Methodology

This analysis draws on a number of methodologies and data sources. We estimated costs by cost categories listed in consensus guidelines.[1] Direct costs include medical care attributable to diagnosed disorders, medical vision aids, undiagnosed vision loss, low vision aids/devices, special education, school screening, and Federal assistance programs. Indirect costs include productivity losses of adults, productivity losses of caregivers, long-term care, transfer payments (not included in total), and deadweight loss from transfer payments. Costs are also reported by payers’ perspective, including government, private insurance and patient costs. All prices and costs were adjusted to 2013 U.S. dollars using the Consumer Price Index (CPI) for nonmedical costs and medical components of the Consumer Price Index for medical expenses. U.S. population values are based on 2011 census estimates.

The prevalence of visual impairment and blindness are based on a meta-analysis of epidemiological studies using gold-standard comprehensive eye examinations for the population aged 40 and older.[2, 3] For the younger population such data are not available. For the population aged 12-39, we estimated prevalence of visual loss based on National Health and Nutrition Examination Survey (NHANES) data from 2005 through 2008 in which autorefractors were used to measure corrected acuity loss. NHANES does not assess acuity in respondents younger than age 12; we imputed prevalence for children younger than age 12 based on the incidence of profound impairment or blindness.

We estimated medical costs attributable to diagnosed disorders, undiagnosed self-reported vision loss, and vision correction using 2003-2008 Medical Expenditure Panel Survey (MEPS) data. MEPS is a nationally representative panel survey of health and healthcare expenditures in which individuals self-report health history, while medical expenditures are measured and confirmed by the respondents’ medical providers. Diagnosed disorders are defined as any diagnosis code related to the eyes, vision, or ocular adnexa, while undiagnosed vision loss is defined as self-reported low vision in the absence of any reported diagnosed disorder. Costs attributable to these two conditions were estimated using a “top-down” econometric approach whereby the incremental costs attributable to these conditions were estimated while controlling for socio-demographic conditions and comorbidities. This approach allows estimation of ancillary costs beyond those directly related to eye care services. Vision correction costs include the costs of optometry visits and the cost of vision aids including eyeglasses and contacts. These costs are measured separately from other medical costs in MEPS; they are not associated with diagnosis codes and are based on non-confirmed, self-reported expenditures. We estimated costs for these conditions using a “bottom-up” accounting approach whereby we measured the weighted sum of these costs.

We estimate productivity losses based on the difference in average incomes among persons self-reporting different levels of visual functionality in the Survey of Income and Program Participation, applied to our estimated prevalent population with impairment and blindness.[4, 5] Costs for other direct and indirect cost categories were estimated based on published parameter values and Federal budgets.