Vision screening for preschoolers with commercial insurance: impact of geography

Abstract

Background

The American Academy of Pediatrics recommends pediatric vision screening to detect and refer vision disorders during the critical early years when intervention yields the greatest benefits. We determined the rate of vision screening for US children 3–5 years of age with commercial insurance and compared rates among those living in rural versus urban areas. Children in rural compared with urban areas were expected to have lower rates of vision screening.

Methods

A cross-sectional study using commercial claims for 3- to 5-year-olds derived from the Merative MarketScan Database (IBM, Armonk, NY), 2011–2020, was conducted. Primary outcome was the proportion of children with a claim for vision screening. Adjusted incident rate ratios (aIRR) of vision screening with 95% confidence intervals were computed for children living in rural compared with urban areas of the United States.

Results

Claims for 2,299,631 children were included. Most children (1,724,923 [75.0%]) were enrolled in preferred provider organization plans and lived in urban areas (2,031,473 [88.3%]). A total of 662,619 (28.8%) had a claim for a vision screening. Children living in rural versus urban areas had a lower adjusted incident rate of vision screening (15.1 vs 30.6%, aIRR 0.57; 95% CI, 0.53–0.61) after adjusting for sex, age, region, and insurance type.

Conclusions

For preschool age children with commercial insurance, vision screening is low, especially in rural compared with urban areas.

Poor vision in young children can cause developmental delays and poor academic outcomes, with negative implications into adulthood.^1–3 About 21% of US children 2–5 years of age have vision disorders that put them at risk of developing amblyopia.⁴ Delayed treatment of these disorders may result in permanent vision loss. ⁵ Treatment of vision disorders is most effective before the age of 8 years.^6,7

To facilitate the early detection of vision disorders in children, routine vision screening for children aged 3–5 years and referrals to eye care professionals for diagnosis and treatment of any identified vision disorders is recommended by the American Academy of Pediatrics (AAP) and other organizations that advocate for children’s health and sight.^5,8–13 The AAP recommends annual vision screening from age 3 to 5 years, whereas the United States Preventive Services Taskforce (USPSTF) recommends vision screening at least once over the same 3-year period. Vision screening is usually performed by primary care providers and trained lay persons as a relatively quick means to identify possible vision impairment.¹⁴ Vision screening tests are conducted primarily using an eye chart or various instrument-based tools (eg, Spot Vision Screener [Welch Allyn, Skaneateles Falls, NY], GoCheck Kids [Gobiquity Mobile Health, Scottsdale, AZ), iScreen [iScreen Vision Inc, Cordova, TN] and Plusoptix S12C [PlusoptiX Inc, Atlanta, GA]). Screening differs from comprehensive eye examinations, which are detailed, gold-standard examinations of the eye and visual system provided by an eye care professional.

The Affordable Care Act (ACA) requires commercial insurers to cover preventive care services recommended by the USPSTF (with a grade of A or B) with no cost sharing. This includes vision screening, which is recommended as effective with a grade B level of evidence (ie, “high certainty that the net benefit is moderate or there is moderate certainty that the net benefit is moderate to substantial”).^15,16 The ACA also considers pediatric vision care an essential health benefit; therefore, all new health insurance plans are required to cover comprehensive eye examinations for children.¹⁶ The Early and Periodic Screening, Diagnostic, and Treatment (EPSDT) benefit is the child health component of Medicaid. It covers a vision screening at each well-child visit, follow-up comprehensive eye examinations, and prescription glasses.¹⁷

Despite recommendations for routine vision screening and the availability of health insurance to cover costs, disparities in access to vision care exist for children based on socioeconomic status and type of health insurance.^18–25 Rural areas have comparatively fewer pediatricians, family physicians, and eye care professionals than urban areas.^{23,24,26–29} Because of these deficits in health care professionals, access to vision screening and comprehensive eye examinations may be more limited in rural areas.

We conducted an analysis to determine the rate of vision screening for 3- to 5-year-olds with commercial insurance and whether disparities exist for children based on rural versus urban residence. We hypothesized that children with commercial insurance who lived in rural compared to urban areas would have a lower rate of vision screening. The results of this study could inform the implementation of targeted, actionable interventions and programs to effectively address possible gaps in access to vision screening for US children living in rural areas.

Methods

Data for the study were obtained from The Merative MarketScan Commercial Claims Database (IBM, Armonk, NY), which contains a large, nationwide sample of claims data from 2011 to 2020 for individuals with commercial insurance.³⁰ Merative MarketScan contains demographic and medical claims data for more than 273 million individuals using data from employers, health plans, and government and public organizations across the United States.³⁰ For each enrollee, the claims database includes demographic information (ie, age, sex, rural versus urban residence) and detailed information about each clinical encounter: date and setting of service, provider type, International Classification of Diseases 9th and 10th revision Clinical Modification (ICD-9-CM and ICD-10-CM) diagnosis and procedure codes, and Current Procedural Terminology 4th edition (CPT-4) codes. The study was exempt from institutional review because it was a secondary analysis of a publicly available deidentified data set.

