Background
The HealthWave program was introduced in January 1999 to provide health insurance to low-income uninsured children in Kansas. How successful has the state been in enrolling eligible children since that time? On the one hand, 49,224 more Kansas children were enrolled in public health insurance in March 2001 than at the end of 1998. On the other hand, prior to HealthWave’s implementation in 1999 it was estimated that there were 60,000 lowincome uninsured children in Kansas. This number quickly became the “target” for enrollment efforts. Should we now assume that there are 11,000 children left in Kansas who could benefit from public health insurance? This Issue Brief discusses the actual and potential enrollment of children in public health insurance some of the challenges of assessing success, and implications for these programs in the future.
Kansas HealthWave Evaluation Project
This three-year, $1.4 million evaluation of the HealthWave program includes a survey of 1,500 HealthWave and Medicaid families, focus groups with HealthWave families, analysis of enrollment and other administrative data, and extensive discussions with program administrators and health care providers from around the state. The project runs through September 2002. KHI serves as the lead organization for the project, where Robert St. Peter, M.D., is the principal investigator and R. Andrew Allison, Ph.D., is the project manager and co-investigator.
Collaborating Organizations
-
- Kansas Health Institute
- Kansas Department of Social and Rehabilitation Services
- Kansas Department of Health and Environment
- University of Kansas Medical Center: Department of Health Policy and Management; School of Nursing; Department of Preventive Medicine, School of Medicine-Wichita
- Kansas State University: Survey Research Laboratory
- Kansas Foundation for Medical Care
Project Funding
This study, funded through a cooperative agreement from the Agency for Healthcare Research and Quality (AHRQ) [HS10536-01], is part of the Children’s Health Insurance Research Initiative (CHIRI), which is co-funded by AHRQ, the David and Lucile Packard Foundation and the Health Resources and Services Administration. The project also receives local funding from the Kansas Health Foundation, the United Methodist Health Ministry Fund, and the Prime Health Foundation.
Enrollment Gains in Public Health Insurance
After 27 months, enrollment in the HealthWave program reached 20,154 children. In addition, the total number of children enrolled in Medicaid increased by more than 29,000 over the same period. In sum, enrollment of children in Kansas’ two public health insurance programs increased by 49,224 — a remarkable 50 percent increase.
While this is an impressive accomplishment, the gains might have been even greater. More than 80,000 children have been newly enrolled in these two programs since January 1, 1999. However, due to the large number of children who have left public health insurance since that time, total enrollment of children in the two programs had increased by the net of only 49,224. This attrition is due to dropout from the programs during the first year of eligibility, coupled with additional losses occurring at the one-year anniversary re-enrollment date. While some of this attrition is expected and appropriate (due to changes in eligibility), much of it is unexplained and poses a significant challenge to program growth [see Issue Briefs Number 10 and Number 11 for a more extensive discussion of attrition].
Is the Prevailing Estimate of Uninsured Kansas Children Useful?
The prevailing estimate is that there were 60,000 low-income, uninsured children in Kansas before HealthWave. However, this estimate is so imprecise and contains so many potential biases that it should not be used to determine how many children were eligible to be enrolled in public health insurance when the HealthWave program began — nor should it be used to determine how many remain unenrolled today.
-
- How the estimate might understate potential enrollment in HealthWave and Medicaid. The 60,000-child estimate is based on data from the Census Bureau’s Current Population Survey (CPS). It represents the number of children in families with annual incomes below 200 percent of the federal poverty level who were without health insurance of any kind for a full calendar year. However, many additional children may be in families whose income drops, or who lose insurance coverage, for part of the year. Including such children would certainly have increased the 60,000-child estimate. Another group of children that could have been included in the 60,000-child estimate is the “under-insured”; i.e., low-income families with insurance that is not adequate to make the care they need affordable. For example, insurance that excludes certain services, has deductibles and copays that are not affordable, or has very low coverage caps.
- How the estimate might overstate potential enrollment. The 60,000-child estimate included some children who were not eligible for either program, even though they were uninsured. For example, children in low-income families who have a parent eligible for the state benefit plan (e.g., state employees and some school district employees) and children who are undocumented aliens or recent immigrants. In addition, the CPS is known to routinely underestimate the number of children covered by Medicaid, mischaracterizing 17-24 percent of Medicaid recipients as uninsured.
- How the estimate is imprecise. Even without these biases and limitations, the number 60,000 itself is too imprecise to be used to assess HealthWave’s enrollment success. In small states like Kansas there are not enough survey participants to get accurate statewide estimates of insurance rates for children; e.g., fewer than 500 Kansas children are included in each year’s sample. To try to make the CPS data useful for small states, the Census Bureau combines data from three years to give a “rolling average.” The 60,000-child number of eligibles in 1999 is based on data collected for the years 1993-95. Even after combining three years of data, the CPS estimates are highly imprecise for small states, easily varying by +/- 12,000 children.
- A better estimate to come. Fortunately, Kansas soon will have better data on the number of uninsured children. A recently completed 8,000household survey, funded by a grant from the federal Health Resources and Services Administration to the Kansas Insurance Department, will support a much more precise estimate of health insurance coverage in the State than has ever before been possible.