that it should increasingly become the primary source of the estimates. This has practical consequences, as the workshop was reminded by David Johnson. The Census Bureau now receives about $20 million annually to pay for the additional work associated with collecting the children’s health insurance data on the CPS and to fund a program of research and modeling for these data. These funds would be put in jeopardy if the user community shifted to sole use of the ACS as the source of these data.
The discussion at the workshop focused on the meaning of this trend for these major household surveys. A few participants speculated that there would be increasing pressure to improve measurement of other factors other than coverage that affect the uninsured, such as health status and access. Moreover, it was noted, these additional data items would be most useful if they were ready in time to measure the coverage and effectiveness of the post-2014 program changes, although it was recognized that the next window of opportunity for adding or adjusting questions in the ACS will not occur until 2018.
In the near term, the upcoming, near-simultaneous release in September 2010 of the ACS and CPS estimates of children’s coverage in 2009 will draw attention to reconciliation and preference issues. Organizations such as the State Health Access Data Assistance Center are focusing on improving understanding of the ACS this year and are engaged in finding out from states what numbers they use so they will be ready for this event.
The workshop discussion focused on the critical factor of time in considering necessary changes to data systems. Officials of the U.S. Department of Health and Human Services reminded everyone that many things will change after 2014, including coverage under Medicaid.
The representatives of the Office of the Assistant Secretary for Policy and Evaluation stressed that the federal government must be flexible during this implementation period and that projects should be put in place to perform some timely infrastructure studies as the implementation activities unfold. Such studies might well learn from the infrastructure studies in Massachusetts that were reported at the workshop, which have helped to assess the effects of the health care changes there.