Utilization of Medical and Health-Related Services Among School-Age Children and Adolescents With Special Health Care Needs (1994 National Health Interview Survey on Disability [NHIS-D] Baseline Data)
Objective. To determine how sociodemographic factors and type of insurance influence use of medical and health-related services by children with special health care needs (CSHCN), after controlling for need.
Methods. A cross-sectional analysis of 1994 National Health Interview Disability Survey was conducted. Children between 5 and 17 years were identified as chronically ill according to the Questionnaire for Identifying Children with Chronic Conditions (n = 3061). Independent variables included child and family characteristics categorized as predisposing, enabling, and need. Dependent variables included use of 4 medical or 7 health-related services.
Results. Most children (88.7%) had seen a physician; 23.9% had an emergency department visit, 11.4% had a mental health outpatient visit, and 6.4% were hospitalized. Health-related service use ranged from <5.0% (transportation and social work) to 65.1% (medical care coordination); 20% to 30% of children used the remaining services (therapeutic, assistive devices, nonmedical care coordination, housing modifications). In fully adjusted logistic models, children with public insurance were significantly more likely than privately insured children to use 2 of the 4 medical services and 5 of the 7 health-related services. Non-Hispanic black children and children from less educated families were significantly less likely to use many of the services examined.
Conclusions. In 1994, factors in addition to need influenced medical and health-related service use by CSHCN. Differences in the scope of benefits covered by public insurance compared with private insurance may influence utilization of medical and especially health-related services. Attention is needed to ensure that CSHCN who are racial/ethnic minorities or are from less educated families have access to needed services. Future studies should determine whether these patterns have changed over time.
Over the past few decades, there has been a growing recognition of the comprehensive service needs of children with chronic and disabling conditions.1,2 The need for a broad array of services has been incorporated in the Maternal and Child Health Bureau’s definition of children with special health care needs (CSHCN)3 and multiple policy statements of the American Academy of Pediatrics.4,5 Besides traditional medical services, CSHCN often require a diverse set of health-related and supportive services such as physical therapy and care coordination to maintain or improve functioning.1–3,6–10 Although medical and health-related services are closely linked to the overall needs of the child, they are largely separated for the purposes of financing and delivery.1,2,11,12 This separation has contributed to a fragmented system of care and has made developing a comprehensive utilization profile of CSHCN difficult.
Greater utilization of routine medical care among CSHCN is well documented.13–15 A number of studies also have shown that the use of medical services by CSHCN is influenced by factors such as health status, insurance status, and sociodemographic characteristics of the child and the family.13,16–21 For example, children with multiple chronic conditions, worse reported health status, or activity limitations are more likely to use physician and hospital services and to use prescription drugs.18,20 The presence of insurance is associated with having a usual source of care; the use and volume of physician services; and fewer reported unmet medical, dental, prescription medication, and mental health needs.16,18 Nonwhite CSHCN are more likely to report having access problems and unmet needs, are less likely to see a physician, and are more likely to be hospitalized than their white counterparts.17 Poor CSHCN are less likely to be insured and use fewer ambulatory services but are more likely to be hospitalized than nonpoor CSHCN.19
Despite an expanding body of research, detailed information on access to and use of services other than physician and hospital care remains limited for CSHCN. This may be attributable, in part, to a lack of available data. Often services provided in schools and other public programs go unreported, and varying eligibility criteria mean that not all CSHCN are captured when data are collected. Moreover, most population-based surveys have not focused specifically on CSHCN. The few studies that have quantified the use of a broader range of services indicate that CSHCN use more physical therapy, mental health services, medical equipment, and respite care services than children without special health care needs.15,20 Less is known about variations in the use of these services by subgroups of CSHCN. At least 1 study suggested that worse levels of functioning increase the likelihood that children with specific chronic conditions receive physical therapy and hospital services and that the number of physical therapy visits increases with family income.20 The same study found no significant associations between levels of child functioning, maternal education, family income, and race on the likelihood or volume of dental and mental health services. This and other studies are limited because of their focus on children with specific conditions or impairments, reliance on small samples drawn from specialty clinics, or a relatively narrow range of child and family characteristics included in their models.18,20–22
This study uses data from the 1994 National Health Interview Survey (NHIS), which included special supplemental surveys on childhood disability (NHIS-D), to examine the use of 4 medical and 7 health-related services among CSHCN between the ages of 5 and 17. In addition to describing utilization, we use a behavioral model to examine the use of medical and health-related services as a function of predisposing, enabling, and need characteristics of the child and the family.23–25 Of particular interest is the impact of insurance on health-related service use. Although insurance has been shown to be an important factor in determining the use of medical services,16,18 its role in the use of other types of services is less clear.
