ARTICLE |
a Department of Social and Behavioral Sciences
b Institute for Health Policy Studies, University of California, San Francisco, California
| ABSTRACT |
|---|
|
|
|---|
METHODS. A systematic literature review was conducted on the effects of insurance status, insurance type, and insurance features on access, utilization, satisfaction, quality, expenditures, and health status.
RESULTS. The strongest evidence emerged for the positive effects of insurance on access and utilization. Limited evidence on the effect of insurance on satisfaction with care showed improved satisfaction ratings for the insured. The studies with findings relevant to out-of-pocket expenditures for insured versus uninsured children with special health care needs all found significantly higher out-of-pocket burden and financial problems among the uninsured. Evidence was mixed for the effects of insurance type (public or private) and insurance characteristics (eg, managed care or fee-for-service payment mechanisms) on outcomes. None of the studies that we reviewed attempted to assess the impact of health insurance on health outcomes.
CONCLUSIONS. Our review of the literature found plentiful evidence demonstrating the positive and substantial impact of insurance on access and utilization. There also is clear evidence that insurance protects families against financially burdensome expenses. The evidence is less conclusive for satisfaction and quality and is nonexistent for health status. These latter outcomes should be the focus of future studies.
Key Words: health insurance health care delivery/access children with special health care needs
Abbreviations: CSHCNchildren with special health care needs MCHBMaternal and Child Health Bureau SCHIPState Children's Health Insurance Program FFSfee-for-service MCmanaged care
Children with special health care needs (CSHCN) constitute a particularly vulnerable subpopulation of children. They experience ongoing health issues that require continuing interaction with the health care system, and they are at risk for health complications that may undermine their normal functioning in the family, at school, and in the community. Families of CSHCN also may incur serious financial consequences as a result of the child's health care needs. Health insurance coverage has the potential to secure access to care and improve the quality of life for these children while protecting their families from financially burdensome health care expenses. Recognizing the importance of health insurance for this population, the federal Maternal and Child Health Bureau (MCHB) established access to adequate public or private health insurance coverage for all CSHCN as 1 of its key priorities or performance objectives, incorporating both continuity of coverage and adequacy of benefits.1 This performance objective proposes that all CSHCN will have continuous public or private coverage as well as adequate coverage (measured by family satisfaction with service benefits, provider networks, and out-of-pocket expenses). This agency-level priority subsequently was incorporated into the President's New Freedom Initiative.2
The Institute of Medicine report Coverage Matters: Insurance and Health Care3 argued the importance of insurance coverage for all populations, highlighting its role in access to care. Within the general population of children, having insurance coverage is associated with improved access and utilization patterns for children, including fewer unmet needs, increased likelihood of having a usual source of care, and greater use of well-child care.4,5 It is reasonable to hypothesize that the impact of insurance is likely to be more pronounced for CSHCN given their increased health care needs. Using the National Survey of CSHCN, the MCHB estimated that only 60% of CSHCN had continuous and adequate health insurance coverage during 2001.6
Early literature on financing health care for children with chronic conditions and disabilities identified some of the problems that are associated with absence of health insurance, including reduced access to care and increased financial burdens for families.4,7,8 Aday et al9 demonstrated insurance-related barriers to access among CSHCN using population-based survey data from the late 1980s. Researchers also demonstrated that cost sharing under private health insurance can result in increased financial burden for families.10,11
Before 1998, chronic condition lists were commonly used to operationalize the concept of special health care needs. Recognizing the limits of the condition-based approach, several leading academicians argued for using a consequence-based approach to defining the population with special needs.12,13 Building on these efforts, the MCHB developed and promulgated a definition of CSHCN as "those who have or are at risk for a chronic physical, developmental, behavioral, or emotional condition and who also require health and related services of a type or amount beyond that required by children generally."14 The adoption of a formal definition by the federal agency that was given lead programmatic responsibility for CSHCN opened the door for a multitude of analyses that profiled these children and evaluated how they and their families fare in the health care system.
