Specific Aims: Prompted by a series of federal initiatives, the health services research community is focused on identifying and documenting disparities in health and health care among racial/ethnic groups as well as across the income distribution. However, little of this growing body of work has addressed disparities of the economic burden of illness. The objective of this project is to examine the extent to which disparities exist in out-of-pocket health care expenses for children and families, and the role that health insurance plays in reducing or eliminating disparities.
The specific aims of this project are: 1) Do disparities exist in out-of-pocket expenditures among children and families with different incomes? 2) Do disparities exist in out-of-pocket expenditures among children and families with different racial/ethnic characteristics? 3) To what extent does health insurance coverage mediate these disparities? and 4) To what extent have there been changes since 1977 in disparities in out-of-pocket expenditures for children and families with different incomes and different race/ethnic characteristics? Research Design and Methods: This study will utilize three national health care expenditures surveys conducted for the Agency for Healthcare Research and Quality and its predecessor agencies: the 1977 National Medical Care Expenditure Survey, the 1987 National Medical Expenditure Survey, and the 1997 Medical Expenditure Panel Survey. We will conduct all analyses using families and then sample children as the unit of analysis. A unique aspect of our study is the analysis of expenditure data at the individual child level and the family level. In all analyses financial burden, the dependent variable, is operationalized in two ways: 1) as the ratio of annual out-of-pocket health care expenses to annual family income multiplied by 100 and 2) whether out-of-pocket health care expenses reach certain catastrophic thresholds (e.g. 15% of family income). To address specific aims 1 and 2 above, a series of cross-sectional analyses, incorporating bivariate and multivariate methods, will be conducted using the 1997 MEPS. Bivariate analysis will be used to assess whether disparities exist on an unadjusted basis. Multivariate analysis, including linear and logistic regression, will be used to assess whether disparities exist on an adjusted basis. Also using the 1997 MEPS, a hierarchical regression analysis will be used to assess whether health insurance is associated with a diminution of income and racial/ethnic disparities in financial burden (aim 3). For the trend analysis (aim 4), a regression-based differences-in-differences approach will be employed to assess changes in disparities over time. This analysis will incorporate pooled data from all three surveys. At least 3 products are expected: a paper describing disparities in financial burden across income and among racial/ethnic groups using family level data; a paper aimed at a pediatric audience that describes disparities using children as the unit of analysis; and a paper describing changes in the distribution of financial burden across income and race/ethnicity from 1977-97.
