OBJECTIVE. The objective was to examine the relationship between underinsurance and other core outcomes for children with special health care needs.
METHODS. This study analyzed data from the National Survey of Children With Special Health Care Needs. Two alternative definitions of underinsurance, designated attitudinal and economic, were investigated. Logistic regression models in which the response variables were the child's status for each of the target core outcomes and underinsurance status was a dichotomous predictor variable were created. In addition to underinsurance status, 10 other predictor variables were included in the model.
RESULTS. Underinsurance is associated with the Maternal and Child Health Bureau core outcomes for children with special health care needs related to satisfaction with care and partnering with families in decision-making, access to a medical home, community-based service delivery that is easy to use, and access to services to make transitions to adulthood. In each case, children with special health care needs who were underinsured had significantly poorer outcomes than did children who were adequately insured.
CONCLUSIONS. Although these results cannot clarify the cause of poorer outcomes, there are clear negative effects associated with the problem of underinsurance. Inadequate health care coverage for children with special health care needs may save dollars in the short-term but, if other outcomes are compromised, then children, their families, and society at large may pay a price in the longer term.
Children with special health care needs (CSHCN) represent ∼13% to ∼18% of children and adolescents in the United States.1,2 Emerging evidence suggests that CSHCN have higher levels of unmet needs for routine and specialty health care, compared with the general pediatric population.1,3,4 CSHCN are less likely than other children to have private insurance, but they are enrolled at higher rates than other children in public insurance programs such as Medicaid or the State Child Health Insurance Program (SCHIP). Overall, the uninsurance rate of CSHCN is lower than that of children without special needs5 but, because of their greater need for health care services, the families of CSHCN often experience their health insurance as inadequate. CSHCN use many more services and have greater health services expenditures than do other children,6 they have more problems accessing medical care, and their families have difficulty paying their medical bills, related to greater need for and use of medical services.5
In recent years, the phenomenon of inadequate insurance, or underinsurance, has begun to gain attention as a potentially important factor in the quality of health care. Although there is no universally accepted definition, the term “underinsurance” is generally used to refer to health care insurance that requires excessive out-of-pocket expenditures, that has significant limits with respect to what health care services are covered, or that fails to cover health care expenses that are perceived by the insured person to be essential for his or her health.7 Underinsurance in vulnerable populations may have disproportionately adverse effects, such as exacerbation of illness because of delayed medical care or longer hospitalizations.8 Underinsurance puts children with conditions such as diabetes mellitus at increased risk for medical complications.9 Limited coverage and inadequate reimbursement can result in significant gaps in services and increased probabilities of compromised health outcomes.10 Those longer-term outcomes of underinsurance rarely have been examined empirically, however.
Adequacy of health care insurance coverage is one of the core outcomes identified as a target for CSHCN by the Maternal and Child Health Bureau (MCHB) national agenda11 and incorporated subsequently into the health care access indicator for Healthy People 2010. The MCHB core outcome measure states that “all families of children with special health care needs will have adequate private and/or public insurance to pay for the services they need.”11 Medicaid expansions since 1980 and the introduction of SCHIP have reduced the family financial burden of children's health care, but significant disparities persist12 and the adequacy of insurance for CSHCN has important policy implications as public insurance continues to evolve.
Measuring the adequacy of health insurance has proved to be a challenging enterprise.13 Ward et al7 summarized the literature and provided what they termed “attitudinal” and “economic” definitions of underinsurance. Attitudinal definitions emphasize consumer perceptions and satisfaction as they relate to health care. Underinsurance is identified when ≥1 health benefit the person prefers to receive is not covered by insurance, when there is ≥1 symptom the person thinks requires treatment for which insurance coverage of treatment is not provided, or when a person is dissatisfied with the insurance plan. Economic definitions focus on a person's ability to pay for health care, including the cost of the insurance premiums, copayments, and deductibles. An economic definition of underinsurance defines a limit above which the expense of health care coverage becomes a burden and interferes with access to care. With this definition, underinsurance is identified when out-of-pocket expenses for necessary medical care exceed a specified percentage of the person's income within a given time frame or when a person delays health care because of out-of-pocket costs associated with the services.