The purpose of this review was to assess and synthesize recent research in the peer-reviewed literature that pertains to the role of insurance for CSHCN. Our literature searches revealed that the volume of research on this topic has grown markedly in recent years. Consequently, now is an opportune time to summarize these contributions and begin the process of formalizing an evidence base that can inform health policy decisions. Our intention then is to help establish such an evidence base by providing a literature-driven assessment of the role of health insurance in influencing access, utilization, satisfaction, quality, expenditures, and health for CSHCN. This review also serves to illuminate aspects of health insurance that require additional study.
| METHODS |
|---|
|
|
|---|
|
Inclusion/Exclusion Criteria
A total of 315 titles were identified using this search strategy. Each subsequently was scanned for relevance. Those that obviously were inappropriate were eliminated, and abstracts were considered for the remaining studies, followed by full article readings to determine final inclusion status. We established a set of inclusion and exclusion criteria to narrow the included articles to those that would contribute to the evidence base on health insurance for CSHCN. Specifically, we excluded studies that were not empirical, addressed populations outside the United States, focused on the adult population, had no apparent relationship to insurance, examined the effect of health on insurance (rather than the reverse), and those that simply examined the prevalence of health insurance. To increase generalizability to all CSHCN, studies that focused on a single illness also were excluded, although we included studies of children with disabilities because these children represent a large and important subpopulation of CSHCN. Similarly, we excluded studies that included multiple conditions when the authors did not include summary measures of chronic conditions grouped together.15 Furthermore, to meet our inclusion criteria for contributing to the evidence base, studies had to be quantitative; incorporate original analysis of primary or secondary data; and have an outcome measure for CSHCN related to access, utilization, satisfaction, quality, expenditures, or health. Moreover, the outcome had to be linked empirically to insurance status, type of insurance, insurance characteristics, or the MCHB's performance objective on health insurance coverage. Reference lists from all included studies also were scanned for possible relevant articles that were not detected by the PubMed Search.
A total of 35 studies met the criteria and are summarized in Table 1, which highlights key elements of the studies, namely the research design, sample characteristics, and key findings. By focusing on these areas, similarities and differences among the studies are distinguished most clearly. For example, 29 of 35 used a cross-sectional design, 21 included samples of >1000 children, and 15 used the 2001 National Survey of CSHCN. Analysis strategies varied greatly, as did outcome variables studied. Only findings that were significant at P < .05 are reported here.
|
| RESULTS |
|---|
|
|
|---|
Impact of Insurance on Access and Utilization
Strong evidence emerged for the importance of insurance in ensuring access to health care for CSHCN. In this literature review, 28 studies examined access and/or utilization outcomes related to insurance status, type, or characteristics.
Insurance Status
Sixteen studies investigated access measures for medical services in relation to presence or absence of coverage, and all but 116 showed that the insured children fared better than the uninsured on a variety of outcomes, including usual source of care, unmet needs, delayed or foregone care, and specialist use. Effect sizes, based on reported odds ratios, ranged from 2 to 7, indicating impressive advantages in access for insured CSHCN over their uninsured counterparts.1721 Two additional studies examined the impact of insurance continuity on access. Both Kane et al22 and Smaldone et al23 found that CSHCN with discontinuous coverage were significantly more likely to report unmet need for routine care and specialty care, as well as delayed or foregone care, than those who were insured all year.
