PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Pediatr Ophthalmol Strabismus. Author manuscript; available in PMC 2007 December 12.
Published in final edited form as:
J Pediatr Ophthalmol Strabismus. 2007; 44(6): 356–362.
PMCID: PMC2132442
NIHMSID: NIHMS21567

Use of Corrective Lenses among Adolescents: Findings from the National Health and Nutrition Examination Survey

Alex R. Kemper, MD, MPH, MS,* James G. Gurney, PhD, Maya Eibschitz-Tsimhoni, MD,†† and Monte A. DelMonte, MD††

Abstract

Purpose

To evaluate demographic variations in the use of corrective lenses among adolescent children.

Methods

Cross-sectional analysis of 3,916 children 12-18 years who participated in the 1999-2002 National Health and Nutrition Examination Survey (NHANES) vision examination component, which included: questions regarding use of corrective lenses; distance visual acuity, with corrective lenses if available; and non-cycloplegic autorefraction. Results reflect population-level estimates.

Results

Overall, 32.2% (95% confidence interval [CI]: 29.5%-35.0%) reported wearing corrective lenses. Girls and those with any private insurance had greater adjusted odds of reporting wearing corrective lenses. In contrast, children aged 15 through 18 years versus those 12 through 14 years, and white children compared to Black or Hispanic children had greater adjusted odds of actually having them available at the time of NHANES participation. Although 12.6% (95% CI: 8.8%-16.3%) of those who did not have their previously prescribed corrective lenses available had 20/25 or better distance visual acuity in both eyes without correction, 26.9% (95% CI:21.6%-32.1%) with their corrective lenses had distance visual acuity of 20/40 or worse in at least one eye when using their corrective lenses.

Limitations

Near visual acuity was not measured and children with corrective lenses available only had their corrected distance visual acuity measured. No data regarding the accuracy of the NHANES assessment of distance visual acuity are available. Autorefraction was performed without cycloplegia.

Conclusions

Many adolescent children report wearing corrective lenses. Variations across demographic characteristics appear to be due to a combination of undertreatment, overtreatment, and compliance with previously recommended corrective lenses.

Keywords: refractive errors, adolescent health services, eyeglasses

Introduction

Previously we described large variations in the receipt of vision care, including the use of corrective lenses, by gender, race/ethnicity, family income, and insurance status.1-3 We found a lower likelihood of vision care among boys, minority children, children with low family income, and the uninsured.

Disparities in healthcare are common.4 A key challenge to overcoming these disparities is understanding the underlying reasons why such differences in care occur. Perceived disparities in the receipt of vision care may simply reflect variations in the prevalence of refractive error. For instance, gender and race differences in the prevalence of childhood visual impairment have been identified in population-based studies in Australia and in several developing countries.5-7 The Australian study also found an association between lower socioeconomic status and the likelihood of visual impairment.8 A school-based study in the US found differences in refractive error by race/ethnicity, but did not evaluate the relationship between visual status and prior receipt of eye care.9

Differences in care may also result from undertreatment, either due to lack of recognition of the problem, incomplete treatment, or difficulty with compliance.10 Children with poor access to medical care or limited resources to purchase or replace corrective lenses may be especially at risk for undertreatment; however, little is known about the degree to which such barriers affect the receipt of eye care. A recent study of 12-year-old Australian children found that undertreatment was uncommon.8

Finally, differences in care may reflect overtreatment of clinically insignificant refractive error.11 The study of 12-year-old Australian children found that 38% of those with corrective lenses had no significant refractive error or other visual impairment.8 Evaluation of overtreatment is difficult because there is no clear consensus among eye care providers on the proper threshold for treatment. Overtreatment is perceived to be exacerbated by the financial reward that some eye care providers receive for prescribing corrective lenses. If true, overtreatment may be more common among populations with very good access to eye care services, such as children with insurance coverage for vision services, those with high family incomes, or those living in communities with high concentrations of eye care professionals.

To gain insight into the sources of variation in the use of corrective lenses among children, we evaluated recently released data from the National Health and Nutrition Examination Survey (NHANES), an ongoing epidemiological study of health in the United States conducted by the Centers for Disease Control and Prevention (CDC).12 NHANES provides the only nationally representative population-based data regarding childhood vision in the United States. NHANES includes data collected on the refractive status of adolescent children, distance visual acuity, use of corrective lenses, and lens strength. Unfortunately, there are important limitations to these data: children less than 12 years of age were not included in this component of NHANES; near visual acuity was not measured; no data are available regarding the accuracy of the distance visual acuity test used; children with corrective lenses available only had their corrected distance visual acuity measured; and refractive status was measured using an autorefractor without cycloplegia, which may lead to misclassification.13 Our primary goal was to determine the degree to which demographic characteristics were associated with use of corrective lenses. Our secondary goal was to gain insight into the degree to which any observed variations were related to differences in visual status as measured in NHANES.