Children 3–5 years of age with continuous enrollment in a commercial health plan from 2011 to 2020 were included. Each child was followed for at least 1 year (if they dropped out of their health plan after a year) and at most 3 years. We chose to focus on 3- to 5-year-olds because USPSTF and AAP recommendations emphasize the importance of vision screening in that age range.^10,9 Children were excluded if they opted out of their commercial plan within a year (ie, before their fourth birthday) because it could not be determined whether they had a vision screening during their nonenrollment. Also excluded were children whose insurance plans were unknown (n = 591,029). There were 60 children who had basic insurance plans, and these were excluded from the analysis because of the small sample. Lastly, we excluded children without a known geographic location (n = 51,783).

Outcome Measures

The primary outcome of interest was the proportion of children that had a vision screening test within a 1- to 3-year time period, indicated by a claim submitted by an individual other than an eye care professional (ie, ophthalmologist or optometrist). The claim could be a visual acuity test using an eye chart (CPT 99173) or an instrument-based test with on-site (CPT 99177), or an instrument-based test with offsite analysis (CPT 99174) of test results. The primary outcome measure was determined using Current Procedural Terminology, Fourth Edition (CPT-4) procedure billing codes submitted by the provider.

The secondary outcome was the proportion of children that had a comprehensive eye examination, indicated by a claim submitted by an eye care professional identified using CPT-4 billing codes, specifically, 99201–5, 99211–5, 92002, 92004, 92012, 92014, 99241–5, 99221–5, 99251–5, and 99281–5.²¹

Predictors and Exposures

The primary exposure variable was the child’s residential location in a rural (metropolitan statistical area) or urban (not a metropolitan statistical area) area, as defined by the US Census Bureau.³¹ Other covariates included in the analysis were selected because of demonstrated and expected associations with the primary outcome and availability of the variable in the database.^19,32 These covariates included sex, age, type of commercial insurance plan (comprehensive, preferred provider organization [PPO], exclusive provider organization [EPO], health maintenance organization [HMO], point-of-service [POS]), and US geographical region (Northeast, South, West, Midwest).

Statistical Analyses

Statistical analyses were conducted using the R computing environment, version 4.3.0. Participant characteristics were summarized for the entire cohort using medians with interquartile ranges for continuous variables and frequencies, and percentages for categorical variables. The cohort was grouped based on a child’s rural or urban residential location. Multivariable negative binomial models were used to determine incident rate ratios (with 95% confidence intervals and P values) of a vision screening. Enrollment time (number of years the child was enrolled in a health plan) was included as an offset. Each model was adjusted for residence, sex, age, US geographic region, and type of commercial insurance coverage. We ran additional sensitivity analyses that included an interaction between commercial insurance type and rurality. For all analyses, P < 0.05 was considered statistically significant.

Results

The study sample included 2,299,631 unique US children. Table 1 provides a summary of the study cohort. Most children lived in urban areas (n = 2,031,473 [88.3%]) and were enrolled in the commercial insurance plan for 2 years. The majority of children (n = 1,724,923 [75.0%]) were enrolled with a PPO, and the least number of children (31,112 [1.4%]) had comprehensive insurance.

Table 1.

Descriptive characteristics of children 3- to 5-years of age with commercial health insurance living in rural versus urban areas of the US from the Merative MarketScan Commercial Claims Database (IBM, Armonk, NY), 2011–2020^a

Patient characteristics	Row total no. (column %)	Residence no. (row %)		P valueb
		Rural	Urban
Children in the cohort	2299631	268158 (11.7)	2031473 (88.3)
Vision screening testc				<0.001
Yes	662619 (28.8)	40467 (6.1)	622152 (93.9)
No	1637012 (71.2)	228673 (13.9)	1419193 (86.1)
Comprehensive eye exam				<0.001
Yes	153822 (6.7)	14505 (9.4)	139317 (90.6)
No	2145809 (93.3)	253653 (11.8)	1892156 (88.2)
Sex				0.38
Male	1183393 (51.5)	138208 (11.7)	1045185 (88.3)
Female	1116238 (48.5)	129950 (11.6)	986288 (88.4)
US geographical region				<0.001
Midwest	508672 (22.1)	82729 (16.3)	425943 (83.7)
South	921375 (40.1)	125147 (13.6)	796228 (86.4)
West	453039 (19.7)	36543 (8.1)	416496 (91.9)
Northeast	416545 (18.1)	23739 (5.7)	392806 (94.3)
Type of commercial insurance coverage				<0.001
Comprehensive	31112 (1.4)	2798 (9.0)	28314 (91.0)
EPO	50252 (2.2)	2468 (4.9)	47784 (95.1)
HMO	304659 (13.3)	19314 (6.3)	285345 (93.7)
POS	188685 (8.2)	15975 (8.5)	172710 (91.5)
PPO	1724923 (75.0)	227603 (13.2)	1497320 (86.8)