Although there is strong evidence for the impact of insurance on medical services, evidence for the impact of insurance on dental care is less consistent. Whereas 1 study reported no difference between the insured and uninsured on access to and use of dental care,24 2 others showed that insurance was associated with reduced unmet dental needs.16,25
Insurance Type
The effect of insurance type (public, private, or mixed) on access and utilization outcomes is less clear. Four studies found that children with public insurance had better access and utilization than those with private insurance (less unmet need for vision care; lower likelihood of health planrelated access issues; and higher utilization rates of various services, including mental health, emergency department, and inpatient and outpatient care).17,2628 Two others20,24 showed that private insurance was superior (for dental care and usual source of care), whereas 418,19,23,29 found no difference (for delayed care, specialist use, and unmet need). Because these studies examined somewhat different outcomes, it would be premature to conclude that publicly funded insurance is associated with better access to care than private insurance for this population. It may be that for certain types of care, Medicaid is superior, whereas for other services, private insurance is better. Of course, variable reimbursement rates and variable coverage of benefits among state Medicaid and State Children's Health Insurance Programs (SCHIPs) as well as private insurers add to the difficulty of making comparisons between private and public insurance. More sophisticated comparisons that take into account such variability are needed to provide a clearer picture of the differences between private and public insurance.30 In general, effect sizes for analyses that compared private and public insurance were much smaller than those that were found for comparisons of insurance versus no insurance.
Insurance Characteristics
The effect of insurance characteristics on access and utilization outcomes was explored in several studies. Mitchell et al31 found no difference in utilization or unmet need between fee-for-service (FFS) and capitated managed care (MC), whereas 2 other studies showed greater unmet need and/or access problems for CSHCN in FFS than for their counterparts in MC.32,33 A large prospective cohort study of children with chronic conditions in Massachusetts examined the effect of removing gatekeeping on access to specialists. The authors found that the rate of visits did not increase significantly, although new patient visits to specialists as a proportion of all specialist visits did increase.34 Another study in Florida found that higher provider reimbursement rates in MC organizations was associated with increased access and satisfaction among families with CSHCN.35
Impact of Insurance on Satisfaction and Quality of Care
There is growing evidence on the effect of insurance on satisfaction with care; 8 studies in this literature review addressed satisfaction and/or quality outcomes.
Insurance Status
Dick et al36 found increased satisfaction ratings (ranging from a 0.4- to a 1.2-point increase on a 10-point scale) for CSHCN after enrollment in SCHIPs in 3 states, and van Dyck et al37 showed that families of uninsured CSHCN nationally were 1.75 times more likely to report dissatisfaction on 1 or more indicators of care than were their insured counterparts. In 1 of the few studies to examine transition services outcomes for adolescents, Scal and Ireland38 reported that insurance status was not significantly associated with adequacy of transition services.
Insurance Type
Newacheck et al,20 using national survey data, found that parents of publicly insured CSHCN were 1.4 times more likely to report dissatisfaction with services than their privately insured peers.
Insurance Characteristics
Only 1 study investigated the effects of insurance characteristics on satisfaction. That study by Mitchell et al31 reported no difference in satisfaction between FFS and MC arrangements in the Oregon Medicaid program.
There is limited published evidence on quality outcomes for CSHCN by insurance status, type of coverage, or coverage characteristics. Among adolescents aged 13 to 17 years, type of insurance was not significantly related to meeting the transition planning core outcome.39 Tippy et al40 showed that CSHCN with adequate health insurance were more likely to have comprehensive care in a medical home than those without adequate health insurance. Szilagyi et al21 examined quality outcomes for all children in their SCHIP cohort study but did not present results separately for the special health care needs subgroup.
Impact of Insurance on Expenditures
Ten studies reported findings for expenditure outcomes. Six of these focused on insurance status, 1 focused on type of insurance, and 5 addressed insurance characteristics.
Insurance Status
Three studies had findings relevant to out-of-pocket expenditures for insured versus uninsured CSHCN, and 2 of these found significantly higher out-of-pocket burden and financial problems among the uninsured.37,41,42 The effect sizes were substantial. For example, insured children with disabilities had 51% lower out-of-pocket expenses as a share of family income compared with uninsured children with disabilities. Similar findings were reported for insured and uninsured CSHCN using national health care expenditure survey data. In contrast, Davidoff et al16 examined the effects of SCHIP expansions for eligible children and reported that out-of-pocket spending was not significantly different before and after SCHIP expansions.
