To describe trends in vision screening based on insurance claims for young children in the United States.
This cross-sectional study used administrative claims data from the 2010–2019 IBM MarketScan Commercial Claims and Encounters Database. We included children aged 1 to <5 years at the beginning of each calendar year. The primary outcome was a vision screening claim within 12 months for chart-based or instrument-based screening. Linear regression was used to evaluate trends over time in vision screening claims and practitioner payment.
This study included a median of 810 048 (interquartile range, 631 523 – 1 029 481) children between 2010 and 2019 (mean [standard deviation] age, 2.5 [1.1] years; 48.7% female). The percentage of children with vision screening claims increased from 16.7% in 2010 to 44.3% in 2019 (difference, 27.5%; 95% confidence interval, 27.4% to 27.7%). Instrument-based screening claims, which were identified in <0.2% of children in 2010, increased to 23.4% of children 1 to <3 years old and 14.4% of children 3 to <5 years old by 2019. From 2013 to 2018, the average of the median practitioner payment for instrument-based screening was $23.70, decreasing $2.10 per year during this time (95% confidence interval, $0.85 to $3.34; P = .009).
Vision screening claims among young children nearly tripled over the last decade, and this change was driven by increased instrument-based screening for children aged <3 years. Further investigation is needed to determine whether the decreasing trends in practitioner payment for screening devices will reduce the adoption of vision screening technology in clinical practice.
Vision screening is an important component of pediatric preventive care, and instrument-based devices may enable vision screening among children unable to perform chart-based evaluations.
Instrument-based vision screening claims have increased over the last decade in the United States, particularly among children <3 years of age. However, practitioner payment for vision screening has decreased during this time interval.
Vision screening is an important component of preventive care delivered to young children in the primary care setting.1 Both chart-based and instrument-based approaches to vision screening are supported by the American Academy of Pediatrics and the American Academy of Ophthalmology for children <5 years old.2 However, screening preschool-aged children by using chart-based techniques is often challenging and may be inconsistently successful.3 Instrument-based screening can provide quicker and simpler identification of risk factors for vision loss, particularly among younger children for whom eye chart evaluations can be tedious or unfeasible.4 The implementation of vision screening devices in the primary care setting has been found to improve the rates of completed vision screening at well-child visits.5
The implementation of instrument-based screening and the impact of these devices on the landscape of preventive vision care delivery has not been well characterized. Among children aged 1 to <3 years, there is evidence of geographic variation in the use of instrument-based screening, and rates of screening vary on the basis of practitioner payment and patient out-of-pocket expenses.6 However, the trends over time in payments received by practitioners for chart-based and instrument-based screening, as well as the proportion of young children receiving each type of vision screening, have not been described by using a nationwide database. In this study, we report the trends in vision screening approach and practitioner payment in the United States over the last decade based on insurance claims data.
Methods
Study Design and Data Source
This cross-sectional study was performed by using administrative claims data from a large database of privately insured patients in the United States (2010–2019 IBM MarketScan Commercial Claims and Encounters Database). The MarketScan database includes several million individuals annually, encompassing employees, their spouses, and dependents who are covered by employer-sponsored private health insurance plans in the United States.7 The study protocol involving the analysis of deidentified insurance claims was exempted by the Harvard Pilgrim Health Care Institutional Review Board and adhered to the tenets of the Declaration of Helsinki. We followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guidelines.8
Study Population
We included children 1 to <5 years of age at the beginning of each calendar year with at least 1 preventive care encounter during the year, identified using Current Procedural Terminology (CPT) codes (CPT 99381-99383 and 99391-99393). We excluded children enrolled in capitated insurance plans (in which a fixed payment per member per unit of time is paid in advance to the physician for health care service delivery) and with <1 year of continuous enrollment (with a 45-day grace period). Demographic data collected included age, sex, and geographic region of residence.
Outcomes
Vision screening-related claims were categorized into chart-based screening (CPT 99173) or instrument-based screening, the latter involving either a device with off-site interpretation (CPT 99174) or on-site interpretation (CPT 99177). Off-site interpretation refers to devices relying on a remote analytic center to interpret screening data, whereas on-site interpretation refers to devices providing an instantaneous report of referral recommendations. The primary outcome was the percentage of children with a vision screening claim during each calendar year (January 1 to December 31). We stratified the children by age (1 to <3 and 3 to <5 years) at the beginning of each calendar year. The secondary outcomes included the percentage of vision screening claims without separate, unbundled practitioner payment and the practitioner payment among vision screening claims with payment.
