To examine racial and ethnic differences in unmet need for vision care among children with special health care needs.
Cross-sectional data from the National Survey of Children with Special Health Care Needs were analyzed. The analytic sample was 14 070 children with special health care needs who needed eyeglasses or vision care in the previous year. Children who did not receive all the eyeglasses or vision care they needed were considered to have unmet need for vision care.
Of the sample, 5.8%, representing approximately 198 600 children with special health care needs in the United States, had unmet need for vision care. Rates of unmet need ranged from 2.5% to 14.3% across the 7 racial and ethnic groups studied. Relative to whites, children of African American, Latino, and multiracial backgrounds had approximately twice the adjusted risk of unmet need, whereas American Indian or Alaskan Native children had a lower adjusted risk. Health care providers, school personnel, insurance coverage, and other factors also contributed to differences in the risk of unmet need, independently of child race/ethnicity.
Further research is needed to explain and address the causes of racial and ethnic differences in unmet need for vision care among children with special health care needs.
Studies1- 3 from several countries indicate that children with vision deficits often have multiple disabling conditions. For instance, among all visually impaired children born prematurely in Finland during a 17-year period, 36% had epilepsy, 54% had cerebral palsy, and 66% had other serious disabilities that would qualify them as children with special health care needs (CSHCN).3 Because CSHCN are often unable to report what they can see during conventional eye examinations, they may require more extensive and individually tailored treatment of their vision than children without special needs.4 Unfortunately, some CSHCN do not receive needed eyeglasses or other types of vision care, which places them at great risk of long-term delays in their educational and social development.5
Racial/ethnic disparities in access to vision care have been documented among children without special needs. In the National Health Interview Survey, rates of vision care use by African Americans (20%) and Latinos (18%) were lower than that of non-Hispanics/non–African Americans (27%).6 There is limited information, however, on unmet need for vision care among minority CSHCN. A study7 of CSHCN and disparities collapsed measures of unmet need for eyeglasses, medications, and mental health care into one variable, which is less informative for interventions targeting vision problems. Although economic factors do not entirely explain disparities, they may have a major effect on minority CSHCN. In families unable to afford part-time home care, adults who quit work to care for CSHCN at home often experience a loss of income and benefits that further limits their ability to pay for care.8
Children of disadvantaged backgrounds whose vision might otherwise go unexamined have traditionally received screening in schools.9 However, a lack of policies for referring schoolchildren for clinical eye examinations has generated concern about the quality of these programs.10,11 Even when referral protocols with vision care providers are established, there is evidence that many schoolchildren do not receive adequate screening and diagnostic services. Among 890 children in Tennessee who received false-positive results in a preschool-based screening program, 20% were inappropriately prescribed eyeglasses in subsequent eye examinations. This rate of misdiagnosis ranged from 2% to 35%, depending on the type of physician performing the examination.12 To our knowledge, previous work has not assessed the impact of school-provider linkages or health care provider type on unmet need for vision care among CSHCN.
In this study, we estimate racial/ethnic differences in unmet need for vision care in a representative sample of CSHCN. We assess potential causes of these differences by examining imbalances in key social and economic resources across racial/ethnic groups. We then test the hypothesis that among CSHCN who need vision care, minorities are more likely than whites to have unmet need, after accounting for variables on families, providers, schools, and other factors relevant to the vision-related health of children.
The National Survey of CSHCN was a cross-sectional telephone survey conducted between October 2000 and April 2002. A random-digit-dial sample of households with CSHCN younger than 18 years was selected from the 50 states and the District of Columbia. Respondents were adults who knew the most about the index child's health. The response rate was 61.1%. Further information on the study design is available elsewhere.13
The definition of special health care needs was based on 5 criteria14 that were applied in a set of questions during the telephone screening of each household.13 A child with special health care needs is an individual younger than 18 years with a health condition lasting 12 months or longer that causes the child the following: (1) to need more health or educational services than most children of the same age; (2) to use prescription medicines; (3) to be more limited in abilities than most children of the same age; (4) to need physical, speech, or other therapies; or (5) to need treatment for developmental or behavioral problems.
Because the target population is CSHCN needing vision care, analyses were restricted to the 14 070 children (36.2% of the total sample of 38 866) who needed eyeglasses or vision care at some point in the previous year.
If the respondent answered no to the question, “Did [child's name] receive all the eyeglasses or vision care [he/she] needed?” the child was categorized as having unmet need for vision care. Similar questions on unmet need have been used in previous surveys.15- 17 Twenty-three observations with missing responses were excluded.