Several recent studies using the NHIS-D have examined unmet medical needs among CSHCN.13,16,17 However, these studies did not quantify the actual use of services other than physician and hospital care and did not explicitly focus on the determinants of medical and health-related service use. To our knowledge, this is the first study to develop a comprehensive, nationally representative utilization profile of CSHCN, something that is necessary to help guide future allocation decisions and to monitor access to care for this population.
Data were drawn from the 1994 NHIS. The NHIS is an ongoing household survey of the civilian, noninstitutionalized population sponsored by the National Center for Health Statistics.26,27 Each year, the survey consists of a core set of questions on demographic characteristics, health status, and use of physician and hospital services. Supplemental questionnaires are used to collect information on topics of special interest. Analyses conducted for this study are based on the 1994 NHIS core questionnaire and supplemental questionnaires on disability.
The 1994 NHIS-D consisted of 2 phases. The first phase of the survey was conducted at the same time as the NHIS core questionnaire and collected basic disability information on all NHIS household members.27 It also served as the primary “screening tool” for the Phase II survey. Phase II of the NHIS-D collected more detailed information on a subset of individuals who were identified as disabled, chronically ill, or impaired on the basis of their responses to certain items in the NHIS core and Phase I questionnaires.28 The NHIS-D, Phase II Child’s Questionnaire gathered information on the use of a variety of services, including home care; child care; specialized services, such as physical therapy and assistive devices; service coordination; and housing modifications and transportation.
For children younger than 17 years, the respondent on all surveys was the parent or adult in the household who was most knowledgeable about the child’s health, usually the child’s mother.16,26–28 Seventeen-year-olds were allowed to complete the surveys for themselves, although the mother was most often the respondent. The household response rate for the 1994 NHIS core survey was 94.1%; the response rate for the NHIS-D, Phase I survey was 92.5% of those who completed the core survey; and the response rate for the NHIS-D, Phase II child’s survey was 91.1% of those who were eligible, resulting in an overall response rate of 79.3%.26–28
Identification of the Study Population
A noncategorical approach was used to identify school-age children and adolescents with chronic conditions. Specifically, 1994 NHIS-D, Phase I and core questions corresponding to items in the Questionnaire for Identifying Children with Chronic Conditions (QuICCC) were used to identify the study population. The QuICCC, an instrument developed and validated by Stein et al,29 identifies children with ongoing conditions that produce consequences in 3 domains: functional limitations, reliance on compensatory modalities (eg, prescription drugs), and service use beyond routine care. This definition is similar to the federal Maternal and Child Health Bureau’s definition of CSHCN3 and is consistent with current conceptualizations of the population. The specific algorithm used to operationalize the QuICCC with the NHIS data was identical to that published previously.30,31
A total of 3728 of the 21 415 children between the ages of 5 and 17 in the Phase I sample met the QuICCC definition. Children younger than 5 were excluded because several of the survey items related to functional limitations were not asked of younger children (eg, activities of daily living); children are often not recognized as having cognitive, emotional, or developmental delays until they enter school; and younger children are more likely to use health services on a regular basis (eg, well-child visits).32,33
The final study population consisted of children who were between 5 and 17 years, had chronic conditions, and were eligible for and completed the Phase II survey (n = 3061). Excluded were children who met the QuICCC definition and were eligible for but did not complete a Phase II questionnaire (n = 307) and children who met the QuICCC definition but were not eligible for the Phase II survey (n = 360) because they did not meet National Center for Health Statistics criteria. There were no major differences between children who met the QuICCC definition and were eligible for but did not complete the Phase II survey (n = 307) and those who did (n = 3061). However, children who met the QuICCC definition but were not eligible for the Phase II survey (n = 360) were much more likely to have been identified on the basis of a single domain (functional limitation only) compared with children who met the QuICCC definition and were eligible for Phase II (70.3% vs 12.6%).
