplace.30 Ophthalmologists are already expressing concern with their patients’ use of generic ophthalmic products.31 Ophthalmic drug formularies of managed care plans may also need to allow for more agents and greater flexibility in prescribing practices.30Author affiliation: Atrius Health/Harvard Vanguard Medical Associates, Watertown, MA.
Funding source: This activity is supported by an educational grant from Allergan, Inc.
Author disclosure: Dr Cardarelli and Mr Smith have no relevant financial relationships with commercial interests to disclose.
Authorship information: Analysis and interpretation of data (WJC, RAS); drafting of the manuscript (WJC, RAS); and critical revision of the manuscript for important intellectual content (WJC, RAS).
Address correspondence to: E-mail: William_Cardarelli@vmed.org.