private physician offices or HMOs (Wood et al., 1994). A national study found, however, that infants were more likely to receive well-child visits and immunizations in private physician offices or HMOs than in public health clinics (Mustin et al., 1994). There is tremendous variation from community to community in the provision of care for low-income groups (Baxter and Mechanic, 1997), and it is very difficult to generalize from the existing studies.
It is well-known that the cost savings under managed care are derived from the increased use of primary care and decreased utilization of specialty and inpatient care. Although this pattern of utilization may be appropriate for children who are generally healthy, its potential impact on children with chronic conditions and special health care needs causes concern (Schlesinger and Mechanic, 1993). Some children's health advocates fear that managed care has an inherent disincentive to provide treatment, although there is little scientific evidence to substantiate this claim. Because the actual costs of inclusion of children with chronic conditions and special needs in managed care plans and networks have not been systematically reported or studied, the issue remains controversial.
Another concern about managed care is that many health plans do not include pediatric specialists in their approved provider networks. As a result, children with complex illnesses may be seen by adult specialists who are trained to care for adults with similar illnesses but who lack the knowledge of the specific manifestations of disease in children and the differences between children and adults (Cartland and Yudkowsky, 1992).
Because at least 36 states now have some form of mandated managed care for at least some children with special needs, several strategies have been proposed to increase the likelihood that their medical needs will be met. These include ensuring that there are a significant number of local pediatric specialists in networks before contracts are negotiated (Newacheck et al., 1994); setting up separate payment systems for children with certain medical conditions or "carving out" these groups (Neff and Anderson, 1995); and developing quality assurance measures that assess quality of care for children with chronic conditions, including clinical practice guidelines and practice parameters, and consumer protections, such as written policies concerning grievances and availability of services (IOM, 1997a; Newacheck et al., 1994).
Long-term strategies include developing better pricing systems, particularly risk adjustment mechanisms that adjust payment rates to reflect the higher expected costs of care for children with chronic conditions. Although health policy and financing experts have been recommending better risk adjustment mechanisms for several years, much work remains to be done to resolve technical and methodological disagreements among proponents of different approaches (Iezzoni, 1997; IOM, 1996).
Despite these technical difficulties and the need for better information involving risk selection and costs of coverage, there is no question that children with special needs typically have additional, nonmedical needs. These include supplemental or adjunctive therapies; specialized transportation, supplies, and equipment; links with schools and other community-based organizations; and family support, including respite care and assistance with out-of-pocket (noncovered) expenses.
Insurance coverage for children is variable and fragmented, and many controversial issues remain unresolved. It is generally agreed that the most comprehensive and child-specific package of benefits is Medicaid, which offers a full range of medically necessary care for children, including children with special needs. In some cases, privately insured children can receive additional assistance from public sources, even Medicaid, to pay for what their private plan does not cover.
Some opponents of Medicaid believe that if private insurers do not guarantee comprehensive coverage, public programs should not do so either. Others take a pragmatic position that it is less expensive for society in the long run to provide preventive services and early interventions for low-