Abstract: SA-PO1044
A Comparison of Death Records Between the USRDS and a Large Health Care System
Session Information
- Hemodialysis and Frequent Dialysis - VI
November 09, 2019 | Location: Exhibit Hall, Walter E. Washington Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 701 Dialysis: Hemodialysis and Frequent Dialysis
Authors
- Sim, John J., Kaiser Permanente Southern California, Pasadena, California, United States
- Shaw, Sally F., Kaiser Permanente Southern California, Pasadena, California, United States
- Zhou, Hui, Kaiser Permanente Southern California, Pasadena, California, United States
- Shi, Jiaxiao, Kaiser Permanente, Pasadena, California, United States
- Jacobsen, Steven J., Kaiser Permanente Southern California, Pasadena, California, United States
Background
The accuracy of mortality data is important when determining the mortality rates for the end stage renal disease (ESRD) population and assessing interventions to improve survivability or quality of life. Little research has been done to examine the completeness and accuracy of the USRDS mortality reports. The purpose of this study is to compare mortality records from a large integrated health system with the USRDS registry.
Methods
A retrospective cohort study (1/1/2007-12/31/2016) of ESRD patients within Kaiser Permanente Southern California (KPSC), an integrated health system, was performed. Patients were linked to the USRDS death records and evaluated. KPSC mortality data are obtained from several sources, but primarily from California state death certificates. USRDS mortality data are similarly obtained from several sources, but primarily from CMS form 2746.
Results
A total of 4827 death records were found between 2007 and 2016. There were 4189 death records found in both KPSC and USRDS databases, 609 found only at KPSC and 29 found only at USRDS. An average of 12.7% of death records per year were captured at KPSC but missing from the USRDS database. Of the 4189 death records, 86.92% of KPSC death records had consistent dates of death (DOD) with the USRDS. A few death records had a DOD that differed by more than a week (1.03%) to more than a year (0.07%).
Conclusion
These data suggest that mortality information from the USRDS could be systematically under-ascertained. Researchers should use caution when using USRDS mortality data because of the potential for incompleteness of the data as currently collected. The use of additional sources of information may supplement and help overcome these challenges.
Funding
- Government Support - Non-U.S.