Statistical Analysis
We reported means, medians, and interquartile ranges for continuous variables and frequency and proportions for categorical variables. We used linear regression to evaluate trends over time in vision screening and practitioner payment. All costs were adjusted to 2018 US dollars by using the Consumer Price Index.9 Analyses were performed by using R, version 4.2.0 (R Core Team, 2022), with a significance level of 0.05 for a two-tailed test.
Results
We analyzed the insurance claims of a median of 810 048 children (interquartile range, 631 523 – 1 029 481) each year. The mean age was 2.5 years (standard deviation, 1.1 years), and 48.7% were female. In our sample, 37.9% of children resided in the South, 24.3% in the Midwest, 19.0% in the Northeast, and 17.1% in the West census regions. The percentage of children with vision screening claims increased from 16.7% in 2010 to 44.3% in 2019 (difference, 27.5%; 95% confidence interval [CI], 27.4% to 27.7%).
Approximately 5% of children 1 to <3 years old and 30% of children 3 to <5 years old had chart-based screening claims each year over the study period. Instrument-based screening claims, which were identified in <0.2% of all children in 2010, increased to 23.4% of children 1 to <3 years old and 14.4% of children 3 to <5 years old by 2019 (Fig 1).
Vision screening claims of privately insured children (aged 1 to <5 years) between 2010 and 2019 in the United States.
Vision screening claims of privately insured children (aged 1 to <5 years) between 2010 and 2019 in the United States.
The percentage of vision screening claims without practitioner payment for chart-based screening did not significantly change over time from 2010 to 2018 (0.1% per year; 95% CI, -0.6% to 2.6%; P = .18) and decreased for instrument-based screening (−2.8% per year; 95% CI, −4.9% to -0.7%; P = .017) (Fig 2A). The average of the median practitioner payment for chart-based screening was $3.72 and did not change over time ($0.03 increase per year; 95% CI, −$0.02 to $0.09; P = .18). From 2013 to 2018, the average of the median payment for instrument-based screening was $23.70, decreasing $2.10 per year during this time (95% CI, $0.85 to $3.34; P = .009; Fig 2B).
(A) The percentage of vision screening claims without separate, unbundled practitioner payment and (B) the median practitioner payment per vision screening claim in 2018 US dollars. Payment data were not available for the 2019 calendar year.
(A) The percentage of vision screening claims without separate, unbundled practitioner payment and (B) the median practitioner payment per vision screening claim in 2018 US dollars. Payment data were not available for the 2019 calendar year.
Discussion
The percentage of children <5 years of age with vision screening claims in the primary care setting increased by approximately 3-fold over the last 10 years. This change has been driven by an increase in the use of instrument-based vision screening devices among children <3 years old. During this decade, payment for chart-based screening did not change. However, there was a decline in payment for instrument-based screening despite advances in vision screening technology and a growing landscape of available devices demonstrating improved performance.
