Correlates of Medicaid Status: An Examination of Children from Low-Income Families in Boston, Chicago, and San Antonio

Ronald Angel, University of Texas at Austin
Sonia Frias, University of Texas at Austin
Terrence Hill, University of Texas at Austin
Maren Andrea Jimenez, University of Texas at Austin

In 1997 the federal government replaced Aid to Families with Dependent Children (AFDC) with Temporary Assistance for Needy Families (TANF). Along with these reforms, the federal government established the State Children's Health Insurance Program (SCHIP) to provide health insurance coverage for children who were uninsured and ineligible for Medicaid through 18 years of age. Since welfare reform, the number of children covered by medical insurance has steadily declined (Committee on Child Health Financing 2001, Hakim, Boben, and Bonney 2000). During this time, SCHIP has been unable to effectively increase the number of children covered by medical insurance (Zuckerman et al. 2001). Considering that children from low-income families are more likely experience poor health than their affluent counterparts (Duncan and Brooks-Gunn 2000), the health insurance status of these children merits special consideration. The aim of this paper, then, is to identify the socio-demographic characteristics of the primary caregiver that predict the Medicaid status of children from low-income families. Given that welfare reform has transferred the responsibility for social programs from the federal government to the states, we also compare the effects of these characteristics across cities. In the end, our goal is to develop a better understanding of why some children are enrolled in Medicaid while others are not. For the purposes of this paper, we analyze data obtained from the Welfare, Children, and Families (WCF) project. The WCF project provides a relatively new source of longitudinal data from 2,400 families in Boston, Chicago, and San Antonio, with waves of collection in 1999 and 2001. In order to identify the socio-demographic characteristics of the primary caregiver that predict the Medicaid status of children from low-income families, we estimate a mutinomial logistic regression model for the log odds of having gained Medicaid by time two, having replaced Medicaid by time two with some other insurance, and having lost Medicaid by time two with no alternative coverage (all versus having kept Medicaid). To account for caregiver characteristics, we include measures of education, ethnicity, marital status, religious affiliation, relationship to focal child, citizenship, workforce participation, and English fluency.