Insurance Type
One study found that type of insurance was not significantly associated with financial problems, need for additional income, or cutbacks in work hours; families of children who were privately insured were more likely to have spent at least $250 to $500 out-of-pocket in the previous year, whereas those with public insurance alone or in combination with private were more likely to have spent less than $250.43
Insurance Characteristics
No consistent pattern of findings emerged on the effect of insurance characteristics on expenditures. Chan and Vanderburg44 reported that CSHCN in Minnesota with preferred provider organization or health maintenance organization coverage incurred higher payments than those with indemnity plans, whereas a second study of children with chronic conditions in Massachusetts found that total expenditures declined over time for a gatekeeping group relative to a comparison indemnity group.45 Shenkman et al46 reported that monthly outpatient charges were highest for CSHCN in managed care health plans that used primarily FFS reimbursement strategies.
Impact of the MCHB's Health Insurance Performance Objective on Outcomes
As indicated at the beginning of this article, the MCHB's health insurance performance objective incorporates a multidimensional conception of insurance continuity and adequacy. Four studies have addressed the impact of achieving all or part of this performance objective. Smaldone et al23 and Baruffi et al47 evaluated state-level data from the 2001 National Survey of CSHCN and found greater difficulty using community-based services, more frequent delays in obtaining care, and a greater prevalence of unmet needs among those with inadequate coverage. Using national data, Kogan et al48 evaluated both continuity and adequacy of insurance and reported that CSHCN with inadequate insurance as well as those with discontinuous coverage were at increased risk for 5 of 7 outcome measures studied. In the study by Honberg et al,6 CSHCN who did not meet 1 or more of the elements of the performance objective had significantly greater rates of unmet needs, access, and financial problems.
Impact of Insurance on Health Outcomes
Health and functional outcomes probably are the most important domains to capture yet the most difficult to measure. The literature described herein demonstrates that health insurance improves access to care. It is reasonable to hypothesize that improved access should translate to improvements in health outcomes and functional ability over the long-term. Using the conceptual model in Fig 1 as a reference, health outcomes related to insurance in turn may affect school attendance, future employment, and productivity. Hence, generating evidence on the impact of health insurance on health outcomes is critical to developing a complete evidence base. None of the studies that we reviewed attempted to assess the impact of health insurance on these types of outcomes.
| DISCUSSION |
|---|
|
|
|---|
A significant limitation to the current evidence base is the heavy reliance on cross-sectional study designs. Although authors attempted to control statistically for confounding variables that could account for the observed differences, threats to internal validity remain. Because nonexperimental designs lack random assignment and a control group, causal inferences concerning the effect of insurance on outcomes cannot be made. A few studies in this review took advantage of natural experiments and used cohort designs in their analyses. By tracking outcomes for treatment and comparison groups over time, these study designs reduce threats to validity. An intensified focus on study design rigor is needed to build a stronger evidence base.
We also noted the absence of studies that examined whether insurance is related to improved health and functional outcomes for CSHCN. In past debates on health insurance expansions for children, policy makers' greatest concerns often were about the potential effects of expansions on the health of children. Collecting the needed data poses significant challenges in measurement. Current tools for measuring health and functional outcomes may not be sensitive enough to measure the modest changes in outcomes that may occur over the short run. In tandem, longer follow-up periods are needed to measure the long-term impacts of health insurance.
Our analysis of the literature on the role of insurance for CSHCN also makes clear that there is more evidence for the effects of insurance on access and utilization than for satisfaction, quality, expenditures, and health. Among the variables assessed in this literature review, satisfaction, quality, and expenditures outcomes are understudied and health outcomes have not been investigated at all. Additional comparisons of public versus private insurance also would be beneficial to the policy dialogue. Using the conceptual model in Fig 1 as a guide, several "downstream" outcomes that potentially are affected by insurance status have not been explored for CSHCN. These include long-term child development, family stress, quality of life, and long-term economic burden on the community. All of these outcomes are important aspects of the health insurance evidence base.
| CONCLUSIONS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Address correspondence to Paul W. Newacheck, DrPH, Institute for Health Policy Studies, University of California, 3333 California St, Suite 265 San Francisco, CA 94118-1944. E-mail: paul.newacheck{at}ucsf.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||