4123.0: Tuesday, November 14, 2000 - 1:00 PM

Abstract #6182

Elderly Medicare and Non-Medicare End of Life Expenditures in the MCBS From 1992-1996

Donald R. Hoover, PhD, MPH1, Cantor Joel, PhD2, Rizie Kumar, MS2, Usha Sambamoorthi, PhD2, and Steven Crystal, PhD2. (1) Statistics and Center for Aging, Health and Health Care Policy, Rutgers University, 473 Hill Center, 110 Frelinghuysen Road, Piscataway, NJ 08854-8019, (732)445-5301, dhoover@stat.rutgers.edu, (2) Center for Aging , Health and Health Care Policy, Rutgers University, 30 College Avenue, Piscataway, NJ 08854-8019

Background: Medical costs are substantially higher during the last year (LY) of life. Lubitz (1993) found that while ~5% of Medicare beneficiaries >=65 died annually in the 1970s-1980s, LY expenditures constituted 27%-31% of Medicare costs. About 40% of LY Medicare expenditures occurred in the last month (LM) of life, largely for inpatient care. Non-Medicare and 1990s LY costs are less well studied.

Methods: The Medicare Beneficiary Survey (MCBS) obtains annual medical expenditures by payer/service for ~12,000 Medicare enrollees selected from a clustered-weighted frame. Linear/logistic polynomial models fit to MCBS expenses of 1992-1996 decedents >=65 years at death estimated mean/probability of LY costs and timing of LY expenditures; both overall and by payer/service. Robust covariance algorithms incorporating cluster/repeated measure effects were used. From these models, death probability and overall costs; mean/probability of annual expenditures for persons >=65 years not in LY of life were estimated.

Results: While 5.5% of 1992-96 MCBS person-years were LY; 26.2% (95% CI: 22.6-30.3%) of Medicare versus 17.8% (16.0-19.9%) of non-Medicare expenses occurred in LY. Among LY expenses, 37.9% (32.1-44.8%) of Medicare compared to 17.7% (14.3-21.8%) of non-Medicare occurred in LM. Costs for all medical payers and services except dental were higher during LY.

Discussion: While non-Medicare costs are higher in LY, the difference is substantially less than that for Medicare costs implying less potential to control non-Medicare costs by reducing End-Of-Life expenditures. LY Medicare costs concentrate in the LM reflecting acute inpatient services while non-Medicare costs spread throughout the LY reflecting chronic treatment of sicker individuals.

Learning Objectives: "At the conclusion of the session, the participant (learner) in this session will be able to:

  1. 1. "State the Relative Changes in Health Care Costs (by Payer and Service) During the Last year of Life and the Portions of Total Medical Costs Spent on End-of Life-Care."
  2. 2. "Identify the Tming Patterns of Health Care Expenditures During the Last Year of Life."
  3. 3. "Understand More About the Potential to Control Differnt Types of Medical Costs Through Reduction of End of Life Expenditures."

Keywords: Cost Issues, End-of-Life Care

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
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.

The 128th Annual Meeting of APHA