Abstract
Annual vs. Monthly Self-Reports of Health Insurance Coverage: Implications for Estimates of the Efficacy of the State Children's Health Insurance Program. Thomas Buchmueller, University of Michigan, and Lara Shore-Sheppard, Williams College.
Description
Over the past two decades there has been a dramatic expansion in public health insurance for children. As a consequence, over a quarter of all children in the US are enrolled in either Medicaid or a separate State Children’s Health Insurance Program (SCHIP) plan and the percentage of children who are uninsured is substantially lower than the rate for adults. A main objective of expanding public health insurance coverage for children is to increase timely and appropriate access to health care. A large literature documents a positive relationship between insurance coverage and various measures of utilization for children. Most studies in this area use data from annual surveys in which insurance coverage is often measured at a point in time and utilization is measured by retrospective reports. An important concern with such data is that in many cases an individual’s insurance coverage at the time of the survey will not match up with his or her coverage during the period for which utilization is recorded. Analyses that combine individuals with partial and full-year coverage are likely to produce biased estimates of the effect of insurance. In addition to this methodological issue, within-year transitions in insurance have potentially important policy implications as children who move on and off public insurance or between public and private coverage will suffer in terms of access to timely and appropriate care.
In this study we will use data from the 1993, 1996, 2000 and 2004 waves of the Survey of Income and Program Participation (SIPP) to analyze the relationship between insurance coverage and health care utilization among children. A main focus of the analysis will be to test for differences between children with continuous insurance over a period of a year and children with changes in coverage. The analysis will be designed to address both substantive policy questions as well as methodological issues of interest to researchers. In particular, the SIPP data will allow us to show how measuring insurance coverage sub-annually affects the estimated relationship between coverage and utilization. The effects of sub-annual measurement will be documented by comparing alternative insurance measures created using the SIPP data and by comparing the SIPP to the National Health Interview Survey.
