Children's Health

Researchers explore children's access to and use of health care services

Access to and use of health care services by children in the United States is the focus of the inaugural issue of the journal Ambulatory Pediatrics. Six articles appearing in the journal were either authored by researchers at the Agency for Healthcare Research and Quality or supported by AHRQ grants. The first study shows that children's use of health care services varies considerably depending on the type of insurance coverage they have, and that children's hospitalization rates vary substantially across States. The second study found few differences between Medicaid and commercially insured children in a Northern California health maintenance organization in access to care, use of health care services, and satisfaction with care.

Four other studies provide a strong methodologic basis for pediatric health services research. In the first study, the researchers review the recent history of child health services research. The other three studies deal with methodologic issues in measuring quality of child health care, the development and use of pediatric outcome measures, and problems faced by investigators involved in pediatric effectiveness research.

McCormick, M.C., Weinick, R.M., Elixhauser, A., and others (2001, January). "Annual report on access to and utilization of health care for children and youth in the United States—2000," Ambulatory Pediatrics 1(1), pp. 3-15.

This article provides an update on insurance coverage, use of health care services, and health expenditures for children based on data gathered through the Medical Expenditure Panel Survey, an ongoing nationally representative survey of medical care use and expenditures and the Healthcare Cost and Utilization Project, a family of administrative databases. Both data collection efforts are sponsored by AHRQ.

The researchers found that there were few changes in insurance coverage of children and adolescents between 1996 and 1998. About two-thirds of American children were covered by private insurance, and 19 percent were covered by public sources. The remaining 15 percent were uninsured. Of the 72 percent of children who had at least one visit to a doctor's office, the average number of visits was 3.9. However, this ranged from 2.7 among the uninsured to 4.2 for those with private insurance.

Slightly more than half of children had a prescription, with an average of 5.4 prescriptions per child with a prescription. The majority of children (85 percent) incurred medical expenditures, and for those children, medical expenditures averaged $1,019. Private health insurance was by far the largest payer of medical care expenses for children (55 percent). However, nearly 21 percent of all expenditures were paid out-of-pocket by children's families.

The researchers also found substantial differences in average length of pediatric hospitalization across States, ranging from 2.7 to 4 days. Rates of hospital admission through the emergency department also varied from 9 percent to 23 percent across States. Injury was the major reason for hospitalization, accounting for one in six hospital stays among children 10 to 14 years of age. In the 10- to 17-year-old age group, one in seven hospital stays was related to a mental disorder. Among 15- to 17-year-old youths, more than one-third of all hospital stays were related to pregnancy and childbirth.

Reprints (AHRQ Publication No. 01-R036) are available from the AHRQ Publications Clearinghouse.

Newacheck, P.W., Lieu, T., Kalkbrenner, A.E., and others (2001, January). "A comparison of health care experiences for Medicaid and commercially enrolled children in a large, nonprofit health maintenance organization." Ambulatory Pediatrics 1(1), pp. 28-35.

Enrollment of children in Medicaid managed care programs has grown dramatically in recent years. To date, however, few studies have attempted to assess whether differences exist in the types of care delivered to Medicaid and commercially enrolled children within the same HMO. This study addressed that issue and found few differences in the quality of care provided to Medicaid and commercially enrolled children in a large non-profit HMO.

The researchers compared access to care, satisfaction, and use of services between Medicaid and commercially sponsored children enrolled in a large California HMO during 1998 through use of a telephone survey and administrative data. They found few differences between the two groups in either of the three measures of care. Where access differences were present, such as problems in finding a personal care provider, the differences favored Medicaid-enrolled children. Medicaid enrollees experienced significantly fewer access problems and barriers to care than commercial enrollees, even after adjustment for confounding factors.

The investigators found a difference only in the volume of emergency department (ED) visits between Medicaid and commercial enrollees across the six use variables they examined; 19 percent of Medicaid enrollees had at least one ED visit, compared with 16 percent of commercial enrollees. They found no differences between the two groups among the four care satisfaction variables and two global assessments of care received. These results suggest that Medicaid-enrolled children in HMOs experience care that is as good as or better than the care provided to their commercially enrolled counterparts.

Reprints (AHRQ Publication No. 01-R039) are available from the AHRQ Publications Clearinghouse.

Lohr, K.N., Dougherty, D., and Simpson, L. (2001, January). "Methodologic challenges in health services research in the pediatric population." Ambulatory Pediatrics 1(1), pp. 36-38.

These authors review the recent history of child health services research (CHSR). They point out that children have long been underrepresented in medical research, and a series of agenda setting meetings held between 1994 and 1997 highlighted the importance of addressing CHSR. Congressional interest has increased budgets at the National Institutes of Health and AHRQ for research on children, and AHRQ has significantly increased its support for CHSR. The Agency funded more than $9 million in new grants focused on child health in FY 1999 and published a strategic plan for children's health services research. AHRQ's reauthorization legislation in 1999 included several provisions to promote CHSR, including naming children a priority population, calling for research networks, and linking CHSR to priorities for training future investigators.

The authors point out that CHSR must address multiple methodologic challenges which affect study design, the research itself, and the reporting of results. Some of these challenges are common to the field of health services research, some stem from the characteristics that children share with other vulnerable populations (for example, the frail or aged elderly), and others are unique to children and adolescents. For example, children's health outcomes may vary depending on their developmental stage. Also, important factors lie within the child's environment, such as the family or community. This article provides an introduction and overview of other articles published in this issue of Ambulatory Pediatrics on methodologic issues in child health services research.