Billing codes for instrument-based vision screening were first introduced in 2008 and expanded in 2013 to specify devices using off-site interpretation (CPT 99174) and on-site interpretation (CPT 99177).10 Most newer screening devices use on-site interpretation and provide instantaneous results; however, practitioner payment is lower for claims using these devices compared with those involving off-site interpretation. The lower payment for devices providing instantaneous results may contribute to the decrease in overall payment for instrument-based screening and discourage further innovation in vision screening technology. Practitioner payment has been shown to vary geographically, and lower payment may discourage screening with these devices.6 Fluctuations in coding practices over time may reflect delayed adoption of new codes by practitioners and response to insurer coverage.10
Many insurance programs argue that vision screening is a mandatory component of a well-child visit and do not pay for additional testing. Approximately 1 in 5 claims in the most recent calendar year did not include a separate, unbundled practitioner payment. Trends in insurance coverage and payment may influence the adoption of vision screening devices among practitioners.11 The 2016 American Academy of Pediatrics and American Academy of Ophthalmology joint policy statement supports the use of instrument-based vision screening starting at 12 months of age, with annual evaluations recommended until the children reach a developmental stage in which they can participate reliably in chart-based visual acuity assessment.2 In 2017, the US Preventive Services Task Force (USPSTF) released an updated statement on visual screening,12 which included a Grade B recommendation for children aged 3 to 5 years, indicating a moderate to substantial net benefit. For children aged 1 to 3 years, the USPSTF provided an “I” statement, issued when there is not enough evidence or when the available evidence is of poor quality or conflicting. As a result, the USPSTF is unable to assess the balance of benefits and harms and cannot make a recommendation for or against providing the preventive service.13 Chart-based screening of preschool children, especially those <4 years of age, is unreliable,14,15 whereas instrument-based screening has been shown to be more reliable than visual acuity screening.15 Despite the USPSTF statement, the percentage of children with vision screening claims has increased more in the <3-year-old population than the 3 to <5-year-old population, which may reflect the role of instrument-based screening devices in children who are otherwise unable to participate in chart-based evaluations. The potential implications of this paradigm shift in vision screening on the diagnosis of vision-threatening conditions and the practice patterns of pediatricians and eye specialists deserve further investigation.
There are several limitations of the data source and analytic approach to consider. First, while the MarketScan database is large and multi-state, it is not nationally representative, and consists of a privately insured population that excludes uninsured children and those enrolled in public insurance plans. As a result, the study cohort may underrepresent socioeconomically disadvantaged children who may be less likely to receive preventive care. Conversely, the underutilization of screening codes by practitioners due to lack of payment by some insurance for submitted codes may contribute to a lower estimated screening rate from this claims-based analysis compared with self-reported practices.11 Given lags in the adoption of new screening codes, the billing codes reported in this dataset may not always reflect the actual device used for screening.10 How individual insurance plans handle vision screening within global payments for preventive care is unclear and cannot be determined in our dataset. Low payment or payment bundled with the preventive visit may disincentivize practitioners from using vision screening billing codes, even when the service may have been provided. In this analysis, we excluded children enrolled in capitated insurance plans in which services such as vision screening may be part of per-member-per-month global payments; however, there may be other non-capitated insurance plans in our study in which payment for the vision screening claim may be part of per-member-per-month global payments and may not be separately captured in the database, leading to an underestimation of physician payment for vision screening. In addition, the insurance claims dataset does not include screening results to be able to measure subsequent referrals to eye specialty care. Finally, this study focuses on vision screening trends before the coronavirus disease 2019 pandemic, which disrupted the delivery of preventive care beginning April 2020.16 The degree to which vision screening in the primary care setting was impacted by the pandemic is not captured in this analysis and warrants further investigation.
In conclusion, vision screening claims among young children have increased significantly over the last decade, and this change has been driven by instrument-based screening device use for children <3 years old. Further investigation is needed to determine whether the decreasing trends in practitioner payment for screening devices will impact further innovation in vision screening technology or the more widespread adoption of vision screening devices in clinical practice.
Acknowledgments
We would like to thank Amy Han, MPH, for assistance with project management.
Dr Oke conceptualized and designed the study, conducted the analyses, drafted the manuscript, and reviewed and revised the manuscript; Dr Lutz reviewed and supervised the analysis, reviewed and revised the manuscript, critically reviewed the manuscript for important intellectual content; Dr Hunter reviewed and supervised the analysis, reviewed and revised the manuscript, provided supervision and administrative support, critically reviewed the manuscript for important intellectual content; Dr Galbraith conceptualized and designed the study, reviewed and supervised the analysis, reviewed and revised the manuscript, provided supervision and administrative support, revised and critically reviewed the manuscript for important intellectual content; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
COMPANION PAPER: A companion to this article can be found online at www.pediatrics.org/cgi/doi/10.1542/peds.2023-062804.
FUNDING: Isdin Oke received grant funding from the Agency for Healthcare Research and Quality (T32HS000063).
CONFLICT OF INTEREST DISCLOSURES: David G. Hunter reports involvement with Rebion, Inc (founder, equity) and Luminopia, Inc (advisor, equity). The remaining authors have indicated they have no potential conflicts of interest relevant to this article to disclose.
Comments