Low Take-Up in Medicaid: Does Outreach Matter and for Whom?

Of the ten million children in the United States who lacked health insurance in 1996, an estimated 4. 7 million were eligible for Medicaid but not enrolled (Thomas Selden et al. , 1998). In response, federal and state governments have recently devoted up to 500 million annually to the development of outreach campaigns to increase take-up among those eligible. However, little is known about the reasons families fail to enroll, how to increase enrollment, or whether outreach can work. There is also no evidence that early enrollment in Medicaid improves outcomes. Children in need of hospitalization who are eligible for Medicaid but not enrolled are typically enrolled in Medicaid at the point of hospitalization. In addition, estimation of the impact of early enrollment in Medicaid on health-care utilization and child health is hindered by the endogeneity of the enrollment decision: children in greater need of medical care are more likely to enroll. Thus, straightforward estimation of the impact of Medicaid enrollment on child health will underestimate the effect of Medicaid enrollment on health. I examine both the causes and consequences of low take-up in Medicaid using data on Medicaid enrollment in California from 1996 to 2000 and the timing and placement of communitybased application assistants that were part of an outreach campaign launched in mid-1998. I � nd the most profound effects of outreach on those with the highest costs of enrolling: Hispanic and Asian children, who have greater language and immigration concerns than other families. Access to bilingual application assistants increases new monthly Medicaid enrollment among Hispanics by 4. 6 percent and among Asian children by 6 percent on average relative to other children in the same neighborhood.