The outcome variables included 4 medical (physician, hospital, emergency department, and mental health/substance abuse outpatient) and 7 health-related services/items (therapeutic, social work, medical care coordination, nonmedical care coordination, assistive devices, transportation, and housing modifications). Each variable was measured dichotomously to indicate whether a child used a given service in the past year.
There are no standard definitions of medical and health-related services. For purposes of the study, services were classified as medical when they often are provided by or involve physicians; health-related services were services that often do not directly involve physicians, recognizing that there is variation in the way services are provided (Appendix A). Information on other services used by CSHCN was either not collected by the NHIS-D or excluded because of small percentages (<1.0%) of reported use (eg, respite care, home health care).
Child and family characteristics, categorized as predisposing, enabling, and need, were included as explanatory factors. Predisposing characteristics included sex, age (categorized to reflect 2 distinct developmental stages: school-age, 5–11 years; adolescence, 12–17 years), race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, other), parental education (less than high school, high school degree or equivalent, some college, college degree or beyond), family size (<5, ≥5), and family structure (2-parent, 1-parent, other). Enabling characteristics, which reflect resources available to obtain services, included poverty status (above, at, or below) and the child’s type of health insurance during the month before the interview (private only, public only, private and public, none). Need characteristics, which reflect the child’s illness level and health status, included the respondent’s perception of the child’s overall health (excellent, very good/good, fair/poor) and the extent to which the child was limited in age-appropriate activities (none, limited in the kind/amount of major activity [attending school] or limited in other activities, unable to perform major activity).
After initial χ2 analyses to examine unadjusted associations between the explanatory factors and each of the utilization variables, we constructed separate logistic regression models to assess associations between child and family characteristics and use of services while simultaneously controlling for the effects of each of the other explanatory factors. Results of the regression analyses are presented as adjusted odds ratios and 95% confidence intervals. Additional models were run to test the significance of several interaction terms (race/ethnicity and health-related need, health insurance and health-related need, age and type of insurance). None of the interaction terms was significant, and they are not presented in the results.
Children were excluded from the logistic regression analyses when their race was classified as other (n = 75) or when information on the other explanatory variables was missing. All analyses were conducted using SUDAAN Version 7.5.3,34 a statistical software package that accounts for the complex, multistage sampling frame of the survey. Analyses were performed on raw survey data and then reanalyzed using a weighting variable to reflect national population estimates. Unless otherwise specified, results reflect weighted data.
