Evaluation of Selected Aspects of the National Medicare Education Program: Final Design Report. 4.4.4 Secondary Outcome Measures


A cha llenge for applying CEA to the NMEP is that different activities operate through different channels and may have different intended purposes. For example, the www.Medicare.gov Web site is not a direct substitute for the community outreach programs in the NMEP. Therefore, as a first step in conducting a CEA of the NMEP, one must identify a primary effectiveness measure that can be compared across activities. Improved beneficiary knowledge is one such measure and may be captured by a previously published index of Medicare beneficiary knowledge derived from questions on the MCBS (Uhrig et al., 2006Uhrig et al., 2006). The extent to which different NMEP activities have unique goals that are not shared by other arms of the NMEP will not be captured in a standard CEA framework but may be accommodated as desired by ASPE using a cost-consequence study.

Data on process outcomes may serve as secondary outcome measures for a cost-consequence study useful for external benchmarking. For each activity of the NMEP, one could gather data not only on costs, but also on the volume of services provided, reported by CMS. This may include the number of calls to the 1-800-MEDICARE helpline, the number of unique hits per day or year to the www.Medicare.gov Web site, and so on. These can be matched with corresponding costs to generate simple cost-consequence metrics.

For example, suppose that the primary outcome establishes that the 1-800-MEDICARE helpline is the most cost-effective NMEP activity. A natural extension of this finding is to know how this compares with other services for older adults. Rather than compare the primary CEA measure (e.g., $1,500 per additional correct item in the MCBS knowledge index), which is specific to the NMEP and has no meaning for other programs, cost-consequence facilitates a more natural comparison. If calls to 1 800-MEDICARE cost $1.00/minute, while calls to Social Security’s 1-800-number cost $0.50/minute, the most “cost-effective” NMEP activity clearly does not fare as well in an external comparison in this example. Thus, the two forms of cost analyses may be used in tandem to maximize the utility of the CEA analysis.

View full report


"nmepreport.pdf" (pdf, 401.55Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®