EPO, exclusive provider organization; HMO, health maintenance organization; POS, point of service plan; PPO, preferred provider organization.

^aData reported as no. (%); row percentages are reported for all patient characteristics, and column percentages are reported for the row total column.

^bχ² test of weighted percent comparing “Yes (Urban)” and “No (Rural).”

^cVision screening test is a claim for a vision screening test with an visual acuity chart (CPT 99173) or an instrument-based tool (CPT 99174, CPT99177) submitted by an individual other than an ophthalmologist or optometrist.

Overall, 662,619 (28.8%) children received a vision screening test. Of those, 498,707 (75.3%) children received a visual acuity test with an eye chart, and 197,815 (29.9%) received an instrument-based screening. Of the latter, 96,251 (48.7%) had onsite analysis and 107,516 (54.4%) had remote analysis. Children could have more than one type of vision screening test or instrument-based test. Most children received a visual acuity test with an eye chart (71.6%) as opposed to an instrument-based test (28.4%), especially in rural areas (80.6%) compared with urban areas (71.0%) (Figure 1). Of the 153,822 children (6.7% of total) who had a comprehensive eye examination (Table 1), 49,404 (32.1%) also had a vision screening test.

FIG 1.

Percentage of US children 3 to 5 years of age with a claim for a visual acuity or instrument-based test by rural or urban residence.

Adjusting for sex, age, region, and insurance plan type, children who lived in rural areas compared with urban areas had a lower adjusted incidence rate of a vision screening test (15.1% vs 30.6%, aIRR, 0.57; 95% CI, 0.53–0.61) (Table 2). Compared to the South, children living in the Midwest (aIRR; 0.71; 95% CI, 0.66–0.77), Northeast (aIRR; 0.90; 95% CI, 0.83–0.96), and West (aIRR, 0.80; 95% CI, 0.73–0.87) had a lower adjusted incidence rate of receiving vision screening (Table 2).

Table 2.

Adjusted incident rate ratios estimated between child’s residence (urban/rural) and commercial insurance coverage based on the Merative MarketScan Commercial Claims Database (IBM, Armonk, NY), 2011–2020^a

Exposure	Vision screening,b no. (%)	Multivariable analysis
		Adjusted IRR (95% CI)	P value
Children in the cohort	662619 (28.8)
Residence
Rural	40467 (15.1)	0.57 (0.53, 0.61)	<0.001
Urban	622152 (30.6)	Reference	—
Sex
Male	336930 (28.5)	0.98 (0.93, 1.04)	0.52
Female	325689 (29.2)	Reference	—
Age	—	1.07 (1.06, 1.08)	<0.001
US geographical region
Midwest	96569 (19.0)	0.71 (0.66, 0.77)	<0.001
Northeast	141672 (34.0)	0.90 (0.83, 0.96)	0.003
West	128622 (28.4)	0.80 (0.73, 0.87)	<0.001
South	295756 (32.1)	Reference	—
Type of commercial insurance coverage
Comprehensive	9242 (29.7)	0.81 (0.73, 0.89)	<0.001
EPO	15648 (31.1)	1.01 (0.91–1.12)	0.87
HMO	97442 (32.0)	1.02 (0.95–1.10)	0.56
POS	59322 (31.4)	0.94 (0.87–1.02)	0.14
PPO	480965 (27.9)	Reference	—

CI, confidence interval; aIRR, adjusted incident ratios; EPO, exclusive provider organization; HMO, health maintenance organization; POS, point of service plan; PPO, preferred provider organization.

^aFor each exposure, the incident rate ratio is adjusted for all other exposures, eg, aIRR for male is adjusted for residence, age, US geographical region, and type of commercial insurance coverage.

^bVision screening test is a claim for a vision screening test with an acuity chart (CPT 99173) or an instrument-based tool (CPT 99174, CPT99177) submitted by an individual other than an ophthalmologist or optometrist.