Sample Population Characteristics
General characteristics of children in the final study sample are shown in Table 1. Previously published epidemiologic estimates of the prevalence of chronic conditions among children 0 to 17 using the same data source and similar definitions indicate that the prevalence of chronic conditions is higher among children who are older (>5 years), male, and more socioeconomically disadvantaged.14,30
Any Use of Medical and Health-Related Services
Table 2 shows the percentage of children using selected medical and health-related services at least once during the year. Direct comparison of service use to survey children without chronic conditions was possible only for physician and hospital services because utilization data on the other types of services were not collected for children without special health care needs (ie, information was collected only for children who were eligible for the Phase II survey). Not surprising, children in the study sample were more likely to have seen a physician and to have been hospitalized during the year compared with their counterparts without chronic conditions (88.7% vs 74.4% and 6.3% vs 1.4%, respectively). On the basis of national estimates of service utilization among the general pediatric population, children in the study sample were approximately twice as likely to use emergency department services than all 5- to 17-year-olds (23.9% vs 12.9%).35,36
Eleven percent of children made an outpatient visit for mental health/substance abuse (MH/SA) services (Table 2). The majority of these children (91.8%) received mental health services, 2.4% of children received substance abuse services, and 5.8% received both mental health and substance abuse services. Previous studies suggest that between 5% and 8% of all children and between 10% and 20% of children identified with mental health problems use mental health services during the year.37–39 Recently, the US Surgeon General reported that 21% of children between the ages of 9 and 17 receive mental health services during the year.40 However, only 9% of children receive mental health services from the health system (general medical care and specialty mental health), whereas 12% receive mental health services exclusively through schools or human services systems (eg, child welfare).40
The 2 health-related services with the highest percentage of reported use were related to care coordination (Table 2). Nearly two thirds (65.1%) of children in the sample reportedly had at least 1 contact with a medical care coordinator, whereas slightly fewer than one third (30.2%) had any contact with a nonmedical care coordinator. Not surprising, physicians were much more likely to be involved in the coordination of medical care than nonmedical care (not shown in Table 2). Fifty-one percent of respondents indicated that a physician was involved in medical care coordination, whereas only 2.6% indicated that a physician was involved in nonmedical care coordination. Parents were the most common nonphysician care coordinator for both medical and nonmedical services. Children could and did report having multiple care coordinators for both medical and nonmedical care. Twenty-eight percent of children (not shown) used both a medical and nonmedical care coordinator at least once during the year.
Similar percentages of children reportedly used therapeutic services, assistive devices/supplies, and home modifications. The 22.6% of children using therapeutic services largely reflects the use of 2 types of therapies: speech (14%) and physical (6%). Fewer than 5% of children had a visit to an occupational therapist, an audiologist, a respiratory therapist, or a recreational therapist. In most cases, children used only 1 or 2 types of therapy during the year; only 10% of children who used any therapeutic services used 3 or more different types.
As with therapeutic services, overall use of assistive devices/supplies was largely driven by the use of 2 devices—inhalers (17%) and nebulizers (6%)—most likely reflecting the high prevalence of asthma. From the data, it cannot be determined how frequently children who used these devices relied on them (eg, occasionally, every day). Fewer than 2% of children reportedly used any of the other 7 devices included in the survey’s checklist (eg, diabetic equipment). The most common housing modifications were accessible parking or a drop-off site (10.5%) and special railings within the home (9.7%).
Only 3% of children in the sample did not use any of the services during the year (not shown in Table 2). The majority of children (82.5%) used multiple (≥2) medical and health-related services; more than one third (34.5%) of children used 4 or more of the 11 services.
Determinants of Medical and Health-Related Service Use
Logistic regression analyses showed that age, race/ethnicity, and parental education were significantly associated with multiple medical (Table 3) and health-related services (Tables 4 and 5). Older children were significantly more likely than younger children to have used 2 of the medical services—emergency department and MH/SA outpatient (Table 3)—and 2 of the health-related services—assistive devices and housing modifications (Table 5). Older children, however, were less likely to have used therapeutic services (Table 4). These findings may reflect developmental differences in care needs. For example, the most commonly used therapeutic service was speech therapy, which is more often used by younger, school-aged children.
Non-Hispanic black children were significantly less likely to have used more than half of the health-related services than non-Hispanic white children, even after controlling for other factors such as poverty, insurance, and health status (Tables 4 and 5). Although there were no significant differences in the use of physician, hospital, or emergency department services by race/ethnicity (Table 3), the direction of the results is similar to previous findings indicating that minority children are less likely to use physician services and more likely to use hospital and emergency department services. Both non-Hispanic black and Hispanic children were significantly much less likely to have had an MH/SA outpatient visit than non-Hispanic white children (Table 3). Children in families headed by someone with less than a high school education were significantly less likely than children from college-educated families to have used each of the medical services except for emergency department visits (Table 3) and 3 health-related services: therapeutic services (Table 4), assistive devices, and transportation (Table 5). The likelihood of use did not significantly differ by sex or family size and structure for most medical and health-related services.
