Everyone who reports or writes on health access policy in Massachusetts should look at this report, Estimates of the Uninsurance Rate in Massachusetts from Survey Data: Why Are They So Different?, from the state’s Division of Health Care Finance and Policy, by Tuesday.
Why Tuesday? On Tuesday the US Census Bureau releases its “Current Population Survey” (CPS), which is the federal government’s attempt to estimate the number of uninsured. The report explains the shortcomings of the CPS uninsured estimate, and compares the various surveys. It will be updated with the CPS data shortly after the findings are released.
Every year there is a lots of angst around the release of the CPS numbers. For the past few years, the administration has hustled out its state survey ahead of or simultaneously with the CPS report, to get ahead of the news cycle. The problem is that for a number of reasons, detailed in the report, the CPS numbers overestimate the number of uninsured. The CPS individual state numbers are particularly problematic, since the sample sizes are fairly small and one needs to average several years to get truly valid data.
The state-sponsored survey has been problematic as well. The 2006 survey originally found only only 372,000 uninsured, which we were quite skeptical of at the time. This low count led in part to low estimates of CommCare eligibles, leading to too-low budget allocations, leading to the need for additional funds and the perception of health reform as suffering from out of control spending. This new study postulates that the survey respondents with Uncompensated Care Pool eligibility may have consided themselves insured, leading to the undercount of the uninsured.
The good news is that the state is revamping its own survey to take into account the best practices of all the various methodologies. The new survey will include a large sample size, include cell-phone only households, and use both phone and web-based responses. The results for 2008 will not be ready in time to compete with the federal CPS release Tuesday.
We would also recommend the DHCFP study to those single-payer advocates who ground their assessment of the progress of chapter 58 using the CPS as their base. For example, despite an expert consensus that the CPS numbers overstate the uninsured, Drs. Steffie Woolhandler and David Himmelstein write frequently that the Census Bureau numbers are the right measure of the uninsured. This leads to them concluding that little progress is being made in covering the uninsured.
We’re pleased at the honest self-assessment contained in DHCFP’s latest study. We hope that reporters and policymakers will examine the study in the next 36 hours.
Brian Rosman