Methods

Data Source

Data were drawn from the 1999-2002 National Health and Nutrition Examination Survey (NHANES). NHANES is based on a complex multistage national probabilistic sample of individuals in the United States. Health interviews are conducted in respondents' homes and examinations are conducted in mobile examination centers. Poststratification sampling weights are provided by NHANES to adjust the demographic data to Census data and to correct for nonresponse.12

Subjects

Children 12 through 18 years of age who participated in the vision examination component of NHANES were included in this analysis. Because we were interested in understanding whether there were differences in prevalence, undertreatment, or overtreatment by race/ethnicity, we included only those individuals who reported themselves as white, black, or Hispanic. Overall, 3,916 children were included in this study. We excluded the “other race” category because there were insufficient numbers (n=157) to allow for reliable evaluation of the relationship between “other race” and the main outcome measures.

Vision Examination

First, subjects were interviewed to determine if they had corrective lenses for distance visual acuity (“Do you wear glasses or contact lenses for distance vision, such as when you watch television?”) and, if so, whether their lenses were currently available at the time of their participation in NHANES. If glasses were available, the prescription of each lens was measured. Visual acuity was assessed with a non-logMAR chart; four correct out of five optotypes were required to pass a line. All subjects had their distance visual acuity assessed, using corrective lenses if available. Next, each subject underwent non-cycloplegic autorefraction with a Nidek Autorefractor/Keratometer (model ARK-760A). Subjects whose distance visual acuity, with corrective lenses if available, was worse than 20/25 in either eye had their distance visual acuity reassessed with correction predicted by the autorefractor. All data were collected by trained technicians. More specific details of the vision examination component of NHANES are available elsewhere.14

Outcome Measures

The main outcome variable was use of corrective lenses. We separately evaluated whether a subject reported wearing corrective lenses and whether the corrective lenses were available at the time of participation in NHANES. We considered corrective lens availability to be a measure of adherence.

Demographic Characteristics

The demographic characteristics we considered were age, gender, race/ethnicity, family income, and health insurance status. Age was categorized as 12 through 14 years or 15 through 18 years to reflect school-level (middle school vs. high school). We dichotomized children by race/ethnicity into non-Hispanic white versus black or Hispanic. Family income was categorized as <200% of the federal poverty level or ≥200% of the federal poverty level.15Health insurance was classified as none, any private insurance, or only public insurance.

Classification of Visual Status

Because autorefraction without cycloplegia can lead to misclassification, we used findings from this component of NHANES to dichotomize subjects by the presence of potential ametropia using a broad and inclusive definition. Based on a previous epidemiological study, we considered children to have absence of evidence of ametropia if the spherical equivalent (SE) refraction in each eye was >−0.50 Diopter (D) to <+2.0 D.7 To be consistent with our definition of evidence of ametropia, we considered corrective lenses to be potentially significant if either lens was ≤ −0.5 D SE or ≥ +2.0 D SE. These criteria were selected to be conservative, so that misclassification would be biased towards the presence of ametropia and significant lenses. We considered poor distance visual acuity to be 20/40 or worse in either eye, and good distance visual acuity to be 20/25 or better in both eyes.

Analysis

We used χ2 tests of association to determine whether potential ametropia varied by the demographic characteristics. The association between use of corrective lenses and the demographic characteristics were evaluated using χ2 tests of association and logistic regression modeling. We used χ2 tests of association to evaluate the relationships visual status and the demographic characteristics among those who did and did not report wearing corrective lenses. Logistic regression modeling was used to evaluate the relationship between the demographic characteristics and poor visual acuity among those without corrective lenses.

Stata 8 software (College Station, TX) was used for all statistical analysis. All results were adjusted for the complex multistage sampling design and by the poststratification weights to reflect population-level estimates. We considered p<0.05 to be statistically significant. Missing values were excluded in the bivariate analyses, and individuals with any missing demographic data were excluded from the logistic regression modeling. To ensure reliable statistical estimates, we only report results if the unweighted sample size was at least 30 individuals per cell. This research was approved by the University of Michigan Medical School Institutional Review Board.

Results

Demographic Characteristics and Potential Ametropia

Table 1 lists the weighted distribution of the demographic characteristics among the children included in this study. Based on the non-cycloplegic autorefraction measures, 49.3% (95% CI: 47.1%-51.6%) of these children had potential ametropia. Potential ametropia was not statistically associated with age (p=0.09), gender (p=0.26), race/ethnicity (p=0.15), family income (p=0.38), or health insurance status (p=0.80).