As a result of the MCHB emphasis on the issue, items related to the inadequacy of insurance, or underinsurance, were included in the National Survey of Children With Special Health Care Needs, henceforth referred to as the CSHCN survey. Five survey items addressed this core outcome. The first 2 items identified children who did not have insurance coverage for some or all of the period in question. The remaining 3 questions addressed the frequency with which the insurance met the child's needs, the reasonableness of costs that were not covered by insurance, and the frequency with which the insurance plan provided access to needed providers. Given their focus on parents' perceptions, the latter 3 items (Table 1) may be conceptualized as operationalizing an attitudinal definition of underinsurance.14
In a previous study,14 we examined the CSHCN survey for items that operationalized the economic definition of underinsurance. An economic definition was constructed from survey items, and an algorithm for defining underinsurance was devised (Table 2). Underinsurance status, as operationalized by the economic and attitudinal definitions, was found to be associated with a small set of child and family variables. The 2 definitions identified approximately equivalent numbers of children as underinsured (attitudinal: 28.9%; economic: 25.6%), but the groups of children identified with the 2 definitions were not identical. Indeed, only approximately one half of the children identified as underinsured with one definition were also identified so with the other definition, which suggests that the 2 approaches to defining underinsurance are distinct.
In addition to adequacy of insurance, the MCHB core outcomes included the following: families of CSHCN will partner in decision-making and will be satisfied with the services they receive; CSHCN will receive coordinated, ongoing, comprehensive care within a medical home; community-based service systems will be organized so that families can use them easily; and youths with special health care needs will receive the services necessary to make the transition to adult life, including adult health care, work, and independence. Little attention has been paid to the question of whether the MCHB core outcomes are interrelated. It is clear that a lack of insurance coverage is associated with increased likelihood of dissatisfaction with health care.15 Furthermore, finance-related problems in the families of CSHCN are associated with reduced likelihoods of meeting 3 of the MCHB core outcomes, namely, medical home, adequate insurance, and organized services.16 However, the question of whether underinsurance is associated with poorer outcomes has not been investigated adequately.
A conceptual model of the effects of underinsurance might suggest that underinsurance status would mediate or set limits on other outcomes; for example, being underinsured reasonably might be expected to decrease the likelihood that CSHCN would receive accessible, comprehensive, and continuous care in a medical home. Alternatively, being underinsured might force families to choose between necessary services because of the financial burden of medical care.
The objective of the present study was to examine the relationship between underinsurance and other core outcomes for CSHCN. We were particularly interested in determining whether differences in the other core outcomes for CSHCN were associated with underinsurance status, with accounting for the effects of other predictor variables. Specifically, the hypothesis under investigation was as follows: after accounting for the effects of correlates of child health outcomes, underinsurance would be associated with a reduced likelihood of meeting other MCHB core outcomes.
We chose to examine the relationship between underinsurance and 4 of the MCHB outcomes. Table 3 presents the text of the MCHB outcomes, with a brief descriptor phrase and the proportion of children in the survey who met the outcome. CSHCN survey items that assessed the core outcome addressing adequacy of insurance constituted one of the definitions of underinsurance; therefore, this outcome was included in the analyses as an independent variable. The remaining MCHB core outcome (“children will be screened early and continuously for special health care needs”) was not measured in the CSHCN survey and was not addressed in the analyses. The correlational design of the present study does not address questions of causality (eg, “does underinsurance lead to poorer outcomes?”); however, it addresses the question of covariation and may set the stage for additional investigations, including causal models.
The present study analyzed data from the CSHCN survey. The CSHCN survey was sponsored by the MCHB and the National Center for Health Statistics.17 The sample was constructed to allow for both national-level and state-level findings. The survey used a random-digit-dial method to sample households with children <18 years of age in each of the 50 states and the District of Columbia. All children in each identified household were screened for special health care needs. The project screened 373055 children for special health care needs and completed 38866 CSHCN interviews, including at least 750 in each state. CSHCN survey data were collected between October 2000 and April 2002. Children in the survey were screened for special health care needs by using the Child and Adolescent Health Measurement Initiative CSHCN screener.18 The present study included 34666 children who were insured for the entire year.
To examine the relationship between underinsurance and other core outcomes, logistic regression models were created in which the response variables were the child's status on each of the target core outcomes (met/did not meet outcome). Underinsurance status was a dichotomous predictor variable indicating whether, according to the attitudinal or economic definitions, the child was underinsured. In addition to underinsurance status, 10 other predictor variables were included in the model (Table 4). For comparison, these predictors were chosen to correspond closely to the predictors in the study by Kuhlthau et al.16 Logistic regression models were created by using the Svylogit procedure in the Stata statistical analysis package (Stata, College Station, TX), a procedure that takes into account the complex sample design for these survey data.