Data on race/ethnicity were collected in 3 steps. First, respondents indicated whether the child was Latino. Respondents then chose 1 or more of the following races for the child: white, black or African American, American Indian, Alaskan Native, Asian, or Native Hawaiian or other Pacific Islander. They could also report verbatim another race. Respondents choosing more than 1 race were asked: “Which do you feel best describes [child name]'s race?” If multiple races were again chosen, the child was recorded as “multiracial.”
We grouped CSHCN who were Latino and African American, multiracial, white, or other as Latino. To maximize counts in the rarer groups, CSHCN of Asian, Native Hawaiian, and other Pacific Islander (ANHOPI) backgrounds were grouped by race, regardless of Latino ethnicity.18 Twenty-one Latino CSHCN with missing race data were grouped as Latino. Fifty-nine observations with missing race/ethnicity data were excluded. Because of small sample sizes, one group was created for ANHOPI children. In summary, each child was grouped as African American, Alaskan Native or American Indian (AIAN), ANHOPI, Latino, multiracial, white, or other.
Age was included in the analysis because of its correlation with childhood visual acuity.19 We also included a variable on sex. The association of childhood vision impairment with maternal education20 and insurance21 likely reflects the effects of socioeconomic status on health. Maternal education was categorized as bachelor's degree or higher, some college, high school, and less than high school. Insurance was categorized as private, Medicaid, State Children's Health Insurance Program (S-CHIP), other, and none. The federal poverty level (FPL) is the annual income threshold below which families risk going without shelter or food.22 Adjusted for family size, this threshold is referred to as “100% of the FPL”; higher annual incomes are expressed as higher percentages of the FPL. We categorized annual household income as follows: less than 100%, 100% to 199%, 200% to 399%, and 400% or greater of the FPL.
Because adults raising more children may have less time to address the vision problems of CSHCN, the analysis included variables on the number of children in the household and whether the child lived with a single adult.23 Language barriers may prevent some parents from developing productive relationships with clinicians,24 so we included a variable for whether the interview was conducted in a non-English language.
Several health status measures were included. Respondents ranked the severity of the child's health condition from 0 (mildest) to 10 (severe). We created a 0- to 17-count variable for the number of other services (excluding vision care) for which the child had unmet needs in the previous year. Respondents indicated the extent to which health problems affected the child's ability to “do things” as follows: a great deal, some, very little, or “don't know.” We included 3 measures of disability: whether an adult quit work to care for the child at home, number of school days missed because of illness in the previous year, and whether the child received supplemental security income.
Evidence that provider type is associated with quality of pediatric vision care12 motivated the inclusion of a variable for whether the child had a pediatrician, generalist physician, nurse practitioner or physician assistant, other clinician, or no regular provider. We also included the number of physician visits provided to the child in the previous year.
Unmet needs may result when health care providers are insensitive to how cultural differences affect their interactions with patients.24,25 We included a variable on how often providers were sensitive to the values and customs of the child's family. Because schools often alert clinicians about a child's vision problems,9 we included a variable on how well clinicians communicated with the child's school, vocational program, or child care provider.
Univariate statistics were calculated to describe the prevalence of unmet need for vision care and other population characteristics. Within each racial/ethnic group, we examined differences in unmet need, annual income, insurance, and school and provider characteristics. We then calculated rates of unmet need for vision care across categories of these social and economic variables. Multivariate logistic regression was used to test the hypothesis that minority CSHCN were more likely than whites to have unmet need. Odds ratios and 95% confidence intervals were calculated. Because 5.8% of the sample had unmet need, we interpret results as risk ratios (ie, as comparisons of probabilities rather than odds).26 We used computer software (Stata) to weight prevalence estimates and to adjust standard errors for the sampling design.27
Because of missing data on maternal education, supplemental security income, adults quitting work to care for a child, provider type, and impact of health on functioning, 7% of observations were dropped. Although there were no significant racial/ethnic differences in missing annual income (F = 1.18, P=.32), we created a “no annual income data” category to avoid losing 9% of observations in the analysis. We replaced missing values with means for these continuous variables (percentage missing): severity (0.5%), age (0.1%), physician visits (1.3%), and missed school days (5.5%).
The subjects of the 14 070 interviews represent approximately 3 310 400 CSHCN needing vision care. Most of this population was white, followed by African American, Latino, multiracial, and other background. Both ANHOPI and AIAN children were less than 1% of the CSHCN needing vision care (Table 1). Of the CSHCN needing vision care, 5.8% (ie, an estimated 198 600 CSHCN in the United States did not receive those services in the previous year.
