Disparities in Receipt of Eye Exams Among Medicare Part B Fee-for-service Beneficiaries With Diabetes — United States, 2017

Elizabeth A. Lundeen, PhD; John Wittenborn; Stephen R. Benoit, MD; Jinan Saaddine, MD

Disclosures

Morbidity and Mortality Weekly Report. 2019;68(45):1020-1023. 

In This Article

Abstract and Introduction

Introduction

Approximately 30 million persons in the United States have diabetes.* Persons with diabetes are at risk for vision loss from diabetic retinopathy and other eye diseases.[1] Diabetic retinopathy, the most common diabetes-related eye disease, affects 29% of U.S. adults aged ≥40 years with diabetes[2] and is the leading cause of incident blindness among working-age adults.[1] It is caused by chronically high blood glucose damaging blood vessels in the retina. Annual dilated eye exams are recommended for persons with diabetes because early detection and timely treatment of diabetic eye diseases can prevent irreversible vision loss§,¶.[3,4] Studies have documented prevalence of annual eye exams among U.S. adults with diabetes;[5,6] however, a lack of recent state-level data limits identification of geographic disparities in adherence to this recommendation. Medicare claims from the 50 states, the District of Columbia (DC), Puerto Rico, and U.S. Virgin Islands (USVI) were examined to assess the prevalence of eye exams in 2017 among beneficiaries with diabetes who were continuously enrolled in Part B fee-for-service insurance, which covers annual eye exams for beneficiaries with diabetes.** This report also examines disparities, by state and race/ethnicity, in receipt of eye exams. Nationally, 54.1% of beneficiaries with diabetes had an eye exam in 2017. Prevalence ranged from 43.9% in Puerto Rico to 64.8% in Rhode Island. Fewer than 50% of beneficiaries received an eye exam in seven states (Alabama, Alaska, Kentucky, Louisiana, Nevada, West Virginia, and Wyoming) and Puerto Rico. Non-Hispanic white (white) beneficiaries had a higher prevalence of receiving an eye exam (55.6%) than did non-Hispanic blacks (blacks) (48.9%) and Hispanics (48.2%). Barriers to receiving eye care (e.g., suboptimal clinical care coordination and referral, low health literacy, and lack of perceived need for care) might limit Medicare beneficiaries' ability to follow this preventive care recommendation. Understanding and addressing these barriers might prevent irreversible vision loss among persons with diabetes.

This analysis was performed using 100% of the Centers for Medicare & Medicaid Services research identifiable files but was restricted to claims for Medicare beneficiaries continuously enrolled in Part B fee-for-service for all of 2017.†† Part B covers outpatient services, including ophthalmologic services. This analysis includes Medicare beneficiaries aged ≥65 years, as well as those aged <65 years who qualify through disability or disease status, in the 50 U.S. states, DC, Puerto Rico, and USVI. Analyses were conducted using SAS Enterprise Guide (version 9.4; SAS Institute).

The outcome measure was the prevalence among Medicare Part B fee-for-service beneficiaries with diabetes of receiving an eye exam during January–December 2017. Beneficiaries received a diagnosis of diabetes if they had at least one diagnosis code (International Classification of Diseases, Tenth Revision) or procedure code (Current Procedural Terminology [CPT] and Healthcare Common Procedure Coding System) defined in the Chronic Conditions Data Warehouse diabetes algorithm on at least one claim during 2016–2017.§§ Prevalence was calculated as the number of continuously enrolled beneficiaries with diabetes who had an eye exam claim in 2017 divided by the number of continuously enrolled beneficiaries with diabetes in that year. Eye exams were defined using CPT codes 92002, 92004, 92014, and 92014 and other evaluation and management visit CPT codes if the provider taxonomy codes indicated an eye care provider.¶¶ Unadjusted percentages are presented nationally and by state and race/ethnicity (white, black, Hispanic, Asian/Pacific Islander, American Indian/Alaska Native, and other). Age-standardized estimates, using direct standardization, were similar, and these data are not presented. Statistical testing was not performed because these data represent 100% of Medicare beneficiaries who met the inclusion criteria.

Among the 30,238,300 continuously enrolled Medicare Part B fee-for-service beneficiaries in 2017, a total of 8,341,000 (28%) had a diabetes diagnosis. The majority (72.4%) of these beneficiaries with a diabetes diagnosis were aged 65–84 years, with fewer aged 40–64 years (14.6%) or ≥85 years (12.1%). Overall, 73.3% of these beneficiaries were white, 13.0% were black, 8.3% were Hispanic, 3.5% were Asian/Pacific Islander, 0.8% were American Indian/Alaska Native, and 1.0% were other racial/ethnic groups.

Nationally, 54.1% of beneficiaries with diabetes had an eye exam in 2017 (Table). The prevalence ranged from 43.9% in Puerto Rico to 64.8% in Rhode Island. In seven states (Alabama, Alaska, Kentucky, Louisiana, Nevada, West Virginia, and Wyoming) and Puerto Rico, <50% of beneficiaries with diabetes received an eye exam (Table) (Figure 1). In nine states (Connecticut, Delaware, Hawaii, Iowa, Maine, Massachusetts, Nebraska, North Dakota, and Rhode Island) ≥60% of beneficiaries with diabetes had an eye exam in 2017.

Figure 1.

Percentage of Medicare Part B fee-for-service beneficiaries with diabetes who had an eye exam, by state — United States, 2017
Abbreviations: DC = District of Columbia; PR = Puerto Rico; USVI = U.S. Virgin Islands.

Nationally, the prevalence of having an eye exam was lower among Hispanic (48.2%) and black (48.9%) beneficiaries with diabetes than it was among whites (55.6%). This was also observed in 46 states and DC. Prevalence was higher among beneficiaries aged ≥85 years (58.6%) and 65–84 years (56.9%) than among those aged 40–64 years (38.0%) or 18–39 years (31.7%) (Figure 2).

Figure 2.

Percentage of Medicare Part B fee-for-service beneficiaries with diabetes who had an eye exam, by age group* — United States, 2017
*Data for beneficiaries aged 0–17 years were suppressed because of small sample size (≤100).

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....