Compared to PPO insurance, children with comprehensive (aIRR 0.81; 95% CI, 0.73–0.89) insurance had a lower adjusted incidence rate of vision screening. Testing for interactions between a child’s residential location and type of commercial health insurance plan, we found that the adjusted incidence rate of a vision screening was lower for rural children who had comprehensive (aIRR 0.28; 95% CI, 0.23–0.32), EPO (aIRR 0.46; 95% CI, 0.39–0.54), and HMO (aIRR 0.68; 95% CI, 0.58–0.78) or POS (aIRR 0.47; 95% CI, 0.42–0.53) insurance compared to urban children with PPOs.

Discussion

We investigated how rural or urban location affects the rate of vision screening among US children with commercial health insurance. The findings from this study could inform the development of effective and equitable policies and programs to improve vision outcomes in childhood. Our results indicate that most preschoolers do not receive the recommended vision screening, even among those with commercial health insurance; children living in rural areas had a lower screening rate than urban children.

A prior study by Oke and colleagues³³ reported similar rates of vision screening for children aged 12–36 months with commercial claims across US census regions. Other studies suggest that rural compared to urban areas have fewer healthcare resources, including fewer facilities, fewer healthcare professionals, lower socioeconomic status, and limited options with regards to health insurance.^34–39 Therefore, the disparities in access to vision screening may be due to limited healthcare resources in rural areas, including a shortage of primary and eye care professionals, which may represent a disincentive for vision screening. Providers may only screen, for example, if parents have specific concerns about their children’s vision or based on the availability of eye care professionals in their area.

Telehealth has made strides in improving access to eye care in rural areas and may prove useful in addressing the rural shortages of primary and eye care professionals that may disincentivize vision screening.^40–44 However, rural providers and patients may be unable to afford digital technology, or they may not have access to the infrastructure and technical expertise to implement and operate these technologies. A more effective approach to bridging the gaps in access to vision screening for children may be incentives and pipeline programs to increase the primary and eye care professional workforce in rural areas.

Children with comprehensive insurance had a lower adjusted incidence rate of a vision screening than children with PPO insurance, which provides some coverage for out-of-network providers, unlike other commercial insurance plans and does not require referrals from primary care physicians to see specialists.^45–47 When we tested for interactions between a child’s residential location and type of commercial health insurance, the negative effect of rurality was exacerbated in children with insurance other than PPOs (ie, EPO, HMO, POS). PPOs may be useful in rural areas that may have fewer health care providers³⁸; yet PPOs have higher premiums and deductibles than other managed care plans.⁴⁸ Thus, an alternative explanation for our finding is that those who enroll in PPOs may have higher socioeconomic status and high health literacy compared with those who opt for other commercial health plans, contributing in turn to higher utilization of vision screening. Similar disparities, even in populations with access to health insurance, have been documented in the literature.^19,20

Compared to children living in the South, children in other regions of the United States had a lower adjusted incidence rate of vision screening. These results are supported by the literature. Vision screening requirements for preschool children vary across US states.^49,50 Children in states with vision screening requirements (legislation, rule, or code) are more likely to receive vision screening compared with states having no requirements.^49,50 Also, higher vision screening rates in the South and Northeast compared with other US regions may be due in part to the higher rates of reimbursement and coverage for instrument-based screening.^33,51

Strengths of this study include its use of a large, multistate sample of claims data for over 2 million commercially insured children, with demographic and claims data from employers, health plans, government, and public organizations across the United States.³⁰ However, our study may underestimate the proportion of children receiving vision screening. Providers may have performed vision screening without using the appropriate screening codes because of human error and low or no reimbursements.^51,52 Some providers are unable to bill for vision screening separately, because it is bundled with the global code for well-child visits.^52,53 Regardless of these limitations, however, there is no reason to believe that the lack of coding for vision screening would vary independently in rural compared to urban areas. It is also important to note that some parents with visually impaired children may already have an established eye care professional and therefore opt out of vision screening with their pediatrician. From our analysis, we found that 6.7% of children had a visit with an eye care professional, and a third of these had also undergone vision screening. Another limitation of our study is that we could not measure some potential confounding variables, for example, race/ethnicity, household income, parental education, and others, because they were not available in the claims dataset. This may have resulted in an exaggeration of the adjusted association between rural versus urban residence and vision screening.

In conclusion, less than one-third of US 3- to 5-year-olds with commercial insurance received a vision screening in line with recommendations. Living in a rural compared to an urban area was associated with less vision screening, more so for children with comprehensive commercial insurance than for children with PPOs. Disparities in the utilization of vision screening for children in rural versus urban areas may lead to delays in the treatment of vision disorders. Targeted, evidence-based interventions are needed to improve the utilization of vision screening for children in rural areas. Effective interventions would ensure early treatment of vision disorders and prevent permanent vision loss, which could otherwise further exacerbate the socioeconomic divide between rural and urban children.