Table 1
Weighted distribution of the demographic characteristics.

Use of Corrective Lenses

Overall, 32.2% (95% CI: 29.5%-35.0%) reported wearing corrective lenses. Of those who reported wearing corrective lenses, most (74.5% [95%CI: 72.2%-77.8%]) had them available during the examination. Table 2 lists the association between the demographic characteristics and the use of corrective lenses.

Table 2
Percentage (%) and adjusted odds ratios (aOR) of reporting wearing corrective lenses and of having corrective lenses available among those who reported wearing them, by demographic characteristic.

The proportion who reported wearing corrective lenses was greater among older children, girls, white children, and those with higher family income. More children with any private insurance reported wearing corrective lenses than uninsured children (p<0.01) or those with only public insurance (p=0.03). However, there was no statistically significant difference between children with public insurance and uninsured children (p=0.20). After adjusting for all demographic factors, girls and those with any private insurance had statistically significant greater odds of reporting they wear corrective lenses.

Among those who reported wearing corrective lenses, a greater proportion of white compared to black or Hispanic children, of those with higher compared to lower family income, and of those with any private compared to those with only public insurance (p<0.01) had their corrective lenses available. The differences were not statistically different between those with any public insurance and those who had no insurance (p=0.22) or between those with no insurance and those with private insurance (p=0.34). After adjusting for all demographic characteristics, older children and white children had statistically significant greater odds of having their corrective lenses available.

Visual Acuity Among Those Who Reporting Not Using Corrective Lenses

Among those who reported not wearing corrective lenses, 19.6% (95%CI: 17.5%-21.6%) had distance visual acuity in at least one eye that was 20/40 or worse. Most (78.1% [95% CI: 73.4% - 82.8%]) children with poor uncorrected distance visual acuity had evidence of potential ametropia based on autorefraction. Providing correction based on findings from autorefraction decreased the proportion with uncorrected poor distance visual acuity by 65%, from 19.6% to 6.8% (95% CI: 5.3%-8.2%)

Table 3 presents the proportion and adjusted odds ratios of having poor distance visual acuity among those who reported not wearing corrective lenses by each of the demographic characteristics. A lower proportion of white children, those with higher family income, and privately insured children (p=0.02 vs. no insurance; p<0.001 vs. only public insurance) had poor visual acuity. However, after adjusting for all demographic characteristics, only lower family income was associated with greater odds of poor uncorrected visual acuity.

Table 3
Percentage (%) and adjusted odds ratios (aOR) of having poor distance visual acuity (20/40 or worse) among those without corrective lenses, by demographic characteristic.

Visual Acuity Among Those Who Reported Using Corrective Lenses

Among those who had corrective lenses available, 26.9% (95% CI: 21.6%-32.1%) had distance visual acuity of 20/40 or worse in at least one eye. After correction based on autorefraction, most (61.7% [95% CI: 52.8%-70.6%]) improved. There were no differences across the demographic characteristics in the likelihood of poor visual acuity among these children with corrective lenses either before or after correction.

Almost all (95.2% [95% CI: 93.2%-97.1%]) children who had corrective lenses available had potentially significant lenses. There were too few children who did not have potentially significant lenses (n=36) to explore reliably the association between the use of such lenses and visual acuity.

Of those who wore corrective lenses but did not have them available at the time of NHANES participation, most (76.7% [95% CI: 70.9%-82.5%]) had poor distance visual acuity, most of whom improved (71.3% [95% CI 63.9%-78.8%]) with correction. However, 12.6% (95% CI: 8.8%-16.3%) of those who did not have their corrective lenses available had 20/25 or better distance vision in both eyes. There were insufficient numbers of cases with good visual acuity among children who wore corrective lenses but did not have them available (n=43) to evaluate differences by demographic characteristics.

Discussion

Based on our findings from this nationally-representative dataset, nearly one-third of adolescent children report wearing corrective lenses for distance vision. The proportion who reported that they wear corrective lenses varied across all of the demographic characteristics we considered, including age, gender, race/ethnicity, family income, and health insurance status. After adjusting for these demographic characteristics, only girls and those with any private insurance had greater odds of wearing corrective lenses. However, among those children who reported wearing corrective lenses, younger children, black or Hispanic children, and those with low family incomes had lower odds of having their corrective lenses available at the time of NHANES participation. No information is available in these data on why the lenses were not available. Importantly, most that did not have their corrective lenses available had poor vision that improved with correction as predicted by autorefraction. Future studies will be needed to understand the reasons children do not use prescribed corrective lenses. Reasons could include issues with the corrective lenses (e.g., inappropriate correction due to changes in ametropia), behavioral and social concerns (e.g., refusal to wear spectacles due to perception of appearance), or economic barriers (e.g., costs of follow-up eye care, costs of replacing lenses to either changed prescription or lost lenses, or costs of more “stylish” frames).