Type of insurance emerged as a significant determinant of use for several medical (Table 3) and health-related services (Tables 4 and 5), after controlling for other factors, including need. Overall, children with any form of public insurance tended to be significantly more likely to have used services than privately insured children. Specifically, children either with public insurance alone or in combination with private insurance were more likely than privately insured children to have used 2 of the medical services (emergency department and mental health outpatient) and 5 of the health related services (therapeutic, social work, nonmedical care coordination, transportation, and housing modifications). Children with both public and private insurance also were more likely to have been hospitalized than privately insured children. Having no insurance meant that children were significantly less likely to have seen a physician (Table 3) or a medical care coordinator (Table 4) and to have used assistive devices compared with privately insured children (Table 5). Uninsured children were, however, significantly more likely to have been hospitalized (Table 3) and to have had a social worker visit (Table 4) than privately insured children.
Need factors were among the strongest predictors of use across multiple medical (Table 3) and health-related services (Tables 4 and 5) in the fully adjusted models. Overall, the direction of the relationship between need factors and the likelihood of service use was as expected, with the likelihood of service use increasing as reported health and functioning declined. For example, children in very good/good health were more than twice as likely to be hospitalized as children in excellent health, whereas those in fair/poor health were >8 times as likely (Table 3). There was some variation in the significance of the need factors by type of service. For example, reported health status was significantly associated with having a physician visit, whereas activity limitations were not (Table 3). Likewise, activity limitations were significantly associated with the use of therapeutic services, nonmedical care coordination, and transportation services, whereas health status was not (Tables 4 and 5). Notably, neither measure of health-related need was significantly associated with the use of MH/SA services.
This analysis of the 1994 NHIS-D indicates that school-age and adolescent CSHCN use a diverse mix of medical and health-related services. However, the percentage of children who are users varies by service type. Although nearly all of the children in the sample had a physician contact during the year and a majority of children had contact with a medical care coordinator, a moderate to small percentage of children used the other services included in the study. Although these data cannot be used to determine the appropriateness or quality of services received, they do provide important baseline measures of the use of a range of services for which data have historically been lacking.
Several studies among the general pediatric population and CSHCN have demonstrated that health-related need, as well as sociodemographic and economic factors, can influence a child’s initial access to ambulatory medical and hospital inpatient services.13,16–20,41–43 The multivariate results of this study extend these earlier findings to an additional set of services that have been identified as important to the care of CSHCN.
As expected, children with worse reported health and functioning were generally more likely to use each of the services than their counterparts. However, the significance of perceived health status and activity limitations varied somewhat by type of service, suggesting that different measures of need identify the need for different services. As policies and programs adopt noncategorical approaches to identify CSHCN and to determine eligibility for benefits, developing and applying appropriate measures of need within the population will be necessary to ensure that the need for certain services is adequately captured. Given that neither measure of need used in this study was associated with the use of mental health outpatient services, it may be important to include measures that more directly capture the mental health status of the child.39 More sensitive and specific measures of severity also may be needed to control for variations across CSHCN.
Although need factors were important in determining service use, factors other than need also played a role, suggesting that access to services may not always be equitable for certain groups of CSHCN. Although the significance of predisposing and enabling factors varied somewhat by type of service, some consistencies did emerge. Three variables in particular were consistently associated with both medical and health-related services after controlling for need: race/ethnicity, parental education, and type of insurance.
Non-Hispanic black children were significantly less likely to use 4 of the 7 health-related services than non-Hispanic white children. Both non-Hispanic black and Hispanic children were much less likely to have used MH/SA outpatient services than non-Hispanic white children, a finding consistent with previous studies of insured adults and children.44,45 Whether these findings are attributable to differences in perceived need, unmeasured differences in socioeconomic or health status, availability of services (eg, a lack of health-related services offered in special education programs in public schools), cultural differences, discrimination, or a combination of such factors cannot be determined from the data. However, it does indicate important areas for future research and monitoring, given the results of another recent study that reported greater levels of unmet need among minority CSHCN.17 Given the large differences in mental health service use, research in the area of cultural perceptions and attitudes toward mental health issues by different racial/ethnic groups may be especially warranted.
