opment of a predictive model of prospective risk. Both the risk variables and the resources variables must be recorded at the person level, so that the variables in each case can be aggregated to the health insurance unit (for aspects of modeling risk) and family levels (for comparing risk with resources). Sufficient information on family relationships must be included to enable the membership of each health insurance unit and family to be identified.
The panel looked closely at three longitudinal surveys: MEPS, the Survey of Income and Program Participation (SIPP), and the Health and Retirement Study (HRS). None of the three surveys collects all of the variables that would be required to develop a prospective measure of MCER, as described in Chapter 4. Most notably, none of the three surveys collects a description of the services and treatments covered by each person’s health insurance plan, and none of them collects sufficient information with which to assess each sample member’s potential liability for out-of-pocket medical costs. MEPS collected detailed information on the health insurance plans of sample members in 1996 but has not done so again. Other survey-specific data gaps exist as well, which limit how fully each survey could support the modeling of MCER. These are discussed in detail in the report.
In summary, none of the surveys is nearly as strong as one would like in its measurement of key baseline characteristics. With its strong measures of chronic health conditions and very high quality expenditure data, MEPS is clearly superior to SIPP. The HRS could provide a supplemental data source for the one-fifth of households that fall into its universe of people over age 50. Estimates from the HRS could be used to validate the model estimates from MEPS for this segment of the population (or perhaps just the elderly).
Once a model of MCER has been developed, the estimates could be used directly (in MEPS), or the predictive model could be applied to another data set that provides measures of the relevant baseline characteristics. The latter approach offers a way to make the measurement of MCER more timely and to extend the measure to a larger and possibly more representative sample. For production, in addition to MEPS, the panel considered the CPS ASEC, the National Health Interview Survey, the American Community Survey, and the Consumer Expenditure series quarterly survey. The CPS ASEC is the source of both the official poverty measure and the SPM, to which the MCER measure is intended as a companion. Producing the two measures from the same survey would enable more direct comparisons than if the two were based on different surveys. The CPS ASEC is the only one of the surveys that can estimate disposable income currently, using imputations for taxes and commuting expenses, but it lacks a measure of liquid assets and has limited information on health conditions. MEPS cannot currently estimate disposable income because it does not collect or