The effects reported here are the results of logistic regression models that include underinsurance plus the 10 predictor variables. Each of the models was significant overall in predicting the outcomes. For this report, we focus on the effect of underinsurance status in the models.
For the satisfaction core outcome, both attitudinal underinsurance status and economic underinsurance status were significant predictors (attitudinal: t = −9.83; P < .0001; economic: t = −8.51; P < .0001). Odds ratios indicated that families of children who were underinsured (according to either definition) were approximately one half as likely to partner in decision-making and be satisfied with the services they received (Table 5).
For the medical home core outcome, both attitudinal underinsurance status and economic underinsurance status were significant predictors (attitudinal: t = −14.88; P < .0001; economic: t = −10.30; P < .0001). Odds ratios indicated that children who were underinsured (according to either definition) were approximately one half as likely to receive coordinated ongoing comprehensive care within a medical home.
For the ease of use core outcome, both attitudinal underinsurance status and economic underinsurance status were significant predictors (attitudinal: t = −10.98; P < .0001; economic: t = −8.91; P < .0001). Odds ratios indicated that families of children who were underinsured (according to either definition) were less than one half as likely to report that community-based service systems were organized so that families could use them easily.
For the transition core outcome, only attitudinal underinsurance status was a significant predictor (t = −1.97; P > .05). The odds ratio indicated that youths who were underinsured, according to the attitudinal definition, were just over one half as likely to receive the services necessary to make transitions to adult life, including adult health care, work, and independence.
The findings demonstrated that underinsurance was associated with the MCHB core outcomes for CSHCN related to satisfaction with care and partnering with families in decision-making, access to a medical home, community-based service delivery that is easy to use, and access to services for transitions to adulthood. In each case, underinsurance was associated with significantly poorer outcomes; this association was observed in a model that took into account the effects of demographic variables that were likely to contribute to those poor outcomes.
For all children, research has demonstrated that discontinued health insurance or no insurance coverage leads to problems with access to care, such as delayed care, unmet medical needs, and unfilled prescriptions.19 CSHCN are as likely as other children to experience gaps in coverage or to be uninsured.20 The lack of insurance has been found to be the strongest indicator of delayed or foregone care for CSHCN and increases the likelihood of experiencing barriers to health care.21 The findings in the present study suggest that the negative outcomes associated with uninsurance may extend to insured children whose coverage is inadequate.
Particularly for low-income families, insurance is not enough to protect families from the financial burden of health care expenses. Low-income families may be “underinsured,” compared with higher-income families, with respect to excess medical costs, and inadequate coverage may be a significant barrier to adequate health care.6 However, there is evidence that commercial insurance plans are more likely to limit coverage and to impose stringent cost requirements, compared with programs provided by Medicaid agencies.22 These findings illustrate the importance of ensuring adequate insurance coverage for all children with respect to access and timeliness of care; because CSHCN as a population have increased medical needs and expenses, their coverage is an even more critical issue for families, health care providers, and policymakers.
Underinsurance may also be a particular concern for immigrant children; immigrant parents without insurance are more likely than nonimmigrant parents to postpone some forms of health care for their children.23 The role of underinsurance in health care disparities among immigrant children, and specifically immigrant CSHCN, is largely unknown.
Because of its design, the present study is limited to demonstrating an association between underinsurance status and children's health outcomes. We are unable to address the extent to which underinsurance is a causal factor in those outcomes. These findings are also limited in scope to the outcomes assessed in the CSHCN survey. Although these outcomes hold special interest because of their visibility in the national agenda for CSHCN, there are other health status and health care quality outcomes that require additional investigation. Finally, we sought to base our operationalizations of underinsurance on an established conceptual framework; however, other definitions of the construct may not be associated with outcomes in the same way.
Research and Policy Implications
There are multiple ways of defining underinsurance. In addition to the attitudinal and economic definitions used in the current study, Ward et al7 proposed what they termed a structural definition, which needs to be operationalized properly in measurable terms. Clear operational definitions of underinsurance should lay the groundwork for additional investigation of the effects of inadequate coverage in terms of outcomes for CSHCN.
The current study showed that underinsurance is a correlate of poor outcomes for CSHCN. An important unanswered question is whether underinsurance is the cause of these poor outcomes or merely is a correlate of other factors that are more centrally causal. Additional investigation involving causal models should help to clarify the role of underinsurance in producing poor outcomes for CSHCN.