Children from less educated families tended to be less likely to use both medical and health-related services, after controlling for other possible confounding factors. Given similar health characteristics, children from families in which the head of the household had less than a high school education were significantly less likely to have seen a physician; to have been hospitalized; and to have used therapeutic services, assistive devices, and transportation services than children from college-educated families. Researchers have suggested that parents who have lower levels of education may be less familiar with “normal” functioning and development of children and therefore less likely to respond to signs or symptoms than highly educated families.46–48 Like race/ethnicity, the education variables may also reflect financial or nonfinancial barriers (eg, hours of provider availability) that were not otherwise captured in the model. It is also possible that parents with lower levels of education are less likely to be aware of the availability and potential value of various health interventions.
We were particularly interested in the impact of insurance on utilization. Overall, we found that children with public insurance were more likely to use services, especially health-related services, than privately insured children. Uninsured children tended to be less likely to use services. Interpreting these findings can be complex. It is possible that these differences may be attributable to unmeasured differences in health status. Children with public insurance may have more severe chronic conditions and more mental health conditions than children with private insurance, with differences remaining even after adjusting for perceived health status and activity limitations. Bivariate analyses did indicate that privately insured children were least likely to have activity limitations. It is also possible that the greater likelihood of use for some services (eg, emergency department) by publicly insured children is the result of inadequate primary care.
Alternatively, these findings may reflect differences in the scope of benefits between public and private insurance. Public insurance often is recognized as having greater breadth and depth than private insurance and, unlike most private plans, have incorporated child-specific standards.1,2,11,49 It also is possible that children in the sample who were covered by public insurance qualified for additional programs or assistance that were not recorded in the survey.
Pediatricians and policy makers should be aware that CSHCN who are covered only by private insurance are potentially at risk for being underinsured.50,51 Mechanisms may be necessary for privately insured children to obtain “wrap around” or “enabling” services. Access to mental health services is an area that may need particularly careful attention. In this study, children who were privately insured were much less likely to have had a mental health outpatient visit than children who were covered by either public insurance alone or both public and private insurance. Although these differences may reflect a higher level of mental health morbidity among children with public insurance that was not adequately controlled for in the models, they may also reflect limited coverage of mental health services by private plans and/or limited participation of mental health providers in plan networks—an issue of ongoing policy debate.50,51
Although CSHCN with public insurance were more likely than privately insured children to have used several of the services, this may have changed over time. At the time of the survey, in 1994, relatively few CSHCN who were publicly insured were enrolled in managed care.52 As more children with public insurance are enrolled in managed care plans, patterns of utilization for publicly insured children may approximate those of privately insured children. To the extent that managed care plans have more flexibility to provide services that are not usually covered in a fee-for-service setting, it is possible that managed care may afford privately insured CSHCN access to services that are not typically covered by private indemnity plans.53 Other changes since 1994, such as erosions in employer-based insurance, increasing premiums and cost sharing requirements, and state budget shortfalls, also may affect children’s use of services.
Lack of insurance among CSHCN is also a concern. Uninsured children were significantly less likely to have seen a physician (yet were more likely to be hospitalized), to use assistive devices, and to have contact with a medical care coordinator. It is possible that uninsured CSHCN are less severely ill than those with insurance, because once a child becomes significantly ill, there is an effort to obtain coverage, particularly Medicaid. However, uninsured children in the study sample did not have significantly better reported health or fewer activity limitations than insured children.