Reprints (AHRQ Publication No. 01-R038) are available from the AHRQ Publications Clearinghouse.

Christakis, D.A., Johnston, B.D., and Connell, F.A. (2001, January). "Methodologic issues in pediatric outcomes research." Ambulatory Pediatrics 1(1), pp. 59-62.

Many of the approaches, models, and techniques used in pediatric outcomes research have been imported uncritically from experience with adult populations. As a result, some of the most interesting and critical aspects of pediatric outcomes research have yet to be fully developed, according to these researchers. They cite several critical challenges in conceptualizing and measuring outcomes for children and adolescents. For example, researchers have to take into account the dynamics of childhood development by identifying outcome measures that discriminate between changes in status or function that are the expected result of normal development and those that might reasonably be credited to the effects of an intervention.

Also, the priority given to different types of health outcomes varies by age, developmental stage, and family and community expectations. For example, what most troubles a 5-year-old girl with acute lymphocytic leukemia may be the pain associated with treatment. A teenager who faces a similar challenge may suffer most from the dependence associated with illness and the threat the illness represents to the developmental task of individuation in adolescence.

Also, most children are healthy. As a result, health supervision or well-child care accounts for a substantial portion of the average pediatrician's day. Yet it is difficult to measure the benefit of well-child care in terms of traditional health status outcomes for most children. What's more, there is a long time delay between health interventions for certain acute childhood conditions such as obesity and significant outcomes of interest, for example, heart disease. Researchers also must take into account the crucial roles that the family system, neighborhood, and community play in fostering or hindering good outcomes for the child.

Kaplan, S.H., Greenfield, S., Connolly, G.A., and others (2001, January). "Methodologic issues in the conduct and interpretation of pediatric effectiveness research." Ambulatory Pediatrics 1(1), pp. 63-70.

These researchers highlight the unique features and methodologic problems of pediatric effectiveness research. The definition of pediatric care has expanded to include problems with genetic origins due to advances in gene therapy, and problems with social behavior, such as domestic violence, teen pregnancy, and guns in households. Providers are being asked to add these problems to routine clinical screening, including identification of high-risk families—for example, those with compromised household safety. Effective management of these problems often involves people other than clinicians. Defining the boundaries of care for these problems is a critical first step in conducting pediatric effectiveness studies, according to these authors.

Researchers also must choose appropriate outcome measures. The time window needed to observe changes in outcomes related to effective care also can affect the choice of appropriate outcome measures. For example, studies of effective pediatric care for a chronic disease such as asthma or juvenile diabetes over a certain time period require a broader array of outcomes, including both clinical and general health status measures. Also, for some types of pediatric interventions, the child is not the sole or even the primary target of the intervention; instead it is the family, parents, or even the community.

Certain requirements must be met to conduct methodologically sound pediatric effectiveness studies. First, development of well-tested outcome measures in pediatrics, which has lagged far behind that for adults, is needed. Second, case-mix adjustment measures that go beyond diagnoses and include multiple clinical severity variables must be developed for children. Third, more methodologic attention is needed to ensure that the target whose care is being evaluated is specified at the earliest design stages of effectiveness studies in order to choose the appropriate sampling units, comparison groups, and analytic methods. Finally, long-term observational studies are needed to understand how care versus child, family, or environmental characteristics affects outcome measures over time.

Palmer, R.H., and Miller, M.R. (2001, January). "Methodologic challenges in developing and implementing measures of quality for child health care." Ambulatory Pediatrics 1(1), pp. 39-52.

Research on methods of quality measurement for child health care is urgently needed, given the limited number of measures currently available to assess care provided to children, assert these authors. One potential shortcut is to adapt existing process-based measures developed for adults. For example, there are multiple measures for routine screening tests in outpatient care for adults such as the HEDIS 2000 measure of breast cancer screening and measures of appropriate use of diagnostic radiology and laboratory services from the Developing and Evaluating Methods to Promote Ambulatory Care Quality Project. This prior work can provide templates for measures that cover multiple aspects of well-child care, such as vision and hearing screens and dental referrals.

For conditions that affect children and adults, measures designed for adults could be adjusted to reflect the specific differences in care process needed for children. For example, the IMSystem measures for patient education on insulin use could capture parental education instead. For the many conditions that affect children only, however, researchers need to develop new quality measures that are unique to childhood.

A major challenge is finding ways to overcome the problem that typifies pediatrics, that of only a few cases of each of a large number of serious chronic illnesses. The researchers suggest two possible approaches: one is to aggregate cases across conditions to create generic measures, such as followup diagnostic tests performed in the emergency department; and the other is to create composites of quality of life, functional status, and routine followup evaluations across multiple chronic diseases.

A particular problem is capturing the consequences of childhood illness for both children and their parents. Further refinement of child survey techniques may address these issues. Last, a different type of challenge is to find techniques to integrate data about the care that children receive across different health care settings and across non-health care agencies that contribute to children's health.

Reprints (AHRQ Publication No. 01-R037) are available from the AHRQ Publications Clearinghouse.

Return to Contents
Proceed to Next Article