The present findings suggest that there are clear negative effects associated with the problem of underinsurance. Inadequate health care coverage for CSHCN may save dollars in the short-term but, if other outcomes are compromised, then children, their families, and society at large may pay a price in the longer term. Economic modeling could help to clarify the relationship between the cost of providing adequate insurance for CSHCN and the benefits in terms of improved outcomes.
CSHCN with public insurance consume a large share of the health care resources. Medicaid and SCHIP are critical and necessary programs for these children. Nevertheless, limited resources and tight budget constraints at the federal and state levels might limit spending in public insurance programs, which would be detrimental to CSHCN. The potential strategies to slow spending in public programs discussed by Tu and Cunningham5 include changes in eligibility criteria, limiting enrollment, and increased cost sharing. Such policy changes would create additional barriers to care for families of CSHCN and decrease the number of children covered by public programs, thus limiting access to care.5
This issue may be of particular concern when health care for CSHCN is delivered via fee-for-service plans (as opposed to capitated health maintenance organization models). Children with disabilities are more likely to be enrolled in such plans, and fee-for-service consumers are more likely to experience an unmet need for some type of health care service.24
It may be fruitful to view adequate insurance as an intermediate outcome that significantly affects other core outcomes for CSHCN. In some cases, efforts directed toward improving other outcomes may be frustrated if underinsurance is not addressed; conversely, improving the insurance coverage for CSHCN may yield unanticipated benefits in terms of other key outcomes. Although such conclusions go well beyond the data presented here, the present study suggests the potential of underinsurance as an intermediate outcome.
Support for this project was provided in part by a MCHB Research Training Program Enhancement Award (project 5 T05 MC 00009-02) and by the Virginia Leadership Education in Neurodevelopmental Disabilities program, with funding from the MCHB (project 5 T73 MC 00040-08).
We thank Virginia Sharp for her thoughtful feedback on an early draft of this manuscript.
- Accepted September 13, 2006.
- Address correspondence to Donald P. Oswald, PhD, Department of Psychiatry, Virginia Commonwealth University, Box 980489, Richmond, VA 23298. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- ↵Mayer ML, Skinner AC, Slifkin RT. Unmet need for routine and specialty care: data from the National Survey of Children With Special Health Care Needs. Pediatrics.2004;113 (2). Available at: www.pediatrics.org/cgi/content/full/113/2/e109
- ↵van Dyck PC. The National Survey of Children With Special Health Care Needs. Available at http://mchb.hrsa.gov/mchirc/dataspeak/events/May_03/index.htm. Accessed December 5, 2006
- ↵Davidoff AJ. Insurance for children with special health care needs: patterns of coverage and burden on families to provide adequate insurance. Pediatrics.2004;114 :394– 403
- ↵Tu HT, Cunninghan PJ. Public Coverage Provides Vital Safety Net for Children With Special Health Care Needs. Washington, DC: Center for Studying Health System Change; 2005. Issue brief 98. Available at: http://hschange.org/CONTENT/778/?PRINT=1. Accessed July 20, 2006.
- ↵Ward A, Beebe TJ, Blewett LA, Smaida S. Issues in Defining and Measuring Adequacy of Coverage Minneapolis, MN: State Health Access Data Assistance Center; 2002
- ↵Newacheck PW, McManus M, Fox HB, Hung Y, Halfon N. Access to health care for children with special health care needs. Pediatrics.2000;105 :760– 766
- ↵Maternal and Child Health Bureau. Achieving and measuring success: a national agenda for children with special health care needs. Available at: http://mchb.hrsa.gov/programs/specialneeds/measuresuccess.htm. Accessed December 5, 2006
- ↵Health Resources and Services Administration. Assessing and Assisting the Underinsured. Rockville, MD: Health Resources and Services Administration; 2001
- ↵Oswald DP, Bodurtha JN, Broadus CH, et al. Defining underinsurance among children with special health care needs: a Virginia sample. Matern Child Health J.2005;9(suppl) :S67– S74
- ↵Ngui EM, Flores G. Satisfaction with care and ease of using health care services among parents of children with special health care needs: the roles of race/ethnicity, insurance, language, and adequacy of family-centered care. Pediatrics.2006;117 :1184– 1196
- ↵Blumberg SJ, Olson L, Frankel M, et al. Design and operation of the National Survey of Children With Special Health Care Needs, 2001. Vital Health Stat 1.2003;(41):1–136
- ↵Satchell M, Pati S. Insurance gaps among vulnerable children in the United States, 1999–2001. Pediatrics.2005;116 :1155– 1161
- Copyright © 2007 by the American Academy of Pediatrics