The Future of Children, Princeton - Brookings: Providing research and analysis to promote effective policies and programs for children.

While most payment policies are likely to create identical incentives for providing care for children and adults, some may have different or opposite effects. For this reason, it is useful to consider whether and how children's needs might differ from those of adults. In some areas, these differences may require only small adjustments if Medicare policies were to be adapted for children, but in other areas, there may be some additional concerns.

Health insurance coverage for children comes from a patchwork of sources, with about 9 million children having no coverage at all.⁸ The majority of children are covered through the employment-based family insurance of their parents. Most low-income children are covered by Medicaid, a public program whose eligibility, coverage, and reimbursement levels vary by state. Children who are uninsured frequently receive care in community-based clinics and hospital emergency rooms. They receive care in physicians' offices only if their parents can pay for it out of pocket.⁹ Thus, providers of care for children are paid by a variety of sources: Blue Cross and Blue Shield and commercial insurance companies (with a wide range of reimbursement rules, service coverages, claims forms, and requirements); Medicaid; and parents. These may be paid on a service-by-service basis or as part of a managed care plan. To this complexity in the insurance system must be added unreimbursed or charity care for the uninsured who cannot pay, the costs of which are shifted to the extent possible to other payers.

Children's needs often differ from those of adults. First, most children need only routine services with little ongoing case management. Thus, annual per capita spending on personal health care for persons through age 18, at $737, is lower than for the adult population, at $1,826.10 Annual per capita spending for children 3 through 12 years of age is much lower, at $525 (all figures are in 1987 dollars). But a small proportion of children are extremely sick and need much more highly specialized treatment.¹¹ Treatment of this kind is available primarily at regional centers, often children's hospitals and major teaching hospitals, because the relatively rare need for it makes local referral centers impractical and not cost-effective.

Second, data from the National Ambulatory Medical Care Survey (NAMCS) show that pediatric specialists devote a larger proportion of office visits to preventive services than do other physicians, but the difference is not dramatic.¹²,¹³ The proportion of visits for which the principal reason is “diagnostic, screening, and preventive” is 22.1% for pediatricians versus 15.7% for all physicians in office practice. In more detail, 5.7% of visits to pediatricians were for general examinations, and 13.5% were for well-baby care. In contrast, 4.0% of visits to all physicians (including pediatricians) were for general examinations, 2.1% for well-baby care, and 3.5% for prenatal care. These three categories of preventive care accounted for 19.2% of visits to pediatricians versus 9.6% of visits to all physicians. Despite these differences in preventive care, the overwhelming majority of visits to pediatricians are devoted to addressing illness.

Third, education and counseling play a more important role in health care for children than they do in adult care.¹⁴,¹⁵ (More discussion of these types of services can also be found in the article by Budetti and Feinson in this journal issue.) These activities account for a substantial part of well-baby care. Counseling about development and about adolescent problems may also be extensive.¹⁴ Because insurers often do not cover many education and counseling activities, data on their magnitude are hard to come by. Activities of this kind are either paid for out of pocket or they are coded as other services and indirectly paid for by insurers.

Fourth, pediatricians point out that most referral care is provided in institutional settings, for example, teaching hospitals. They explain that referral care comprises a smaller portion of physician services in pediatric care than in adult care and that the volume of service is often too small to support a community specialty practice.¹⁶

Finally, because inpatient care may be traumatic for children, many pediatricians use it only as a last resort.¹⁷ For example, children are far less likely to be hospitalized than adults. Thus, discouraging hospitalization may not be an appropriate incentive for inclusion in a payment system for children's health care, because most pediatricians already use this method of treatment only as a last resort.

If payment reform recognizes special needs of children, a fundamental issue is whether age is the most appropriate delimiting factor, or whether special pediatric adjustments ought to be centered on medical conditions and type of facility instead. The obvious, and simplest, approach would be to base payment on an age limit, such as 18, regardless of where or for what care is received. For children receiving inpatient services in general acute care hospitals, this approach would make sense. But an 18-year-old may be more like an adult than like an infant. Further, an age definition could be inappropriate for the types of illness and care that characterize the specialty hospitals in which very sick children are treated. Persons in their early twenties may be treated in these facilities for conditions such as leukemia, for example. Thus, if payments have been developed that effectively capture the costs of treating childhood leukemia, they may be appropriate for these older patients as well. Alternatively, facility type (such as children's hospitals) may be the relevant criteria for making special payment adjustments instead of age or condition. It may make sense to take age as the primary defining characteristic but to add secondary facility-defined or treatment-defined dimensions on the grounds that, under some circumstances, strict age limits do not make substantive sense.