The prevalence of unmet need for vision care was highest among multiracial CSHCN, followed by children of other backgrounds, Latinos, African Americans, and whites. The AIAN and ANHOPI children had approximately 3% rates of unmet need (Table 2). The socioeconomic variables associated with unmet need in Table 2 were also associated with race/ethnicity (Table 3). Racial/ethnic differences in unmet need remained after controlling for differences in health status and other child and family characteristics (Table 4). African Americans, Latinos, and multiracial CSHCN were 2 to 3 times more likely to have unmet need than whites. Conversely, AIAN children had one fifth the risk of unmet need of whites (Table 4).
Compared with the privately insured, uninsured CSHCN had nearly twice the risk of unmet need for vision care, whereas Medicaid and S-CHIP recipients had decreased risks. Children in the lower 3 annual household income groups had higher risks of unmet need than children at 400% or greater of the FPL. Unmet need was also associated with more children in the household, number of missed school days, and number of other unmet needs. In households in which an adult had stopped working, CSHCN were more likely to have unmet need. Children whose health affected their ability to do things either some or very little were more likely to have unmet need than CSHCN without impaired functioning.
Compared with pediatric patients, CSHCN with generalist physicians and nurse practitioners or physician assistants were more likely to have unmet need for vision care. Increased risks of unmet need were found among CSHCN whose providers were never culturally competent (vs always-competent providers), among CSHCN with schools and providers in fair or poor communication, and among CSHCN whose parents did not know about the quality of school-provider communication (compared with providers and schools in excellent communication). Increased physician visits were associated with lower risk of unmet need.
Health care providers and schools have essential roles in ensuring that CSHCN are screened for vision problems and, if necessary, referred for clinical eye examinations. Our analysis showed that CSHCN of African American, Latino, and multiracial backgrounds were more likely than whites to have unmet need for vision care, even after controlling for differences in health and several other important variables. Further research is needed to understand the causes of these racial/ethnic differences and to design programs to address them.
In the multivariate analysis, AIANs were less likely than whites to have unmet need. In the bivariate analysis, however, there was no significant difference between these 2 groups (Table 2). After controlling for only 1 variable, health insurance, the comparison of AIAN and whites became significant at P=.06 (odds ratio, 0.44; 95% confidence interval, 0.19-1.02) (data not shown). This change is likely due to a large difference in Medicaid coverage: 44.0% of AIANs had Medicaid vs 11.9% of whites (Table 3). Although Medicaid recipients had lower adjusted risk of unmet need than privately insured CSHCN (Table 4), the fact that only the AIAN effect estimate changed after adding insurance made us suspect that the Medicaid effect was different for AIANs. For this reason, we restricted a bivariate analysis of insurance and unmet need to AIANs (n = 159). The result showed that AIANs with Medicaid had greater risk of unmet need than privately insured AIANs (odds ratio, 15.86; 95% confidence interval, 1.48-169.52) (data not shown). Studies using larger samples of AIANs may help further explain this intragroup difference.
Bivariate analysis showed S-CHIP recipients had a higher prevalence of unmet need for vision care than the privately insured (Table 2), probably because CSHCN who are eligible for S-CHIP have worse health. After multivariate adjustment, however, the needs of CSHCN for vision care were apparently better served through S-CHIP. In most states, S-CHIP and Medicaid do not impose substantial cost sharing on families. By contrast, private plans are increasingly charging separate premiums for optometric services,28 which may lead families to forego coverage for eyeglasses entirely. Children with cost-sharing insurance in the Rand Health Insurance Experiment had worse functional far vision at follow-up than did children receiving free care.29 Virtually all states include vision care in their Medicaid programs,30 and many vision care providers are willing to accept children with Medicaid,31 despite concern among advocates that low reimbursements reduce physician participation in the program. Our finding that uninsured CSHCN had increased risk of unmet need underscores the importance of these safety net programs.
Children in households in which an adult quit work to care for them had increased risk of unmet need for vision care. Previous work8 suggests that the costs of caring for CSHCN are an obstacle to economic self-sufficiency for families, which may account for the association of lower annual income with unmet need. We did not expect that CSHCN in families with more children would have lower risk of unmet need for vision care. Perhaps children without special needs provide practical support at home, thus freeing up adults to tend to CSHCN.