Uninsured CSHCN may be particularly vulnerable if they do not have a medical home. Foregoing visits to a physician may prevent continuity of care and result in an unrecognized need for additional services. Insurance expansions among children, such as the State Children’s Health Insurance Plan, may have improved insurance coverage and access to care for uninsured CSHCN. However, case studies of State Children’s Health Insurance Plan indicate varying degrees of coverage for mental health services, ancillary therapies, and other specialized therapies for CSHCN.54
Several limitations of this study should be noted. First, because this is a cross-sectional study, causal relationships cannot be assessed. Second, no statements can be made regarding the appropriateness or the quality of the services used. Third, survey content limited explanatory factors to traditional characteristics of the child and the family. Environmental and provider-related factors that better reflect the supply and availability of services may also influence service use and should be examined in future research. Previous studies also have shown that additional child and family factors, such as mother’s mental health status, utilization, and employment status, may influence the child’s use of health services.41,42 Also, better measures of severity level may be needed to control for variations in health status across CSHCN. Fourth, the NHIS is based on parental reports of health status and utilization of services. Parental reports may differ from assessments made by providers or children themselves. Moreover, assessment of similar symptoms or health characteristics may vary across parents. One recent study also suggests that screening instruments, including the QuICCC, may underidentify CSHCN who are black or Hispanic.55 Parental reported data also raise recall bias concerns. Few previous studies have addressed recall bias among parents reporting on behalf of their children. One study suggested a high level of agreement between parental reports and actual events.56 Recall may be less problematic when utilization is measured dichotomously rather than as the number of services used. However, parents may not be fully aware of health-related services provided in schools, and certain parents may experience language barriers or may not understand certain questions, such as those related to activity limitations. Fifth, although the NHIS-D provided considerable breadth, depth in some areas was limited. For example, in the case of some services (eg, transportation), respondents were asked whether the child used these services, without determining the purpose or content of the services used. Sixth, the data are almost 10 years old and may not reflect current utilization patterns. However, the NHIS-D remains the most comprehensive source of nationwide information on children with chronic conditions. More recent data are now available with the release of the National Survey on Children With Special Health Care Needs.57
Providing comprehensive, coordinated care to CSHCN challenges pediatricians, policy makers, and families. Before the fielding of the NHIS-D, national estimates of the utilization of services beyond basic medical care were largely unavailable for CSHCN. The data presented in this study provide baseline measurements of the use of a variety of services that are recognized as important in the care of CSHCN. Future studies should assess whether there have been substantial changes in the use of services and in the factors that determine use.
The differentials in service use by type of insurance found in this study suggest that providing adequate insurance coverage for all CSHCN remains an appropriate and important priority. Continued efforts also are needed to identify and remove financial and nonfinancial barriers to comprehensive medical and health-related services for particularly vulnerable subgroups of CSHCN, including minority children and those from families with lower levels of parental education.
This work was funded by the Partnership for Solutions: Better Lives for People with Chronic Conditions, a national program of the Robert Wood Johnson Foundation.
- Received October 14, 2002.
- Accepted February 13, 2003.
- Reprint requests to (W.E.W.) Department of Health Policy, Management, and Behavior, University at Albany, State University of New York, One University Place, Rm 167, Albany, NY 12144. E-mail:
Presented at the Pediatric Academic Societies Annual Meeting; May 4, 2002; Baltimore, MD.