Nearly 20% of the children who reported not wearing corrective lenses and about 25% of those who reported that they wear corrective lenses and had them available had visual acuity 20/40 or worse in at least one eye. Although the distance acuity test may have misclassified some children, most children with poor vision improved with correction by autorefraction. After adjusting for the demographic characteristics, lower family income was associated with poor visual acuity as measured in NHANES among those without corrective lenses, perhaps reflecting the economic barriers to the receipt of vision care.

NHANES provides only limited insight into potential overtreatment because children with corrective lenses available had measurement only of their corrected distance visual acuity, instead of measuring both corrected and uncorrected distance visual acuity. Although all children underwent autorefraction, there is likely misclassification because cycloplegia was not used. However, we did find that nearly 13% of children who reported wearing corrective lenses but who did not have them available had 20/25 or better distance vision in both eyes. At least some of this may be accounted for by correction for near vision problems.

Although there are important limitations in the design of the childhood vision component of NHANES, the data suggest that both undertreatment and overtreatment contribute to the variations we have observed in the use of corrective lenses across demographic characteristics. To develop effective strategies to ensure appropriate use of corrective lenses, future studies are needed to comprehensively understand barriers to appropriate care and the functional harm of undertreatment, such as the development of amblyopia or poor school performance, versus the consequences of overtreatment, including wasting limited healthcare resources. Equally important will be understanding the social and behavioral pressures that discourage adolescents from compliance with corrective lenses.

Footnotes

Financial Support: National Eye Institute (K23-EY14023).

References

1. Kemper AR, Bruckman D, Freed GL. Receipt of specialty eye care by children. Ambul Pediatr. 2003;3:270–274. [PubMed]
2. Kemper AR, Bruckman D, Freed GL. Prevalence and distribution of corrective lenses among school-age children. Optom Vis Sci. 2004;81:7–10. [PMC free article] [PubMed]
3. Kemper AR, Cohn LM, Dombkowski KJ. Patterns of vision care among Medicaid-enrolled children. Pediatrics. 2004;113:e190–e196. [PMC free article] [PubMed]
4. Agency for Healthcare research and Quality 2004 National healthcare disparities report. U.S. Department of Health and Human Services; Rockville, MD: 2004.
5. Maul E, Barroso S, Munoz SR. Refractive error study in children: results from La Florida, Chile. Am J. Ophthalmol. 2000;129:445–454. [PubMed]
6. Pokharel GP, Negrel AD, Munoz SR, Ellwein LB. Refractive error study in children: results from Mechi Zone, Nepal. Am J Ophthalmol. 2000;129:436–444. [PubMed]
7. Robaei D, Rose K, Ojaimi E, Kifley A, Huynh S, Mitchell P. Visual acuity and the causes of visual loss in a population-based sample of 6-year-old Australian children. Ophthalmology. 2005;112:1275–1182. [PubMed]
8. Robaei D, Kifley A, Rose KA, Mitchell P. Refractive error and patterns of spectacle use in 12-year-old Australian children. Ophthalmology. 2006;113:1567–1573. [PubMed]
9. Kleinstein RN, Jones LA, Hullett S, et al. Refractive error and ethnicity in children. Arch Ophthalmol. 2003;121:1141–1147. [PubMed]
10. Fiscella K, Franks P, Gold MR, Clancy CM. Inequality in quality: addressing socioeconomic, racia, and ethnic disparities in health care. JAMA. 2000;283:2579–2584. [PubMed]
11. Donahue SP. How often are spectacles prescribed to “normal” preschool children? J AAPOS. 2004;8:224–229. [PubMed]
12. National Center for Health Statistics National Health and Nutrition Examination Survey. Available at: http://www.cdc.gov/nchs/nhanes.htm, accessed Sept. 14, 2006.
13. Zhao J, Mao J, Luo R, Li F, Pokharel GP, Ellwein LB. Accuracy of noncycloplegic autorefraction in school-age children in China. Optom Vis Sci. 2004;81:49–55. [PubMed]
14. National Center for Health Statistics Vision procedures manual. Available at: http://www.cdc.gov/nchs/data/nhanes/vi.pdf, accessed Sept. 14, 2006.
15. US Census Bureau Poverty Thresholds. Available at: http://www.census.gov/hhes/www/poverty/threshld.html, accessed Sept. 14, 2006.