Children whose health affected their ability to do things were more likely to have unmet need; however, there were unexpected differences across the disability categories. Specifically, CSHCN whose health conditions had only very little or some effect on their functional abilities had increased risk of unmet need, but CSHCN whose conditions had a greater effect on their functioning did not (compared with CSHCN with unimpaired functioning). It is possible that need for vision care seemed less important to respondents whose CSHCN had other severe health conditions.32
Patients of pediatricians had lower risk of unmet need for vision care than CSHCN with other types of providers. Nonpediatricians may not be adequately prepared for evaluating and treating the vision problems of CSHCN, because services for children require space, equipment, and other clinical resources that differ from those needed for adults.19 The lack of a difference between pediatric patients and CSHCN without regular providers may reflect similarities in the health of the 2 groups (ie, CSHCN in better health may not use services frequently enough to obtain regular providers). The inverse association of physician visits with unmet need likely reflects the health benefits of service use.
When parents perceived health care providers as respectful of their cultures, CSHCN were more likely to have their vision care needs met. Relationships between parents, providers, and schools will promote the exchange of information on the daily visual functioning of CSHCN,24,27 which may explain why CSHCN whose health care providers communicated well with schools were less likely to have unmet need. The increased risk of unmet need among CSHCN whose parents did not know about the quality of provider-school communication suggests that providers and schools have important roles in encouraging parents to obtain needed vision care for CSHCN.
We acknowledge that racial/ethnic differences in unmet need may reflect differences in the tendencies of parents to report need.33 We assessed for bias in reported need for vision care by examining the racial/ethnic distribution of responses to the question, “During the past 12 months, was there any time your child needed eyeglasses or vision care?” in the total sample of 38 866 households. We found no significant racial/ethnic differences in reported need (F = 0.86, P=.51). In the future, cognitive testing may help determine how respondents of diverse backgrounds interpret questions on need for vision care.34
This study was limited by the complexities of racial/ethnic self-identification in surveys. For example, respondents may have reported CSHCN as “not Latino,” but then offered verbatim responses of various Latino backgrounds as other types of race.13 In addition, CSHCN in aggregated racial/ethnic categories may have differed substantially from each other in their risk of unmet need; the null finding for ANHOPI children may be due to such heterogeneity (and low statistical power). Another limitation is the cross-sectional study design. Longitudinal studies would help clarify the association of unmet need with health status, poverty, and other variables.
In terms of policy implications, the possibility that Medicaid and S-CHIP provide better access to vision care than private insurance is promising for CSHCN in those programs; however, recent reform proposals would make Medicaid similar to private insurance by increasing patient cost sharing.35 Further analysis should assess how such reforms could affect CSHCN, and whether Medicaid functions differently for AIANs or other minorities. Because schools provide CSHCN with services that were previously delivered by health care or home care providers, input from clinicians and parents may help ensure the quality of school-based vision programs.36 Finally, racial/ethnic differences persisted after accounting for school and health care provider characteristics, which suggests that these resources should be leveraged to increase the availability of vision care for all CSHCN.
Correspondence: Kevin C. Heslin, PhD, Research Centers in Minority Institutions, Charles R. Drew University of Medicine and Science, 2594 Industry Way, Lynwood, CA 90262 (firstname.lastname@example.org).
Submitted for Publication: July 26, 2005; final revision received October 17, 2005; accepted November 2, 2005.
Financial Disclosure: None.
Funding/Support: This study was supported by grant 1R24-HS014022-01A1 from the Agency for Healthcare Research and Quality; grant G12-RR03026-18-S1 from the National Eye Institute; grant G12-RR03026 from the National Center for Research Resources; and grant 1 P20MD00148-01 from the National Center on Minority Health and Health Disparities.
Acknowledgment: We thank Naihua Duan, PhD, and Susan L. Ettner, PhD, for comments on earlier drafts of the manuscript.
Country-Specific Mortality and Growth Failure in Infancy and Yound Children and Association With Material Stature
Use interactive graphics and maps to view and sort country-specific infant and early dhildhood mortality and growth failure data and their association with maternal
Thank you for submitting a comment on this article. It will be reviewed by JAMA Ophthalmology editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Register and get free email Table of Contents alerts, saved searches, PowerPoint downloads, CME quizzes, and more
Subscribe for full-text access to content from 1998 forward and a host of useful features
Activate your current subscription (AMA members and current subscribers)
Purchase Online Access to this article for 24 hours
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 9
Customize your page view by dragging & repositioning the boxes below.
and access these and other features:
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.