- ↵McPherson M, Arango P, Fox H, et al. A new definition of children with special needs. Pediatrics.1998;102 :137– 140
- ↵American Academy of Pediatrics, Committee on Children With Disabilities. Care coordination: integrating health and related systems of care for children with special health care needs. Pediatrics.1999;104 :978– 981
- ↵American Academy of Pediatrics, Committee on Children With Disabilities. Managed care and children with special health care needs: a subject review. Pediatrics.1998;102 :657– 660
- ↵Pless IB, Pinkerton P. Chronic Childhood Disorder: Promoting Patterns of Adjustment. Chicago, IL: Yearbook Publishers; 1975
- ↵Halfon N, Inkelas M, Wood DL, et al. Health care reform for children and families: refinancing and restructuring the US child health system. In: Andersen RM, Rice TH, Kominski GF, eds. Changing the US Health Care System. San Francisco, CA: Jossey-Bass, Inc; 1996:229–254
- ↵Grason H, Morreale M. Health services for children and adolescents: a non-system of care. In: Stein REK, ed. Health Care for Children: What’s Right, What’s Wrong, What’s Next? New York, NY: United Hospital Fund; 1997:107–133
- ↵Newacheck PW, Strickland B, Shonkoff JP, et al. An epidemiologic profile of children with special health care needs. Pediatrics.1998;102 :117– 123
- ↵Newacheck PW, McManus MA. Financing care for disabled children. Pediatrics.1988;81 :385– 394
- ↵Newacheck PW, McManus M, Fox HB, et al. Access to health care for children with special health care needs. Pediatrics.2000;105 :760– 766
- Aday LA, Andersen R. A framework for the study of access to medical care. Health Serv Res.1974;Fall :208– 220
- ↵Adams PF, Marano MA. Current Estimates From the National Health Interview Survey, 1994. Hyattsville, MD: National Center for Health Statistics; 1995:10(193). US Department of Health and Human Services PHS Publ. No. 96-1521
- ↵National Center for Health Statistics. Data File Documentation, National Health Interview Survey of Disability, Phase 1, 1994 (machine readable data file and documentation). Hyattsville, MD: National Center for Health Statistics; 1996
- ↵National Center for Health Statistics. Data File Documentation, National Health Interview Survey Disability Followback on Children, 1994 (machine readable data file and documentation). Hyattsville, MD: National Center for Health Statistics; 1998
- ↵Stein REK, Westbrook LE, Bauman LJ. The Questionnaire for Identifying Children With Chronic Conditions: a measure based on a noncategorical approach. Pediatrics.1997;99 :513– 521
- ↵American Academy of Pediatrics, Committee on Practice and Ambulatory Medicine. Recommendations for preventative pediatric health care. Pediatrics.2000;105 :645– 646
- ↵SUDAAN [statistical software package]. Version 7.5.3. Research Triangle Park, NC: Research Triangle Institute; 1997
- ↵US Congress, Office of Technology Assessment. Children’s Mental Health: Problems and Services: A Background Paper. Washington, DC: US Government Printing Office; 1996. Publ. No. OTA-BP-H-33
- ↵US Department of Health and Human Services. Mental Health: A Report of the Surgeon General. Rockville, MD: US Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Center for Mental Health Services, National Institutes of Health, National Institute of Mental Health; 1999
- ↵Woodward CA, Boyle MH, Offord DR, et al. Ontario Child Health Study: patterns of ambulatory medical care and their correlates. Pediatrics.1988;82 :425– 434
- ↵Fox HB, McManus MA, Reichman MB. The Strengths and Weaknesses of Private Health Insurance Coverage for Children With Special Health Care Needs. Report prepared for the Health Resources and Services Administration/Maternal and Child Health Bureau. Washington, DC: Maternal and Child Health Policy Research Center; 2002
- ↵Fox HB, Newacheck PW. Private health insurance of chronically ill children. Pediatrics.1990;85 :50– 57
- ↵US General Accounting Office. Medicaid Managed Care: Serving the Disabled Challenges State Programs. Washington, DC: US General Accounting Office; 1996. Publ. No. GAO/HEHS-96-136
- ↵Fox HB, Graham RR, McManus MA, Chen CY. Issue Brief, Number 5. An Analysis of States’ CHIP Policies Affecting Children With Special Health Care Needs. Washington, DC: Maternal and Child Health Policy Research Center; 1999
- ↵US Department of Health and Human Services, Maternal and Child Health Bureau, Division of Services for Children With Special Needs. A National Agenda for Children With Special Health Care Needs. Measuring Success for Healthy People 2010. Washington, DC: US Department of Health and Human Services, Maternal and Child Health Bureau, Division of Services for Children With Special Needs; 1999
- Copyright © 2003 by the American Academy of Pediatrics