BACKGROUND: The State Children's Health Insurance Program (SCHIP) was enacted in 1997 to provide health insurance coverage to uninsured low-income children from families who earned too much to be eligible for Medicaid. OBJECTIVES: To develop a "baseline" portrait of SCHIP enrollees in 5 states (Alabama, Florida, Kansas, Indiana, and New York) by examining: 1) SCHIP enrollees' demographic characteristics and health care experiences before enrolling in SCHIP, particularly children with special health care needs (CSHCN), racial and ethnic minority children, and adolescents; 2) the quality of the care adolescents received before enrollment; and 3) the changes in enrollee characteristics as programs evolve and mature. METHODS: Each of 5 projects from the Child Health Insurance Research Initiative (CHIRI) surveyed new SCHIP enrollees as identified by state enrollment data. CHIRI investigators developed the CHIRI common core (a set of survey items from validated instruments), which were largely incorporated into each survey. Bivariate and multivariate analyses were conducted to ascertain whether there were racial and ethnic disparities in access to health care and differences between CSHCN and those without. Current Population Survey data for New York State were used to identify secular trends in enrollee characteristics. RESULTS: Most SCHIP enrollees (65% in Florida to 79% in New York) resided in families with incomes < or =150% of the federal poverty level. Almost half of SCHIP enrollees lived in single-parent households. A majority of SCHIP parents had not had education beyond high school, and in 2 states (Alabama and New York) approximately 25% had not completed high school. The vast majority of children lived in households with a working adult, and in a substantial proportion of households both parents worked. Children tended to be either insured for the entire 12 months or uninsured the entire 12 months before enrolling in SCHIP. Private insurance was the predominant form of insurance before enrollment in SCHIP in most states, but 23.3% to 51.2% of insured children had Medicaid as their most recent insurance. HEALTH CARE USE AND UNMET NEEDS BEFORE SCHIP: The vast majority of all SCHIP enrollees had a usual source of care (USC) during the year before SCHIP. The proportion of children who changed their USC after enrolling in SCHIP ranged from 29% to 41.3%. A large proportion of SCHIP enrollees used health services during the year before SCHIP, with some variability across states in the use of health care. Nevertheless, 32% to almost 50% of children reported unmet needs. CSHCN: The prevalence of CSHCN in SCHIP (between 17% and 25%) in the study states was higher than the prevalence of CSHCN reported in the general population in those states. In many respects, CSHCN were similar to children without special health care needs, but CSHCN had poorer health status, were more likely to have had unmet needs, and were more likely to use the emergency department, mental health care, specialty care, and acute care in the year before enrolling in SCHIP than children without special health care needs. RACE AND ETHNICITY: A substantial proportion of SCHIP enrollees were black non-Hispanic or Hispanic children (Alabama: 34% and <1%; Florida: 6% and 26%; Kansas: 12% and 15%; and New York: 31% and 45%, respectively). Minority children were poorer, in poorer health, and less likely to have had a USC or private insurance before enrolling in SCHIP. The prevalence and magnitude of the disparities varied among the states. QUALITY OF CARE FOR ADOLESCENTS: Seventy-three percent of adolescent SCHIP enrollees engaged in one or more risk behaviors (ie, feeling sad or blue; alcohol, tobacco, and drug use; having sexual intercourse; and not wearing seat belts). Although almost 70% of adolescents reported having had a preventive care visit the previous year, a majority of them did not receive counseling in each of 4 counseling areas. Controlling for other factors, having a private, confidential visit with the physician was associated with an increased liked likelihood (2-3 times more likely) that the adolescent received counseling for 3 of 4 counseling areas. TRENDS OVER TIME: New York SCHIP enrollees in 2001, compared with 1994 enrollees in New York's SCHIP-precursor child health insurance program, were more likely to be black or Hispanic, older, from New York City, and from families with lower education, income, and employment levels. A greater proportion of 2001 enrollees was uninsured for some time in the year before enrollment, was insured by Medicaid, and lacked a USC. Secular trends in the low-income population in the state did not seem to be responsible for these differences. Program modifications during this time period that may be related to the shift in enrollee characteristics include changes to benefits, outreach and marketing efforts, changes in the premium structure, and the advent of a single application form for multiple public programs. CONCLUSIONS: SCHIP enrollees are a diverse group, and there was considerable variation among the 5 study states. Overall, SCHIP enrollees had substantial and wide-ranging health care needs despite high levels of prior contact with the health care system. A sizable minority of SCHIP enrollees has special health care needs. There is racial and ethnic diversity in the composition of enrollees as well, with racial and ethnic disparities present. The quality of care adolescents received before enrollment in SCHIP was suboptimal, with many reporting unmet health care needs and not receiving recommended counseling. The characteristics of SCHIP enrollees can be expected to change as SCHIP programs evolve and mature. POLICY IMPLICATIONS: 1) Benefits should be structured to meet the needs of SCHIP enrollees, which are comparable to Medicaid enrollees' needs in many respects. 2) Provider networks will have to be broad if continuity of care is to be achieved. 3) Multiple outreach strategies should be used, including using providers to distribute information about SCHIP. 4) The quality of care delivered to vulnerable populations (eg, minority children, CSHCN, and adolescents) should be monitored. 5) States and health plans should actively promote quality health care with the goal of improving the care received by SCHIP enrollees before enrollment. 6) States will have to craft policies that fit their local context. 7) Collecting baseline information on SCHIP enrollees on a continuous basis is important, because enrollee characteristics and needs can change, and many vulnerable children are enrolling in SCHIP.