The 130th Annual Meeting of APHA

4168.0: Tuesday, November 12, 2002 - Board 5

Abstract #40893

Methodological considerations in using the Medical Expenditure Panel Survey to estimate medical expenditures for decedents and survivors

Steven R. Machlin, Center for Cost and Financing Studies, Agency for Healthcare Research and Quality, 2101 East Jefferson St., Suite 500, Rockville, MD 20852, 301-594-0968, smachlin@ahrq.gov, William Yu, Division of Statistical Research and Methods, CCFS, Agency for Healthcare Research and Quality, 2101 E. Jefferson Street, Suite 500, Rockville, MD 20852, and Janet L. Valluzzi, MBA, OTR/L, The George Washington University, 2151 Jamieson Avenue, #609, Alexandria, VA 22314.

Several studies have shown that individuals consume a disproportionate amount of medical resources toward the end of life. Many of these studies are based on Medicare administrative data, which are inherently restricted to persons and services covered by Medicare. This study uses data on the civilian noninstitutionalized population from the first two panels (1996-97 and 1997-98) of the Medical Expenditure Panel Survey Household Component (MEPS-HC) to compare expenditures for medical care among persons in their last months of life relative to persons that survived. Expenditures include payments made for inpatient care, ambulatory care, prescribed medicines, dental care, and home health care. The analysis highlights methodological issues as well as results obtained when using MEPS-HC data to analyze medical expenditures over time. Analyses are conducted for both the elderly (65 years and over) and non-elderly populations. In addition to comparing overall expenditures on a per capita per unit time basis, differences in expenditures by type of service and source of payment are examined. Multivariate analyses are used to explore the extent to which differences in expenditures between decedents and survivors are explained by other factors (e.g. age, perceived health status) that are generally associated with expenses for medical care.

Learning Objectives:

  • At the end of this session, the participant will be able to

    Presenting author's disclosure statement:
    I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

    Issues in Health Care Statistics

    The 130th Annual Meeting of APHA