Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

Health Services Research & Development

Go to the ORD website
Go to the QUERI website

HSR&D Citation Abstract

Search | Search by Center | Search by Source | Keywords in Title

Controlling for patient case mix at the end of life: Issues in identifying cause of death

Richardson S, Yu W. Controlling for patient case mix at the end of life: Issues in identifying cause of death. Poster session presented at: AcademyHealth Annual Research Meeting; 2005 Jun 1; Boston, MA.




Abstract:

Research Objective: Researchers investigating health care at the end of life often control for patient case mix using the death certificate underlying cause of death. It is unclear whether the reported underlying cause of death is consistent with the reason to control for case mix. The purpose of this study was to examine whether the reported underlying cause of death was consistent with the medical condition that required the most resource use or the last important principal diagnosis during the final year of life. Study Design: We obtained underlying cause of death as reported on the death certificate for 12,808 VA patients who died in California between October 1999 and September 2001. Using administrative data from both VA and Medicare, we compared the reported underlying cause of death with: the major principal diagnosis closest to death and the diagnosis responsible for the plurality of health care costs in the final year of life. We measured agreement between methods at three different levels of identification: the top two causes of death, the top four causes of death, and the top 10 causes of death, using CDC cause of death categories. Population Studied: Our study sample consisted of veteran decedents in California, but the results should be generalizable to a broader population of decedents. Principal Findings: All three methods classified similar numbers of patients into each cause of death category. However, neither of the methods using administrative data was consistent with the death certificate underlying cause of death at the patient level. For the top 10 causes of death, the patient-level consistency was slightly above 50%. As a whole, these methods were more consistent in identifying cancer deaths than deaths due to other causes. Conclusions: The three methods generate considerable discrepancies in patient classification. The principal diagnosis and cost-based methods are relatively consistent with one another, but not with death certificate data. Implications for Policy, Delivery or Practice: Our results suggest that methods using health utilization data should not be used as a proxy for death certificate data. However, for many research purposes, methods based on costs in the final year of life or principal diagnoses near death may in fact be more appropriate than death certificate data.





Questions about the HSR&D website? Email the Web Team.

Any health information on this website is strictly for informational purposes and is not intended as medical advice. It should not be used to diagnose or treat any condition.